AABB Technical Manual 20th Edition

20TH EDITION 20TH EDITION
nil TECHNICAL
ANUAL
Advancing Transfusion and
. Cellular Therapies Worldwide
4550 Montgomery Avenue. 700N • Bethesda,MD "lInII'I,&_'lIY,
Phone: + 1.301.907.6977• www.aabb.org
CLAUDIA S. COHN
MEGHAN DELANEY
SUSAN T. JOHNSON
LOU IS M. KATZ
203013
ISBN 978-1-56395-370-5
I1111
iJiJ.
T ch ic I
Manual
20TH EDITION
Other related publications available from the AABB:
Transfusion Therapy: Clinical Principles and Practice, 4th edition

Edited by Marisa Marques, MD; Joseph Schwartz, MD, MPH; and Yan Yun Wu, MD, PhD
Transfusion Medicine: Self Assessment and Review, 3rd edition
By Douglas P. Blackall, MD, MPH, and Justin D. Kreuter, MD
Blood Transfusion Therapy: A Physician's Handbook, 12th edition
Edited by Nicholas Bandarenko, MD, and Karen King, MD
Judd's Methods in Immunohematology, 3rd edition
ByJohn W. Judd, FIBMS;Susan T. Johnson, MSTM, MT(ASCP)SBB;and Jill Storry, PhD, FIBMS
Antibody Identification: Art or Science? A Case Study Approach
ByJanis R. Hamilton, MS, MT(ASCP)SBB;Susan T. Johnson, MSTM, MT(ASCP)SBB;
and SallyV. Rudmann, PhD, MT(ASCP)SBB
To purchase books or to inquire about other book services, including digital downloads and large-quantity sales,
please contact our sales department:
• 866.222.2498 (within the United States)
• +1301.215.6499 (outside the United States)
• +1301.951.7150 (fax)
• www.aabb.org-Resources-Marketplace
MBB customer service representatives are available by telephone from 8:30 am to 5:00 pm ET, Monday
through Friday, excluding holidays.
Tee ical
al
20TH EDITION
Edited by
Claudia S. Cohn, MD, PhD

Associate Professor/Director, Blood Bank Laboratory
University of Minnesota
Minneapolis, MN
Meghan Delaney, DO, MPH

Chief, Pathology and Laboratory Medicine Division, and
Director, Transfusion Medicine
Children's National Hospital
Washington, DC
Susan T. Johnson, MSTM, MT(ASCP)SBB

Director, Clinical Education
Versiti
Milwaukee, WI
Louis M. Katz, MD
Chief Medical Officer
MississippiValley Regional Blood Center
Davenport, IA
Mention of specific products or equipment by contributors to this AABBpublication does not represent an
endorsement of such products by the AABBnor does it indicate a preference for those products over other similar
competitive products. Product listings, descriptions, and references are not intended to be comprehensive. Any forms
and/or procedures in this book are examples. AABBdoes not imply or guarantee that the materials meet federal, state,
or other applicable requirements. It is incumbent on the reader who intends to use any information, forms, policies, or
procedures contained in this publication to evaluate such materials for use in light of particular circumstances
associated with his or her institution.
AABBauthors are requested to comply with a conflict of interest policy that includes disclosure of relationships with
commercial firms. A copy of the policy is located at http://www.aabb.org.
Efforts are made to have publications of the AABBconsistent in regard to acceptable practices. However, for several
reasons, they may not be. First, as new developments in the practice of blood banking occur, changes may be
recommended to the Standards jor Blood Banks and Transfusion Services. It is not possible, however, to revise each
publication at the time such a change is adopted. Thus, it is essential that the most recent edition of the Standards be
consulted as a reference in regard to current acceptable practices. Second, the views expressed in this publication
represent the opinions of authors. The publication of this book does not constitute an endorsement by the AABBof any
view expressed herein, and the AABBexpressly disclaims any liability arising from any inaccuracy or misstatement.
Copyright © 2020 by AABB.All rights reserved. No part of this book may be reproduced or transmitted in any form or
by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval
system, without permission in writing from the Publisher.
The Publisher has made every effort to trace the copyright holders for borrowed material. If any such material has
been inadvertently overlooked, the Publisher will be pleased to make the necessary arrangements at the first
opportunity.
AABB ISBNNo. 978-1-56395-370-5

4550 Montgomery Avenue Printed in the United States
Suite 700, North Tower
Bethesda, Maryland 20814-3304
Cataloging-in-Publication Data
Technical manual / editor, Claudia S. Cohn-20th ed.

p.; cm.
Including bibliographic references and index.
ISBN 978-1-56395-370-5
1. Blood Banks-Handbooks, manuals, etc. I. Cohn, Claudia S. II. AABB.
[DNLM: 1. Blood Banks-laboratory manuals. 2. Blood Transfusion-
laboratory manuals. WH 25 T2548 2020]
RMI72.T432020
615'.39-dc23
DNLMlDLC
Technical Manual
Authors
Jason Acker, MBA, PhD Paul M. Mansfield, MT(ASCP)CMSBB
Caroline R. Alquist, MD, PhD Irina Maramica, MD, PhD, MBA
Chester Andrzejewski Jr, PhD, MD Cami Melland, MLS(ASCP)CMSBB
J. Wade Atkins, MS, MT(ASCP)SBB,COA(ASQ) Yunchuan Delores Mo, MD
Debra J. Bailey, MT(ASCP)SBB Sandra Nance, MS, MT(ASCP)SBB
P. Dayand Borge Jr, MD, PhD Martin L. Olsson, MD, PhD
Colleen Bowman, MT(ASCP)SBB,COA(ASQ) Thierry Peyrard, PharmD, PhD, EurSpLM
Gwen Clarke, MD Rowena C. Punzalan, MD
Laura S. Connelly-Smith, MBBCh, DM Eva D. Quinley, MS, MT(ASCP)SBB,COA(ASQ)
Lauren A. Crowder, MPH Fran Rabe, MS
Melissa M. Cushing, MD Anna Razatos, PhD
Arthur B. Eisenbrey III, MD, PhD Susan N. Rossmann, MD, PhD
Lay See Er, MSTM, SBB(ASCP)CM,COA(ASQ) Jeremy Ryan Pena, MD, PhD
Richard O. Francis, MD, PhD Annette J. Schlueter, MD, PhD
Steven M. Frank, MD Joseph Schwartz, MD, MPH
James D. Gorham, MD, PhD Nadine Shehata, MD, FRCP
Patricia C. Grace, RN, BSN Ira A. Shulman, MD
Nicole R. Guinn, MD Whitney R. Steele, PhD, MPH
Sarah K. Harm, MD Sean R. Stowell, MD, PhD
Eldad A. Hod, MD Susan L. Stramer, PhD
Jay Hudgins, DO David F. Stroncek, MD
Orieji Illoh, MD Annika M. Svensson, MD, PhD
Melanie Jorgenson, RN, BSN, LSSGB Yvette C. Tanhehco, PhD, MD, MS
Cassandra D. Josephson, MD Lynne Uhl, MD
Margaret A. Keller, PhD Ralph R. Vassallo, MD, FACP
James M. Kelley, MD, PhD Franz F. Wagner, MD
Debra A. Kessler, RN, MS Jennifer Webb, MD, MSCE
Patricia M. Kopko, MD Julia S. Westman, PhD
Kevin J. Land, MD Barbee l. Whitaker, PhD
Lani Lieberman, MD Edward C.C. Wong, MD
Michael L. Linenberger, MD, FACP Sandy Wortman, MT(ASCP)CMSBB
Acknowledgments
HE 20TH EDITION OF theAABBTechni-
T cal Manual is the collaborative result of many
dedicated volunteers. My thanks go out to all the
Interorganizational Task Force on Domestic Disasters
and Acts of Terrorism
Molecular Testing Laboratories Standards
chapter authors, and my three Associate Editors for Program Unit
this edition-Meghan Delaney, SueJohnson, and Lou
Patient Blood Management Education
Katz. They have been a "dream team" and I greatly
appreciate their guidance in the subjects covered, and Committee
the many hours spent reviewing and revising each Patient Blood Management Standards
chapter with the authors. Committee
We, in turn, would like to thank the many mem- Quality Systems Accreditation Committee
bers of the following AABBcommittees and task forc- Relationship Testing Standards Program Unit
es who reviewed every word of the chapters, meth- Transfusion Medicine Section Coordinating Commit-
ods, and appendices in the 20th edition. Their tee - Subgroup, Donor and Blood Component
participation makes this volume unique in the litera- Management
ture and contributes to its excellent reputation.
Transfusion Medicine Section Coordinating Commit-
tee - Subgroup, Pediatric Transfusion
REVIEWING GROUPS Transfusion Medicine Section Coordinating Commit-
MBB Representative to ASFA tee - Subgroup, Technical Practices and Serology
MBB Representative to ISBTWorking Party for Transfusion Medicine Section Coordinating Commit-
Transfusion Transmitted Infectious Disease tee - Subgroup, Transfusion Safety and Patient
Cellular Therapy Section Coordinating Blood Management
Committee - Subgroup, Cellular Therapy Product Transfusion-Transmitted Disease Committee
Collection and Clinical Practices
Cellular Therapy Section Coordinating We would be remiss if we did not also mention
Committee - Subgroup, Product Manufacturing the valuable material drafted by contributors to earli-
and Testing er editions that we included in this new edition. The
Circular of Information Task Force selected tables, figures, methods, and narrative that
we kept could not have been improved. Finally, we
Clinical Transfusion Medicine Committee
would like to thank AABBstaffwho supported our ef-
Donor History Task Force forts.
FDALialsonCommittee
Immunohematology Reference Laboratories Accredi- Claudia S. Cohn, MD, PhD
tation Program Unit Editor in Chief
Immunohematology Reference Laboratories Stan-
dards Program Unit
Preface
N BEHALF OF THE EDITORS, AU- survey responses and feedback from AABB
thors, and many reviewers I am membership. For the 19th edition several key
pleased to introduce the 20th edition measures were taken to focus the content and
of the AABB Technical Manual. The Techni- reduce redundancy. These measures were
cal Manual conveys the latest information in very successful and, as a result, few structural
blood banking/transfusion medicine along changes were necessary for this edition.
with well-established material. Using the However, improvements in content contin-
model developed by previous editorial teams, ued unabated. Sue Johnson and the authors
when possible we paired new authors with of the chapters on blood groups worked dili-
seasoned veterans to bring a fresh perspective gently to bring the book's terminology for
to the content. This also allows for continuity blood group antigens more closely in line
as new authors in the current edition move with official ISBT nomenclature. In fact, their
into the role of primary author for the next work is already being used as an exemplar for
edition. I am grateful for the steady guidance other AABBpublications. Chapters addressing
and hard work of Associate Editors Meghan topics that have evolved significantly since
Delaney, DO, MPH; Sue Johnson, MSTM, publication of the 19th edition underwent
MT(ASCP)SBB;and Louis Katz, MD. All of us major revision. These expanded updates
were first-time editors of the Technical Man- include content on massive transfusion,
ual and learned a great deal from AABBstaff patient blood management, molecular testing,
and the editors of past editions. and relevant transfusion-transmitted disease.
The Technical Manuals excellence is Other key updates involving several chapters
ensured by a stringent peer-review process. resulted from the April 2020 release of new
Each chapter underwent an initial review by federal regulations on donor deferrals and
the editors for content and style, followed by subsequent revision of the donor history ques-
reviews from numerous AABB committee tionnaire.
members who provided focused expertise as The Technical Manual was in the final
well as comments on readability. Compliance stages of production when the COVID-19
experts then conducted a final check to make pandemic swept across the world. Not sur-
sure we provided the most up-to-date infor- prisingly, the pandemic had an impact on the
mation on the standards, guidance, and regu- book. First, we considered adding material
lations that help govern the activities readers about disaster preparedness, blood shortages,
engage in. These extra steps provide a high and convalescent plasma. However, the infor-
level of confidence that the TechnicalManual mation was changing so rapidly that we felt
is a reliable resource for our community. pandemic-related content would be out of
The overall structure and content of the date by the time we went to press. Second,
Technical Manual is adjusted periodically by the final review meeting at the AABB office
ix
x AABB TECHNICAL MANUAL
was converted to a virtual remote process. It tion on the virus, treatment, blood collection
took longer, but I believe the results are equal challenges, and lessons learned can be
in quality of outcome. The third impact also included in that edition.
relates to remote working. Our community as It has been a privilege to work with the tal-
a whole has developed and improved ways to ented and dedicated staff at AABB and the
benefit from access to online resources. transfusion medicine/blood banking commu-
Accordingly, the methods and appendices are nity that helped to produce this edition. On
now posted online, rather than loaded onto a behalf of the editors, authors, and reviewers
USB flash card in a pocket adhered to the we hope you enjoy using the 20th edition of
inside back cover. Readers who turn to the the Technical Manual and find it a useful ref-
table of contents page for the methods will erence tool.
find the access code there. We hope that the
pandemic will be resolved well before publi- ClaudiaS. Cohn, MD, PhD
cation of the 21 st edition, and that informa- Editor in Chief
Contents in Print
Preface ix
QUALITY AND RELATED ISSUES
1. Quality Management Systems: Principles and Practice •••••••• 1

Eva D. Ouinley, MS, MT(ASCP)SBB, COA(ASO), and Patricia C. Grace, RN, BSN
Background .
Concepts in Quality .. , . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Quality Management Systems Approach 4
Evaluation of the Quality Management System 5
The Quality Management System in Practice .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 5
Key Points. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Appendix 1-1. Glossaryof Commonly Used Quality Terms. . . . . . . . . . . . . . . . . . . . 26
Appendix 1-2. Code of Federal Regulations Quality-Related References 28
Appendix 1-3. Suggested Quality Control Performance Intervals for Equipment
and Reagents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2. Facilities, Work Environment, and Safety •••••• • • •• • • • • • • •• 33

]. Wade Atkins, MS, MT(ASCP)SBB, COA(ASO), and
Colleen Bowman, MT(ASCP)SBB, COA(ASO)
Safety Program 35
Fire Prevention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 39
Electrical Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 40
Biosafety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Chemical Safety 48
Radiation Safety 53
Shipping Hazardous Materials ...................... 56
General Waste Management 56
Key Points. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
References 57
Appendix 2-1. Safety Regulations and Recommendations Applicable to
Health-Care Settings 61
Appendix 2-2. General Guidance for Safe Work Practices, Personal Protective
Equtpment, and Engineering Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
xi
xii A A B B TE C H N I CAL MAN UAL
Appendix 2-3. BiosafetyLevel 2 Precautions 66

Appendix 2-4. Sample List of Hazardous Chemicals That May Be Encountered
in a Blood Bank 67
Appendix 2-5. Chemical Categories and How to Work Safelywith Them 69
Appendix 2-6. Incidental Spill Response 71
Appendix 2-7. Managing Hazardous Chemical Spills 74
3. Regulatory Considerations in Transfusion Medicine and

Cellular Therapies •••••••••••••••••••••••••.••••••••••• 77
Joseph Schwartz, MD, MPH; Otteji Illoh, MD; and
Yvette C. Tenhehco, PhD, MD, MS
FDA Oversight of Blood Establishments 78
Medical Laboratory Laws and Regulations 84
Local Laws, Hospital Regulations, and Accreditation 85
Human Cells, Tissues, and Cellular and Tissue-Based Products (HCT/PS) 86
Immune Effector Cells 89
Key Points 90
References 90
4. National Hemovigilance: The Current State ••••..•••.•••••. 9S

Kevin]. Lend, MD; Barbee I Whitaker, PhD; and Lynne Uhl,MD
International Hemovigilance 96
US Hemovigi1ance 97
Recipient Hemovigi1ancein the United States 100
Blood Donor Hemovigilance in the United States 103
Next Steps in US Hemovigilance 106
Key Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Appendix 4-1. Protocol for Hospital Reporting Adverse Events to Blood Suppliers .. 116
Appendix 4-2. Severity Grading Tool for Blood Donor Adverse Events 124
BLOOD COLLECTION AND TESTING
S. Allogeneic and Autologous Blood Donor Selection .•••..•. 127

Debra A. Kessler, RN, MS, and Susan N Rossmann, MD, PhD
Overview of Blood Donor Screening 127
Selection of Allogeneic Blood Donors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Abbreviated DHQ for Frequent Donors 132
Blood-Center-Defined Donor EligibilityCriteria 133
Recipient-Specific"Designated" or "Directed" Blood Donation 135
Key Points 137
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Table of Contents xiii
6. Whole Blood and Apheresis Conection of Blood

Components Intended for Transfusion •••••••••••••••••• 141
Jason Acker, MBA, PhD, and Anna Razatos, PhD
Donor Preparation and Care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 141
Blood Collection " 145
Blood Component Storage 153
Postcollection Processing/Blood Component Modification 158
Quarantine of Blood Components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 164
Labeling of Blood Components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 164
Key Points ;. ~'.; ; ...•.... ; ;. 165
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 166
7. Infectious DiseaseScreening••••••••••••••••••••••••••••• 173

Lauren A. Crowder, MPH,. Whitney R. Steele, PhD, MPH,.and
Susan L. Stramer, PhD
Historical Overview of Blood Donor Screening. . . . . . . . . . . . . . . . . . . . . . . . . . .. 173
Donor Screening Tests ..................................•.......... 177
Residual Infectious Risksof Transfusion . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. 196
Screening for SpecificAgents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 199
Pathogen Inactivation Technology 215
Summary 217
Key Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 218
BLOOD GROUPS
8. Molecular Biology and Immunology in Transfusion

Medicine •••••••••••••••••••.••••••••••••••••••••••••• 229
Sean R. Stowell, MD, PhD, end Iemes D. Gorham, MD, PhD
Analysis of DNA 229
Analysis of Protein 238
BasicImmunology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 243
Key Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
References , , " •...... "" '.........• ,. . . . . . . . 251
9. Blood Group Genetics •••.••.•••••.••••••.•.•..•..••••.•• 255

Margaret A. Keller, PhD, and Sandy Wortman, MT(ASCP/MSBB
Genomic Organization and Gene Regulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 256
Genetic Variation 262
Inheritance of Genetic Traits 266
Structural Variation 270
Chimerism 273
Gene Position Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
xiv A A B B TE C H N I CAL MAN U A L
Genetic Modifiers of Blood Group Antigen Expression 274

Population Genetics 275
Relationship Testing 278
Blood Group Gene Mapping 279
Gene, Protein, and Blood Group Terminology 280
Blood Group Genomics 281
Summary 291
Key Points 292
References 293
10. ABO and Other Carbohydrate Blood Group Systems ••••••• 297
Martin L. Olsson, MD, PhD, and]ulia S. Westman, PhD
The ABO system (001) 297
The H System (018) 310
The LE System (007) 313
I and i Antigens of the I Blood Group System (027) and I Blood Group Collection .. 315
PIPK (003) and GLOB(028) Blood Group Systems 318
The FORSBlood Group System (031) 323
The SID Blood Group System (038) 323
Key Points 323
References 324
11. The Rh System • • . • • . . . . • . • • • • • • • • • . . . • • • • • • • • • . • • . • • • • •. 329

Thierry Peyrerd, PharmD, PhD, EurSpLM, and Franz F. Wagner, MD
Historical Perspective 329
Terminology 332
Rh Locus 333
RHD Genotype 334
Antigens 337
Rh Genotyping 346
Rhnull Syndrome and the RhAG (030) Blood Group System 347
Antibodies to Rh Blood Group System Antigens 347
Technical Considerations for Rh Typing 348
Key Points 349
References 350
12. Other Blood Group Systems and Antigens ••••••••••••••••• 355

Cami Melland, MLS(ASCPlMSBB, and Sandra Nance, MS, MT(ASCP)SBB
The MNS System (002) 359
The LU System (005) 363
The KEL(006) and Xk (019) Systems 364
The FYSystem (008) 368
The JK System (009) 370
The DI System (010) 372
The YT System (011) 373
The XG System (012) 374
Table of Contents xv
The SC System (013) 374

The DO System (014) 374
The CO System (015) 375
The LW System (016) 375
The CH/RG System (017) • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. . . . . .. 377
The GE System (020) 377
The CROMSystem (021) 378
The KN System (022) 378
The IN System (023) 379
The OK System (024) 379
The RAPHSystem (025) 379
The JMH System (026) 379
The GIL System (029) 380
The RHAGSystem (030) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 380
The JR System (032) 380
The LANSystem (033) 380
The VELSystem (034) 381
The CD59 System (035) 381
The AUG System (036) 381
The KANNOSystem (037) 382
The SID System (038) 382
The CTL2 System (039) 382
Antigens that Do Not Yet Belong to a Blood Group System 382
Erythroid Phenotypes Caused by Mutations in Transcription Factor Genes . . . . . . . 384
Key Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 385
13. Identification of Antibodies to Red Cell Antigens •••••••••• 389

Lay See Er, MSTM, SBB(ASCPlM, COA(ASO), and
Debra]. Bailey, MT(ASCP)SBB
Basic Concepts in Red Cell Antigen Expression. . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
Initial Antibody Identification Considerations 391
BasicAntibody Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
Complex Antibody Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 402
Selected Procedures.. .. . .. .; .. . 414
Considerations FollowingAntibody Identification. . . . i . . . . . . •.. . . . . .. 421
Immunohematology Reference Laboratories . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 424
Key Points ; ; . . . . . . . .••. . . . . 424
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 425
Suggested Readings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 428
14. The Positive Direct Antiglobulin Test and Immune-

Mediated Hemolysis •••••••••••••••••••••••••••••••••• 429
P. Dayand Borge Jr, MD, PhD, and Paul M. Mensfield, MT(ASCPlMSBB
The DAT 430
Autoimmune HemolyticAnemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 434
Drug-Induced Immune Hemolytic Anemia 445
xvi A A B B TE C H N I CAL MAN U A L
Key Points 449

References 449
Appendix 14-1. Drugs Associated with Immune Hemolytic Anemia 452
15. Platelet and Granulocyte Antigens and Antibodies •••••••• 457

David F. Stroncek, MD, and Ralph R. Vassallo,MD, FACP
Platelet Antigens and Antibodies 457
Granulocyte Antigens and Antibodies 469
Key Points 473
References 473
16. The HLA System•.•••••..••••••••••••..•••.•..••••••••.•• 479

Jeremy Ryan Petie, MD, PhD; Arthur B. Eisenbrey III, MD, PhD; and
Patricia M. Kopko, MD
Biochemistry, Tissue Distribution, and Structure 479
Genetics of the MHC 484
HLATyping 488
Other Non-HLAHistocompatibility Determinants 489
Crossmatching and Detection of HLAAntibodies 490
The HLASystem and Transfusion 491
HLATesting and Transplantation 493
Other Clinically SignificantAspects of HLA 495
Clinical Consultation in HLA 497
Regulatory Aspects of Clinical Histocompatibility 498
Future Directions 498
Summary 499
Key Points 499
References 499
ESSENTIALS OF TRANSFUSION PRACTICE
17. Transfusion-Service-Related Activities: Pretransfusion

Testing and Storage, Monitoring, Processing,
Distribution, and Inventory Management of
Blood Components •••••••••••••••••••••••••••••••••••• 503
Caroline R. Alquist, MD, PhD, and Sarah K Harm, MD
Samples and Requests 503
Pretransfusion Testing of Recipient Blood 504
Blood and Blood Component Storage and Monitoring 509
Pretransfusion Processing 518
Distribution 522
Issuing of Components 523
Inventory Management 527
Key Points 529
Table of Contents xvii
References 530
Appendix 17-1. Sources of False-PositiveResults in Antiglobulin Testing 533
Appendix 17-2. Sources of False-Negative Results in Antiglobulin Testing 534
Appendix 17-3. Causes of Positive Pretransfusion Test Results 535
18. Administration of Blood Components •••••••••••••••••••• 537

Melanie Jorgenson, RN, 13SN,LSSGB
Events and Considerations Before Dispensing Components 537
Blood Component Transportation and Dispensing 542
Blood Administration ;........ 543
Documentation of the Transfusion 547
Unique Transfusion Settings :....................... 548
Conclusion 549
Key Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549
References 550
19. Hemotherapy Decisions and Their Outcomes •.•••••••••••• 553

Nadine Shehsts, MD, FRep, and Yunchuan Delores Mo, MD
Red Blood Cell Transfusion 553
Platelet Transfusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
Plasma Transfusion 567
Cryoprecipitate Transfusion 569
Granulocyte Transfusion 570
Massive Transfusion Protocols 571
Key Points. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573
References 573
20. Patient Blood M&ll'Iilg4!rnent •••••.••••••••••••••••••••••••

.: ••• ~.' "_. ': : • ":: •• _--- -; - -.' - _'oe .:_ - -, .: _-_: : _ - _ : :. .' .' ':, •• - - - - - •
583 • •
Steven M. Frank, MD, and Nicole R. Guinn, MD

Definition and Scope of Patient Blood Management . . . . . . . . . . . . . . . . . . . . . . .. 583
Resources to Support a PBM Program 584
Patient Blood Management Standards and Certification .. . . . . . . . . . . . . . . . . . .. 585
Methods of Patient Blood Management 585
Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 600
Extremes of Transfusion 605
Summary 606
Key Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. 607
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607
21. Approaches to Blood Utilization Auditing ••••••••••••••••• 613

Ira A. Shulman, MD; Jay Hudgins, DO; and Irina Maramica, MD, PhD, MBA
The Auditing Process ; . . . . . . . . . . . . . . . . . . . . . . .. 614
Defining Audit Criteria 615
Types of Blood Utilization Review 617
Blood Utilization Review of Transfusions to High-RiskPatients 620
xviii A A B B TE C H N I CAL MAN U A L
The Role of a Computerized Provider Order Entry System in Blood Utilization

Review 620
Use of "Big Data" to Assess Performance and Progress Measures in
Transfusion Medicine 621
Key Points 623
References 624
22. Noninfectious Complications of Blood Transfusion •••••••• 627

Eldad A. Hod, MD, and Richard O. Francis, MD, PhD
Hemovigilance 627
Recognition and Evaluation of a Suspected Transfusion Reaction 627
Acute or Immediate Transfusion Reactions 634
Delayed Transfusion Reactions 648
Fatality Reporting Requirements 652
Key Points 653
References 653
SPECIAL PATIENTS AND SITUATIONS
23. Perinatal Issues in Transfusion Practice ••••••••••••••••••• 659

Lent Lieberman, MD; Gwen Clarke, MD; and Annika M. Svensson, MD, PhD
Hemolytic Disease of the Fetus and Newborn 659
Pregnancy-Related Thrombocytopenia 665
Key Points 668
References 668
24. Neonatal and Pediatric Transfusion Practice •••••••••••••• 673

Edward C. C. Wong, MD, and Rowena C. Punzalan, MD
Hematopoiesis, Coagulation, and Physiology 673
RBCTransfusion in Neonates 675
RBCTransfusion in Infants Older than 4 Months and Children 682
Platelet Transfusion in Neonates and Children 685
Plasma and Cryoprecipitate Transfusion in Neonates and Children 687
Granulocyte Transfusion in Neonates and Children 688
Other Considerations Common to Transfusion of Neonates and Children 689
Adverse Effectsand Prevention 694
Key Points 695
References 696
25. Therapeutic Apheresis •••.••.•••••••••••••••••••••••••••• 705

Jennifer Webb, MD, MSCE, and Chester Andrzejewski Jr, PhD, MD
General Principles 705
Device Modalities 706
Patient Evaluation and Management 707
Table of Contents xix
Vascular Access. . . . . . . .. . . . . . . . . . .. . ; 709

Anticoagulation 711
Adverse Effects 711
Pediatric Apheresis 713
Therapeutic Apheresis Indications ................... 713
Therapeutic Apheresis Procedure Documentation, Payment, and Provider
Credentialing 727
Key Points. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 729
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 730
26. The Collection and Processing of Hematopoietic

Progenitor Cells ••••••••••••••••••••••••••••.••••••••• 737
Laura S. Connelly-Smith, MBBCh, DM, and Michael L. Linenberger, MD, FACP
Clinical Utility 737
Histocompatibility, Donor Type, and Graft Source 741
HPC Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 743
Processing Human Progenitor Cells .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 748
Specialized Cell-ProcessingMethods 750
Cryopreservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 751
Quality Control 752
Shipping and Transporting HPC Cellular Products 754
Patient Care 754
Regulatory and Accreditation Considerations 756
Conclusion 757
Key Points 757
References 758
27. Transfusion Support for Hematopoietic Stem Cell

Transplant Recipients •••••••••••.••••••••••••••••••••• 767
james M. Kelley, MD, PhD, and Melissa M. Cushing, MD
ABO Compatibilityfor Blood Component Selection Following HSCT 768
Blood Component Support for HSCT Patients 772
Pediatric Considerations 773
Key Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 774
References 774
28. Human Tissue Allografts and the Hospital

Transfusion Se.rvice •••••••••••••••.••••••••••••••••••• 777
Annette]. Schlueter, MD, PhD; Fran Rebe, MS; and Cassandra D. josephson, MD
Tissue Donation and Transplantation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 777
Federal Regulations, State Laws, and ProfessionalStandards. . . . . . . . . . . . . . . . .. 782
Hospital Tissue Services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 784
Key Points. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 789
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 790
Index 793
xx AABB TECHNICAL MANUAL
Contents Online
www.aabb.org/methods
METHODS
1. General Laboratory Methods
Method 1-1. Shipping Hazardous Materials

Method 1-2. Monitoring Temperature During Shipment of Blood
Method 1-3. Treating Incompletely Clotted Specimens
Method 1-4. Solution Preparation Procedure
Method 1-5. Serum Dilution Procedure
Method 1-6. Dilution of Percentage Solutions Procedure
Method 1-7. Preparing a 3% Red Cell Suspension
Method 1-8. Preparing and Using Phosphate Buffer
Method 1-9. Reading and Grading Tube Agglutination
2. Red Cell Typing Methods
Method 2-1. Determining ABO Group of Red Cells-Slide Test

Method 2-2. Determining ABO Group of Red Cells and Serum-Tube Test
Method 2-3. Determining ABO Group of Red Cells and Serum-Microplate Test
Method 2-4. Initial Investigation of ABO Grouping Discrepancies Procedure
Method 2-5. Detecting Weak A and BAntigens and Antibodies by Cold Temperature
Enhancement
Method 2-6. Confirming Weak A and B Antigens Using Enzyme-Treated Red Cells
Method 2-7. Confirming Weak A or B Subgroup by Adsorption and Elution
Method 2-8. Testing Salivafor A, B, H, Lea,and Le"Antigens
Method 2-9. Confirming Anti-At in an Az or Weak A Subgroup
Method 2-10. ResolvingABO Discrepancies Caused by Unexpected Alloantibodies
Method 2-11. Determining Serum Group Without Centrifugation
Method 2-12. Determining Rh(D) Type-Slide Test
Method 2-13. Determining Rh(D) Type-Tube Test
Method 2-14. Determining Rh(D) Type-Microplate Test
Method 2-15. Testing for Weak D
Method 2-16. Preparing and Using Lectins
Method 2-17. RemovingAutoantibody by Warm Saline Washes
Method 2-18. Using SulfhydrylReagents to Disperse Autoagglutination
Method 2-19. Using Gentle Heat Elution to Test Red Cells with a Positive DATResult
Method 2-20. DissociatingIgG by Chloroquine for Antigen Testing of Red Cellswith a
Positive DAT Result
Method 2-21. Using Acid Glycine/EDTA to Remove Antibodies from Red Cells
Method 2-22. Separating Transfused from Autologous Red Cells by Simple Centrifugation
Method 2-23. Separating Transfused from Autologous Red Cellsin Patients with Hemoglobin
S Disease
Table of Contents xxl
3. Antibody Detection, Identification, and Compatibility Testing
Method 3-1. Using Immediate-Spin Compatibility Testing to Demonstrate ABO

Incompatibility
Method 3-2. Saline Indirect Antiglobulin Test Procedure
Method 3-3. Albumin or LISS-AdditiveIndirect Antiglobulin Test Procedure
Method 3-4. LISS-SuspensionIndirect AntiglobulinTest Procedure
Method 3-5. PEG Indirect Antiglobulin Test Procedure
Method 3-6. Prewarming Procedure
Method 3-7. Detecting Antibodies in the Presence of Rouleaux-Saline Replacement
Method 3-8. Preparing Ficin Enzyme Stock, 1% wlv
Method 3-9. Preparing Papain Enzyme Stock, 1% wIv
Method 3-10. Standardizing Enzyme Procedures
Method 3-11. Evaluating Enzyme-Treated Red Cells
Method 3-12. One-Stage Enzyme Procedure
Method 3-13. Two-Stage Enzyme Procedure
Method 3-14. Performing a Direct Antiglobulin Test
Method 3-15. Antibody Titration Procedure
Method 3-16. Using SulfhydrylReagents to DistinguishIgM from IgGAntibodies
Method 3-17. Using Plasma Inhibition to Distinguish Anti-Ch and -Rgfrom Other Antibodies
with Similar Characteristics
Method 3-18. Treating Red Cells Using DTT or AET
Method 3-19. Neutralizing Antl-Sd"with Urine
Method 3-20. Adsorption Procedure
Method 3-21. Using the American Rare Donor Program
4. Investigation of a Positive DAT Result
Method 4-1. Cold-AcidElution Procedure

Method 4-2. Glycine-HClIEDTAElution Procedure
Method 4-3. Heat Elution Procedure
Method 4-4. Lui Freeze-Thaw Elution Procedure
Method 4-5. Cold Autoadsorption Procedure
Method 4-6. Determining the Specificityof Cold-ReactiveAutoagglutinins
Method 4-7. Cold Agglutinin Titer Procedure
Method 4-8. Adsorbing Warm-ReactiveAutoantibodies Using Autologous Red Cells
Method 4-9. Adsorbing Warm-ReactiveAutoantibodies Using Allogeneic Red Cells
Method 4-10. Polyethylene GlycolAdsorption Procedure
Method 4-11. Performing the Donath-Landsteiner Test
Method 4-12. Detecting Drug Antibodies by Testing Drug-Treated Red Cells
Method 4-13. Detecting Drug Antibodies by Testing in the Presence of Drug
5. Hemolytic Disease of the Fetus and Newborn

Method 5-1. Testing for Petomaternal Hemorrhage-The Rosette Test
Method 5-2. Testing for Fetomaternal Hemorrhage=Modified Kleihauer-BetkeTest
Method 5-3. Using Antibody Titration Studies to Assistin Early Detection of Hemolytic
Disease of the Fetus and Newborn
xxii A A B B TE C H N I CAL MAN UA L
6. Blood Collection, Component Preparation, and Storage
Method 6-1. Screening Female Donors for Acceptable Hemoglobin Level-Copper Sulfate
Method
Method 6-2. Preparing the Donor's Arm for Blood Collection
Method 6-3. Collecting Blood and Samples for Processing and Testing
Method 6-4. Preparing Red Blood Cells from Whole Blood
Method 6-5. Preparing Prestorage Red Blood Cells Leukocytes Reduced from Whole Blood
Method 6-6. Using High-Concentration Glycerol to Cryopreserve Red Cells-Meryman
Method
Method 6-7. Using High-Concentration Glycerol to Cryopreserve Red Cells-Valeri Method
Method 6-8. Checking the Adequacy of Deglycerolization of Red Blood Cells
Method 6-9. Preparing Fresh Frozen Plasma from Whole Blood
Method 6-10. Preparing Cryoprecipitated AHF from Whole Blood
Method 6-11. Thawing and Pooling Cryoprecipitated AHF
Method 6-12. Preparing Platelets from Whole Blood
Method 6-13. Removing Plasma from Platelets (Volume Reduction)
7. Transplantation of Cells and Tissue
Method 7-1. Infusing Cryopreserved Hematopoietic Cells

Method 7-2. Processing Umbilical Cord Blood
Method 7-3. Investigating Adverse Events and Infections Following Tissue AllograftUse
8. Quality Control Methods
Method 8-1. Validating Copper Sulfate Solution

Method 8-2. Calibrating Liquid-in-GlassLaboratory Thermometers
Method 8-3. Calibrating Electronic Oral Thermometers
Method 8-4. Testing RefrigeratorAlarms
Method 8-5. Testing Freezer Alarms
Method 8-6. Calibrating Centrifuges for Platelet Separation
Method 8-7. Calibrating a SerologicCentrifuge for Immediate Agglutination
Method 8-8. Calibrating a SerologicCentrifuge for Washing and Antiglobulin Testing
Method 8·9. Testing Automatic Cell Washers
Method 8-10. Monitoring Cell Counts of Apheresis Components
Method 8-11. Counting Residual White Cells in Leukocyte-Reduced Blood and
Components-Manual Method
APPENDICES
Appendix 1. Normal Values in Adults
Appendix 2. Selected Normal Values in Children
Appendix 3. Typical Normal Values in Tests of Hemostasis and Coagulation (Adults)
Appendix 4. Coagulation Factor Values in Platelet Concentrates
Appendix 5. Approximate Normal Values for Red Cell, Plasma, and BloodVolumes
Appendix 6. Blood Group Antigens Assigned to Systems
Table of Contents xxiii
Appendix 7. Examples of Gene, Antigen, and Phenotype Symbolsin Conventional and

International Society of Blood Transfusion Terminology
Appendix 8. Examples of Correct and Incorrect Terminology
Appendix 9. Distribution of ABO/Rh Phenotypes by Race or Ethnicity
Appendix 10. Abbreviations Used in the Technical Manual, 20th Edition
ACKNOWLEDGMENT
Special thanks to Maureen Beaton, MT(ASCP)BB,COA(ASQ),
for review of the methods and appendices.
(!uaUt)'. ana~ ..f!ment .Systems:
Principles and Practice
Eva D. Quinley, MS, MT(ASCP)SBB, CQA(ASQ), and Patricia C. Grace, RN, BSN
QUALITY MANAGEMENT SYSTEM an organization will spend fewer resources to

(OMS)is a collection of business process- achieve the same operational and quality out-
es focused on achieving quality while comes. An effective OMS provides confidence
meeting customer requirements. It is expressed to the customer, the organization, and other in-
as the organizational structure, policies, proce- terested parties that the organization will pro-
dures, processes, and resources needed to im- vide products and services that consistently
plement quality management. Why is this im- meet or exceed requirements or customer ex-
portant in the fields of transfusion medicine and pectations, and it increases efficiencies,thus re-
cellular therapies? The answer to this is sim- ducing costs.
ple-the customers served, whether they are
other health-care providers or patients, depend
on the assurance that the products and services UND
produced and provided are safe and effectivefor
their intended use. A OMS is the framework for Quality has been central to transfusion medicine
continual improvement by enhancing customer from its inception, the opening of the first blood
satisfaction. In a OMS, customer requirements bank in the United States.at the Cook County
are defined, processes are designed to meet Hospltaltn Chicago in 1937. Continuous scien-
those requirements, and processes are in place tific progress in many aspects of transfusion
to manage and improve the level of service that medicine has contributed to the quality and
is provided. safety of blood components and transfusion ser-
It is also the expectation of regulators that or- vices, and now cellular therapies. During the
ganizations have processes in place to ensure 1990s, after the discovery of the human immu-
that products and services are safe and success- nodeficiency virus .(HIV) and the advent of
ful in producing the desired results..The imple- AIDS, a very sensitized and informed public de-
mentation of an effective OMS will help to en- manded that the highest level. of quality be
sure these outcomes. . .; \. achieved and .maintained in. all processes in-
Finally, with decreases in utilization and involved in the provision of all blood components
creased costs of operations, it is also important and services. The Food and Drug Administra-
that organizations involved in transfusion medi- tion (FDA)introduced the concept of a "zero
cine and cellular therapies operate in the most risk blood supply" in an effortto decrease the in-
cost-effectivemanner possible.A good OMS will herent risk of the transmission of infectious
reduce rework, waste, and inefficiencies; thus, agents and transfusion complications. Regulate-
Eva D. Quinley, MS, MT(ASCP)SBB,COA(ASQ),Regional Director, Vitalant-Illinois, Chicago, Illinois; and Patri-
cia C. Grace, RN, BSN, Senior Quality Director, Vitalant-California, Mather, California
The authors have disclosed no conflicts of interest.
1
2 AABB TECHNICAL MANUAL
ry agencies such as the FDA, the Centers for Standards (NCCLS),a global organization head-
Medicare and Medicaid Services (CMS), and quartered in the United States, is a member of
state departments of health, and accrediting or- ANSI. The FDA and AABB incorporate many
ganizations such as AABB,the College ofAmeri- ISO principles into their regulations and stan-
can Pathologists (CAP), The Joint Commission, dards. For example, AABB'sOSEs are rooted in
and the Foundation for the Accreditation of Cel- the 20 clauses of the ISO 9000 series and are
lular Therapies (FACT)require facilities operat- compatible with ISO standards.
ing in transfusion medicine and cellular thera-
pies to establish and follow a quality assurance
program, including quality control (OC)process- CONCEPTS iN QUALITY
es and procedures, as part of their licensing, cer-
Quality Assurance
tification, and/or accreditation programs. Every
laboratory must comply with the Clinical Labo- The concept of quality assurance is broad, and
ratory Improvement Amendments of 1988 the goals of quality assurance are to significantly
(CLIA), quality requirements implemented by decrease errors; ensure the credibility of results;
CMS that have a primary focus on laboratory implement safe and effectivemanufacturing pro-
quality. In 1995, the FDAreleased its Guideline cesses and system controls; and ensure contin-
for Quality Assurance in Blood Establishments. ued improvement in product safety and quality.
This guideline, along with other FDA-issued A quality assurance program is defined as a sys-
guidance, assistsfacilitiesin maintaining compli- tem designed and implemented to ensure that
ance with the current good manufacturing prac- manufacturing is consistently performed in such
tice (cGMP)requirements found in the Code of a way as to yield a product of consistent quality.J
Federal Regulations (CFR), Title 21, Parts 200 A good quality assurance program includes ways
and 600. CFR Title 21, Part 820, formerly to detect, investigate, assess, prioritize, and cor-
known as the GMP requirements for medical rect errors, with the ultimate goal of error pre-
devices, provides regulations applicable to man- vention. Ouality assurance activities also in-
ufacturers of medical devices, including blood clude retrospective reviews and analyses of
establishment computer systems (BECS). operational performance data to determine
AABB's Ouality System Essentials (OSEs), whether the overall process is in a state of con-
minimum requirements for blood banking and trol and to detect shifts or trends that require at-
cellular therapy operations, are based on all of tention. Ouality assurance provides information
these specificationsand provide additional guid- to process managers regarding levels of perfor-
ance in implementing practices that ensure mance that can be used in setting priorities for
quality and compliance with cGMP and current process improvement.
good tissue practice (cGTP) regulations. AABB
and CAPare granted "deemed status" as accred- Quality Control
iting organizations under the CLlA '88 program OC is one aspect of a quality assurance program.
by CMS, as are The Joint Commission and some Its purpose is to determine, through testing or
state regulatory bodies. The International Orga- observation, whether a process or particular task
nization for Standardization (ISO) has estab- within a process is working as expected at a giv-
lished international standards in most fields, en time. OC involves sampling and testing. His-
which represent minimal requirements. These torically, transfusion services and donor centers
standards are generic in content and can be ap- have used many OC measures as standard prac-
plied to any organization, large or small, what- tices in their operations. Examples include re-
ever its product may be. The United States is agent OC; product OC; clerical checks; visual
represented in ISO by the American National inspections; and regular measurements, such as
Standards Institute (ANSI). The Clinical and temperature readings on refrigerators and vol-
Laboratory Standards Institute (CLSI),formerly ume or cell counts on finished blood compo-
the National Committee for Clinical Laboratory nents. If OC is not within specifications,it may
C HAP T E R 1 Quality Management Systems:Principlesand Practice 3
indicate a problem, either with the process itself Quality Systems

or with how the process is being executed.
Trends in QC may indicate the potential for a A system is defined as an organized, purposeful
problem in the future. structure that consists of interrelated and inter-
dependent elements (components, processes,
Quality Management entities, factors, members, parts, etc). These ele-
ments continually influence one another (direct-
Quality management considers interrelated pro- ly or indirectly) to maintain their activity and
cesses in the context of the organization and its the existence of the system, in order to achieve
relations with customers and suppliers. It ad- the goal of the system. The quality system is
dresses the leadership role of executive manage- made up of a set of interrelated processes that
ment in creating a commitment to quality work together to ensure quality. (See Fig 1-1.)
throughout the organization, the understanding It is important to understand what a process
of suppliers and customers as partners in quality,
is. Basically,a process can be defined as a set of
the management of human and other resources,
activities that uses resources to transform in-
and quality planning. An important goalin quali-
puts to outputs. The whole blood collection pro-
ty management is to establish a set of controls
that ensure process and product quality but are cess, for example, has many inputs, such as a
not excessive. Controls that do not add value trained phlebotomist, an approved collection
should be eliminated to conserve limited reo set, an approved arm-scrub solution, a calibrated
sources and allow staff to focus attention on scale, and phlebotomy standard operating proce-
those controls that are critical to the operation, dures (SOPs), all working together to produce
Statistical tools, such as process capability the output, a unit of whole blood. The quality of
measurement and control charts, allow a facility the output is determined by the quality and con-
to evaluate process performance during the trol that is in place with the inputs and with the
planning stage and in operations. These tools process itself. Validation of a process is key in
help determine whether a process is stable ensuring the process is consistent and produces
(ie, in statistical control) and is capable of meet· the desired output. Validation is discussed more
ing product and service specifications. fullylater in this chapter.
FIGURE1-1. Systems and processes.

Control of the Process Quality Planning

Strategies for managing a process should address A necessary activity to ensure success of the
all of its components, including its interrelated OMS is quality planning. This is defined as "a
activities, inputs, outputs, and resources. Suppli- systematic process that translates quality policy
er qualification, formal agreements, supply veri- into measurable objectives and requirements,
fication, and inventory control are strategies for and lays down a sequence of steps for realizing
ensuring that the inputs to a process meet speci- them within a specified timeframe."! A written
fications. Personnel training and competency as- quality plan provides the framework for imple-
sessment, equipment maintenance and control, menting and maintaining an effective OMS.
management of documents and records, and im- This should be a living document that is re-
viewed and edited as needed.
plementation of appropriate in-process controls
provide assurance that the process will operate
as intended. End-product testing and inspection,
outcome measurement, and customer feedback
provide data to evaluate product quality and im-
prove the process. These output measurements To develop and implement a OMS, it is import-
and quality indicators are used to evaluate the ant for organizations to follow a planned path.
effectiveness of the process and process con- The steps of this path include:
trols.
To manage a system of processes effectively, 1. Determining the needs and expectations of
the facility must understand how its processes the customer and other interested parties.
interact and what cause-and-effectrelationships 2. Establishing a quality policy and quality
exist between them. As an example, the conse- objectives.
quences of accepting a donor who is not eligible 3. Determining the processes needed to obtain
reach into almost every other process in the fa- those quality objectives and who is responsi-
cility. If a donor with a history of high-risk be- ble for those processes.
havior is not identified as such during the selec- 4. Ensuring adequate resources are available to
tion process, the donated unit(s) may return execute those processes.
positive test results for one of the viral marker 5. Determiningand applyingmethods to evaluate
assays, triggeringfollow-up testing, look-backin- those processes, including malting a determi-
vestigations, and donor deferral and notification nation of the effectiveness and efficiency of
procedures. Components must be quarantined each process.
and their discard documented. Personnel in- 6. Designingways to prevent nonconformances
volved in collecting and processing the unit(s) and ways to correct nonconformances that
are at risk of exposure to infectious agents. Part are not prevented.
of quality planning is to identify these relation- 7. Establishinga process for continual improve-
ships so that quick and appropriate corrective ment of the OMS.
action can be taken when process controls fail.
It is important to remember that operational Such an approach can be used to develop a
processes include not only product manufac- OMS or to maintain and improve an existing
ture or service creation but also the distribution OMS. The needs and expectations of the cus-
of a product or service. Distribution generallyin- tomer or interested parties must be defined and
volves interaction with the customer. The quali- documented as fully as possible. The voice of
ty of that transaction is critical to customer satis- the customer is critical to success. Once an orga-
faction and should not be overlooked in the nization understands what customers want, a
design and ongoing assessment of the OMS. quality policy and quality objectives should be
developed with that information in mind. It is issues of greatest concern. For example, one
important to consider those who regulate or ac- area of concern to blood establishments and to
credit the organization in the development of their customers is product availability. An evalu-
the policy and objectives. Although.some do not ation of product availabilitywould provide infor-
consider these bodies as customers, they cer- mation as to whether. the right product is avail-
tainly have a vested interest in an organization able at the right time, and opportunities to
that operates in transfusion medicine or cellular improve where this falls short. The audience for
therapies. Resources to achieve the objectives the evaluation is determined by the stakeholders
must be determined, and then there must be a and usually would include senior management,
way to ensure that they. are adequate. As de- inventory control, sales, and marketing,
scribed further in the next section, once the pol- Information needed for the evaluation de-
icy, objectives, and procedures are in place, pends on the purpose and the audience. Once
methods to evaluate the effectiveness and effl-
these are decided, then information can be gath-
cien~y of these are necessary. A major goal is to
ered to support decision-making or simply to in-
find ways to prevent nonconformances from
form. The information may come from a num-
happening in the first place, but because of the
nature of the work, nonconformances will oc- ber of sources: production reports, error reports,
cur. When they do, it is imperative to have a audits or inspections, and customer feedback, to
method to detect and correct nonconformances name a few.
and prevent them from happening again. Finally, A number of tools exist to evaluate the infor-
because a quality system is somewhat dynamic, mation. A tool is any chart, device, software,
a philosophy of continual improvement needs to strategy, or technique that supports quality man-
be developed and executed. agement efforts. Many of the tools are easy to
use, but it is important that the audience be con-
sidered when choosing which tools to use. A
F THE number of software vendors produce software
IE ENT that is designed specificallyto monitor and eval-
uate the OMS.
It is important to evaluate the OMS routinely to

determine if it is working as expected. The eval- THE QUALITY AN EMENT
uation should include the followingitems: TEM IN P TI IE
.. Engagement of the stakeholders. Several elements comprise a OMS, and the appli-
1/1 Purpose of the evaluation. cation of those elements in transfusion medicine
1/1 Audience for the evaluation. and cellular therapies is described in the text that
s Information needed for the evaluation. follows.Basicelements of the OMS include:
II Sources for information related to the evalua-
tion. III Organization and leadership.

@ Tools. III Customer foCUS.
I[> Human resources.
Evaluation of the OMS begins with the en- Equipment management.
gagement of those who have vested interest in lID Suppliers and materials management.
the results of the evaluation. In blood establish- Process control and management.
ments and cellular therapy facilities, stakehold- Documents and records.
ers most often include quality, operations, and Information management.
management but might include other areas such Management of nonconforming events.
as recruitment and human resources, depending Monitoring and evaluation.
on what processes are being evaluated. The pur- @ Process improvement.
pose of the evaluation should be aligned with @ Facilities,work environment, and safety.
Organization and Leadership Monitoring of training effectiveness and

competency.
An organization must be structured such that
ill Review and approval of validation protocols
the OMS can be well implemented. The struc-
and results.
ture should facilitate communication through-
Review, validation, and approval of OMS
out the organization. It is also important that
software.
clear descriptions of authority and the responsi-
lib Development of evaluation criteria for sys-
bilities of each role are defined in writing. The
tems.
role of senior management is fundamental to the
Reviewand approval of suppliers and mainte-
success of any OMS. It is the responsibility of
nance of an approved supplier list.
leadership to create an environment where indi-
Review of product specifications.
viduals are fully engaged in the OMS and to
Determination of the suitability of products.
monitor it to ensure that the system operates ef-
Monitoring and trending.
fectively. Specificduties assigned to top manage-
Review of reports of adverse reactions, error
ment include:
reports, and complaints.
lib Audit of operational functions.
Establishing, implementing, and maintaining
Inspection oversight and management.
a quality policy and associated quality goals
Reporting to regulators, accrediting bodies,
and objectives.
customers, or others as necessary.
Providing adequate resources to carry out
the operations of the facilityand the OMS.
Although traditionally the quality depart-
Ensuring appropriate design and effectiveim-
ment has had responsibility for the majority of
plementation of new or modified processes
and procedures. these activities, it may be wise to have opera-
Participating in the review and approval of tions participate in some of these activities,
policies, processes, and procedures. again with the caveat that one does not review
ill Enforcing adherence to operational and qual- one's own work. This reinforces the concept
ity policies, processes, and procedures. that quality is everyone's responsibility.
Overseeing operations and regulatory and ac-
creditation compliance. Customer Focus
Periodicallyreviewing and assessing OMS ef- To obtain true quality, it is imperative for an or-
fectiveness. ganization to understand the needs and require-
Identifying designees and defining their re- ments of the customer. Organizations that
sponsibilitieswhen assisting executive man- provide blood components or other cellular
agement in carrying out these duties.
products and services have a variety of custom-
ers, and each should be considered. Processes
The individual who is assigned to oversee an
organization's quality activities should report to and services should be designed and developed
top management. This individual may perform with the customer requirements in mind. Cus-
some of the tasks but does not have to personally tomer requirements need to be documented', of-
perform all the quality functions. It is desirable tentimes the documentation is contained in a
for this individualto operate totally separate from supplier agreement or contract. Once the re-
operations, although in smaller organizations the quirements are established, there should be a
individual may be involved in operational activi- mechanism to receive feedback from the cus-
ties as well. The key here is that the individual tomer at regular intervals to determine if the re-
should never review his or her own work. Quali- quirements are being met. Such feedback may
ty functions include the following: be obtained from an analysis of key metrics de-
veloped in conjunction with the customer (eg,
Review and approval of SOPs. fill rates, on-time delivery, customer complaint
Review and approval of training plans. rates) or may be gleaned from customer surveys.
Human Resources dividuals are selected for hire based on their

ability to meet those job qualifications, including
The human resources department is focused on
training, education, and experience.
activities relating to employees. These activities
normally include recruiting, hiril1g,and training
of new employees; training of current employ- Orientation and Training
ees; employee benefits; management of employ- Orientation is critical for a new employee to get
ee concerns; oversight of performance reviews the right start. Each employee needs to under-
and necessary corrective actions; retention; and stand his or her role, aswell as how that role fits
everyday staffneeds. Staffingmust be adequate into the organization. Orientation training gen-
to perform the work and to support the OMS. erally will include an overview of the organiza-
tion and its customers, benefits training, an in-
Job Descriptions troduction to cGMP and/or cGTP regulations,
Organizations should have well-written job de- privacy training, and safety training.
scriptions for all personnel. The job descriptions Specific training for tasks that are performed
should identify the key role and responsibilities as part of an individual's actual job usually oc-
of a particular position, as well as educational curs in the operational department where the
and experiential requirements. In some cases, individual is hired. Training on SOPs that the in-
the job description also contains physical re- dividual will need to perform those tasks is re-
quirements such as lifting a certain amount of quired. Additionally, each employee needs to
weight or the ability to stand for long hours. fully understand thecGMP/cGTP require-
Certain requirements in a job description may ments applicablein the performance of a job.All
be the result of regulatory requirements or in- training must be documented, and initial and
dustry standards. For example, in some states ongoing assessments of competence are re-
individuals must have certain licenses to per- quired.
form laboratory testing. Such requirements
should be well defined within each job descrip- Competency Assessments
tion. Job descriptions should be periodically re-
To ensure that staff maintain the ability to per-
viewed to ensure that they are truly.reflectrve of
form their jobs well, routine competency assess-
what an individual does in a particular job. Em-
ments should be conducted to determine their
ployees should review their job descriptions ini-
level of competence in performing the work.
tially, following revision, and periodically to un-
derstand and acknowledge the responsibilities of Organizations need to have a written plan for
the position. Often, regulators or accrediting conducting competency assessments, including
agencies request to see current Signed job de- what will be done if an individual does not pass.
scriptions during their inspections or assess- CMS has specific.requirements for competency
ments of an organization. An additional benefit assessments of testing personnel. The minimal
of a well-wrltten job description is that it serves regulatory requirements for assessment of com-
as an aid to the development of a training curric- petencyfor such individualsinclude:
ulum.
1. Direct observations of routine patient test
Hiring performance, including patient preparation,
if applicable, specimen handling, processing,
Human resources oversees the hiring process,
and testing.
which includes activities such as contacting can-
didates, setting up interviews, and ensuring new 2. Monitoring the recording and reporting of
employees are oriented. It may also include pre- test results.
employment screening such as drug testing. 3. Review of intermediate test results or work-
During the hiring process, job qualifications are sheets, QC records, proficiency testing (PT)
matched against applicant qualifications, and in- results, and preventive maintenance records.
4. Direct observations of the performance of for calibration that lists what should be calibrat-
instrument maintenance and function checks. ed, the frequency of calibration, and procedures
5. Assessment of test performance through test- for performing the calibration. The manufactur-
ing previously analyzed specimens, internal er should recommend the frequency of calibra-
blind testing samples, or external PT samples. tion, and regulations can be found in the CFR.4
6. Assessment of problem-solvingskills. However, if no guidance is available, the organi-
zation should follow standard practice in the in-
dustry or, if none exists, should establish a rea-
Competency assessment, which includes all
six items, must be performed for testing person- sonable frequency based on the criticality of the
nel for each test the individual is approved to measurement.
perform by the director of the CLlA-certifiedlab- The actual performance of calibration can be
oratory.3 outsourced to an approved outside vendor, but it
The competency program should be docu- is the organization's responsibility to maintain
mented and administered to all staff as required. calibration records and to ensure that the ven-
There should be a defined schedule for the ad- dor performs the activities in compliance with
ministration of the assessments. Documenta- applicableregulations and standards. Calibration
tion of the results of competency assessments records and procedures need to be available
should be available for inspection by regulatory during inspections and assessments.
or accrediting bodies. Equipment OC, performed routinely, is also
important in ensuring that equipment is operat-
Equipment Management ing as expected. Documentation of any OC that
is performed should be evaluated in a timely
Equipment used in processes must be installed manner, and the results should be evaluated to
as directed by the manufacturer and qualified to determine if there are trends over time that may
ensure that it is working as intended. This quali- indicate the equipment is beginning to drift. The
fication should be accomplished according to frequency of OC is dependent on the criticality
written procedures and documented. Installa- of the function of the equipment. Equipment
tion qualification is necessary as part of valida- used in donor eligibilitydetermination and test-
tion activities and will be addressed later in this ing, for example, may require daily OC because
chapter. Organizations must ensure that they of the criticality of its use. Review of OC records
operate equipment in line with manufacturers' needs to be timely to limit the scope of investi-
recommendations, which may include require- gation, should the review reveal a problem.
ments for temperature, humidity, surrounding
space, or other environmental conditions for op- Selection of Equipment
eration.
Equipment must be maintained to ensure Organizations should select equipment based on
proper working conditions. Organizations its ability to meet preestablished and document-
should have written programs for equipment ed specifications. Other factors that should be
cleaning and maintenance in line with manufac- considered are cost, service, history with others
turers' recommendations. Preventive mainte- in the industry, and support. Usually, organiza-
nance should be established and well docu- tions have several vendors from which to
mented. Records of this work must be available choose, and thus the additional factors become
for inspectors or assessorsto view. even more important. It is key that organiza-
For equipment used in measurement, routine tions establish criteria on the front end of the se-
calibration is required. Calibration involves lection process; the equipment should fit the or-
comparing a measurement device to a known ganization's needs. Workflow should be well
standard and then adjusting it, if necessary, to defined before selection of equipment. The orga-
measure the same as the standard. Routine cali- nization should not have to alter its process to fit
bration is a requirement for some equipment. the equipment unless there is only one supplier
An organization should have a written program and no other choice. The manufacturer of the
equipment should be qualified according to the suppliers, through the process of supplier qualifi-
organization's supplier qualificationprocess. cation, that meet those requirements.
Equipment Identification Supplier Qualification
EquiPment should be uniquely identified, and a Supplier qualification is a process whereby an

list of equipment should be maintained. This list organization determines whether or not ..a sup-
should be kept up to date, and when equipment plier can meet its requirements. Such require-
is moved from one location to another or re- ments usually include the abilityto meet regula-
moved from service, the action taken should be tions, the availabilityof supply, the timeliness of
recorded. Because of the amount of equipment delivery, responsiveness to issues and problems,
cost, and support. Other requirements may be
in an organization that performs blood banking,
specificto the organization. It is a common prac-
transfusion medicine, or cellular therapy activi- tice for organizations to participate in buying
ties, tracking equipment can be a daunting task. groups that perform qualificationof suppliers for
Software vendors have developed automated those participating in the group.
solutions to assist organizations with this, but Supplier qualification may include written
even if it must be done manually, the tracking of surveys or on-site audits of the supplier and re-
equipment is necessary. Equipment that is out of view of written surveys completed by current
service should be removed from operational ar- customers. Surveys may be more cost effective,
eas, if at all possible, and clearlylabeled as out of but on-site audits are generally considered best
service so it will not be used in the manufactur- if the supplier is providing materials or services
ing process. that are critical to operations. In fact, the more
critical the materials supplied, the more strin-
Suppliers and Materials Management gent the qualification should be. See Table 1-1
for a list of factors that may be considered
Ensuring that a supplier can provide what is during supplier qualification.
needed to perform the work and that the sup- An organization should maintain a list of ap-
plies meet preestablished specificationsis a criti- proved suppliers. This list should be reviewed
cal aspect of the OMS. Organizations must de- routinely for each supplier's ability to consistent-
termine and document requirements and seek ly meet the needs of the organization. Suppliers
TABLEH. Factorsto Consider in Supplier Qualification
Supplier-relevant quality documents Quality manual, complaint-handling methodology

Results of audlts ()rin~pections Previous FDA inspections, supplier qualification audit
Supply or product requirements Ability to meet functional requirements
Costof materials and services Product cost, maintenance fees, parts costs
Delivery arrangements Standing orders, turnaround time for stat
Financial security and market position How long the organization hasbeen in business, IRS 990
Support after sale Training, validation guidance, contract/agreement review meetings
FDA = Food and Drug Administration; ISO = International Organizationfor Standardization; EU = European Union; IRS 990 =
Internal RevenueService Form 990.
can be added or removed from the list as neces- ing, the product must be quarantined, either
sary. Management of the list usually falls to the physically or with clear labeling, until disposi-
quality department, although it could be placed tion is determined by the quality department.
in an area such as purchasing, with quality over- Supplies not meeting the preestablished criteria
sight. should remain in quarantine, and the supplier
should be notified of the issue. The inspection
Contracts and Agreements should be conducted according to a written pro-
cedure, and there should be documentation of
It is common practice to develop a written con-
the suitability of the supplies or their disposition,
tract or agreement with a supplier that stipulates
if found unsuitable. Most often these supplies
the organization's requirements and expecta-
are returned to the supplier, but they may be dis-
tions. The document should define the role of
both the organization and the supplier in the re- carded if the supplier does not need them for
lationship and should also stipulate the manner further investigation.
in which the supplier will operate to meet the
organization's needs. It is a good idea to estab- Process Control and Management
lish and document metrics that can be moni- Process control is the sum of activities involved
tored on a regular basis. If metrics indicate that in ensuring a process is predictable, stable, and
there is a problem, corrective actions should be consistently operating at the target level of per-
taken and documented. If the problem cannot formance with only normal variation. Important
be corrected, it may mean that the supplier aspects of process control include:
should be removed from the approved list.
AABBstandards stipulate that organizations SOPs.
should monitor their agreements. Other organi- Process validation.
zations have similar requirements. For exam- Computer system validation.
ple, The Joint Commission requires that hospital Test method validation.
blood banks establish metrics with their blood QC.
suppliers that are evaluated routinely. The eval- Training.
uation of metrics should be documented, as well Tracking and trending.
as any corrective actions required of the supplier
as a result of failure to meet the expectations. Standard Operating Procedures
Receipt and Inspection of Incoming Supplies SOPs provide instruction on how to perform a
task and are key to achieving consistency and
When supplies are initially received, it is import- control in operations. A full discussion of SOPs
ant that they are physically separated from sup- is provided in the "Documents and Records"
plies that are in use until they can be inspected section (further below).
for suitability for use. Some organizations have
caged areas where incoming supplies are quar- ProcessValidation
antined until inspected; others use shelving and
labeling, often with color coding, to quarantine One of the most important aspects of process
incoming supplies. The incoming inspection and control is the initial establishment that a process
release is most commonly performed by the consistently works to produce a desired result;
quality department, but in some instances, oper- that is, the validation of the process. Process val-
ations may conduct the inspection for quality, idation is defined as the collection and evalua-
Organizations should develop criteria for action of data, from the process-design stage
ceptance, and incoming supplies should be in- through commercial productton, that establish
spected against such criteria. Both external scientific evidence that a process is capable of
packaging and the contents of that packaging consistently delivering a high-quality product."
should be inspected for acceptance. If there is Validation establishes that a process has consis-
something wrong with the product or packag- tent results that meet predetermined require-
ments. Validations should be performed for all do a tremendous amount of testing of the soft-
critical processes according to a written valida- ware to determine limitations, etc; yet, the user
tion protocol. The protocol should contain the of that software must validate the software in
following: the user's environment with the user's staff and
SOPs. Consultants may be used to assist with
II> System description. validation, but final validation and the results of
<1/1 Purpose of the validation. that validation are the responsibility of the end
II> Riskassessment. user. The amount of validation work needed is
@ Responsibilities. dependent on the process, its criticality, and the
e Test cases. ability to test the end result 100% of the time or
@ Acceptance criteria. not. (See Fig 1-2.)
@ Problem-reporting mechanism. A risk assessment aids in the determination
e Approval signatures. of how much testing must be done. The more
II Supporting documentation. risk a process introduces, the more testing an or-
ganization normally does. This is especially im-
The system description identifies the compo- portant when there is no way to test the end re-
nents of the system used for the process and insult of a process other than destructive testing. If
cludes a description of how those components the process is not a high-risk or critical process,
work together during the process. It should also then less testing may be done, as the organiza-
include environmental conditions under which tion is willing to accept the risk should the pro-
the system operates, as applicable,and any utili- cess not work as expected.
ty specifications. Within a validation there are multiple roles.
The purpose of the validation is usually The individual who writes the validation proto-
straightforward. Validations may be conducted col is responsible for ensuring it is complete and
because a process is new or something signifi- contains all the necessary information and suffi-
cant has changed within the process, and assur- cient test cases to obtain the degree of assurance
ance is needed that the process still remains in a desired. Those who execute the validation pro-
validated state. A process validation has three tocol must have training in the process and may
phases: installation qualification, operational often be individualswho will perform the pro-
qualification, and performance qualification. In- cess routinely, although this is not always the
stallation qualification ensures that any equip- case. The quality department and others, as ap-
ment used within the process is installed appro- propriate, approve the validation protocol and
priately and is qualified to perform as the the final results of the validation before a pro-
manufacturer states. This evaluation should en- cess is implemented.
sure that the environment, including utilities, is Test cases should be developed to test the
appropriate for its operation as defined by the various parameters of the process and to chal-
manufacturer. It also ensures that necessary lenge the process as much as reasonable. The
SOPs are written, training is developed, and more testing that is performed, the more assur-
staff are trained in the execution of related ance that the process works consistently, but it
SOPs. Operational qualification demonstrates is not always possibleto perform enough tests to
that the process operates as intended, and it fo- get lOO%assurance. Usually an organization
cuses on the process capability(worst-case chal- seeks a comfort level of testing that is reasonable
lenges). The final phase, performance qualifica- and in line with industry standards. Each test
tion, demonstrates that the process works as case should have expected results. If those re,
expected in a normal working environment. sults are not obtained, a problem report must be
Although a manufacturer usually does a sig- executed, and there should be a resolution be-
nificant amount of validation work before bring- fore proceeding. Failure of a test case could be
ing equipment or software to market, the end the result of improper installation qualification,
user still has to perform the user's own valida- a poorly written test case, an unrealistic expect-
tion. For example, computer software vendors ed result, or poor execution of the test case
Verify and
Is the output YES Is verification sufficient YES
~ control the
verifia ble? and cost effective?
process.
NO NO
Accept risk
and control as
much as
LOW possible.
Determine level of risk. I
I
HIGH Fully validate
-~
Fully
t for business
reasons.
Redesign
validate. the process.
I
FIGURE 1·2. Process validation tree. Note: If low risk is not acceptable, a process design is usually
required to remove residual risk.
itself. If investigation does not produce a cause ifications. During the validation, an organization
and a resolution for the issue, then the process may uncover the following:
itself may have to be changed, or the process
may be implemented with limitations that are ® Design flaws.
documented within the validation summary. ~ Inadequate requirements.
The acceptance criteria for a validation must @ Errors in SOPs.
be documented before the validation work be- Errors in user manuals.
gins. This criteria should not be changed in the Training deficiencies.
middle of a validation unless there is a good rea- Incompatibilitieswith interfaces.
son to change it. If change does occur, the vali- Incompatibilities with the physical environ-
dation protocol must be amended, and the ment.
amended protocol must be approved again. Misconceptions about process capabilities.
Once the validation protocol is written, it
needs approval from operations and quality, at a Followingcompletion of the test cases, a vali-
minimum. In a CLlA laboratory setting, the dation summary is normally written. This sum-
CLlAlaboratory director must also approve vali- mary describes the expected and observed re-
dation work. Approval must occur before any sults and whether or not those results are
execution of test cases. As stated above, if there acceptable. It also delineates any problems en-
is good reason to modify the protocol, it is nec- countered during the validation and what was
essary to amend it and have it approved again. done to resolve those problems. It defines any
The protocol may include supporting documen- process limitations, either known before begin-
tation such as user manuals or pertinent techni- ning the validation or discovered during the vali-
cal articles. dation. Finally,it contains a conclusion based on
The validation proves the process is consis- the results. Before implementing the process,
tent and produces an end result that meets spec- the validation summary should be reviewed and
approved, again by operations, quality, and the proval or clearance, or makes modifications to
medical director, as appropriate. Supporting doc- an FDA-approved or -cleared test system, or if
umentation may accompany the summary, as the manufacturer does not provide perfor-
well as a timeline for implementation of the pro- mance specifications, then the laboratory must
cess, although this is often a separate document. establish the test system performance specifica-
It is important to remember that although a vali- tions before reporting patient results." At a mini-
dation gives an organization confidence in its mum, the following must be established for the
processes and significantlyreduces the need for test system:
end-product testing, no validation, no matter
how extensive, can test every possibilityor con- Il) Accuracy.
trol the human element. Continual process Il) Precision.
monitoring is needed to ensure the process re- Reportable range of test results for the test
mains in a validated state. system.
Reference intervals (normal values).
Computer System Validation Analyticalsensitivity.
A computer system is composed of hardware, Analytical specificity, including interfering
software, peripheral devices, networks, person- substances.
nel, and documentation. Validation of the com- Any other performance characteristic re-
puter system in the environment where it will quired for test performance (eg, specimen or
be used by those who will use it is required. reagent stability).
This also includes validation of interfaces be-
tween systems. For example, a blood establish- Based on performance speclflcations,the lab-
ment would need to validate the interface be- oratory must also establish calibration and con-
tween its BECS and a testing instrument. trol procedures and document all activities for
Although much work is done by the vendors test method validation. Title 42 CFR, Part
who develop software, the end user must still 493.1253 provides additional information on
perform a validation and may even repeat work this.
that the vendor has already done to ensure that
testing is performed under the environmental Quality Control
and process conditions experienced by the end OC is an important aspect of process control. It
user. An important part of validating a computer ensures the proper functioning of materials, re-
system is to ensure that the system can still op- agents, equipment, and methods. OC is an
erate when stressed. The FDAhas issued guid- event that is different from validation in that it is
ance to assist organizations in the performance not repeated to gain assurance of consistency,
of computer system validation." but rather it is repeated at a given frequency to
ensure results are within acceptable ranges, and
Test Method Validation also over time to determine if any trends are de-
When laboratories wish to implement a non- veloping that might indicate something is even-
waived test using an FDA-approvedor -cleared tually going to fail. The frequency of OC testing
test system, CLlArequires that the performance is usually determined by the criticality of what is
specifications established by the manufacturer being tested. Some OC frequency is dictated by
be verified by the laboratory before it reports pa- regulatory or accrediting bodies such as the
tient results'? At a minimum, the laboratory FDA.4Additional information on OC frequency
must demonstrate that it can obtain perfor- is provided in Appendix 1-3. All OC should be
mance specificationscomparable to those of the well documented to include who did the test-
manufacturer for accuracy, precision, reportable ing, the date it was done, the results, and
range, and reference intervals (normal values). whether or not the testing was acceptable.
If the laboratory develops its own method, Documentation should be concurrent with the
introduces a test system not subject to FDA ap- performance of the testing, and records should
be available for future inspections or assess- Document Creation

ments.
Documents should be created in a consistent
Unacceptable QC results should be evaluat-
manner. An SOP should be in place to define the
ed, and the process should not continue until format of documents as well as the review and
the issue is resolved. Corrective actions may be approvalprocess, both initially and at routine in-
necessary before acceptable QC can be ob- tervals. Documents should have unique identifi-
tained. Items that fail QC should be marked cation (eg, numbering system), and changes to
"not for use" until the issue is resolved. Because documents should be made in a controlled man-
QC is performed on a schedule, if a failure oc- ner (change control). Document control is a key
curs, it may be necessary to assess product pro- element of process control. Many organizations
duced since the last acceptable QC result. This now have computerized document control sys-
is clearlywhy determining the criticalityof what tems that are validated for activities such as doc-
is being tested is important; the more frequent ument development, routing for review and
the QC, the less product is involved in the eval- approval, controlled printing, revision, and ar-
uation. chival. Most organizations are moving away
from paper documentation as much as possible.
Back-up documents are required, however, for
Training
those instances when computers are down.
Orientation training is critical for new employ-
ees and is discussed along with specificjob train- Quality Manual
ing and competency assessment in the "Human The quality manual is one of the most important
Resources" section. Workplace safety training is documents in a blood or cellular therapy estab-
detailed in Chapter 2. lishment. The quality manual describes the or-
ganization's quality policy, quality objectives,
Tracking and Trending and overall approach to quality in all aspects of
Tracking is an integral part of record-keeping the business. It defines how the organization is
and is noted in the following section ("Docu- structured (organization chart) to ensure imple-
mentation of the quality system and defines the
ments and Records"). Trending is a concept em-
roles of staff, including line staff all the way up
bodied in many quality system activities.
to senior management. It points out how the
quality system integrates with operational tasks
Documents and Records
and how those tasks are monitored to ensure
Documentation is important in that it provides quality outcomes.
evidence and details of what was done. Good
documentation can provide full traceability (de- Policies and ProcessDocuments
tails) and trackability (a logical sequence of Policies describe the manner in which an orga-
steps) in the execution of processes. Organiza- nization operates. They are high-level docu-
tions involved in the production of blood and ments describing the position that an organiza-
cellular products create many documents and tion takes on a particular topic. Not all policies
records. These include: are regulated. For example, a dress code policy
or a tobacco use policy is not required by any
'" Quality manuals. regulatory or accrediting bodies, but if an orga-
'" Policiesand process documents. nization wants staff to dress a particular way or
SOPs. to avoid the use of tobacco in the work place, a
@ Work instructions. policyis a good way to document and communi-
O!}Job aids. cate the information. As necessary, policies are
'" Forms. supported by other forms of documentation
Labels. within an organization, such as SOPsand forms.
Process documents are also high level and structions need to be made in a controlled man-
describe the inputs for a process, the conversion ner that allows for the changes to be approved,
that takes place, and the output of a process. A made, validated, and communicated to all stake-
process document provides the big picture and holders before implementation. Retraining may
may take the form of a flow chart. This is partic- be necessary for significantSOPchanges.
ularly helpful when trying to understand a pro-
cess at a high level and when introducing con- Job Aids
cepts to personnel in training.
A job aid is an excerpt from an approved proce-
Standard Operating Procedures and Work dure or work instruction. These are often used
Instructions when a portion of the SQP has a table or infor-
mation that must be frequently referenced. Job
SOPs describe who does what and when (in se- aids must be controlled just as procedures and
quence or order); they describe the steps of a work instructions, and there should be a way to
process. Well-written SOPs provide the "how" reference a job aid to the procedure it rep-
in performing a process. They should be detalled resents. Uncontrolled job alds should not be al-
enough for a trained individual to perform the lowed. Job aids may be included as hyperlinks in
task, but not so detailed that they are unneces- computerized documents.
sarily restrictive. SOPs should be written with
input from subject-matter experts and should be Forms
validated to ensure that they are effective. SOP
validation usually involves an individual per- Blank forms provide templates for the capture of
forming the task using the SOP as written. The information. Once completed, a form becomes a
individual notes whether the steps In the SOP record providing objective evidence of work per-
make sense, whether the steps can be performed and subject to record retention require-
formed as written, and if the desired outcome is ments. Forms should be managed within docu-
achieved. Finalized SOPs should be reviewed ment control and should be designed by
and approved by the appropriate department individuals with experience; it is not true that
personnel and the medical director, as appropri- anyone can design a form. Often mistakes can
ate, and then approved by quality before becom- be avoided by careful form design. If the form is
ing effective and released. Staff should receive not self-explanatory, instructions on how to
training on all SOPs applicableto their jobs, and complete it should be available, and individuals
SOPs must be accessible at all times to staff per- should receive training on completion of the
forming the work. form. This will reduce the likellhood of errors.
SOPsneed to be periodicallyreviewed to en- Forms may be included as part of SOPs and may
sure that they are current and are reflective of be hyperlinked in computerized documents.
the work as it is being done. Some organizations
review a portion of their SOPs each quarter to Labels
ensure that the entire collection of SOPs is re-
viewed at the required interval. Although not always thought of as a document,
Work instructions provide step-by-step in- labels need to be created and maintained within
structions for how a task is performed. They are the document control system to ensure that the
more specific and more detalled than proce- label is correct, meets regulatory requirements,
dures. Not all organizations have work instruc- and is current. Changes to labels need to be
tions; some organizations choose to just use the managed just as changes to documents are:
term procedure for all step-by-stepdocuments. within a controlled system, reviewed for accura-
Whatever term an organization chooses, the cy and compliance, and approved. Certain labels
documents describing how the work is done still must be submitted to the FDA for approval."
need to be controlled and managed in a consis- Organizations must maintain a current master
tent manner. Changes to SOPs and/or work in- set of labels for reference.
Document Maintenance Policies,process documents, procedures, and

completed forms are also examples of records
As previously stated, documents should be cre-
found in an organization involved in transfu-
ated and maintained in a controlled manner.
sion medicine and/or cellular therapy, and de-
Version control is critical. Organizations also
scribe how the work was being performed at
need to have a mechanism whereby changes to
any particular time in the organization's history.
documents can be requested and communicat-
The records may be in paper or electronic form
ed once those changes have occurred. A docu-
and must be easilyidentified. There should also
ment history that records changes to a docu-
be information as to who created the record. Be-
ment should be developed and maintained.
cause organizations may use both signatures
When a document is revised, and the revised
and initials, it is necessary to maintain a current
copy is approved and released, an archived copy
signature sample and list of initials for all em-
of the originalversion should be retained for fu-
ployees. If the identity of the record creator is
ture reference.
captured electronically by entry into the com-
Organizations should prepare a master list of
puter system or by electronic badge swipe, this
all the various types of documents in use. This
needs to be in compliance with electronic
list should define the most current version, how
record-keepingrules.
many copies are out, and where those copies are
Because of the nature of the records created
located. This aids in document control; when by transfusion medicine or cellular therapy orga-
revisions occur, it helps to ensure that all old
nizations, many records are confidential, espe-
copies of the documents are removed and re- ciallythose containing donor or patient informa-
placed with the revised version. tion. Records should never be left where they
can be viewed by individualswho have no need
Records
to view them. If records containing confidential
Records are the evidence of what was done and information are made available to those outside
prove that procedures were followed and docu- the organization, any confidential information
mentation of the work performed was captured. should be redacted.
Records should be created concurrently with Records, whether in paper or electronic
the performance of the work, documenting each form, must be protected from unauthorized
critical step. Good records provide the details changes, from inadvertent destruction, and from
(traceability-who, what, when, where, how) damage caused by rodents, fire, or water. Re-
and a logicalsequence of steps taken (trackabili- cord storage should be designed to accomplish
ty). It is also important that records are perma- these goals. It also should allow the records to
nent, which means that indelible ink should be be retrieved easily. Access to records should be
used. Any corrections should be made in a man- restricted, particularly if the records contain
ner that allows one to see what the error was, confidentialinformation.
who performed the correction, and when. Re- Organizations need to have a written record-
cords should be managed so that the following retention policy that is compliant with regula-
aspects are addressed: tions and standards; records should be retained
in accordance with that policy. Once a record
<l! Creation and identification of records. has reached its "end of life," it should be dis-
o Confidentiality. carded in a manner that protects any confiden-
'" Protection of the integrity of records. tial information. Such destruction methods in-
Protection from inadvertent destruction. clude shredding, burning, and degaussing (a
II> Protection from damage from rodents, fire, method to magnetically erase data from elec-
and water. tronic storage media, eg, hard drives).
III Storage and retrieval. Many organizations outsource record stor-
@ Retention. age, retrieval, retention, and destruction to off-
@ Destruction. site vendors. Such vendors should be qualified,
and organizations need to ensure timely access Backup of critical electronic information is
to their records for inspections and assessments. important. Backups should be routinely run,
If records are stored electronically, an organi- and there should be written procedures to re-
zation must ensure that the integrity of the elec- store any data that may be inadvertently lost.
tronic data is protected from unauthorized Records should be retained according to writ-
changes. Additionally,the data must be stored in ten policies that ensure the timely destruction of
a manner that would not cause inadvertent loss documents that are no longer necessary.
of data from overwriting, physical damage, or
system crashes. Data integrity should be as- Management of Nonconforming Events
sessed periodically.
Organizations must have a documented Because nonconformances are inevitable in any
mechanism for error correction for paper docu- system that involves human involvement, the
ments and for electronic documents. In both OMS should contain processes and procedures
cases, it is important that the error is not obliter- to detect, document, investigate, correct, and
ated..The common industry practice for correct- follow up on nonconforming events such as the
ing errors on paper documents is to draw a sin- production of a product that does not meet spec-
gle line through the error, write the correction ifications. Such processes and procedures must
above it and then initial and date the correction. be in line with regulations and applicable stan-
If an explanation for the correction is needed, dards and should include the following:
this can be.written alongside the correction. If
there is insufficient room, an asterisk can be Documentation of the event, either electron-
used and the explanation written elsewhere on icallyor on paper, with some sort of classifi-
the document, even on the back. Electronic cation, usually based on the severity of the
document maintenance should allow an audit risk.
trail to show what the error was, what correc- Determination of the effect, if any, on the
tion was made, who made it, and when it was quality of products or services.
made. Evaluation of the effect on interrelated activi-
ties.
Information Management
e Investigation and root cause analysis.
Selection of appropriate corrective action.
Organizations have a tremendous amount of in- Implementation of corrective action, as ap-
formation that must be managed as part of the propriate.
quality system, and much of that information, as Notification and recall.
previously stated, must be confidentially main- Implementation of appropriate preventive ac-
tained. Access to information should be limited tion.
to those who need the information for work Reporting to external agencies when re-
purposes. The integrity of data, both written and quired.
electronic, must be guarded. Unauthorized Evaluation of the effectiveness of the correc-
copying of information, whether paper or elective and preventive action (CAPA)taken.
tronic, should not be allowed. If documents are
maintained in a paper state and contain highly Staff should be trained to find and report
confidential information, they should be in nonconformances, which include errors, acci-
locked file cabinets, and if stored electronically, dents, and adverse reactions in donors and re-
they should be protected by access rights. Al- cipients. It is important to capture the facts of
though the topic is not within the scope of this the event in sufficient detall to allow a complete
chapter, organizations that store protected and thorough investigation.
health information must be compliant with the It is critically important, once a nonconfor-
Health Insurance Portability and Accountability mance is discovered, to determine the impact of
Act (HIPM). More information on HIPM may that nonconformance on products and/or
be found at www.hhs.gov/hipaa. services. If the nonconformance has a negative
impact on product quality, it may be necessary ing causes and contributing factors; a failure
to quarantine the product(s) or to perform a re- mode effects analysis(FMEA),a step-by-stepap-
call if the product has been distributed. The proach for identifying all possible failures in a
sooner an organization can gain control of the design, a manufacturing or assemblyprocess, or
affected product, the better. Organizations a product or service; and the five whys, useful
should always consider the impact on their for drilling down to the true cause. For the last
products and/or services as soon as possible af- one, there is no magic to the number five; one
ter discoveryof a nonconformance. may ask fewer or more whys to get to the true
It is also a good idea to determine if the non- cause of a nonconformance. (See Fig 1-3.)
conformance has impact on other areas of the Once the root cause has been identified, it is
organization's operations. It may be necessary to then important to select an appropriate correc-
involve more than one department in the inves- tive action. The action should fix the issue, but
tigation to fully understand what occurred and at the same time, the corrective action must be
its impact. reasonable. For example, if the true fix for a
Not all nonconformances require a full inves- problem is a new computer system to capture
tigation, but there is normally some level of in- specific data accurately, this cannot be accom-
vestigation required for most nonconformances. plished quickly,so a more reasonable alternative
A thorough investigationmay involve interview- may need to be chosen until the new computer
ing staff, reviewing training records, direct ob- system can be implemented. Also, although it
servation of the process, and reviewing SOPs. seems intuitive, it should be noted that an orga-
Also, investigationsmay involve other record re- nization must ensure the corrective action actu-
view or review of data to determine the extent
ally addresses the true issue and is not merely
of the nonconformance.
addressing a symptom of the problem. In the ex-
Root cause analysis is a collective term that
ample given, the organization may choose to im-
describes a wide range of approaches, tools, and
plement a newly designed form to reduce the
techniques used to uncover causes of problems.
likelihood of error or may implement a second
A root cause analysis to determine the cause or
causes of the nonconformance is often neces- review for accuracy. A note of caution here,
sary. It is important to continue to ask why, to however, is that simply adding more review sel-
determine the true root causers). Without find- dom fixes the problem. In fact additionalreview
ing the true cause of a nonconformance, it is can sometimes make things worse.
possible that the problem will recur. If the root As part of corrective action, notification of
cause is fixed, the problem should not recur. Al- customers about the nonconformance may be
though there is substantial debate on the defini- necessary, and it may be necessary to recall addi-
tion of root cause, the followingare considered tional product, depending on the scope of the is-
true": sue and the results of the investigation. The or-
ganization should have a documented process
® Root causes are specificunderlying causes. for these actions.
® Root causes are those that can reasonably be Correcting the nonconformance is import-
identified. ant, but equally important is determining if
III Root causes are those management has con- there are actions that can be taken to prevent
trol to fix. the issue from recurring ever again. There are
® Root causes are those for which effectiverec- short-term corrective actions, which fix the
ommendations for preventing recurrences problem temporarily, and long-term corrective
can be generated. actions, which normally include preventive ac-
tion and permanently fix the problem. Finding
There are several tools that are useful in per- the best way to minimize the likelihood of the
forming a root cause analysis. These include problem recurring while maintaining the ability
brainstorming, useful in generating potential to operate within the constraints of limited re-
causes; a fishbone diagram, useful in determin- sources is key.
FIGURE 1·3.The five whys. Note: There is no magic to asking five times; it may take more or less to
get to the root cause.
Preventive actions should be implemented should include a description of any new proce-
whenever possible. For example, proper training dures implemented to avoid recurrence. MBB
might be considered a preventive action if it is Association Bulletin #04-06 provides additional
determined that a nonconformance resulted information on these reporting requirements, in-
from a lack of training. In this instance, training cluding a form for reporting donor fatalities.12
might be considered both a corrective and a pre- Regardlessof their licensure and registration
ventive action. It fixes the current problem, and status with the FDA, all donor centers, blood
it prevents future occurrences at the same time. banks, and transfusion services must promptly
Depending on the nature of the nonconfor- report biological product deviations (BPDs)-
mance, regulatory bodies or accrediting agen- and information relevant to these events-to
cies may need notification as well. Processes the FDA13,14 using Form FDA 3486 when the
and procedures should include information on event 1) is associated with manufacturing (ie,
whom to notify and when. Avoluntary recall is collecting, testing, processing, packing, label-
instituted for nonconforming product that has ing, storing, holding, or distributing); 2) rep,
been distributed. In-house products can be dealt resents a deviation from cGMP, established
with directly. specifications, or applicable regulations or stan-
Fatalities related to blood collection or trans- dards, or that is unexpected or unforeseen; 3)
fusion or to cellular therapy products must be may affect the product's safety, purity, or poten-
reported as soon as possible to the FDA Center cy; 4) occurs while the facilityhad control of, or
for Biologics Evaluation and Research (CBER). was responsible for, the product; and 5) involves
[See21 CFR 606.170(b) and 1271.350(a)(i), re- a product that has left the facility's control (ie,
spectively.] Instructions for reporting to CBER has been distributed).
are availablein published guidance" and on the Using the same form, facilities must also
FDAwebsite. I I A written follow-upreport.must promptly report BPDs associated with a distrib-
be submitted within 7 days of the fatality and uted cellular therapy product if the event
20 A A B B TE C H N I CAL MAN UA L
represents a deviation from applicable regula- quate, particularly for those events that are rare
tions, standards, or established specifications (eg, severe syncope or machine failure). The as-
that relate to the prevention of communicable sessment should cover the OMS and, at a mini-
disease transmission or contamination of the mum, processes that are critical to the organiza-
product. This requirement pertains to events tion's operations. When issues are found during
that are unexpected or unforeseeable but may the assessment, the process should include a
relate to the transmission or potential transmis- mechanism to respond to those issues, ensuring
sion of a communicable disease or may lead to that important stakeholders are aware of the is-
product contamination." More information sues and what corrective actions, if any, are
concerning BPD reporting can be found on the planned. The quality department should take re-
FDAwebsite." sponsibility for oversight of these assessments
There must also be a mechanism to report and to ensure that actions are taken as warranted.
medical device adverse events to the FDA and
the device manufacturer." The Joint Commis- Quality Indicators
sion encourages reporting of sentinel events, in- Quality indicators are statistical measures that
cluding hemolytic transfusion reactions involv- give an indication of output quality. They are
ing the administration of blood or components useful in the evaluation of customer require-
having major blood group incompatibilities.18,19 ments, personnel, inventory management, and
Hemovigilance reporting provides an oppor- process control and stability (this list is not all in-
tunity to detect, investigate, and respond to ad- clusive).Quality indicators may be based on out-
verse transfusion reactions and events that comes such as quantity-not-sufficient (ONS)
result in nonconformances. A number of organi- rates, or they may be based on the process's abil-
zations monitor such data, including FDA and ity to deliver an expected result consistently. As
the Centers for Disease Control and Prevention an example of a process quality indicator, if a
(CDC). customer requires stat deliveries to arrive within
1 hour, the percentage of deliveries that meet
Monitoring and Evaluation the customer's requirement is a measurement of
Organizations should have a system for monitor- the abilityof the process to deliver within the re-
ing and evaluating the effectiveness of the orga- quired time frame. Organizations should estab-
nization's processes. This system should be de- lish alert limits for quality indicators; involving
fined as part of the OMS. Monitoring can occur the customer is important in malting sure that
at various levels: the level of input to the pro- alert limits are appropriately set.
cess, the in-process activities, the results, or the Organizations should communicate quality
process, or even the system in which the pro- indicator results frequently so stakeholders are
cess resides. Although record review and analy- aware of how the organization is performing.
sis are ongoing forms of monitoring, the use of Customers may want to be included in this re-
porting. Run charts, control charts, and bar
internal and external assessments of the process-
charts are often useful in displayingquality indi-
es is very useful. Assessments may include com-
cator data. Control charts allow an organization
parison of actual to expected results and can
to see if the process is operating as expected; if
consist of quality assessments, peer reviews,
not, corrective actions are indicated.
self-assessments,and PT.
Organizations should have a process that de-
Blood Utilization
scribes how internal assessments are conducted.
Each assessment should be well planned and In recent years, organizations have become
conducted according to the plan. Assessorsmay even more focused on blood utilization patterns.
look at data such as quality indicators and other This is driven by the desire to decrease costs and
quality records, observe processes as they are to provide better patient care. Patient blood
performed, or interview staff to determine if management (PBM)as a discipline has come to
knowledge of processes and procedures is ade- the forefront; many organizations now have a
staff member devoted to transfusion safety: the External Assessments

transfusion safety officer (TSO).Utilization com-
External assessments are those that are conduct-
mittees review physician ordering and transfu- ed by agencies and organizations that are not af-
sion practices routinely. The committees also re- filiated with the facility being assessed. They
view sample collection and labeling, adverse may be voluntary, as in the case of MBB assess-
events in patients, near-miss events, outdates, ments, or mandatory, as in the case of FDA in-
discard, appropriateness of use, and compliance. spections. Organizations that assess or inspect
Many hospitals have set up order alerts in the blood banks, transfusion services, and cellular
hospital computer system to appear when physi- therapy facilitiesinclude the following:
cians order outside of established guidelines.
MBB has published clinical practice guidelines '$ MBB.
for red cell and platelet transfusion.P'" College of American Pathologists (CAP).
Alternatives to red cell transfusion, such as Commission on Office Laboratory Accredita-
tion (COLA).
preoperative anemia treatment, are under study
Centers for Medicare and Medicaid Services
or have been incorporated into routine practice
(CMS).
in many health-care settings in efforts to de- The Joint Commission.
crease the need for transfusions. Physicians are Foundation for the Accreditation of Cellular
asked to use data to determine if a second trans- Therapy (FACT).
fusion is warranted instead of issuing a blanket Food and Drug Administration (FDA).
order for transfusion of 2 units, a common prac- State health departments.
tice among transfusing physicians. (See Fig 1-4
for an example of this kind of improvement seen The list includes both accrediting organiza-
as a result of the implementation of a PBM pro- tions and regulatory entities. Other agencies
gram in a large teaching hospital.) The use of such as the Department of Transportation
thromboelastography (TEG)or thromboelastom- (DOT) and the Nuclear Regulatory Commission
(NRC)may assess the organization as well.
etry (TEM) to guide physicians on when to
Accreditation is voluntary, whereas regula-
transfuse to correct coagulation factor levels is tion is the law. The external assessments con-
now common practice. In addition to providing ducted by these entities are usually against stan-
better patient care, PBM limits needless transfu- dards or regulations promulgated by the
sions, thereby saving dollars and ensuring com- organization that is performing the assessment.
ponents are available for those who need them Although with any assessment or inspection
most. there is some level of angst, these assessments
85 ,------------
65 1-------------
!55
'.:i
~ 45 ~l-unit orders
c!! -2-unit orders
15
FIGURE1·4. Results of patient blood management in reducing 2-unit blanket transfusions (lower line)
at the University of Tennessee Medical Center Blood Bank, Knoxville, Tennessee. (Courtesy of Dr.
Chris Clark and Anna Rains.)
are usually beneficial to the facility and are op- patient safety. Such processes should include a
portunities for learning and improving opera- method to determine root causes of issues along
tions. It is important that staff are trained on with appropriate corrective and preventive ac-
how to conduct themselves during an assess- tions and an evaluation of the effectiveness of
ment or inspection, both to decrease their anxi- those actions. Information gleaned from the
ety and to ensure they understand the role and nonconformance management system should be
scope of authority for the assessor. used to improve operations. This is a primary
If issues are discovered during an external as- benefit of an effectivenonconformance manage-
sessment or inspection, these are usually document process. Other sources of improvement
mented and provided to the facility in an exit opportunities include:
meeting. The facility should perform root cause
analysis and implement corrective actions as Customer-supplier established metrics.
required. Normally the facility will submit a Complaints.
written response to the entity performing the QC records.
external assessment. Just as with internal assess- PT.
ments, management needs to be well informed Internal audits.
of the findings and corrective actions related to Quality indicators.
those findings. External assessments.
II> Financialanalysis of operations.
Proficiency Testing
PT is the testing of samples previously unknown Many organizations combine the principles
to a laboratory that are sent by a CMS-approved of Lean manufacturing and Six Sigma as part of
PT program. There are a number of organiza- their continuous improvement processes. Lean
tions that provide PT samples; AABB,for exam- Six Sigma is "a managerial approach that com-
ple, provides PT for immunohematology refer- bines Six Sigma methods and tools of the Lean
ence laboratories. Normally a facility will manufacturing/lean enterprise philosophy, striv-
perform at least three testing events each year. ing to eliminate waste of physical resources,
PT samples should be managed as any other time, effort,and talent, while ensuring quality in
samples that the laboratory tests, and the testing production and organizational processes.v"
should be rotated among staff so that different Lean Six Sigmahas two objectives: l) a focus on
individuals are tested. Submitted results are eliminating non-value-added steps in processes
compared to other laboratories, and a pass/fail and 2) eliminating defects and improving the
determination is assigned. Accrediting organiza- overall process. Lean Six Sigma uses the define,
tions monitor the PT results for facilitiesthey ac- measure, analyze, improve, and control (DMA-
credit. When failures occur, it is expected that IC) method-a five-stepapproach to process im-
the facilitywill conduct an investigation, try to provement. (See Table 1-2.) This approach can
find the root cause of the issue, and implement be used not only for problem-solvingbut also for
any CAPAthat is needed. Additionally, the ac- process improvement. Organizations that have
crediting body may require subsequent submis- implemented Lean Six Sigma have made signifi-
sion of one or more surveys following a failure cant improvements in their processes while sav-
to ensure the testing process remains in control ing valuable resources.
and the CAPAwas effective.
Facilities, Work Environment, and
Process Improvement Safety
Continuous improvement is a tenet of any quali- Facilities must be adequate for the work per-
ty program, and organizations that manufacture formed and must be maintained to provide a
blood components or cellular therapy products safe environment for staff, patients, donors, and
should have processes in place that allow for visitors. The facility itself must be clean and or-
continuous improvement in operations and in derly so as not to jeopardize staff or product
TABLE1·2. DMAIC Process

" Define the problem, improvementactivity, opportunity for improvement, project goals, and customer (inter-
nal and external) requireme~ts
• Measure process performance
• Analyzethe process to determine root causes of variation, poor performance (defects)
• Improve process performance by add~essingand eliminating the root causes
" Control the improved process and future process performance
safety. Sufficient space is necessary within the A written disaster plan that addresses both
facility to prevent mix-ups. during the. perfor- internal and external disasters is crucial for the
mance of processes, and .buildmgutilities, venti- organization, addressing what to do in the event
lation, sanitation, trash, and hazardous sub- of a disaster to maintain safety for staff and to
stance disposal must support the organization's maintain business continuity as much as possi-
operations. Safetyconcerns include general safe-
ble. Staffmust have training in safety and on the
ty components such as the use of nonslip surfac-
es and proper lifting techniques, as well as.fire disaster plan itself. A routine test of the disaster
safety, biologic and chemical safety, radiation plan, includini theVari()lls scenarios that might
safety, and disaster preparedness, response, and produce disaster Situations, needs to be exer-
recovery. cised under the oversight of management.
24 A A B B TE C H N I CAL MAN UAL
high degree of confidence that processes are reliable and reproducible by ensuring appropriate
installation qualification of equipment, operational qualification (including personnel training
and SOP development), and performance qualification to achieve stated outcomes.
7. Documents and Records. Documents include policies, process descriptions, procedures,
work instructions, job aids, forms, and labels. Records provide evidence that the process was
performed as intended and allow assessment of product and service quality.
8. Information Management. Unauthorized access to, or modification or destruction of, data
and information must be prevented, and confidentiality of patient and donor records main-
tained. Data integrity should be assessed periodically, and backup devices, alternative systems,
and archived documents maintained.
9. Management of Nonconforming Events. Deviations from facility-defined requirements,
standards, and regulations must be addressed by identifying, documenting, and classifying oc-
currences and assessing effects on quality. Root cause analysis is critical in determining appro-
priate corrective action. Nonconforming events that create potential risk for the public require
reporting to external agencies.
10. Monitoring and Evaluation. Assessment of facility processes includes internal and external
.assessments, monitoring of quality indicators, blood utilization assessment, proficiency testing,
and analysis of data.
11. Process Improvement. Opportunities for improvement may be identified from deviation re-
ports, nonconforming products and services, customer complaints, QC records, proficiency
testing results, internal audits, quality indicator monitoring, and external assessments. Process
improvement includes determination of root causes, implementation of corrective and preven-
tive actions, and evaluation of the effectiveness of these actions. The implementation of Lean
Six Sigma can significantly increase efficiencies and reduce opportunities for error.
12. Facilities, Work Environment, and Safety. Procedures related to general safety; biologic,
chemical, and radiation safety; fire safety; and disaster preparedness are required. Space alloca-
tion, building utilities, ventilation, sanitation, trash, and hazardous substance disposal must
support the organization's operations.
REFERENCES
1. Food and Drug Administration. Guideline for 5. Food and Drug Administration. Guidance for in-
quality assurance in blood establishments. (July dustry: Process validation: General principles
11, 1995) Silver Spring, MD: CBER Office of and practices. (January 2011) Silver Spring,
Communication, Outreach, and Development, MD: CBER Office of Communication, Out-
1995. reach, and Development, 2011.
2. Business dictionary. Quality planning. Fairfax, 6. Food and Drug Administration. General princi-
VA:WebFinance,Inc, 2019. [Availableat http:// ples of software validation; final guidance for in-
www.businessdictionary.com/definition/quality- dustryand FDA staff. (January 11, 2002) Silver
planning.html (accessedSeptember8, 2019).] Spring, MD: CBEROffice of Communication,
3. Centers for Medicare and Medicaid Services. Outreach, and Development, 2002.
What do I need to do to assess personnel com- 7. Code of federal regulations. Title 42, CFR Part
petency? Baltimore,MD: CMS,2012. [Available 493. Washington, DC: US Government Publish-
at https://www.cms.gov/Regulations-and- ing Office,2019 (revisedannually).
Guidance/Legislation/CLIA/Downloads/ 8. Food and Drug Administration. Guidance for in-
CLIA_CompBrochure_508.pdf.] dustry: Changes to an approved application:Bio-
4. Code of federal regulations. Title 21, CFR Part logical products: Human blood and blood com-
606.60. Washington, DC: US Government Pub- ponents intended for transfusion or for further
lishingOffice,2019 (revisedannually). manufacture. (December 2014) Silver Spring,
MD: CBER Office of Communication, Out- 15. Code of federal regulations. Title 21, CFR Parts
reach, and Development, 2014. 1270 anq1271. Washington, DC: US Govern-
9. RooneyJI, Vanden Heuvel LN. Root cause anal- ment Publishing Office, 2019 (revised annual-
ysis for beginners. Quality Progress 2004;37:45- ly).
53. 16. Food and Drug Administration. Biological prod-
10. Food and Drug Administration. Guidance for in- uct deviations: Includes human tissue and cellu-
dustry: Notifying FDA of fatalities related to lar and tissue-based product (HCT/P) deviation
blood collection or transfusion. (September reporting. Silver Spring, MD: CBER Office of
2003) Silver Spring, MD: CBEROffice of Com- Communication, Outreach, and Development,
munication, Outreach, and Development, 2019. [Availableat https:llwww.fda.gov/vac
2003. cines-b100d-biologics/report-problem~center-
11. Food and Drug Administration. Transfusionl biologics-evaluatlon-research/biological-prod
donation fatalities: Notificationprocess for trans- uct-deviatlons.]
fusion related fatalities and donation related 17. Code of federal regulations. Title 21, CFR Part
deaths. Silver Spring,MD: CBEROfficeof Com- 803. Washington, DC: US Government Publish-
munication, Outreach, and Development, ing Office,2019 (revised annually).
2019. [Available at https:/Iwww.fda.gov/ 18. Hospital accreditation standards. Oakbrook Ter-
vaccines-blood-biologics/report-problem-center- race, IL:Joint Commission Resources, 2019.
biologics-evaluation-research/transfusiondona 19. Laboratory accreditation standards. Oakbrook
tion-fatahties.] Terrace, IL:Joint Commission Resources, 2017.
12. Reporting donor fatalities. Association bulletin 20. Carson JL, Grossman BJ, Kleinman S, et al. Red
#04-06. Bethesda, MD: MBB, 2004. blood cell transfusion: A clinical practice guide-
13. Code of federal regulations. Title 21, CFR Parts line from MBB. Ann Intern Med 2012;157:49-
606,610,630, and 640. Washington, DC: US 58.
Government Publishing Office, 2019 (revised 21. Kaufman M, Djulbegovic B, Gernsheimer T, et
annually). al. Platelet transfusion: A clinical practice guide-
14. Food and Drug Administration. Guidance for in- line from the MBB. Ann Intern Med 2015;
dustry: Biological product deviation reporting 162:205-14.
for blood and plasma establishments. (March 22. Investopedia. Lean Six Sigma. New York: In-
2020) Silver Spring, MD: CBEROffice of Com- vestopedia LLC, 2018. [Available at http://
munication, Outreach, and Development, www.investopedia.com/terms/l/lean-six-
2020. sigma.asp (accessedSeptember 9, 2019).J
APPENDIX 1-1
Glossary of Commonly Used Quality Terms
Biovigilance Collection and analysis of adverse-eventdata for the purpose of improving outcomes in
the collection and use of blood components, organs, tissues, and cellular therapies.
Calibration Comparison of measurements performed with an instrument to those madewith a more
accurate instrument or standard for the purpose of detecting, reporting, and eliminating
errors in measurement.
Changecontrol Establishedprocedures for planning, documenting, communicating, and executing
changesto infrastructure, processes,products, or services. Such procedures include the
submission, analysis, approval, implementation, and postimplementation review of the
change and decisions made about the change. Formal changecontrol provides a mea-
sure of stability and safety and avoids arbitrary changesthat might affect quality.
Control chart A graphic tool usedto determine whether the distribution of data values generatedby a
process is stable over time. A control chart plots a statistic vs time and helps to deter-
mine whether a process is in control or out of control according to defined criteria
(eg, a shift from a central line or a trend toward upper or lower acceptancelimits).
End-product test Verification through observation, examination, or testing (or a combination) that the fin-
and inspection ished product or service conforms to specified requirements.
Near-miss event An unexpectedoccurrence that did not adverselyaffect the outcome but could have
resulted in a serious adverse event.
Process An action that takes input(s) and transforms it into output.
Process control Activities intendedto minimize variation within a processto produce a predictable output
that meets specifications.
Qualification Demonstration that an entity is capableof fulfilling specified requirements and verifica-
tion of attributes that must be met or complied with for a person or thing to be consid-
ered fit to perform a particular function. For example,equipment may be qualified for an
intended use by verifying performance characteristics, such as linearity, sensitivity, or
easeof use.An employee may be qualified on the basis of technical, academic,and prac-
tical knowledge and skills developedthrough training, education, and on-the-job perfor-
mance.
Quality assurance Activities involving quality planning, control, assessment, reporting, and improvement
necessaryto ensure that a product or service meets defined quality standards and
requirements.
Quality control Operationaltechniques and activities usedto monitor and eliminate causesof unsatis-
factory performance at any stage of a process; involves sampling and testing.
Quality indicators Measurableaspects of processesor outcomes that provide an indication of the condition
or direction of performance over time. Quality indicators are used to monitor progress
toward stated quality goals and objectives.
Quality The organizational structure, processes,and procedures necessaryto ensurethat the
management overall intentions and direction of an organization's quality program are met and that the
quality of the product or service is ensured. Quality managementincludes strategic plan-
ning, allocation of resources, and other systematic activities, such as quality planning,
implementation, and evaluation.
APPENDIX 1-1
Glossary of Commonly Used Quality Terms (Continued)
Quality planning A systematic process that translates qualitypolicy into measurableobjectives and
requirements, and lays down a sequence of steps for realizing them within a specified
time frame.
Requirement A stated or obligatory need or expectation that can be measured or observed and that is
necessaryto ensure quality, safety, ettectlveness, or customer satisfaction. Hequlre-
ments can include things that the system or product must do, characteristics that it must
have,and levels of performance that it must attain.
Specification Description of a set of requirements to be satisfied by a product, material, or process
indicating, if appropriate, the procedures to be used to determine whether the require-
ments are satisfied. Specifications are often in the form of written descriptions, draw-
ings, professional standards, and other descriptive references.
System An organized, purposeful structure that consists of interrelated and interdependent ele-
ments (components, processes, entities, factors, members, parts, etc).
Validation Demonstration through objective evidencethat the requirements for a particular applica-
tion or intended use have been met. Validation provides assurance that new or changed
processesand procedures are capable of consistently meeting specified requirements
before implementation.
Verification Confirmation, by examination of objective evidence,that specified requirements have
been met.
APPENDIX 1-2
Code of Federal Regulations Quality-Related References
Personnel 600.10,606.20 211.25, 211.28 1271.170

Facilities 600.11, 606.40 211.42-58 1271.190
Environmental control 211.42 1271.195
and monitoring
Equipment 606.60 211.63-72,211.105, 1271.200
211.182
Supplies and reagents 606.65 211.80 1271.210
Standard operating 606.100 211.100-101 1270.31,1271.180
procedures
Process changesand 211.100-101 1271.225,1271.230
validation
Quality assurance/quality 211.22
control unit
Label controls 610.60-68, 606.120-122 211.122-130 1271.250, 1271.370
Laboratory controls 606.140 211.160

Records and record 600.12,606.160 211.192,211.194, 1270.33, 1271.55,
reviews 211.196 1271.270
Receipt, predistribution, 606.165 211.142,211.150 1271.265
and distribution
Adverse reactions 606.170 211.198 1271.350
Tracking 211.188 1271.290
Complaints 606.170-171 211.198 1271.320
Reporting deviations 600.14,606.171 1271.350
Storage 640.2, 640.11, 640.25, 211.142 1271.260
640.34, 640.54, 640.69
HCT/Ps= human cells, tissues, and cellular and tissue-based products.
APPENDIX 1-3
Suggested Quality Control Performance Intervals for Equipment and Reagents *
Refrigerators
@ Recorder Daily
III Manualtemperature Daily
@ Alarm system board (if applicable) Daily
@ Temperaturecharts Daily (review and change weekly)
II Alarm activation Quarterly
Freezers
® Recorder Daily
@ Manualtemperature Daily
e Alarm system board (if applicable) Daily
<II Temperaturecharts Daily (review and change weekly)
III Alarm activation Quarterly
Platelet Incubators
* Recorder Daily
III Manualtemperature Daily
® Temperaturecharts Daily (review and change weekly)
@ Alarm activation Quarterly
® Ambient platelet storage Every4 hours
Laboratory equipment
Centrifuges/cell washers
41 Speed Quarterly
@ Timer Quarterly
II Function Yearly
0) Tubefill level (serologic) Day of use
110 Salinefill volume (serologic) Weekly
@ Volume of antihuman globulin dispensed (if applicable) Monthly
" Temperaturecheck (refrigerated centrifuge) Day of use
011 Temperatureverification (refrigerated centrifuge) Monthly
(Continued)
APPENDIX 1-3
Suggested Quality Control Performance Intervals for Equipment and Reagents*
(Continued)
Equipmellt or Reagellt Frequency of Quality Control
Heating blocks/waterbaths/view boxes

It Temperature Day of use
1& Quadrant/areachecks Periodically
Component thawing devices Day of use
pH meters Dayof use
Blood irradiators
'" Calibration Yearly
® Turntable (visual check eachtime of use) Yearly
® Timer Monthly/quarterly
., Source decay Dependenton source type
<!I> Leaktest Twice yearly
$ Dosedelivery check (with indicator) Eachirradiator use
iii Dosedelivery verification
- Cesium-137 Yearly
- Cobalt-60 Twice yearly
- Other source As specified by manufacturer
Thermometers (vs NIST-certified or traceable thermometer)
'II Liquid-in-glass Yearly
<!I> Electronic As specified by manufacturer
Timers/clocks Twice yearly
Pipette recalibration Quarterly
Sterile connection device
• Weld check Eachuse
@ Function Yearly
Blood warmers
e Effluenttemperature Quarterly
® Heatertemperature Quarterly
Alarm activation Quarterly
APPENDIX 1·3
Suggested Quality Control Performance Intervals for Equipment and Reagents *
(Continued)
Blood collection equipment

Whole blood equipment
OIl Agitators Day of use
• Balances/scales Day of use
OIl Gram weight (vs NIST-certified) Yearly
Microhematocrit centrifuge
• Timer check Quarterly
@ Calibration Quarterly
• Packedcell volume Yearly
Cell counters/hemoglobinometers Day of use
Blood pressure cuffs Twice yearly
Apheresis equipment
'f; Checklist requirements As specified by manufacturer
Reagents
Redcells Day of use
Antisera Day of use
Antiglobulin serum Day of use
Transfusion-transmissible diseasemarker testing Eachtest run
Miscellaneous
Copper sulfate Day of use
Shipping containers for blood and component transport
Twice yearly
(usually at temperature extremes)
'The frequencies listed are suggested intervals, not requirements. For any new pieceof equipment, installation, operational,
and performance qualifications must be performed. After the equipment has been suitably qualified for use, ongoing QC
testing should be performed. Dependingon the operationaland performancequalification methodology,the ongoing QCmay
initially be performed more often than the ultimately desired frequency. Once a record of appropriate in-range QC results
has been established (during either equipment qualification or the ongoing QC),the frequency of testing can be reduced. At
a minimum, the frequency must comply with the manufacturer's suggested intervals; if no such guidance is provided by the
manufacturer, the intervals given in this table are appropriate to use. Recalibration of equipment may be required when
there is reasonto believean unusual event might have affected calibration.
NIST = National Institute of Standards and Technology, QC = quality control.
Facilities, ork Environment,
and Safety
J. Wade Atkins, MS, MT(ASCP)SBB, CQA(ASQ), and
Colleen Bowman, MT(ASCP)SBB, CQA(ASQ)
HE PHYSICAL WORK ENVIRON- stitute, and The Joint Commission, have similar
ment can have a significant impact on the or more detailed safety program requtrements."
safety, efficiency, and. effectiveness of US federal regulations and recommenda-
work processes and on the quality of work. It tions intended to protect the safety of workers
should be designed and managed in a way that and the public in health-care settings are listed
meets operational needs and provides for the in Appendix 2-1. Appendix 2-1 also lists rele-
safety of staff and visitors. The layout of the vant safety recommendations of trade and pro-
physical space; management of utilities; flow of fessional associations. The contents of these reg-
personnel, materials, and waste; and ergonomic ulations and guidance are discussed in more
factors should all be considered in the facility detail in each section of this chapter. US state
management plan. and local government regulations may have ad-
In addition to providing adequate facilities, ditional safety requirements, including architec-
the organization should develop and implement tural and construction safety considerations.
a safety program that defines policies and proce-
dures for safe work practices and emergency re-
sponses. Such a program also includes require- ILI I
ments for training, hazard communication, use Facility Design and Workflow
of engineering controls, and protective equip-
ment. All employees are responsible for protect- Effective design and maintenance of facilities,
ing their own safety and the safety of others by along with the physical organization of work ac-
adhering to policies and procedures set forth in tivities, can help reduce or eliminate many po-
the facility safety program. tential hazards. Facility design, workflow, and
AABBrequires its accredited facilities to plan, maintenance also affect process efficiency, pro-
implement, and maintain a program to mini- ductivity, error rates, employee and customer
mize risks to the health and safety of donors, pa- satisfaction, and the quality of products and ser-
tients, volunteers, and .employees from biologi- vices.
cal, chemical, and radiological hazards.l-' Other During the design phase for a new or reno-
professional and accrediting organizations, in- vated space, the location and flow of personnel,
cluding the College of American Pathologists materials, and equipment should be considered
(CAP), the Clinical and Laboratory Standards In- in the context of the processes to be performed.
J. Wade Atkins, MS, MT(ASCP)SBB,COA(ASQ), Quality Assurance Supervisor, and Colleen Bowman,
MT(ASCP)SBB,COA(ASQ),Quality Assurance Specialist,National Institutes of Health, Bethesda, Maryland
The authors have disclosedno conflictsof interest.
33
Adequate space must be allotted for personnel or work-areas. Exits and fire safety equipment
movement, location of supplies and large equip- must not be blocked or obstructed in any way.
ment, and private or distraction-free zones for Receptacles and disposal guidelines for nonhaz-
certain manufacturing tasks (eg, donor inter- ardous solidwaste and biohazard, chemical, and
viewing, record review, and blood component radiation waste should be clearly identified.
labeling). The facility must offer designated Housekeeping responsibilities, methods, and
"clean" and "dirty" spaces and provide for con- schedules should be defined for every work ar-
trolled movement of materials and waste in and ea. Written procedures, initial training and con-
out of these areas. Chemical fume hoods and bi- tinuing education of personnel, and ongoing
ological safety cabinets (BSCs)should be located monitoring of housekeeping effectivenessare es-
away from drafts and high-traffic areas. The sential to safe operations.
number and location of eyewash stations and
emergency showers must also be considered in Clean Rooms
planning. In some cases, additional special wa- Sterile manufacturing by aseptic technique
ter sources for reagent preparation must be pro- should be considered for open processing activi-
vided. The location of very heavy equipment, ties. Usuallyan ISO (International Standards Or-
such as irradiators, should be taken into account ganization) Class 5 BSC can accommodate this
to ensure that the flooring has sufficient load- requirement. Laboratories that process cellular
bearing capacity. therapy products may choose to adopt clean-
Laboratoriesmust be designed with adequate room specifications and maintenance practices
illumination and electrical power and conve- to meet the requirements of the Food and Drug
niently located electrical outlets. Emergency Administration (FDA)current good tissue prac-
backup power sources, such as uninterruptible tice regulations." International standards for
power supplies and backup generators, should clean rooms are published by ISO and provide
be considered to ensure that blood components, specificationsfor general manufacturing applica-
cellular therapy products, and critical reagents tions to limit airborne particulates, contami-
are not compromised during power failures. The nants, and pollutants.'? These standards also
National Electrical Code is routinely used as a provide guidance for pharmaceutical and bio-
national guideline for the design of essential technology applications that include methods to
electrical distribution systems, with modifica- assess,monitor, and control blocontamlnation.'!
tions approved by the local building authority
that has junsdiction." Restricted Areas
Heating, ventilation, and air handling must
be adequate for the needs of the facility. Envi- Hazardous areas should be clearly and uniform-
ronmental monitoring systems should be consid- ly identified with warning signs in accordance
ered for laboratories that require positive or neg- with federal Occupational Safetyand Health Ad-
ative air pressure differentials or where air ministration (OSHA) and Nuclear Regulatory
filtration systems are used to control particle lev- Commission (NRC) standards so that personnel
els. The nationally accepted specifications for entering or working around them are aware of
ventilation are published by the AmericanSociety existing biological, chemical, or radiation dan-
of Heating, Refrigerating, and Air-Conditioning gers.":" Adequate training must be provided for
Engineers." regular staff in these areas. Staff members not
normally assigned to these areas should receive
adequate training to avoid endangering them-
Housekeeping
selves. Riskareas can be stratified. For example,
The workplace should be kept clean and free of high-risk areas might include those that contain
clutter. Work surfaces and equipment should be chemical fume hoods, BSCs, and storage areas
regularly cleaned and disinfected. Items that forvolatile chemicals or radioisotopes.Technical
may be hazardous or may accumulate dust and work areas might be considered moderate risk
debris should not be stored above clean supplies and restricted to laboratory personnel. Adminis-
C HAP T E R 2 Facilities,WorkEnvironment, and Safety 35
trative and clerical areas are generally consid- Sections 12101-12213, 1990]. Several factors
ered low risk and not restricted. Guidance for may contribute to employee fatigue, musculo-
restricted access based on biosafetylevels is pub- skeletal disorder syndromes, or injury, including
lished by the US Department of Health and Hu- the following":
man Services (DHHS).16
Organizations should consider establishing Awkward postures-positions that place
specific safety guidelines for visitors with busi- stress on the body, such as reaching over-
ness in restricted areas and verifying that safety head, twisting, bending, kneeling, or squat-
guidelines have been reviewed before the visi- ting.
tors enter the area. Casual visitors should not be @> Repetition-performance of the same mo-
allowed in restricted areas. Children should not tions continuously or frequently.
be allowed in areas where they could be ex- Force-the amount of physical effort used to
posed to hazards and should be closely super- perform work.
vised in those areas where their presence is per- Pressure points=pressing of the body against
mitted. hard or sharp surfaces.
Vibration-continuous or high-intensity
Mobile Sites hand/arm or whole-body vibration.
@ Other environmental factors-extreme high
Mobile blood-collection operations can present
or low temperatures or lighting that is too
special challenges. An advance safety survey of
the proposed collection site helps ensure that dark or too bright.
hazards are minimized.
Actions to correct problems associated with
Responsibility for site safety should be as-
signed to an individual with adequate knowl- ergonomics may include the following:
edge to recognize safety concerns and the au-
Engineering improvements to reduce or
thority to address them in a timely manner. All
eliminate the underlying cause, such as mak-
mobile personnel should be trained to recognize
ing changes to equipment, workstations, or
unsafe conditions and understand how to effec-
tively implement infection-control policies and materials.
.. Administrative improvements, such as pro-
procedures in a variety of settings.
Hand-washing accessis essential at collection viding variety in tasks; adjusting worksched-
sites. Carpeted or difficult-to-cleansurfaces may ulesand work pace; providing recoveryor re-
be covered using an absorbent overlay with wa- laxation time; modifying work practices;
terproof backing to protect them from possible ensuring regular housekeeping and mainte-
blood spills. Portable screens and crowd-control nance ofwork spaces, tools, and equipment;
ropes ....
are. helpful. in directing traffic flow to and encouraging exercise.
maintain safe work areas. Food-service areas Provision of personal protective equipment
should be physically separated from areas for (PPE),such as gloves, knee and elbow pads,
blood collectionand storage. Blood-contaminated protective footwear, and other items that em-
waste must be either returned-to a central1oca- ployees wear to protect themselves against
tlon for disposal or packaged and decontaminat- injury.
ed in accordance with local regulations for med-
ical waste.
F TV PH
Ergonomics An effective safety program starts with a well-
Consideration in physical design should be giv- thought-out safety plan. This plan identifies the
en to ergonomics and to accommodations for in- applicable regulatory requirements and de-
dividuals covered under the Americans with scribes how they will be met. A safety plan in-
DisabilitiesAct 142 United States Code (USC), cludes procedures to:
Provide a workplace free of recognized haz- Safe work practices, including waste dispos-
ards. al.
Evaluate all procedures for potential expo- Emergency response plan.
sure risks.
Evaluate each job duty for potential exposure Management controls should be established
risks. to ensure that these elements are implemented
ill Identify hazardous areas or materials with and maintained and that they are effective.
appropriate labels and signs. Management is responsible for the following:
Educate staff, document training, and moni-
tor compliance. Developing and communicating the written
Apply standard precautions (including uni- plan.
versal and blood and body fluid precautions) Ensuring implementation and providing ade-
to the handling of blood, body fluids, and tis- quate resources.
sues. 'Ii! Providing access to employee health services
Dispose of hazardous waste appropriately. related to prevention strategies and treat-
Report incidents and accidents; provide treatment of exposures.
ment and follow-up. Monitoring compliance and effectiveness.
Provide ongoing review of safety policies, Evaluatingand improving the safety plan.
procedures, operations, and equipment.
Prepare for and respond to disasters, includ- Basic Elements of a Safety Program
ing the testing of these facility-specificproce-
dures at defined intervals. Training
Prepare for and respond to threats to person- Employees must be trained to recognize the
al safety such as active shooters or bomb hazards in their workplace and take appropriate
threats, at a facility-specificlevel. precautions. Supervisors are responsible for as-
sessing and documenting each employee's un-
Safetyprograms shouid consider the needs of derstanding of and abilityto apply safetyprecau-
all persons affected by the work environment. tions before independent work is permitted.
Most obvious is the safety of technical staff Safety training must precede even temporary
members, but potential risks for blood donors, work assignments if significant potential for ex-
ancillary personnel, volunteers, visitors, house- posure exists. Staff members who do not
keeping staff, and maintenance and repair work- demonstrate the requisite understanding and
ers must also be evaluated. Laboratories should skills must receive additional training. Training
consider appointing a safety officerwho can pro- should be provided not only to laboratory staff
vide general guidance and expertise." Typical but also to housekeeping and other personnel
duties of a safety officerare to develop the safety who may come into contact with hazardous
program, oversee orientation and training, per- substances or waste. Table 2-llists topics to cov-
form safety audits, survey work sites, recom- er in work safety training programs.
mend changes, and serve on or direct the activi-
ties of safety committees. Facilities using Hazard Identification and Communication
hazardous chemicals and radioactive materials
often assign speciallytrained individuals to over- Employers are required to provide information
see chemical and radiation protection programs about workplace hazards to their staffto help re-
as needed.F'" Five basic elements must be ad- duce the risk of occupational illnesses and inju-
dressed for each type of hazard covered in the ries. Staff need to know what hazardous sub-
safety program: stances they are working with and where those
materials are located in the facility.This commu-
Training. nication is achieved by means of signage, labels
Hazard identification and communication. on containers, written information, and training
'Ii! Engineering controls and PPE. programs.
C HAP T E R 2 Facilities, WorkEnvironment, and Safety 37
TABLE2-1. Topics to Cover in a Work Safety Training Program

Work safety training programs should ensure that all personnel:
(ij Haveaccess to a copy of pertinent regulatory texts and an explanation of the contents.
Understand the employer's exposure control plan and how to obtain a copy of the written plan.
Understand how hepatitis and human immunodeficiency virus (HIV) are transmitted and how often; and are
familiar with the symptoms and consequences of hepatitis B virus (HBV), hepatitis Gvirus (HGV), and HIV
infection.
(ij Know that they are offered vaccination against HBV.
Recognizetasks that pose infectious risks and distinguish them from other duties.
(ij Know what protective clothing and equipment are appropriate for the procedures they will perform and how
to usethem.
(ij Know and understand the limitations of protective clothing and equipment (eg, different types of gloves are
recommended according to the permeability of the hazardous material to be used).
Know where protective clothing and equipment are kept.
(ij Become familiar with and understand.all requirements for work practices specified in standard operating
procedures torthe tasks they perform, including the meaning of signs and labels.
Know how to remove, handle, decontaminate, and dispose of contaminated material.
Know the appropriate actions to take and the personnel to contact if exposed to blood or other biological,
chemical, or radiologic hazards.
Know the corrective actions to take in the event of spills or personal exposure to fluids, tissues, and contam-
inated sharp objects; know appropriate reporting procedures; and know what medical monitoring is recom-
mended when parenteral exposure may have occurred.
Know their right to accessto medical treatment and medical records.
Know fire safety procedures and evacuation plans.
Recognizefacility-specific verbal announcements and how to respond.
Engineering Controls and PPE laboratory area. General guidance on the use of
If the physical work space cannot be designed to
engineering controls and PPE is included in Ap-
eliminate the potential for exposure to hazards, pendix 2-2.
appropriate protective gear must be provided.
Engineering controls are physical plant controls Safe Work Practices
or equipment, such as sprinkler systems, chemi- Employeesmust be trained in how to work with
cal fume hoods, and needleless systems that iso- hazardous materials in ways that protect them-
late or remove the hazard from the workplace.
selves, their coworkers, and the environment.
PPE is specialized clothing or equipment,
such as gloves, masks, and laboratory coats, Safe work practices are defined .as tasks per-
worn by employees for protection against a haz- formed in a manner.that reduces the likelihood
ard. Employees should remove their PPE, such of exposure to workplace hazards. General rec-
as gloves and laboratory coats, and should wash ommendations for safe work practices are in-
their hands with soap and water when leaving a cluded in Appendix 2-2.
Emergency Response Plan refuses the vaccine, that the refusal be docu-
mented."
When engineering and work practice controls
fail, employees must know how to respond
Monitoring Programs
promptly and appropriately. The purpose of ad-
vance planning is to control a hazardous situa- Employersmust provide a system for monitoring
tion as quickly and safely as possible. Regular exposure to certain substances as defined in the
testing of the emergency response plan identi- OSHAstandard if there is reason to believe that
fies areas for improvement and builds staff confi- exposure levels routinely exceed the recom-
dence in their ability to respond effectively in a mended action level."
real emergency. OSHA requires facilities with
more than 10 employees to have a written Medical First Aid and Follow-Up
emergency response plan. Verbal communica- When requested by a worker who has sustained
tion of the plan is acceptable for facilities with known or suspected blood exposure, monitoring
10 or fewer employees.18 for HBV, hepatitis C virus (HCV), and human
immunodeficiency virus (HIV)infection should
Management Controls be provided with appropriate counseling. In
Supervisorypersonnel must monitor safety prac- some states, consent is required for this volun-
tices in their areas of responsibility. Continuing tary testing; rejection of offered testing must be
attention to safety issues should be addressed in documented. The usual schedule includes im-
routine staff meetings and training sessions. Pe- mediate tests of the worker and the source of
riodic audits performed by a safety professional the potentially infectious material, with follow-
increases safety awareness. Management should up testing of the worker at intervals after expo-
seek staff input on the design and improvement sure.13,14 All aspects of accident follow-up
of the facility's safetyplan. should be appropriatelydocumented.
The safety program policies, procedures, The Centers for Disease Control and Preven-
guidance, and supporting references should be tion (CDC) has published recommendations for
documented in writing and made availableto all both preexposure and postexposure prophylaxis
personnel at risk. These documents should be if the contaminating material is HBVpositive or
reviewed on a regular basis and updated as tech- if this information is unknown." HBVimmune
nology evolves and new information becomes globulin is usually given concurrently with HBV
available. Riskmitigation studies should be con- vaccine in cases of penetrating injuries. Postex-
ducted periodically. Strategies or other safety posure prophylaxis for HIVis continually evolv-
provisions should be updated or implemented ing; policies are generally based on Public
with safetyimprovements. Work sites and safety Health Service recommendations and current
equipment should be inspected regularly to en- standards of practice.
sure compliance and response readiness. Check-
lists may be helpful for documenting safety in- Reporting Accidents and Injuries
spections and assessingsafety preparedness.3,4, 19 When an injury occurs, relevant information
should be documented, including the date and
Employee Health Services time of injury and the place where it occurred;
the nature of the injury; descriptions of what
Hepatitis Prophylaxis
happened from the injured person and any wit-
All employees routinely exposed to blood must nesses; and the first aid or medical attention pro-
be offeredhepatitis Bvirus (HBV)vaccine if they vided. The supervisor should complete anyacci-
do not already have HBV-protectiveantibodies dent reports and investigation forms required by
(ie, antibodies to hepatitis B surface antigen). the institution's insurer and worker's compensa-
OSHArequires that the vaccine be offered at no tion agencies. Employers must report fatalities
cost to all employees and, if any employee and injuries resulting in the hospitalization of
three or more employees to OSHA within 8 est alarm and ftre-containment equipment are
hours of the accldent.f located and how to use it, and what the evacua-
OSHA requires health-service employers tion policies and routes are.
with .:11.or more workers to maintain records of All staff members in facilities accredited by
occupational injuries .and illnesses requiring a CAP or The Joint Commission are required ..to
level of care that exceeds the capabilities of a participate in fire drills at least annually.3,.5 ~I).ar-
person trained in first aid.23)pitial documenta- eas where patients are housed or treated, The
tion must be completed within6 days.of the in· Joint Commission.requires q_uarterlydril1~.J)l?:
cident. Records of first aid provided by a non- each shift. Staffparticipation and understanding
physician for minor injuries, such as cuts or should be documented.
burns, do not need to be retained. Alllogs, sum-
maries, and supplemental records must be pre- Hazard Identification and
served for at least 5 years beyond the calendar Communication
year of occurrence. Medical records of employ-
ees should be preserved for the duration of em- Emergency exits must be clearly marked with
ployment plus 30 years, with few exceptions." an exit sign. Additional signage must be posted
along the exit route to show the direction of
Latex Allergies travel if it is not immediately apparent. All flam-
mabie materials should be labeled with appro-
Adverse reactions associated with latex, pow- priate hazard warnings, and flammable storage
dered gloves,or both include contact dermatitis, cabinets should be clearlymarked.
allergic dermatitis, urticaria, and anaphylaxis.
Medical devices that contaln latex must bear a Engineering Controls and PPE
caution label. The National Institute for Occupa-
tional Safety and Health (NIOSH)offers recom- Laboratoriesstoring large volumes of flammable
mendations to prevent these allergic reactions." chemicals are usually built with.z-hour fire sepa-
Most health-care facilitieshave adopted a latex- ration walls, or with l-hour separation walls if
free glove policy to prevent latex allergy reac- there is an automatic fire-extinguishing system."
tions. Fire detection and alarm systems should be pro-
vided in accordance with federal, state, and 10'
cal regulations. All fire equipment should be in-
spected on a regular basis to ensure that it is in
good working order. Fire extinguishers should
Fire prevention relies on a combination of faclll- be readily available, and the staff should be
ty design that is based on the National Fire Pro- trained to use them properly. Housekeeping and
tection Association (NFPA) Life Safety Code, inventory management plans should be de-
which identifies processes to maintain fire pro- signed to control the accumulations of flamma-
tection systems in good working order, and fire ble and combustible materials stored in the facil-
safe-work pracnces." The Life Safety Code in- ity. In areas where sprinkler systems are
cludes both active and passive fire-protection installed, all items should be stored at least' 18
systems (eg, alarms, smoke detectors, sprinklers, inches below the .sprlnkler heads. Facilities
exit lights in corridors, and fire-ratedbarriers). should consult local fire codes, which may re-
quire greater clearance.
Training
Safe Work Practices
Fire safety training is recommended at the start
of employment and at least annually thereafter. Emergencyexit routes must be clear of anything
Training should emphasize prevention and an that would obstruct evacuation efforts... Exit
employee's awareness of the work environment, doors must not be locked in such a way asto im-
including how to recognize and report unsafe pede egress. Permanent exit routes must be de-
conditions, how to report fires, where the near- signed to allow free and unobstructed exit from
all parts of the facility to an area of safety. Sec- with receptacles and connectors. This training
ondary exits may berequlred for areas larger should also help them recognize potential prob-
than 1000 square feet; facilities should consult lems, such as broken receptacles and connec-
local safety authorities with jurisdiction, such as tors, improper electrical connections, damaged
the local fire marshal and NFPA, for guidance on cords, and inadequate grounding.
secondary exits.
Hazard Identification and
Emergency Response Plan Communication
The fire emergency response plan should en- The safetyplan should address the proper use of
compass both facility-wideand area-specificsitu- receptacles and connectors. Equipment that
ations. It should describe reporting and alarm does not meet safety standards should be
systems; location and use of emergency equip- marked to prevent accidental use.
ment; roles and responsibilities of the staff
during the response; "defend-in-place" strate- Engineering Controls and PPE
gies; and conditions for evacuation, evacuation OSHA requires that electrical systems and
procedures, and exit routes.v" equipment be constructed and installed in a way
When a fire occurs, the general sequence for
that minimizes the potential for workplace haz-
immediate response should be to 1) rescue any-
ards. When purchasing equipment, the facility
one in immediate danger; 2) activate the fire
should verify that it bears the mark of an OSHA
alarm system and alert others in the area; 3)
approved independent testing laboratory, such
confine the fire by closing doors and shutting off
as Underwriters Laboratories.f Adequate work-
fans or other oxygen sources if possible; and 4)
ing space should be provided around equipment
extinguish the fire with a portable extinguisher
to allow easy access for safe operation and main-
if the fire is small, or evacuate if it is too large to
tenance. Ground-faultcircuit interrupters should
manage.
be installed in damp or wet areas.
ELECTRICAL SAFETY Safe Work Practices

Electrical safety practices focus on two factors:
Electricalhazards, including fire and shock, may 1) proper use of electrical equipment and 2)
arise from the use of faulty electrical equipment; proper maintenance and repair of this equip-
damaged receptacles, connectors, or cords; or ment. Staff should not plug equipment into or
unsafe work practices. Proper use of electrical unplug equipment from an electrical source
equipment, periodic inspection and mainte- with wet hands. Overloading circuits with too
nance, and hazard recognition training are es- many devices may cause the current to overheat
sential to help prevent accidents that may result the wiring and potentially generate a fire. Dam-
in electric shock or electrocution. The severity aged receptacles and faulty electrical equipment
of shock depends on the path that the electrical must be tagged and removed from service until
current takes through the body, the amount of they have been repaired and checked for safety.
current flowing through the body, and the Flexiblecords should be secured to prevent trip-
length of time that current is flowing through ping and should be protected from damage from
the body. Even low-voltage exposures can lead heavy or sharp objects. Flexible cords should be
to serious injury." kept slackened to prevent tension on electrical
terminals, and cords should be checked regular-
Training
ly for cut, broken, or cracked insulation. Exten-
Safety training should be designed to make em- sion cords should not be used in lieu of perma-
ployees aware of electrical hazards associated nent wiring.
Emergency Response Plan of whether or not they contain visible blood; 3)

nonintactskin; or 4) mucous membranes.
In case of an emergency in which it is not possi-
The OSHA Blood-Borne Pathogens Standard
ble to decrease the power or disconnect equip-
refers to the use of universal precautions. How-
ment,' the power supply should be shut offfrom
ever, OSHA recognizes the more recent guide-
the circuit breaker. If it is not possible to inter-
lines from the CDC and, in Directive CPL 02-
rupt the power supply, a nonconductive materi-
al, such as dry wood, should be used to pry a 02-069, allows hospitals to use acceptable alter-
victim from the source of current." Victims natives, including standard precautions, as long
must not be touched directly. Emergency first as all other requirements in the standard are
aid for victims of electrical shock must be met,"
sought. Water-based fire extinguishers should
not be used on electrical fires. Biosafety Levels
Recommendations for biosafety in laboratories
are based on the potential hazards pertaining to
B~ A specific infectious agents and the activities per-
formed." Biosafety recommendations include
The facility must define and enforce measures to
guidance on both engineering controls and safe
minimize the risk of exposure to biohazard ma-
terials in the workplace. Requirements pub- work practices. The four biosafety levels are des-
lished by OSHA (Blood-BornePathogens Stan- ignated in ascending order, with increasing pro-
dard) and recommendations published by the tection for personnel, the environment, and the
US DHHS provide the basis for an effective bio- community:
safetyplan. 13,14,16
BiosafetyLevel I (BSL-I)involves work with
Blood-Borne Pathogens Standard
agents of no known or of minimal potential
hazard to laboratory personnel and the envi-
The OSHA Blood-Borne Pathogens Standard is ronment. Activities are usually conducted on
intended to protect employees in all occupations open surfaces, and no containment equip-
where there is a risk of exposure to blood and ment is needed.
other potentially infectious materials. It requires €!> BSL-2work involves agents of moderate po-
the facility to develop an exposure controlplan tential hazard to personnel and the environ-
and describes appropriate engineering controls, ment, usually from contact-associated expo-
PPE, and work practice controls to minimize the sure. Most blood bank laboratory activities
risk of exposure. The standard also requires em- are considered BSL-2.
ployers to provide HBVvaccinations to any staff II!J BSL-3includes work with indigenous or ex-
members with. occupational exposure, provide otic agents that may cause serious or.poten-
medical follow-up care in case of accidental ex- tially lethal disease as a result of exposure to
posure, and keep records related to accidents aerosols (eg, Mycobacterium tuberculosis) or
and exposures. by other routes (eg, HN) that would result in
grave consequences to the infected host.
Standard Precautions Recommendations for BSL-3 work are de-
Standard precautions represent the most current signed to contain aerosols and minimize the
recommendations by the CDC to reduce the risk of surface contamination.
risk of transmission of blood-borne pathogens BSL-4appli~sto work with dangerousorex.
and other pathogens in hospitals. Standard pre- otic agents that pose high individual risk of
cautions apply to all patient-care activities, re- life-threatening disease from aerosols (eg,
gardless of diagnosis, in which there is a risk of agents of hemorrhagic fevers or filoviruses).
exposure to I) blood; 2) any body fluids, secre- BSL-4is not applicableto routine blood-bank-
tions, and excretions, except sweat, regardless related activities.
The precautions described in this section fo- when work is in progress and BSCs or other
cus on BSL-2requirements. Laboratoriesshould containment equipment for work that may in-
consult the CDC or National Institutes of Health volve infectious aerosols or splashes. Hand-
guidelines for precautions appropriate for high- washing sinks and eyewash stations must be
er levels of containment. available.The work space should be designed so
that it can be easily cleaned, and bench tops
Training should be impervious to water and resistant to
OSHArequires annual training for all employees chemicals and solvents.
whose tasks increase their risk of infectious ex- To help prevent exposure or cross-contami-
posure.14,29 Training programs must be tailored nation, work area telephones can be equipped
to the target group both in level and content. with speakers to eliminate the need to pick up
General background knowledge of biohazards, the receiver. Computer keyboards and tele-
understanding of control procedures, or work phones can be covered with plastic. Such equip-
experience cannot meet the requirement for ment should be cleaned on a regular basis and
specifictraining, although an assessment of such when visiblysoiled.
knowledge is a first step in planning program BSCs are primary containment devices for
content. Workplace volunteers require at least handling moderate-risk and high-risk organ-
as much safety training as paid staff members isms. There are three types-Classes I, II, and
who perform similarfunctions. III-with Class III providing the highest protec-
tion to workers. In addition to protecting
Hazard Identification and personnel during the handling of biohazard ma-
Communication terials, a BSCmay be used to prevent contami-
nation of blood and cellular therapy products
The facility's exposure control plan communi- during open-processing steps. A comparison of
cates the risks present in the workplace and de- the features and applicationsof the three classes
scribes controls to minimize exposure. BSL-2 of cabinets is provided in Table 2-2.
through BSL-4facilities must have a biohazard BSCs are not required by standard precau-
sign posted at the entrance when infectious tions, but centrifugation of open blood samples
agents are in use. The sign notifies personnel or manipulation of units known to be positive
and visitors of the presence of infectious agents, for HBVsurface antigen or HIV are examples of
provides a point of contact for the area, and indi- blood bank procedures for which a BSC could
cates any special protective equipment or work be useful. The effectivenessof the BSCis a func-
practices required. tion of directional airflow inward and down-
Biohazard warning labels must be placed on ward through a high-efficiencyfilter. Efficacyis
containers of regulated waste; refrigerators and reduced by anything that disrupts the airflow
freezers containing blood or other potentially in- pattern. Care should be taken not to block the
fectious material; and other containers used to front intake and rear exhaust grills. BSCperfor-
store, transport, or ship blood or other potential- mance should be certified annually."
ly infectious materials. Blood components that In 2001, OSHA revised the Blood-Borne
are labeled to identify their contents and have Pathogens Standard and required that employ-
been released for transfusion or other clinical ers implement appropriate control technologies
use are exempted. and safer medical devices in exposure control
plans and that employers solicitinput from their
Engineering Controls and PPE
employees to identify, evaluate, and select engi-
OSHArequires that hazards be controlled by en- neering and work practice controls. Examplesof
gineering or work practices whenever possi- safer devices are needleless systems and self-
ble." Engineering controls for BSL-2laborato- sheathing needles in which the sheath is an inte-
ries include limited access to the laboratory gralpart of the device.
CHAPTER 2 Facilities, Work Environment, and Safety 43
, c:
:::l Q)
0.. 0..
'c 0
C<! c::
E ro
en
en
ro
(3
Q)
Q)
en
~
Q)
E c: '§ '§
c: o Q)
c:
Q)
e~
'S: Q)
Q)
0>
c:
Q)
0>
c:
Q) "20.. en (/)
cti ~ ~
o
c: ts
:::l (3 (3
~
Q)
"0
e Q)
Q)
Q)
Q)
c, 0.. en en
(/)
a.:>
(/)
(/)
ce
-
C3
o
c::
o
(/)
'C
rn
Cl..
E
o
(.)
«
en en
en en
<tI <tI
zs
Q)
zs
Q)
Q) Q)
(/) (/)
N
co
en en
en en
<tI ro
zs c::5
Disinfectants Storage
The EnvironmentalProtectionAgency (EPA)main- Hazardous materials must be segregated, and ar-

tains a list of chemicai products that have been eas for different types of storage must be dearly
shown to be effectivehospitalantimicrobialdisinfec- demarcated. Blood must be protected from un-
tarns." The Associationfor Professionalsin Infection necessary exposure to other materials and vice
ControlandEpidemiologyalsopublishesa guidance versa. If transfusion products cannot be stored
to assist health-care.professionalswith decisionsin- in a separate refrigerator from reagents, speci-
volvingjudiciousselectionand proper use of specific mens, and unrelated materials, areas within the
dlslnfectants."Forfacilitiescoveredunder the Blood- refrigerator must be clearly labeled and extra
Borne PathogensStandard,OSHAallowsthe use of care must be taken to reduce the likelihood of
EPA-registeredtuberculocidal disinfectants; EPA- spills and other accidents. Storage areas must be
registereddisinfectantsthat are effectiveagainstboth kept clean and orderly; food or drink is never al-
HIVand HBV;or, more commonlyused, a diluted lowed where biohazard materials are stored.
bleachsolution,usually 10%by volume;or a combi-
nationofthese, to decontaminatework surfaces." PPE
Before selecting a disinfectant product, sever- When hazards cannot be eliminated, OSHA re-
al factors should be considered. Among them quires employers to provide appropriate PPE
are the type of material or surface to be. treated and clothing and to clean, launder, or dispose of
and the hazardous properties of the disinfectant PPE at no cost to their employees." Standard
product, such as corrosiveness and the level of PPE and clothing include uniforms, laboratory
disinfection required. After a product has been coats, gloves, face shields, masks, and safety
selected, procedures need to be written to en- goggles. Indications and guidance for their use
sure effective and consistent cleaning and treat- are discussed in Appendix 2-2.
ment of work surfaces. Some factors to consider
for effective decontamination include contact Safe Work Practices
time, type of microorganisms, presence of or-
ganic matter, and concentration of the chemical Safework practices appropriate for standard pre-
agent, Laboratory personnel should review the cautions include the following:
basic information on decontamination and fol-
low the manufacturer's instructions. Wash hands after touching blood, body flu-
ids,secretions, excretions, and contaminated
Decontamination items, whether or not gloves are worn.
Wear gloves when touching blood, body flu-
Reusable equipment and work surfaces that may ids, secretions, excretions, and contaminated
be contaminated with blood require daily clean- items, and change gloves between tasks.
ing and decontamination. Obvious spills on Wear a mask and eye protection or a face
equipment or work surfaces should be cleaned shield during activities that are likely to gen-
up immediately; routine wipe-downs with disin- erate splashes or sprays of blood, body fluids,
fectant should occur at the end of each shift or secretions, and excretions.
on a different frequency that provides equiva- Wear a gown during activities that are likely
lent safety. Equipment that is exposed to blood to generate splashes or sprays of blood, body
or other potentially infectious material must be fluids, secretions, or excretions.
decontaminated before it is serviced or shipped. Handle soiled patient-care equipment in a
When decontamination of all or a portion of the manner that prevents exposure; ensures that
equipment is not feasible, a biohazard label stat- reusable equipment is not used for another
ing which portions remain contaminated shouid patient until it has been cleaned and repro-
be attached before the equipment is serviced or cessed appropriately; and ensures that single-
shipped. use items are discarded properly.
@ Ensure that adequate procedures are defined Procedures should be assessed for risks of
and followed for the routine care, cleaning, biohazard exposures and risks inherent in work-
and disinfection of environmental surfaces ing with a donor or patient during the screening
and equipment. and donation processes. Some techniques or
Handle soiled linen in a manner that pre- procedures are more likely to cause injury than
vents exposure. others, such as using lancets for finger puncture,
@ Handle needles, scalpels, and other sharp in- handling capillary tubes, crushing vials for arm
struments or devices in a manner that mini- cleaning, handling any unsheathed needles,
mizes the risk of exposure. cleaning scissors, and giving cardiopulmonary
(I Use mouthpieces, resuscitation bags, or oth- resuscitation.
er ventilation devices as an alternative to In some instances, it may be necessary to col-
mouth-to-mouth resuscitation methods. lect blood from donors known to pose a high
risk of infectivity (eg, collection of autologous
Laboratory Biosafety Precautions blood or source plasma for the production of
Several factors need to be considered when as- other products, such as vaccines). The FDApro-
sessing the risk of blood exposures among labo- vides guidance on collecting blood from such
ratory personnel. Some of these factors include "high-risk" donors.35,36 The most recent regula-
the number of specimens processed, personnel tions and guidance should be consulted for
behaviors, laboratory techniques, and types of changes or additions.
equipment. 34 The laboratory director may wish
to institute BSL-3practices for procedures that Emergency Response Plan
are considered to be higher risk than BSL-2. Table 2-3 lists steps to take when a spill occurs.
When there is doubt whether an activity is BSL-
Facilities should be prepared to handle both
2 or BSL-3, the safety precautions for BSL-3
small and large blood spills. Good preparation
should be followed. BSL-2precautions that are
for spillcleanup includes several elements:
applicable to the laboratory setting are summa-
rized in Appendix 2-3.
Work areas designed so that cleanup is rela-
tively simple.
Considerations for the Donor Room
® A spill kit or cart containing all necessary
The Blood-BornePathogens Standard acknowl- supplies and equipment with instructions for
edges a difference between hospital patients and their use placed near areas where spills are
healthy donors, in whom the prevalence of in- anticipated.
fectious disease markers is significantly lower. Responsibilityassigned for kit or cart mainte-
The employer in a volunteer blood donation fa- nance, spillhandling, record-keeping, and re-
cilitymay determine that routine use of glovesis view of significantincidents.
not required for phlebotomy as long as the fol- Personnel trained in cleanup procedures and
lowing conditions exist": procedures for reporting significant inci-
dents.
fl> Glovesare made availableto those who want
to use them, and their use is not discour- Biohazard Waste
aged.
III Gloves are required when an employee has Medical waste is defined as any waste (solid,
cuts, scratches, or breaks in skin; when there semisolid, or liquid) generated in the diagnosis,
is a likelihood that contamination will occur; treatment, or immunization of human beings or
while an employee is drawing autologous animals in related research, production, or test-
units; while an employee is performing thera- ing of biologics. Infectious waste includes dis-
peutic procedures; and during training in posable equipment, articles, or substances that
phlebotomy. may harbor or transmit pathogenic organisms or
10 The policyis periodicallyreevaluated. their toxins. In general, infectious waste should
C HAP T E R 2 Facilities, Work Environment, and Safety 47
TABLE 2-3. Blood Spill CleanupSteps

Evaluatethe spill.
Wear appropriate protective clothing and gloves. If sharp objects are involved, gloves must be puncture resis-
tant, and a broom or other instrument should be used during cleanup to avoid injury.
Remove clothing if it is contaminated.
Post warnings to keep the area clear.
Evacuatethe area for 30 minutes if an aerosol has been created.
Contain the spill if possible.
If the spill occurs in the centrifuge, turn the power off immediately and leavethe cover on for 30 minutes. The
use of overwraps helps prevent aerosolization and contain the spill.
Useabsorbent material to mop up most of the liquid contents.
Cleanthe spill area with detergent.
Flood the area with disinfectant and use it as described in the manufacturer's instructions. Allow adequate
contact time with the disinfectant.
Wipe up residual disinfectant if necessary.
Dispose of all materials safely in accordance with biohazard guidelines. All blood-contaminated items must be
autoclaved or incinerated.
be either incinerated or decontaminated before not be released into the environment during
disposalin a sanitary landfill. handling.
If state law allows, blood and components,
suctioned fluids, excretions, and secretions may Guidance for Biohazard Waste Disposal
be carefullypoured down a drain connected to a
Employees must be trained before handling or
sanitary sewer. Sanitary.sewers may also be.used
disposing of biohazard waste, even if it is pack-
to dispose of other potentially infectious wastes aged. The following disposal guidance are rec-
that can be ground and flushed into the sewer. ommended":
State and local health departments should. be
consulted about laws and regulations pertaining Identify biohazard waste consistently; red
to disposalof biologicalwaste into the sewer. seamless plastic bags (at least 2 mm thick) or
In the blood bank, all items contaminated containers carrying the biohazard symbol are
with liquid or semi-liquidblood are to be consid- recommended.
ered hazardous materials. Items contaminated ® Place bags in a protective container with clo-
with dried blood are considered hazardous if sure upward to avoid breakage and leakage
there is potential for the dried material to flake during storage or transport.
offduring handling. Contaminated sharp objects Prepare and ship waste transported over pub-
are always considered hazardous because of the lic roads according to US Department of
risk for percutaneous injury. However, items Transportation (DOT)regulations.
such as used gloves, swabs, plastic pipettes with Discard sharps (eg, needles, broken glass,
excess liquid removed, or gauze contaminated glass slides, and wafers from sterile connec-
with small droplets of blood may be considered tion devices) in rigid, puncture-proof, leak-
nonhazardous if the material is dried and will proof containers.
III Put liquids in leak-proof, unbreakable con- of as nonhazardous solid wastes. The staff
tainers only. should check with the local solid-waste authori-
® Do not compact waste materials. ty to ensure that the facility is in compliance
with regulations for the area. Waste containing
Storage areas for infectious material must be broken glass or other sharp items should be dis-
secured to reduce accident risk. Infectious posed of using a method consistent with policies
waste must never be placed in the public trash for the disposalof other sharp or potentially dan-
collection system. Most facilities hire private gerous materials.
carriers to decontaminate and dispose of infec-
tious or hazardous waste. The facility should
disclose all risks associated with the waste in HE leAL SAF T
their contracts with private companies. The car-
rier is responsible for complyingwith all federal, One of the most effective preventive measures
state, and local laws for biohazard (medical) that a facility can take to reduce hazardous
waste transport, treatment, and disposal. chemical exposure is to choose alternative non-
hazardous chemicals whenever possible. When
Treating Infectious or Medical Waste the use of hazardous chemicals is required, pur-
chasing these supplies in small quantities reduc-
Facilities that incinerate hazardous waste must
es the risks associated with storing excess chem-
comply with EPA standards of performance for
icals and then dealingwith their disposal.
new stationary sources and emission guidelines
OSHArequires that facilitiesusing hazardous
for existing sources." In this regulation, a
chemicals develop a written chemical hygiene
hospital/medical/infectious waste incinerator is
plan (CHP)and that the plan be accessibleto all
any device that combusts any amount of hospi-
employees. The CHP should outline procedures,
tal waste or medical/infectious waste.
equipment, PPE, and work practices that are ca-
Autoclaving is another common method for
pable of protecting employees from hazardous
decontamination of biohazard waste, used for
chemicals used in the facility.15,20 The CHP must
blood samples and blood components. The fol-
lowing elements are considered in determining also provide assurance that equipment and pro-
processing time for autoclaving: tective devices are functioning properly and that
criteria to determine implementation and main-
5 Size of load being autoclaved. tenance of all aspects of the plan are in place.
® Type of packaging of item(s) being auto- Employees must be informed of all chemical
claved. hazards in the workplace and be trained to rec-
<j Density of items being autoclaved. ognize chemical hazards, protect themselves
Number of items in a single autoclave load. when working with these chemicals, and know
® Placement of items in the autoclave to allow where to find information about particular haz-
for steam penetration. ardous chemicals. Safety audits and annual re-
views of the CHP are important control steps to
It is useful to place a biological indicator in help ensure that safety practices comply with
the center of loads that vary in size and contents the policies set forth in the CHP and that the
to evaluate optimal steam penetration times. CHP is up to date.
The EPA provides detailed information about Establishinga clear definition of what consti-
choosing and operating such equipment. 37 tutes hazardous chemicals is sometimes diffi-
For decontamination, material should be au- cult. Generally, hazardous chemicals pose a sig-
toclaved for a minimum of 1 hour. For steriliza- nificant health risk if an employee is exposed to
tion, longer treatment times are needed. A gen- them or a significantphysicalrisk, such as fire or
eral rule for decontamination is to process for 1 explosion, if handled or stored improperly. Cate-
hour for every 10 pounds of waste. Usually, de- gories of health and physical hazards are listed
contaminated laboratory wastes can be disposed in Tables 2-4 and 2-5. The NJOSH Pocket Guide
TABLE2·4. Categories of Health Hazards
Irritants Agents causing irritation (eg, edema or burning) to skin or mucous

membranes upon contact
Corrosives Agents causing destruction of human tissue atthe site of contact
Toxic or highly toxic agents Substances causing serious biologic effects following inhalation,
ingestion, or skin contact with relatively small amounts
Reproductive toxins Chemicals that affect reproductive capabilities, including chromo-
somal damages and effects on fetuses
Othertoxins Hepatotoxins; nephrotoxins; neurotoxins; agents that act on the
hematopoietic system; and agents that damagethe lungs, skin, eyes,
or mucous membranes
TABLE2·5. Categories of Physical Hazards
or flammable Chemicals that can burn (including combustible and flammable liq-
chemicals uids, solids, aerosols, and gases)
Compressed gases Gasesor mixtures of gases ina container under pressure
Explosives Unstable or reactive chemicals that undergo violent chemical
changes at normal temperatures and pressure
Unstable (reactive) chemicals Chemicals that could be self-reactive under certain conditions
(shocks, pressure, or temperature)
Water-reactive chemicals Chemicals that react with water to releasea gas that either is flam-
mable or presents a health hazard
to Chemical Hazards provides a quick reference work in an area in which hazards exist. If a new
for many common chemicals." employee has received prior training, it may not
The facilityshould identify a qualified chemi- be necessary to retrain the individual, depend-
cal hygiene officer to be responsible for developing on the employer's evaluation of the new em-
ing guidelines for hazardous materials." The ployee's level of knowledge. New employee
chemical hygiene officer is also accountable for training is likely to be necessary regarding such
monitoring and documenting accidents and for specifics as the location of each relevant safety
initiating process change as needed. data sheet (SDS),details on chemical labeling,
PPE to be used, and site-specificemergency pro-
Training cedures.
Training must be provided for each new
Employees who may be exposed to hazardous physical or health hazard when it is introduced
chemicals must be trained before they begin into the workplace but not for each new chemi-
cal that falls within a particular hazard class.IS able to control the release of, or exposure to,
For example, if a new solvent is brought into the the chemical.)
workplace and the solvent has hazards similar to
existing chemicals for which training has al- Hazardous Chemical Labeling and Signs
ready been conducted, then the employer need
The Hazard Communication Standard requires
only make employees aware of the new sol-
vent's hazard category (eg, corrosive or irri- manufacturers of chemicals and hazardous ma-
tant). However, if the newly introduced solvent terials to provide the user with basic informa-
is a suspected carcinogen and carcinogenic haz- tion about the hazards of these materials
ard training has not been provided, then new through product labeling and the SDS.15 Em-
training must be conducted for employees with ployers are required to provide employees who
potential exposure. Retraining is advisable as of- are expected to work with these hazardous ma-
ten as necessary to ensure that employees un- terials with information about what the hazards
derstand the hazards linked to the materials of the materials are, how to read the labeling,
with which they work, particularly any chronlc how to interpret symbols and signs on the la-
and specific target-organ health hazards. bels, and how to read and use the SDS.
At a minimum, hazardous-chemical contain-
Hazard Identification and er labels must include the name of the chemical,
Communication name and address of the manufacturer, hazard
warnings, symbols, designs, and other forms of
Hazard Communication warning to provide visual reminders of specific
Employers must prepare a comprehensive haz- hazards. The label may refer to any SDS for ad-
ard communication program for all areas in ditional information. Labels applied by the man-
which hazardous chemicals are used to comple- ufacturer must remain on containers. The user
ment the CHP and "ensure that the hazards of may add storage requirements and dates of re-
all chemicals produced or imported are classi- ceipt, opening, and expiration. If chemicals are
fied, and that information concerning the classi- aliquoted into secondary containers, the second-
fied hazard is transmitted to employers and em- ary container must be labeled with the name of
ployees."!' The program should include labeling the chemical and appropriate hazard warnings.
of hazardous chemicals, instructions on when Additional information, such as precautionary
and how to post warning labels for chemicals, measures, concentration if applicable, and date
directions for managing SDS reports for hazard- of preparation, are helpful but not mandatory.
ous chemicals in the facilities, and employee It is a safe practice to label all containers with
training. Safety materials made available to em- their content, even water. Transfer containers
ployees should include the following: used for temporary storage need not be labeled
if the person performing the transfer retains con-
" The facility's written CHP. trol and intends the containers to be used imme-
@ The facility's written program for hazard diately. Information regarding acceptable stan-
communication. dards for hazard communication labeling is
@ Identification of work areas where hazardous provided by the NFPA40 and the American Coat-
chemicals are located. ings Association."
@ Required list of hazardous chemicals and the Signs meeting OSHA requirements must be
relevant SDSs. (It is the responsibility of the posted in areas where hazardous chemicals are
facility to determine which chemicals may used. Decisions about where to post warning
present a hazard to employees. This determi- signs are based on the manufacturer's recom-
nation should be based on the quantity of mendations regarding the chemical hazards, the
chemical used; physical properties, potency, quantity of the chemical in the room or labora-
and toxicity of the chemical; manner in tory, and the potency and toxicity of the chemi-
which the chemical is used; and means avail- cal.
Safety Data Sheet Engineering Controls and PPE

The SDS identifies the physical and chemical Guidelines for laboratory areas in which hazard-
properties of a hazardous chemical (eg, flash ous chemicals are used or stored must be estab-
point or vapor pressure), its physical and health lished. Physical facilities, and especiallyventila-
hazards (eg, potential for fire.. explosion, and tion, must be adequate for the nature and
signs and symptoms of exposure), and precau- volume of work conducted. Chemicals must be
stored according to chemical compatibility (eg,
tions for the chemical's safe handling and use.
corrosives, flammables, and oxidizers) and in
Specific instructions in an individual SDS take minimal volumes. Bulk chemicals should be
precedence over generic information in the haz- kept outside work areas. NFPA standards and
ardous materials program. others provide guidance for proper storage,
Employers must maintain copies of each re- sometimes in storage cabinets.4,40,42
quired SDSin the workplace for each hazardous Chemical fume hoods are recommended for
chemical and ensure that SDScopies are readily use with organic solvents, volatile liquids, and
accessible during each work shift to employees dry chemicals with a significant inhalation haz-
when they are in their work areas. When house- ard." Although constructed with .safety giass,
hold consumer products are used in the work- most fume hood. sashes are not designed to
place in the same manner that a consumer serve as safety shields. Hoods should be posi-
tioned in an area where there is minimal foot
would use them (ie, when the duration and fre-
traffic to avoid disrupting the airflow and com-
quency of use, and therefore exposure, are not promising the containment field.
greater than those that the typical consumer PPE that may be provided, depending on the
would experience), OSHAdoes not require that hazardous chemicals used, includes chemical-
an SDS be provided to purchasers. However, if resistant gloves and aprons, shatter-proof safety
exposure to such products exceeds that normal- goggles,and respirators.
ly found in consumer applications, employees Emergency showers should be accessible to
have a right to know about the properties of areas where caustic, corrosive, toxic, flamma-
such hazardous chemicals. OSHA does not re- ble, or combustible ..chemicals are used.v"
quire or encourage employers to maintain an There should be unobstructed access, within 10
seconds, to. the showers from the areas where
SDSfor nonhazardous chemicals.
hazardous chemicals are used. Safety showers
SDSforms typicallyinclude the following:
should be periodically flushed and tested for
function, and associated floor drains should be
Identification of properties.
checked to ensure that drain traps remain filled
'" Hazardls) identification.
with water.
Composition/information on ingredients.
First-aidmeasures.
'" Fire-fightingmeasures. Safe Work Practices
e Accidental release measures. Hazardous material should not be stored or
e Handling and storage considerations. transported in open containers. Containers and
Exposure controls/personal protection infor- their lids or seals should be designed to prevent
mation. spills or leakage in all reasonably anticipated
e Physical and chemical properties. conditions. Containers should be able to safely
Stabilityand reactivity. store the maximum anticipated volume and be
Toxicologyinformation. easy to clean. Surfaces should be kept clean and
Ecologicinformation. dry at all times.
Disposalconsiderations. When an employee is working with a chemi-
Transport information. cal fume hood, all materials should be kept at
Regulatoryinformation. least six inches behind the face opening.: The
Other information. vertical sliding sash should be positioned at the
height specified on the certification sticker. The ate area. The response typically comes from
airfoilbaffles and rear ventilation slots must not outside the immediate release area by per-
be blocked. Appendix 2-5 lists suggestions for sonnel trained as emergency responders.
working safelywith specificchemicals. These spillsinclude those that involve imme-
diate danger to life or health, serious threat
Emergency Response Plan of fire or explosion, and high levels of toxic
substances.
The time to prepare for a chemical spill is before
it occurs. A comprehensive employee training
Appendix 2-7 addresses the management of
program should provide each employee with all
hazardous chemical spills. Spill cleanup kits or
tools necessary to act responsibly at the time of a carts tailored to the specific hazards present
chemical spill. The employee should know re-
should be available in each area. The kits or
sponse procedures, be able to assess the severity carts may contain rubber gloves and aprons,
of a chemical spill, know or be able to quickly shoe covers, goggles, suitable aspirators, gener-
look up the basic physical characteristics of the al absorbents, neutralizing agents, a broom, a
chemicals, and know where to find emergency dust pan, appropriate trash bags or cans for
response telephone numbers. The employee waste disposal, and cleanup directions. Chemi-
should be able to assess, stop, and confine the
cal absorbents, such as clay absorbents or spill
spill; either clean up the spill or call for a spill
blankets, can be used for cleaning up a number
cleanup team; and follow procedures for report- of chemicals and thus may be advantageous for
ing the spill. The employee must know when to employees to use in spillsituations.
ask for assistance, when to isolate the area, and With any spill of a hazardous chemical, but
where to find cleanup materials. especially of a carcinogenic agent, it is essential
Chemical spillsin the workplace can be cate-
to refer to the SDSand contact a designated su-
gorized as follows": pervisor or designee trained to handle these
spills and hazardous waste disposal." Facilityen-
01;Incidental releases are limited in quantity
vironmental health and safety personnel can
and toxicity and pose no significant safety or
also offer assistance. The employer must assess
health hazard to employees. They may be
the extent of the employee's exposure. After an
safely cleaned up by employees familiarwith
exposure, an employee must be given an oppor-
the hazards of the chemical involved in the
tunity for medical consultation to determine the
spill. Waste from the cleanup may be classi-
need for a medical examination.
fied as hazardous and must be disposed of in
Another source of workplace hazards is the
the proper fashion. Appendix 2-6 describes
unexpected release of hazardous vapors into the
appropriate responses to incidental spills.
environment. OSHA has set limits for exposure
* Releases that may be incidental or may re-
to hazardous vapors from toxic and hazardous
quire an emergency response may pose an
substances." The potential risk associated with
exposure risk to employees depending on the
a chemical is determined by the manufacturer
circumstances. Considerations such as the
and listed on the SDS.
hazardous substance properties, circumstanc-
es of release, and mitigating factors play a
Chemical Waste Disposal
role in determining the appropriate response.
The facility's emergency response plan Most laboratory chemical waste is considered
should provide guidance on how to deter- hazardous and is regulated by the EPAthrough
mine whether a spill is incidental or requires the Resource Conservation and Recovery Act
an emergency response. (42 USC §6901 et seq, 1976). This regulation
'" Emergency response releases pose a threat specifiesthat hazardous waste can be legallydis-
to health and safety regardless of the circum- posed of only at an EPA-approveddisposal facili-
stances surrounding their release. These ty. Disposal of chemical waste into a sanitary
spills may require evacuation of the immedl- sewer is regulated by the Clean Water Act (33
C HAP T E R 2 Facilities,Work Environment, and Safety 53
usc §1251 et seq, 1977), and most US states chemical bonds. Molecules and atoms become
have strict regulations concerning disposal of ionized or excited (or both) by absorbing this en-
chemicals in the water system. Federal and ap- ergy. The direct action path leads to radiolysisor
plicable state regulations should be consulted formation of free radicals that, in turn, alter the
when a facility is setting up and reviewing its structure and function of molecules in the cell.
waste disposal policies. Molecular alterations can cause cellular or
chromosomal changes, depending on the
amount and type of radiation energy absorbed.
RADIATI N FETY Cellular changes can be manifested as a visible
somatic effect (eg, erythema). Changes at the
Radiationcan be defined as energy in the form of chromosome level may be manifested as leuke-
waves or particles emitted and .propagated mia or other cancers or possibly as germ-cell de-
through space or a material medium. Gamma fects that are transmitted to future generations.
rays are electromagnetic radiation, whereas al- Several factors influence the level of biologi.
pha and beta emitters are examples ofparticulate cal damage from exposure, including the type of
radiation. The presence of radiation in the blood radiation, part of the body exposed, total ab-
bank, such as self-containedblood irradiators, re- sorbed dose, and dose rate. The total absorbed
quires additional precautions and training.4,46 dose is the cumulative amount of radiation ab-
sorbed in the tissue, The greater the dose, the
Radiation Measurement Units greater the potential for biological damage. Ex-
The measurement unit quantifying the amount posure can be acute or chronic. The low levels
of energy absorbed per unit mass of tissue is the of ionizing radiation likely to occur in blood
gray (Gy)or radiation absorbed dose (rad); 1 Gy = banks should not pose any detrimental risk.47-50
100 rad.
Dose equivalency measurements are more RegUlations
useful than simple energy measurements be- The NRC controls the use of radioactive materi-
cause dose equivalency measurements take into als by establishing licensure requirements.
account the ability of different types of radiation States and municipalities may also have require-
to cause biological effects. The ability of radia- ments for inspection, licensure, or both. The
tion to cause damage is assigned a number, type of license for using radioisotopes or irr~dia-
called a quality factor (OF). For example, expo- tors depends on the scope and magnitude of the
sure to a given amount of alpha particles (OF = use of radioactivity. US facilities should contact
20) is far more damaging than exposure to an the NRC and appropriate state agencies for in-
equivalent amount of gamma rays (OF = 1). The formation on .license.requirements. and applica-
common unit of measurement fordose equiva-
tions as soon as such activities are proposed.
lency is the roentgen or rad eqwyalent man BachNltOlicensed establishment must have
(rem). Rem is the dose from any type of radiation
a qualified radiation safety officerwho is respon-
that produces biologicaleffects in humans equiv-
sible for establishing personnel protection re-
alent to 1 radof x-rays, gamma ray~,or beta rays.
quirements and for ensuring proper disposal and
To obtain the dosefrom a particular type of radi-
handling of radioactive materials. Specificradia-
ation inrem, the number of rad should be multi-
tion safety policies and procedures should ad-
plied by the OF (rad x OF = rem). Because the
dress dose limits, employee training, warning
OF for gamma rays, x-rays, and most beta parti-
signs and labels, shipping and handling guid-
cles is 1, the dose in rad is equal to the dose in
ance, radiation monitoring, and exposure man-
rem for these types of radiation.
agement. Emergency procedures must be clear-
ly defined and readily available to the staff.
Biological Effects of Radiation
In 2005, the NRC imposed additional securi-
Any harm to tissue begins with the absorption of ty requirements for high-risk radioactive sourc-
radiation energy and subsequent disruption of es, including those used in blood irradiators.
The purpose of the increased controls is to re- Bioassays, such as thyroid and whole body
duce the risk of unauthorized use of radioactive counting or urinalysis, may be used to deter-
materials that may pose a threat to public health mine whether there is radioactivity inside the
and safety. These 2005 measures include con- body and if so, how much. If necessary, bioas-
trolled access, approval in writing of individuals says are usually performed quarterly and after an
deemed trustworthy and reliable to have unes- incident where accidental intake may have oc-
corted access, a system of monitoring to imme- curred.
diately detect and respond to unauthorized Survey meters are sensitive to low levels of
access, and documentation of authorized per- gamma or particulate radiation and provide a
sonnel and monitoring activtties." In 2007, a re- quantitative assessment of radiation hazard. Sur-
quirement for fingerprintingwas added." vey meters can be used to monitor storage areas
for radioactive materials or wastes, testing areas
Exposure Limits during or after completion of a procedure, and
The NRC sets standards for protection against packages or containers of radioactive materials.
radiation hazards arising from licensed activi- Surveymeters must be calibrated annually by an
ties, including dose limits.12Such limits, or max- authorized NRC licensee. Selection of appropri-
imal permissible dose equivalents, are a measure ate meters should be discussed with the radia-
of the radiation risk over time and serve as tion safety officer.
standards for exposure. The occupational total In areas where radioactive materials are han-
effective-dose-equivalentlimit is 5 rem/year, the dled, all work surfaces, equipment, and floors
shallow dose equivalent limit (skin) is 50 rem/ that may be contaminated should be checked
year, the extremity dose equivalent limit is 50 regularly with a wipe test. In the wipe test, a
rem/year, and the eye dose equivalent limit is moistened absorbent material (the wipe) is
15 rem/year.l'r" Dose limits for an embryo or passed over the surface and then measured for
fetus must not exceed 0.5 rem during pregnan- radiation.
cy.12,47,53Employers are expected not only to
maintain radiation exposure below allowable
Training
limits, but also to keep exposure levels as far be-
low these limits as can reasonably be achieved. Personnel who handle radioactive materials or
work with blood irradiators must receive radia-
Radiation Monitoring tion safety training before beginning work. This
training should address the presence and poten-
Monitoring is essential for early detection and
prevention of problems resulting from radiation tial hazards of radioactive materials in the em-
exposure. Monitoring is used to evaluate the fa- ployee's work area, general health protection is-
cility's environment, work practices, and proce- sues, emergency procedures, and radiation
dures and to comply with regulations and NRC warning signs and labels in use. Instruction in
licensing requirements. Monitoring is accom- the followingareas is also suggested:
plished with the use of dosimeters, bioassays,
survey meters, and wipe tests." <@ NRC regulations and license conditions.
Dosimeters, such as film or thermo-lumines- <@ The importance of observing license condi-
cent badges, rings, or both, measure personnel tions and regulations and of reporting viola-
radiation doses. The need for dosimeters de- tions or conditions of unnecessary exposure.
pends on the amount and type of radioactive @ Precautions to minimize exposure.
materials in use; the facilityradiation safety offi- Interpretation of results of monitoring devic-
cer determines individual dosimeter needs. Film es.
badges must be changed at least quarterly and in Requirements for pregnant workers.
some instances monthly, be protected from high @ Employees' rights.
temperature and humidtty, and be stored at Documentation and record-keeping require-
work away from sources of radiation. ments.
The need for refresher training is deter- when working with certain radioactive mate-
mined by the license agreement between the rials. These requirements are usually stipulat-
NRC and the facility. ed in the license conditions.
Using appropriate shielding such as a lead
Engineering Controls and PPE barrier for gamma rays or plexiglass for beta
Although self-contained blood irradiators pres- particles.
ent little risk to laboratory staff and film badges Using good housekeeping practices to mini-
are not required for routine operation, blood es- mize the spread of radioactivity to uncon-
tablishments with irradiation programs must be trolled areas.
licensed by the NRC;.48
The manufacturer of the blood irradiator usu- Emergency Response Plan
ally accepts responsibility for radiation safety re- Radioactivecontamination is the dispersal of ra-
quirements during transportation, installation, dioactive material into or onto areas in which it
and validation of the unit as part of the purchase
is not intended-for example, the floor,work ar-
contract. The radiation safety officer can help
eas, equipment, personnel clothing, or person-
oversee the installation and validation processes
and should confirm that appropriate training, nel skin. The NRC regulations state that gamma
monitoring systems, procedures, and mainte- or beta radioactive contamination cannot ex-
nance protocols are in place before use and that ceed 2200 disintegrations per minute (dpm) per
they reflect the manufacturer's recommenda- 100 em' in the posted (restricted) area or
tions. Suspected malfunctions must be reported 220 dpm/100 em' in an unrestricted area, such
immediately so that appropriate actions can be as a corridor. Foralpha emitters, these values are
initiated. 220 dpm/100 crrr' and 22 dpm/100 crrr', re-
Blood irradiators should be located in secure spectively."
areas so that only trained individuals have ac- If a spill occurs, employees' contaminated
cess. Fire protection for the unit must aiso be skin surfaces must be washed several times, and
considered. Automatic fire detection and con- the radiation safety officer must be notified im-
trol systems should be readily available in the mediately to provide further guidance. Others
immediate area. Blood components that have must not be allowed to enter the area until
been irradiated are not radioactive and pose no emergency response personnel arrive.
threat to the staffor the general public.
Radioactive Waste Management
Safe Work Practices
Policies for the disposal of radioactive waste,
Each laboratory should establish policies and whether liquid or solid, should be established
procedures for the safeuse of radioactive materi- with input from the radiation safety officer.
als. These policies and procedures should in- Liquid radioactive waste may be collected
clude requirements for followinggenerallabora- into large sturdy bottles labeled with an appro-
tory safety principles, appropriate storage of
priate radiation waste tag. The rules for separa-
radioactive solutions, and proper disposal of ra-
tion by chemical compa,tibility apply. Bottles
dioactive wastes. Radiation safety can be im-
proved with the followingprocedures: must be carefully stored to protect against spill-
age or breakage.. Dry or solid waste may be
Minimizing the time of exposure by working sealed in a plastic bag and tagged as radiation
as efficientlyas possible. waste. The isotope, its activity, and the date on
Maximizing the distance from the source of which the activity was measured should be re-
the radiation by staying as far from the corded on the bag. Radioactivewaste must nev-
source as possible. er be discharged into the facility's drain system
III Maximizing shielding (eg, by using a self- without prior approval by the radiation safety
shielded irradiator or wearing a lead apron) officer.
Radioisotopic Irradiator Removal and the proper shipping name is "biological sub-
Replacement stance, Category B" (UN3373). HIVor HBVin
culture is classified as a Category A infectious
Advances in technology have created alterna- substance, but if these viruses are present in a
tives to cesium irradiators that are comparable specimen that is not an active culture, then they
to, or even more effective than, those in use for are still classifiedas Category B.
blood irradiation. These non-radioisotopicirradi- Patient specimens with minimal likelihood of
ators mitigate security risks, eliminate liability containing pathogens are exempt from hazard-
risk, and provide longer lasting consistent ous materials regulations if the specimens are
throughput. properly packaged and marked. Blood compo-
The Department of Energy's Office of Radio- nents, cellular therapy products, and tissue for
logical Security offers a program to assist facili- transfusion or transplantation are not subject to
ties that wish to make the switch from a cesium hazardous material regulations. Method 1-1 pro-
irradiator to a safer alternative. Financial incen- vides additional shipping instructions for safe
tives include no-cost removal and disposalof the transport of these materials. However, the most
cesium irradiator and partial payment toward a recent revision of the lATA or US DOT regula-
new non-radioisotopic device. Detailed informa- tions should be consulted for the most current
tion is availableat ORSinfo@nnsa.doe.gov. classlfication, packaging, and labeling require-
ments as well as for limitations on the volumes
of hazardous materials that can be packaged to-
u gether in one container.
Hazardous materials commonly shipped by

transfusion medicine, cellular therapy, and clini-
cal diagnostic services include infectious sub-
stances, biological substances, liquid nitrogen, Those responsible for safety at a facilitymust be
and dry ice. concerned with protecting the environment as
The US DOT regulations for transportation of well as all staffmembers. Every effort should be
hazardous materials are harmonized with the made to establish facility-wide programs to
international standards published annually by reduce solid wastes, including nonhazardous
the International Air Transport Association and, especially,hazardous wastes (ie, biohazard,
(IATA).55,56 These regulations provide instruc- chemical, and radioactive wastes). 57
tions for identifying, classifying, packaging, A hazardous-waste-reduction program insti-
marking, labeling, and documenting hazardous tuted at the point of use of the material achieves
materials to be offered for shipment on public several goals. It reduces the institutional risk for
roadways or by air. occupational exposures to hazardous agents, re-
Specimens are classifiedas CategoryA if they duces "cradle-to-grave"liabilityfor disposal, and
are known or likely to contain infectious sub- enhances compliance with environmental re-
stances in a form that is capable of causing per- quirements to reduce pollution generated from
manent disability or life-threatening or fatal dis- dailyoperations of the laboratory.37,58,59
ease in otherwise healthy humans or animals Facilities can minimize pollution of the envi-
when an exposure occurs. The proper shipping ronment by practicing the "three R's": reduce,
name for Category A specimens is "infectious reuse, and recycle. Seeking suitable alternatives
substances, affectinghumans" (UN2BI4) or "in- to materials that create hazardous waste and
fectious substances, affecting animals only" separating hazardous waste from nonhazardous
(UN2900). waste can reduce the volume of hazardous
Specimens that may contain infectious sub- waste and decrease costs for its disposal.
stances but do not have the level of risk de- Changes in techniques or materials to reduce
scribed above are classified as Category B, and the volume of infectious waste or render it less
hazardous should be carefully considered, and ty for disposal of the final containers. State and
employees should be encouraged to identify saf- local health departments must be involvedin re-
er alternatives whenever possible. viewing transportation and disposal practices
Facilities should check with state and local where this is an issue, and procedures must be
health and environmental authorities about cur- developed in accordance with state. and local
rent requirements for storage and disposal of a regulations as well as those of the US DOT. An
particular multihazardous waste before creating
example of a·risk mitigation strategy would be
that waste. If creating the multihazardous waste
cannot be avoided, the volume of waste generat- implementing the use of approved devices to
ed should be minimized. For example, in some measure hemoglobin using a cuvette testing sys-
states, copper sulfate contaminated with blood tem. The cuvettes can then be discarded much
is considered a multihazardous waste. The dis- like other sharps, therefore mitigating the risk of
posal of this waste poses several problems with both chemical and blood-borne sources of expo-
transportation from draw sites to a central facili- sure.
REFERENCES
1. Gammon R, ed. Standards for blood banks and 4. Clinical laboratory safety: Approved guideline.
transfusion services. 32nd ed. Bethesda, MD: 3rd ed. NCCLS document GP17-A3. Wayne,
MBB,2020. PA: Clinical and Laboratory Standards Institute,
2. Haspel RL,ed. Standards for cellular therapy ser- 2012.
vices. 9th ed. Bethesda, MD: MBB, 2019. 5. Hospital accreditation standards. Oakbrook Ter-
3. Laboratory Accreditation Program laboratory race, IL: The Joint Commission, 2019.
general checklist. Northfield, IL: College of 6. Laboratory accreditation standards. Oakbrook
American Pathologists, 2018. Terrace, IL:The Joint Commission, 2019.
7. NFPA 70-National electrical code. Quincy, 20. Code of federal regulations. Title 29, CFR Part
MA:National Fire Protection Association,2017. 1910.1450. Washington, DC: US Government
8. ANSIIASHRAE Standard 62.1·2016. Ventila- PublishingOffice,2019 (revised annually).
tion for acceptable indoor air quality. Atlanta, 21. Centers for Disease Control and Prevention.
GA:American Societyof Heating, Refrigerating, Public Health Serviceguidelinesfor the manage-
and Air-ConditioningEngineers, Inc, 2016. ment of occupational exposures to HBV, HCV,
9. Code of federal regulations. Title 21, CFR Part and HIV and recommendations for postexpo-
1271.190. Washington, DC: US Government sure prophylaxis. MMWR Morb Mortal Wkly
PublishingOffice,2019 (revisedannually). Rep 2001;50:1-52.
10. ISO-14644: Cleanrooms and associated con- 22. Code of federal regulations. Title 29, CFR Part
trolled environments, Parts 1-9. ISO/TC 209. 1904.39. Washington, DC: US Government
Geneva, Switzerland: International Organiza- PublishingOffice,2019 (revised annually).
tion for Standardization, 1999-2015. 23. Code of federal regulations. Title 29, CFRParts
11. ISO-14698: Cleanrooms and associated con- 1904.1 and 1904.7. Washington, DC: US Gov-
trolled environments-bio-contamination con- ernment PublishingOffice, 2019 (revised annu-
trol, Part 1: General principles and methods. ally).
ISO/TC 209. Geneva, Switzerland: Internation- 24. Code of federal regulations. Title 29, CFR Part
al Organization for Standardization, 2015. 1910.1020. Washington, DC: US Government
12. Code of federal regulations. Title 10, CFR Part PublishingOffice,2019 (revised annually).
25. NIOSH Alert: Preventing allergic reactions to
20. Washington, DC: US Government Publish-
natural rubber latex in the workplace. (June
ing Office,2019 (revisedannually).
13. SiegelJD, Rhinehart E, Jackson M, et al for the 1997) NIOSH Publication No. 97-135. Wash-
Healthcare Infection Control Practices Advisory ington, DC: National Institute for Occupational
Safety and Health, 1997. [Availableat http://
Committee. 2007 Guideline for isolation pre-
www.cdc.gov/niosh/docs/97-135/.]
cautions: Preventing transmission of infectious
26. NFPA 101: Life safety code. Quincy, MA: Na-
agents in healthcare settings. Atlanta, GA: Cen-
tional Fire Protection Assoctation,2018.
ters for Disease Control and Prevention, 2007. 27. Fowler TW, Miles KK. Electrical safety: Safety
[Available at https://www.cdc.gov/infection and health for electrical trades student manual.
controll guidelines/isolation/index.html.] (January 2009) NIOSH Publication No. 2009-
14. Code of federal regulations. Title 29, CFR Part 113. Washington, DC: National Institute for Oc-
1910.1030. Washington, DC: US Government cupational Safetyand Health, 2002.
PublishingOffice,2019 (revisedannually). 28. OSHA technical manual: TED 1-0.15A. Wash·
15. Code of federal regulations. Title 29, CFR Part ington, DC: US Department of Labor, 1999.
1910.1200. Washington, DC: US Government 29. Enforcement procedures for the occupational
PublishingOffice,2019 (revisedannually). exposure to bloodborne pathogens. Directive
16. US Department of Health and Human Services. CPL 02-02-069. Washington, DC: US Depart-
Biosafetyin microbiologicaland biomedical lab- ment of Labor,2001.
oratories. 5th ed. Washington, DC: US Govern- 30. US Department of Health and Human Services.
ment PublishingOffice,2009. Appendix A: Primary containment for biohaz-
17. Bernard B, ed. Musculoskeletal disorders and ards: Selection,installation, and use of biological
workplace factors:A critical review of epidemio- safety cabinets. In: Biosafetyin microbiological
logic evidence for work-related musculoskeletal and biomedical laboratories. 5th ed. Washing-
disorders of the neck, upper extremity, and low ton, DC: US Government Publishing Office,
back. NIOSHpublication no. 97-141. Washing- 2009. [Available at http://www.cdc.gov/bio
ton, DC: National Institute for Occupational safety/publications.]
Safetyand Health, 1997. 31. Richmond JY. Safe practices and procedures for
18. Code of federal regulations. Title 29, CFR Part working with human specimens in biomedical
1910.38. Washington, DC: US Government research laboratories. J Clin Immunoassay
PublishingOffice,2019 (revisedannually). 1988;11:115-19.
19. Wagner KD, ed. Environmental management in 32. US Environmental Protection Agency. Pesticide
healthcare facilities.Philadelphia:WB Saunders, registration: Selected EPA-registered disinfec-
1998. tants. Washington, DC: EPA,2016. [Availableat
https://www.epa.gov/pesticide-registration/ 2014. New York: American National Standards

selected-epa-regtstered-disinfectants.] Institute, 2014.
33. RutalaWA. APIC guideline for selection and use 44. Inspection procedures for 29 CFR 1910.120
of disinfectants. Am J Infect Control 1996;24: and 1926.65, paragraph (q): Emergency re-
313-42. sponse to hazardous substance releases. OSHA
34. Evans MR, Henderson DK,BennettJE. Potential Directive CPL02-02-073. Washington, DC: Oc-
for laboratory exposures to biohazardous agents cupational Safety and Health Administration,
found in blood. Am J Public Health 1990;80: 2007.
423-7. 45. Code of federal regulations. Title 29, CFR Part
35. Food and Drug Administration. Memorandum: 1910.1000. Washington, DC: US Government
Guideline for collection of blood or blood prod- PublishingOffice, 2019 (revisedannually).
ucts from donors with positive tests for infec- 46. Cook SS. Selection and installation of self-con-
tious disease markers ("high risk" donors). (Sep- tained irradiators. In: Butch S, Tiehen A, eds.
tember, 1989) Silver Spring, MD: CBEROffice Bloodirradiation: A user's guide. Bethesda, MD:
of Communication, Outreach, and Develop- MBB Press, 1996: 19-40.
ment, 1989. 47. Beir V. Health effects of exposure to low levels
36. Food and Drug Administration. Memorandum: of ionizing radiation. Washington, DC: National
Revisionto 26 October 1989 guidelines for col- Academy Press, 1990:1-8.
lection of blood or blood products from donors 48. Regulatory guide 8.29: Instruction concerning
with positive tests for infectious disease markers risks from occupational radiation exposure.
("high-risk" donors). (April 17, 1991) Silver Washington, DC: Nuclear Regulatory Commis-
Spring, MD: CBEROffice of Communication, sion, 1996.
49. NCRP report no. 115: Risk estimates for radia-
Outreach, and Development, 1991. [Available
tion protection: Recommendations of the Na-
at https://www.fda.gov/vaccines-blood-biolog-
tional Council on RadiationProtection and Mea-
ics/ other-recommendations-biologics- manufac
surements. Bethesda, MD: National Council on
turers/memoranda-blood-establfshments.]
RadiationProtection and Measurements, 1993.
37. US Environmental Protection Agency. EPA
50. NCRP report no. 105: Radiation protection for
guide for infectious waste management. EPN medical and allied health personnel: Recom-
530-SW-86-014. NTIS #PB86-199130. Wash- mendations of the National Council on Radia-
ington, DC: National Technical Information Ser- tion Protection and Measurements. Bethesda,
vice, 1986. MD: National Council on Radiation Protection
38. Code of federal regulations. Titie 40, CFR Part and Measurements, 1989.
264. Washington, DC: US Government Publish- 51. EA-05 090. Enforcement action: Order impos-
ing Office, 2019 (revisedannually). ing increased controls (licensees authorized to
39. NIOSH pocket guide to chemical hazards. possess radioactive material quantities of con-
Washington, DC: National Institute for Occupa- cern). (November 14, 2005) Rockville,MD: US
tional Safety and Health, 2010. [Available at Nuclear Regulatory Commission, 2005.
https://www.cdc.gov/niosh/npg.] 52. RIS 2007-14. Fingerprinting requirements for
40. NFPA 704-Standard system for the identifica- licensees implementing the increased control or-
tion of the hazards of materials for emergency der. (June 5, 2007) Rockville,MD: US Nuclear
response. Quincy, MA: National Fire Protection RegulatoryCommission, 2007.
Association,2017. 53. US Nuclear Regulatory Commission regulatory
41. American Coatings Association. HMIS imple- guide 8.13: Instruction concerning prenatal ra-
mentation manual. 4th ed. Neenah, WI: JJ diation exposure. Washington, DC: NRC, 1999.
Kellerand Associates, Inc, 2014. 54. Nuclear RegulatoryCommissionregulatoryguide
42. Lisella FS, Thomasston SW. Chemical safety in 8.23: Radiation surveys at medical institutions.
the microbiology laboratory. In: Fleming DO, Washington, DC: NRC, 1981.
RichardsonJH, TulisJJ, Vesley D, eds. Laborato- 55. Code of federal regulations. Title 49, CFR Parts
ry safety, principles, and practices. 2nd ed. 171.22. Washington, DC: US Government Pub-
Washington, DC: American Societyfor Microbi- lishing Office, 2019 (revisedannually).
ologyPress, 1995:247-54. 56. Dangerous goods regulations manual. 54th ed.
43. American national standards for emergency eye- Montreal, PQ, Canada: International Air Trans-
wash and shower equipment. ANSI Z358.1- port Association, 2019 (revisedannually).
57. United States Code. Pollution prevention act. Clinical and Laboratory Standards Institute,
42 USC §§13101 and 13102 et seq. 2011.
58. Clinical laboratory waste management. Ap- 59. Code of federal regulations. Title 21, CFRPart
proved guideline.3rd ed. GP05-A3.Wayne, PA: 606.40(d)(1.)Washington,DC: USGovernment
PublishingOffice,2019 (revisedannually).
APPENDIX 2-1
Safety Regulations and Recommendations Applicable to Health-Care Settings
FederalRegulations and
Recommendations
NuclearRegulatory Commission 10 CFR20 St~ndardsfor Protection Against Radia-
tion
10 CFR36 Licenses and Radiation Safety Require-
ments for Irradiators
Guide8.29 Instructions Concerning Risks from
Occupational Radiation Exposure
OccupationalSafety and Health 29 CFR1910.1030 Occupational Exposureto Blood-
Administration borne Pathogens
29 CFR1910.1020 Access to EmployeeExposureand
Medical Records
29 CFR1910.1096 Ionizing Radiation
29 CFR1910.1200 HazardCommunication Standard
29 CFR1910.1450 Occupational Exposureto Hazardous
Chemicals in Laboratories
Department of Transportation 49 CFR171-180 HazardousMaterials Regulations
Environmental Protection Agency EPAGuidefor Infectious Waste Manage-
(EPA) ment
Centersfor DiseaseControl and Guidelinefor Isolation Precautions in
Prevention Hospitals
Foodand Drug Administration 21 CFR606.3-606.171 Current Good Manufacturing Practicefor
Blood and Blood Components
21 CFR630.40 GeneralRequirementsfor Blood, Blood
Components, and Blood Derivatives
21 CFR640.1-640.130 Additional Standards for Human Blood
and Blood Products
21 CFR211.1-211.208 Current Good Manufacturing Practicefor
Finished Pharmaceuticals
21 CFR1270 Human Tissue Intendedfor Transplanta-
tion
21 CFR1271 Human Celis,Tissues, and Cellularand
Tissue-BasedProducts
(Continued)
APPENDIX 2-1
Safety Regulations and Recommendations Applicable to Health-Care Settings
Tradeand Professional
Organizations
National Fire Protection NFPA70 National Electrical Code
Association (NFPA)
NFPA70E Electrical Safety Requirements for
EmployeeWorkplaces
NFPA101 Life Safety Code
NFPA99 Standards for Health Care Facilities
NFPA704 Standard for Identification of the Hazards
of Materials for Emergency Response
National Paint and Coatings Hazardous Materials Identification Sys-
Association tem Implementation Manual
International Air Transport Dangerous Goods Regulations
Association
CFR = Code of Federal Regulations.
APPENDIX 2-2
General Guidance for Safe Work Practices, Personal Protective Equipment, and Engineering
Controls
UNIFORMS AND LABORATORY COATS
Personnel should wear closed laboratory coats or full aprons over long-sleeved uniforms or gowns when they
are exposed to blood, corrosive chemicals, or carcinogens. The material of required coverings should be
appropriate for the type and amount of hazard exposure. Plastic disposable aprons may be worn over cotton
coats when there is a.high probability of large spills or splashing of blood and body fluids; nitrile rubber
aprons may be preferred when caustic chemicals are poured.
Protective coverings should be removed before the employee leavesthe work area and should be discarded or
stored away from heat sources and clean clothing. Contaminated clothing should be removed promptly,
placed in a suitable container, and laundered or discarded as potentially infectious. Home laundering of gar-
ments worn in Biosafety Level 2 areas is not permitted because unpredictable methods of transportation and
handling can spread contamination, and home laundering techniques may not be ettectlve.'
GLOVES
Gloves or equivalent barriers should be used whenever tasks are likely to involve exposure to hazardous
materials.
Types of Gloves
Glove type varies with the task:
Sterile gloves: for procedures involving contact with normally sterile areas of the body.
Examination gloves: for procedures involving contact with mucous membranes, unless otherwise indicated,
and for other patient care or diagnostic procedures that do not require the use of sterile gloves.
e Rubber utility gloves: for housekeeping chores involving potential blood contact, instrument cleaning and
decontamination procedures, and handling concentrated acids and organic solvents. Utility gloves may be
decontaminated and reused but should be discarded if they show signs of deterioration (eg, peeling, cracks,
or discoloration) or if they develop punctures qr tears.
Insulated gloves: for handling hot or frozen material.
The following situations require the use of gloves1:
The following guidelines should be used to determine when gloves are necessary':
@ For donor phlebotomy when the health-care worker has cuts, scratches, or other breaks in his or her skin.
For phlebotomy of autologous donors or patients (eg, therapeutic apheresis procedures or intraoperative red
cell collection).
For persons who are receiving training in phlebotomy.
@ When handling open blood containers or specimens.
When collecting or handling blood or specimens from patients or donors known to be infected with a blood-
borne pathogen.
When examining mucous membranes or open skin lesions.
When handling corrosive chemicals and radioactive materials.
(Continued)
APPENDIX 2-2
General Guidance for Safe Work Practices, Personal Protective Equipment, and
Engineering Controls (Continued)
When cleaning up spills or handlingwaste materials.

Whenthe likelihood of exposurecannot beassessedbecauseof lack of experiencewith a procedureor situa-
tion.
The Occupational Safety and HealthAdministration (OSHA)does not require the routine use of gloves by phle-
botomists working with healthy prescreeneddonors or the changing of unsoiled gloves betweendonors if
gloves are worn." Experiencehas shown that the phlebotomy process is low risk becausedonors typically
have low rates of infectious diseasemarkers. Also, exposure to blood is rare during routine phlebotomy, and
other alternatives can be usedto provide barrier protection, such as using a folded gauzepad to control any
blood flow when the needleis removedfrom the donor's arm.
Employerswhose policies and procedures do not require routine gloving should periodically reevaluatethe
potential needfor gloves. Employeesshould never be discouragedfrom using gloves, and gloves should
always be available.
Guidance on Use
The safe use of gloves by employees includes the following3.4:

'" Securely bandageor cover open skin lesions on handsand arms before putting on gloves.
Changegloves immediatelyif they are torn, punctured, or contaminated;after handling high-risk samples;or
after performing a physicalexamination(eg, on an apheresisdonor).
Removegloves by keepingtheir outside surfaces in contact only with outside and by turning the glove inside
out while taking it off.
Use gloves only when needed,and avoid touching clean surfaces such as telephones, doorknobs, or com-
puter terminals with gloves.
Changegloves betweenpatient contacts. Unsoiledgloves neednot be changedbetweendonors.
Wash handswith soap or other suitable disinfectant after removing gloves.
Do not wash or disinfect surgical or examination gloves for reuse. Washing with surfactants may cause
"wicking" (ie, enhanced penetration of liquids through undetectedholes in the glove). Disinfecting agents
may causedeterioration of gloves.
Useonly water-basedhand lotions with gloves,if needed;oil-basedproducts causeminute cracks in latex.
FACE SHIELDS, MASKS, AND SAFETY GOGGLES
Where there is a risk of blood or chemical splashes,the eyesand mucous membranesof the mouth and nose
should be protected." Permanentshields fixed as a part of equipment or bench design are preferred (eg,
splash barriers attached to tubing sealers or centrifuge cabinets). All barriers should be cleanedand disin-
fected on a regular basis.
Safety glasses alone provide impact protection from projectiles but do not adequatelyprotect eyesfrom bio-
hazardor chemical splashes. Full-faceshields or masks and safety goggles are recommended when perma-
nent shields cannot be used. Many designs are commercially available;eliciting staff input on comfort and
selection can increase use.
Masks should be worn wheneverthere is dangerfrom inhalation. Simple, disposable dust masks are adequate
for handling dry chemicals, but respirators with organic vapor filters are preferred for areaswhere noxious
APPENDIX 2-2
General Guidance for Safe Work Practices, Personal Protective Equipment, and
Engineering Controls (Continued)
- - -
fumes are produced (eg, for cleaning upspills of noxious materials). Respirators should be fitted to their
wearers and checked annually.
HAND WASHING
Frequent, thorough hand washing is the first line of defense in infection control. Blood-borne pathogens gen-
erally do not penetrate intact skin, so immediate removal reduces the likelihood of transfer to a mucous mem-
brane or broken skin area or of transmission to others. Thorough washing of hands (and arms) also reduces
the risks from exposure to hazardous chemicals and radioactive materials.
Employees should always wash their hands before leaving a restricted work area or using a biosafety cabinet,
between medical examinations, immediately after becoming soiled withblood or hazardous materials, after
removing gloves, or after using the toilet. Washing hands thoroughly before touching contact lenses or apply-
ing cosmetics is essential.
OSHAallows the use of waterless antiseptic solutions for hand washing as an interim method.' These solu-
tions are useful for mobile donor collections or in areas where water is not readily available for cleanup pur-
poses. If such methods are used, however, hands must be washed with soap and funning water as soon as
possible thereafter. Becausethere is no listing or registration of acceptable hand-wipe products similar to the
one that the Environmental Protection Agency maintains for surface disinfectants, consumers should request
data from the manufacturer to support advertising claims.
EYEWASHES
Laboratory areas that contain hazardous chemicals must be equipped with eyewash stations." Unobstructed
access within a 10-second w(llk from the location of chemical use must be provided for these staflons, Eye-
washes must operate so.that both of the user's hands are free to hoidopen the eyes. Procedures andlndlca-
tions for use must be posted, and routine function checks mustbe per,forrned.Testing eyewashtountalns
weekly helps ensure proper function and flushes out stagnant water. Portable eyewash systems are allowed
only if they can deli\lerflushing fl~idto the eyesata rate of at least 1.5 liters per minute for 15 minutes.They
should be monitored routinely to ensure the purity of their contents.
Employees should be trained in the proper use of eyewash devices, although prevention-through consistent
and appropriate use of safety glasses or shields-is preferred. If a splash occurs, the employee should be
directed to keep nls or her eyelidsopen and to. use the eyewash according to procedures, or the employee
should go to the nearest sink and direct a steady, tepid stream of water into his or her eyes. solunons other
than water should be used only in accordance with a physician's direction.
After eyes are adequately flushed (many facilities recommend 15 minutes), follow-up medical care sho~ld be
sought,especiallyif pain or redness develops. Whether washing the eyes is effective in preventing infecti()n
has hot been demonstrated, but it is considered desirable when acCidentsoccur. ....
1. Code of federal regulations. Title 29, CFR Part 1910.1030. Washington, DC: US Government Publishing Office, 2019
(revised annually).
2. OccupationalSafety and Health Administration. Enforcement procedures for the occupational exposure to bloodborne
pathogens. OSHA Instruction CPL 02-02-069. Washington, DC: US Government Publishing Office, 2001. [Available at
htlps:llwww.osha.gov/enforcemenVdirectives/cpl-02-02-069.1
3. Clinical laboratory safety: Approved guideline. 3rd ed (GP17-A3). Wayne, PA: Clinical and Laboratory Standards
Institute, 2012.
4. CAPaccreditation checklists: Laboratory general.Chicago:Collegeof American Pathologists, 2018.
5. American national standards for emergencyeyewashand shower equipment. ANSI Z358.1-2009. New York: American
NationalStandards Institute, 2009.
APPENDIX 2-3
Biosafety Level 2 Precautions
Biosafety Level 2 precautions as applied in the blood establishment setting include at least the
"
ill High-risk activities are appropriately segregated from lower-risk activities, and the boundaries are clearly
defined.
@ Benchtops are easily cleanedand are decontaminateddaily with a hospital disinfectant approved by the Envi-
ronmental Protection Agency.
" Laboratory rooms have closable doors and sinks. An air system with no recirculation is preferred but not
required.
€I Workers are required to perform proceduresthat createaerosols (eg, opening evacuatedtubes, centrifuging,
mixing, or sonication) in a biological safety cabinet or equivalentor to wear masks and goggles in addition to
gloves and gowns during such procedures. (Note: Opentubes of blood should not be centrifuged. If whole
units of blood or plasma are centrifuged, overwrapping is recommendedto contain leaks.)
l!l Gowns and gloves are used routinely and in accordancewith general safety guidelines. Faceshields or their
equivalentsare used where there is a risk from splashing.
* Mouth pipetting is prohibited.
'Ill No eating, drinking, smoking, applying cosmetics, or manipulating contact lenses occurs in the work area.
All food and drink are stored outside the restricted area,and laboratory glassware is never used for food or
drink. Personnelare instructed to avoid touching their face, ears, mouth, eyes, or nose with their hands or
other objects, such as pencils and telephones.
Needles and syringes are used and disposed of in a safe manner. Needles must never be bent, broken,
sheared, replacedin a sheath, or detachedfrom a syringe before being placed in puncture-proof, leak-proof
containers for controlled disposal. Proceduresare designedto minimize exposureto sharp objects.
® All blood specimens are placed in well-constructed containers with secure lids to prevent leaking during
transport. Blood is packagedfor shipment in accordancewith regulatory agency requirements for etiologic
agents or clinical specimens, as appropriate.
III Infectious waste is not compactedand is decontaminatedbefore its disposal in leak-proof containers. Proper
packaging includes double, seamless,tear-resistant, orange or red bagsthat are enclosed in protective car-
tons. Both the cartons and the bags inside display the biohazardsymbol. Throughout delivery to an incinera-
tor or autoclave, waste is handled only by suitably trained persons. If a waste management contractor is
used,the agreement should clearly define the respectiveresponsibilities of the staff and the contractor.
II> Equipment to be repaired or submitted for preventive maintenance,if potentially contaminated with blood,
must be decontaminatedbefore its releaseto a repairtechnician.
@ Accidental exposure to suspected or actual hazardous material is reported to the laboratory director or
responsible person immediately.
1. Clinical laboratory safety: Approved guideline. 3rd ed (GP17-A3). Wayne, PA: Clinical and laboratory Standards
Institute, 2012.
2. FlemingDO.laboratory biosafetypractices.In: FlemingDO,RichardsonJH, Tulls JJ, VesleyDO,eds. laboratory safety,
principles,and practices.2nd ed. Washington,DC:AmericanSocietyfor Microbiology Press,1995:203-18.
APPENDIX 2-4-
Sample List of Hazardous Chemicals That May Be Encountered in a Blood Bank
Ammonium chloride
Bromelin Irritant, sensitizer
Calcium chloride Irritant
Carbon dioxide (frozen dry ice) Caustic
Carbonyl iron powder Oxidizer
Chloroform Toxic, suspected carcinogen
Chloroquine Irritant, corrosive
Chromium-111 chloride hexahydrate Toxic, irritant, sensitizer
Citric acid Irritant
Copper sulfate (cupric sulfate) Toxic, irritant
Dichloromethane Toxic, irritant
Digitonin Toxic
Dimethyl sulfoxide Irritant
Dry ice (carbon dioxide, frozen) Caustic
Ethidium bromide Carcinogen, irritant
Ethylenediaminetetraaceticacid Irritant
Ethyl ether Highly flammable and explosive, toxic, irritant
Ficin (powder) Irrltant, sensitizer
Formaldehydesolution (34.9%) Suspected carcinogen, combustible, toxic
Glycerol Irritant
Hydrochloric acid Highly toxic, corrosive
Imidazole Irritant
Isopropyl (rubbing) alcohol Flammable, irritant
Liquid nitrogen Caustic
Lyphogel Corrosive
2-Mercaptoethanol Toxic, stench
Mercury Toxic
Mineral oil Irritant, carcinogen, combustible
Papain Irritant, sensitizer
Polybrene Toxic
Sodium azide Toxic, irritant, explosive when heated
(Continued)
APPENDIX 2-4
Sample List of Hazardous Chemicals That May Be Encountered in a Blood Bank (Continued)
Sodium ethylmercurithiosalicylate (thimerosal) Highly toxic, irritant

Sodium hydrosulfite Toxic, irritant
Sodium hydroxide Corrosive,toxic
Sodium hypochlorite (bleach) Corrosive
Sodium phosphate Irritant, hygroscopic
Sulfosalicylic acid Toxic, corrosive
Trichloroacetic acid Corrosive,toxic
Trypsin Irritant, sensitizer
Xylene Highly flammable, toxic, irritant
APPENDIX 2-5
Chemical Categories and How to Work Safely with Them
Acids, alkalis, and Irritation, severe burns, During transport, protect Store concentrated acids
corrosive compounds tissue damage large containers with in acid safety cabinets.
plastic or rubber Limit volumes of
bucket carriers. concentrated acids to
During pouring, wear eye 1 liter per container.
protection and Post cautions for
chemical-resistant- materials in the area.
rated gloves and Report changes in
gowns as appearance to chemical
recommended. safety officer.
Always add acid to water; (Perchloric acid may be
never add water to explosive if it becomes
acid. yellowish or brown.)
When working with large
jugs, have one hand on
the neck and the other
hand at the base, and
position them away
from the face.
Acrylamide Neurotoxic, carcinogenic, Wear chemically rated Store in a chemical
absorbed through the gloves. cabinet.
skin Wash hands immediately
after exposure.
Compressed gases Explosive Label contents. Transport using hand
Leavevalve safety covers trucks or dollies.
on until use. Placecylinders in a stand
Open valves slowly for or secure them to
use. prevent tipping over.
Label empty tanks. Store in well-ventilated
separate rooms.
Do not store oxygen close
to combustible gas or
solvents.
Checkconnections for
leaks using soapy water.
(Continued)
APPENDIX 2-5
Chemical Categories and How to Work Safely with Them (Continued)
Flammable solvents Classifiedaccording to Useextreme caution Make every attempt to

flash point-see when handling. replace hazardous
material safety data Post "No Smoking" signs materials with less
sheet, classified in working area. hazardous materials.
according to volatility Keepa fire extinguisher Store containers larger
and solvent cleanup kit than 1 galion in a
in the room. flammable-solvent
Pour volatile solvents storage room or a fire
under a suitable hood. safety cabinet.
Useeye protection and Ground metal containers
chemical-resistant by connecting the can to
neoprene gloves when a water pipe or ground
pouring. connection. If the
No flame or other source recipient container is
of possible ignition also metal, it should be
should be in or near electrically connectedto
areaswhere flammable the delivery container
solvents are being during pouring.
poured.
Label as "flammable."
Liquid nitrogen Freezeinjury, severe Use heavy insulated The tanks should be
burns to skin or eyes gloves and goggles securely supported to
when working with avoid being tipped over.
liquid nitrogen. The final container of
liquid nitrogen (freezing
unit) must be securely
supported to avoid
being tipped over.
APPENDIX 2-6
Incidental Spill Response*
Acid-resistant gloves
Acetic Apron and coveralls absorbent material
Contact causes burns to Goggles and face shield Absorbent boom
Hydrochloric
skin and eyes. Acid-resistant foot covers Leak-proof containers
Nitric Spills are corrosive. Absorbent pillow
Perchloric Fireor contact with metal Mat (cover drain)
Sulfuric may produce irritating Shovel or paddle
Photographic or poisonous gas.
chemicals (acidic) Nitric, perchloric, and
sulfuric acids are
water-reactive
oxidizers.
Basesand caustics Spills are corrosive. Gloves Basecontrol/neutralizer
Potassium hydroxide Fire may produce Impervious apron or Absorbent pillow
irritating or poisonous coveralls Absorbent boom
Sodium hydroxide
gas. Goggles or face shield Drain mat
Photographic Impervious foot covers Leak-proof container
chemicals (basic) Shovel or paddle
Chlorine Inhalation can cause Gloves (double set of 4H Chlorine control powder
Bleach respiratory irritation. undergloves and butyl Absorbent pillow
Liquid contact can or nitrile overgloves) Absorbent material
Sodium hypochlorite
produce irritation of the Impervious apron or Absorbent boom
eyes or skin. coveralls Drain mat
Toxicity is caused by Goggles or face shield Vapor barrier
alkalinity, possible Impervious foot covers Leak-proof container
chlorine gas (neoprene boots for Shovel or paddle
generation, and oxidant emergency response
properties. releases)
Self-contained breathing
apparatus (emergency
response releases)
Cryogenic gases Contact with liquid Full face shield or Handtruck (to transport
Carbon dioxide nitrogen can produce goggles cylinder outdoors if
frostbite. Neoprene boots necessary)
Nitrous oxide
Releasecan create an Gloves (insulated to Soap solution (to check
Liquid nitrogen oxygen-deficient provide protection from for leaks)
atmosphere. the cold) Putty (to stop minor pipe
Nitrous oxide has and line leaks)
anesthetic effects.
(Continued)
APPENDIX 2-6
Incidental Spill Response* (Continued)
Flammablegases Simple asphyxiant Faceshield and goggles Handtruck (to transport

Acetylene (displaces air). Neopreneboots cylinder outdoors if
Oxygen gases Inhaledvapors havean Double set of gloves needed)
Butane anesthetic potential. Coverallswith hood and Soap solution (to check
Propane Flammablegasesposean feet for leaks)
extremefire and
explosion hazard.
Releasecan create an
oxygen-deficient
atmosphere.
Flammableliquids Vapors are harmful if Gloves (double set of 4H Absorbent material
Acetone inhaled (central underglovesand butyl Absorbent boom
nervous system or nitrile overgloves) Absorbent pillow
Xylene
depressants). Impervious apron or Shovel or paddle
Methyl alcohol Liquid is harmful if coveralls (nonmetal,
toluene absorbed through the Goggles or face shield nonsparking)
Ethyl alcohol skin. Impervious foot covers Drain mat
Other alcohols Substancesare extremely Leak-proof container
flammable.
Liquid evaporatesto form
flammable vapors.
Formaldehydeand Vapors are harmful if Gloves (double set of Aldehyde neutralizeror
glutaraldehyde inhaled; liquids are 4H underglovesand absorbent
4% formaldehyde harmful if absorbed butyl or nitrile Absorbent boom
through skin. overgloves) Absorbent pillow
37% formaldehyde
Substancesare irritants Impervious apron or Shovel or paddle
10% formalin to skin, eyes, and coveralls (nonsparking)
2% glutaraldehyde respiratory tract. Goggles Drain mat
Formaldehydeis a Impervious foot covers Leak-proof container
suspected human
carcinogen.
37% formaldehyde
should be kept away
from heat, sparks, and
flames.
APPENDIX 2-6
Incidental Spill Response* (Continued)
Mercury Mercury and mercury Gloves (double set of 4H Mercury vacuum or spill
Cantor tubes vapors are rapidly undergloves and butyl kit
absorbed in respiratory or nitrile overgloves) Scoop
Thermometers
tract, gastrointestinal Impervious apron or Aspirator
Barometers (GI) tract, or skin. coveralls Hazardouswaste
Sphygmomanometers Short-term exposure may Goggles containers
Mercuric chloride cause erosion of Impervious foot covers Mercury indicator powder
respiratory or GI tracts, Absorbent material
nausea,vomiting, Spatula
bloody diarrhea, shock, Disposabletowels
headache,or metallic Sponge with amalgam
taste. Vapor suppressor
Inhalation of high
concentrations can
cause pneumonitis,
chest pain, dyspnea,
coughing, stomatitis,
gingivitis, and
salivation.
Avoid evaporation of
mercury from tiny
globules by quick and
thorough cleaning.
*This list of physical and health hazardsis not intended as.a substitute for the material safety data sheet (SDS) information.
In case of a spill or if any questions arise, always refer to the chemical-specific SDSfor more complete information.
APPENDIX 2-7
Managing Hazardous Chemical Spills
Deenergize. Liquids: For 37% formaldehyde, deenergizeand remove all sources of ignition
within 10 feet of spilled hazardous material. For flammable liquids, remove all
sources of ignition.
Gases:Removeall sources of heatand ignition within 50 feet for flammable gases.
Removeall sources of heat and ignition for nitrous oxide release.
Isolate, evacuate,and Isolate the spill area and evacuateeveryone from the area surrounding the spill
secure the area. exceptthose responsible for cleaning up the spill. (For mercury, evacuatewithin
10 feet for small spills or 20 feet for large spills.) Securethe area.
Havethe appropriate SeeAppendix 2-2 for recommended PPE.
personal protective
equipment (PPE).
Contain the spill. Liquids or mercury: Stop the source of spill if possible.
Gases:Assess the scene; consider the circumstances of the release(quantity,
location, and ventilation). If circumstances indicate that it is an emergency
response release,make appropriate notifications; if the releaseis determined to
be incidental, contact the supplier for assistance.
Confine the spill. Liquids: Confine the spill to the initial spill area using appropriate control equip-
ment and material. For flammable liquids, dike off all drains.
Gases:Follow the supplier's suggestions or request outside assistance.
Mercury: Useappropriate materials to confine the spill (seeAppendix 2-6). Expel
mercury from the aspirator bulb into a leak-proof container, if applicable.
Neutralizethe spill. Liquids: Apply appropriate control materials to neutralizethe chemical (see
Appendix 2-6).
Mercury: Use a mercury spill kit if needed.
Cleanup the spill. Liquids: Scoop up solidified materials, booms, pillows, and any other materials.
Put used materials into a leak-proof container. Labelthe container with the name
of the hazardous materials. Wipe up residual material. Wipe the spill area sur-
face three times with a detergent solution. Rinse the areaswith clean water. Col-
lect the supplies used (eg, goggles or shovels) and remove gross
contamination; place equipment to be washed and decontaminated into a sepa-
rate container.
Gases: Follow the supplier's suggestions or request outside assistance.
Mercury:Vacuum upthe spill using a mercury vacuum, or scoop up mercury paste
after neutralizationand collect the paste in a designatedcontainer. Usea sponge
and detergentto wipe and cleanthe spill surfacethreetimes to removeabsorbent.
Collectall contaminateddisposal equipment and put it into a hazardouswastecon-
tainer. Collectsuppliesand removegross contamination; placeequipmentthat will
bethoroughly washedand decontaminatedinto a separatecontainer.
APPENDIX 2-7
Managing Hazardous Chemical Spills (Continued)
Dispose. Liquids: Dispose of material that was neutralizedas solid waste. Follow the facil-
ity's procedures for disposal. Forflammable liquids, check with the facility safety
officer for appropriate waste determination.
Gases:The manufacturer or supplier will instruct the facility about disposal if
applicable.
Mercury: Labelwith appropriate hazardous waste label and Department of
Transportation diamond label.
Report. Follow appropriate spill documentation and reporting procedures. Investigate
the spill; perform a root cause analysis if needed.Act on opportunities for
improving safety.
RegulatQ.r.y.C:().n5.id,rations
in Transfusion Medicine and
Cellular Therapies
Joseph Schwartz, MD, MPH; Grieji lIIoh MD; and Yvette C. Tanhehco, PhD, MD, MS
HE FIELDSOF TRANSFUSIONMEDI- accreditation organizations such as AABB or

cine and cellular therapy are highly regu- The Joint Commission publish specific sets of
lated disciplines. Over the years, different standards that need to be met in order for ac-
regulatory bodies have provided oversight at creditation to be granted. Some regulatory agen-
both the state and federal levels in the United cies will grant deeming authority to select ac-
States. The Food and Drug Administration creditation organizations. For example, CMS
(FDA)and the Centers for Medicare and Medic- regulates laboratory testing through the Clinical
aid Services (CMS) are the primary regulatory Laboratory Improvement Amendments (CLlA).
agencies providing federal oversight. In addi- CMS accepts certain accreditation organization
tion, state health departments and other agen- inspections, meaning that the organizations
cies may provide some degree of regulatory have been approved by CMS as having stan-
oversight. Individuals and establishments in- dards and an inspection process that meet or ex-
volved with transfusion medicine and cellular ceed the CMS requirements. Table 3-1 summa-
therapies should be familiar with the different rizes agencies and organizations involved in
requirements of these agencies. regulation and accreditation of blood bank,
It is important to distinguish between regula- transfusion medicine, and cellular therapy facili-
tion and accreditation. Regulations have the ties. The scope of their regulatory oversight
force of law, while accreditation standards are and/or accreditation is detailed on these organi-
not legallybinding. Bloodbanks, transfusion ser- zations' respective websites. In addition, some
vices, and cellular therapy facilitiesmust follow states may have regulations pertaining to blood
the rules set by regulatory agencies. In contrast, banks and cellular therapy facilities.
Joseph Schwartz, MD, MPH, Director, Transfusion Medicine and Cellular Therapy, Columbia University Medical
Center and New York-PresbyterianHospital, and Professor of Pathology and Cell Biology,Columbia University,
New York, New York; Orieji Illoh, MD, Director, Division of Blood Components and Devices, Office of Blood
Research and Review, Center for Biologics Evaluation and Research, Food and Drug Administration, Silver
Spring, Maryland; and Yvette C. Tanhehco, PhD, MD, MS, Assistant Director, Transfusion Medicine, Director,
Apheresis, and Director, Cellular Therapy Laboratory, Columbia University Irving Medical Center and New York-
Presbyterian Hospital, and Assistant Professor of Pathology and Cell Biology, Columbia University, New York,
New York
The authors have disclosed no conflicts of interest. This chapter reflects the views of the authors and should not
be construed to represent their employers' views or policies.
TABLE 3-1. Regulatory and Accreditation Bodies Involved in Blood Banking and CellularTherapies
Foodand Drug Administration (FDA) AABB
Centersfor Medicare and Medicaid Services (CMS) Collegeof American Pathologists (CAP)
Department of Homeland Security The Joint Commission
Nuclear Regulatory Commission (NRC) Foundationfor the Accreditation of Cellular

Therapy (FACT)
Environmental Protection Agency (EPA) National Marrow Donor Program (NMDP)
Occupational Safety and HealthAdministration (OSHA) World Marrow Donor Association (WMDA)
Local state departments of health American Association for Laboratory

Accreditation (A2LA)
US Department of Transportation (US DOT)
National Fire Protection Association (NFPA)
turer to obtain a biologicslicense before placing

a product in interstate commerce.' Further-
more, the US Department of Health and Human
In the United States, when federal laws are en- Services (DHHS)has broad authority to prevent
acted by Congress, they are published as stat- communicable disease transmission under Sec-
utes and placed into the appropriate subject ar- tion 361 of the PHSAct (42 USC264).
eas (titles) of the United States Code (USCV The FDAregulates drugs and medical devic-
Regulations created by federal agencies to en- es under the FD&CAct, which was first passed
force laws are placed (by title) in the Code of in 1938 and amended in 1976 to include medi-
Federal Regulations (CFR).The FDAis the fed- cal devices. Under this act, blood and blood
eral agency that enforces the federal laws relat- components are considered to be drugs because
ed to drugs and biologics, which include blood they are intended to cure, mitigate, treat, or pre-
and blood components, related devices, and vent disease in humans. Manufacturers of drugs
manufacturing facilities. and certain devices must demonstrate to the
Section 351 of the Public Health Service
FDAthe safety and efficacyof a product before
(PHS)Act (USCTitle 42, Section 262) and the
it can be marketed. The FD&C Act requires
Food, Drug, and Cosmetic (FD&C)Act (21 USC
301 et seq) are two statutes that govern the reg- blood product manufacturers to register with
ulation of blood and blood components. The the FDA, obtain biologicslicenses to ship in in-
PHSAct defines blood and blood components as terstate commerce, and follow current good
biological products. This law was first estab- manufacturing practice (cGMP) regulations. It
lished in 1944 as an expansion of the Biologics also prohibits adulteration and misbranding of
Control Act of 1902. In addition to requiring products, authorizes inspection of manufactur-
that biological products be manufactured in a ing facilities,and defines civil and criminal pen-
manner to ensure the safety,purity, and potency alties for violations. The act establishesrequire-
of the product, the PHSAct requires a manufac- ments for the use of unapproved drugs and
C HAP T E R 3 Regulatory Considerationsin TransfusionMedicine and Cellular Therapies 79
devices in their investigational phases and in provals are periodically published on the FDA's
public health emergencies." website."
Within the FDA, the Center for Biologics In addition to regulations, which are legally
Evaluation and Research (CBER)regulates blood binding, the FDA may publish recommenda-
products and most other biological therapies." tions in guidance documents. These guidance
CBER uses multiple overlapping safeguards to documents generally explain FDA's current
ensure that recipients of blood products or cellu- thinking on an issue. The guidance may clarify
lar therapies are protected. This FDA blood- or explain how manufacturers may comply with
safety system includes measures in the following the statute or regulations, or establish good
areas: donor screening, donor testing, donor de- manufacturing standards for blood products.
ferral records, product quarantine, and investi- FDA guidance documents generally do not es-
gation of manufacturing deficiencies. The Cen- tablish legally enforceable responsibilities unless
ter for Devices and RadiologicalHealth (CDRH) specific regulatory or statutory requirements are
regulates most medical devices, but CBER re- cited. The FDA may consider requests for alter-
tains primary jurisdiction over medical devices native approaches to the recommendations stat-
used for blood donation, blood processing, ed in guidance documents if such approaches
transfusion, and cellular products. The FDA'sOf- satisfy the requirements of the applicable law or
fice of Regulatory Affalrs (ORA)has responsibili- statute."
ty for all field operations, which include inspec- As part of the development process for FDA
tions and investigations of blood and device regulations and guidance documents, several fo-
manufacturers.' rums are offered for input from the public and
The FDA promulgates applicable regulations regulated industry. Proposed rules and draft
for blood and blood components and related de- guidance documents are published in the Feder-
vices under both the PHS and FD&CActs. Regu- al Register with an invitation for written com-
lations for blood products are found in Parts ments, which are filed in public dockets. When
210,211, and 600-680 of CFRTitle 21.6 These final rules are published in the Federal Register,
regulations are intended to ensure blood donor the accompanying preamble responds to key
safety and the safety, purity, and potency of questions and comments submitted by the pub-
blood and blood components. In addition, blood lic. The FDA also receives petitions to write or
establishments are required to report fatalities change regulations. Expert opinions on current
associated with blood donation or transfusion to issues are sought from several advisory commit-
the FDA. Table 3-2 provides a summarized list tees, including the FDA Blood Products Adviso-
of relevant regulations applicable to blood estab- ry Committee (BPAC); the FDA Cellular, Tis-
lishments. On May 22, 2015, the FDA pub- sue, and Gene Therapies Advisory Committee
lished a final rule, "Requirements for Blood and (CTGTAC); and the DHHS Advisory Commit-
Blood Components Intended for Transfusion or tee on Blood and Tissue Safety and Avallability
for Further Manufacturing Use," revising 21 (ACBTSA). Public meetings and workshops
CFR Parts 606-660 and updating the FDA's pre- hosted by the FDAon selected topics provide an
vious requirements." The new requirements in- additional opportunity for input. The FDA web-
clude a determination of donor eligibility and site provides links to relevant regulations and
donation suitability, as well as regulations to guidance documents.
help protect donor health.
Manufacturers of blood and blood compo- Registration of Blood Establishments
nents may submit written requests to the FDA
and DeviceManufadurers
for approval of exceptions or alternative proce-
dures to any requirement in the regulations [21 Blood establishments include blood and plasma
CFR 640.120 (all. When the FDA grants ap- donor centers, blood banks, transfusion ser-
provals of exceptions or alternative procedures, vices, other blood product manufacturers, and
the circumstances for these approvals may not independent laboratories that engage in testing
necessarily apply to other facilities. These ap- of donors and blood and blood components. to
TABLE 3-2. Regulations of Interest in Title 21 of the CFR (Food and Drugs)
Topic Section ! Topi Section

FDA general Donor eligibility 630.10, 630.15
Enforcement 1-19 Donation suitability 630.30
Research and development 50-58 Donor notification 630.40
cGMP for drugs 210-211 Blood product stan- 640
dards
Biological products 600-690 Blood collection 640.4

General 600 Blood testing 640.5, 610.40
Licensing 601 Red Blood Cells 640.10-.17
cGMP for blood components 606 Platelets 640.20-.25, 606.145
Personnel, resources 606.20-.65 Plasma 640.30-.34
Standard operating proce- 606.100 Cryoprecipitated AHF 640.50-.56
dures
Labeling 606.120-.122 Exceptions, alterna- 640.120

tives
Compatibility testing 606.151 Medical devices 800-898

Records 606.160-.165 Device adverse events 803
Adverse reactions 606.170 Hematology and 864
pathology
Product deviations 606.171 Tissues
Establishment registration 607 Human cells, tissues, 1271*

and cellular and
tissue-based prod-
ucts
General standards 610 General provisions 1271.1-.20

Donation testing 610.40 Procedures for regis- 1271.21-.37
tration and listing
Donor deferral 610.41,630.10 Donor eligibility 1271.45-.90

Look-back 610.46-.47 cGTP 1271.145-.320
Dating periods 610.50, 610.53 Additional require- 1271.330-.440
ments and inspection
and enforcement
*The following citations represent Subparts A, B, C, D, and E-F,respectively.
CFR = Code of Federal Regulations; FDA = Food and Drug Administration; cGMP = current good manufacturing practice;
AHF = antihemophilic factor; cGTP = current good tissue practice.
C HAP T E R 3 Regulatory Considerations in TransfusionMedicine and Cellular Therapies 81
The FDA has promulgated regulations that re- Licensure of Blood and Blood
quire blood establishments (21 CFR 607) and Component Manufacturers
device manufacturers (21 CFR 807) to register Blood.and blood component manufacturers who
their manufacturing facilities and list the prod- distribute blood products in interstate com-
ucts they manufacture. All establishments that merce must be registered and licensed. The
manufacture blood products are required to reg- blood establishment obtains approval for ltcen-
ister with the FDA, unless they are exempt un- sure by submitting a BiologicsLicense Applica-
der 21 CFR 607.65. Registrants must provide a tion (BLA)to the FDA.The FDA's evaluation of
list of every blood product manufactured, pre- BLAstypicallyincludes the review of supporting
pared, or processed for commercial distribution. documents, such as standard operating proce-
Manufacturers must register and list their prod- dures, labels, quality control data, and a preli-
ucts within 5 days of beginning operations and cense facility inspection. Once a license is is-
annually. sued, the license number is placed on the label
Facilities that routinely collect and process for those products approved to be distributed in
interstate commerce. In addition, licensed man-
blood (including autologous units) or perform
ufacturers are required to inform the FDA of
such procedures as irradiation; washing; prestor- changes in the manufacturing process described
age leukocyte reduction; pooling; or freezing, in their approved BLA.13The reporting category
deglycerolization, and rejuvenation must regis- for such changes depends on the potential of the
ter with the FDA.Transfusion services acting as change to adversely affect the safety, purity, and
depots that forward products to other hospitals potency of the product.
must register as distribution centers. If blood ir- The FDA has published specific guidance
radiation is performed outside the blood bank or ("Changes to an Approved Application: Biologi-
transfusion service, such as in a nuclear medical Products: Human Blood and Blood Compo-
cine. department, that facility must register as nents Intended for Transfusion or for Further
well. Manufacture," December 2014) to assist blood
Transfusion services that do not collect or establishments in determining the appropriate
process blood and blood components are ex- reporting mechanism." As described in the
guidance, the three reporting categories into
empt (21 CFR 607.65) from the registration re-
which a change to an approved application may
quirement in 21 CFR 607. In order to be ex-
be placed are defined in 21 CFR601.12 and are
empt, they must be part of a facility certified as follows:
under CLlA (1988; 42 USC 263a and 42 CFR
493) or certified for reimbursement by CMS.11 <II Major Change: A change that has a substan-
Their manufacturing activities are basic, such as tial potential to have an adverse effect on the
compatibility tests, preparing Red Blood Cells safety or effectiveness of the product. Major
from whole blood, converting unused plasma to changes require the submission of a Prior Ap-
Recovered Plasma; pooling certain blood com- proval Supplement (PAS)to the FDA,which
ponents immediately before transfusion, reduc- the FDAmust approve before the .product is
ing leukocytes in blood components with distributed in interstate commerce [21 CFR
bedside filters, or collecting blood only in emer- 60 1.12(b)].
Moderate Change: A change that has a mod-
gency situations. Under the memorandum of
erate potential to have an adverse effect on
understanding in 1980 between the FDA and the safety or effectiveness of the product.
CMS, the responsibility for routine inspections Moderate changes require the submission of
of these transfusion services was assigned to a Changes Being Effected in 30 Days Supple-
CMS.12The FDA,however, still has jurisdiction ment (CBE30) to the FDA at least 30 days
over transfusion services and may conduct its before interstate distribution of the product
own inspections if warranted. made using the change [21 CFR 601.12(c)].
In certain circumstances, the FDAmay deter- Class III medical devices carry the greatest
mine that the product made using the risk of the three device classifications.These
change may be distributed immediately upon are devices for which general controls, by
receipt of the Changes Being Effected themselves, are insufficient and for which
Supplement (CBE) by the FDA [21 CFR there is insufficient information to establish
601.12 (c)(5)]. special controls to provide reasonable assur-
Minor Change:A change that has a minimal ance of their safety and efficacy. For exam-
potential to have an adverse effect on the ple, tests used to determine red cell antigen
safety or effectiveness of the product. Minor type by molecular methods are regulated as
changes do not need prior approval from the Class III devices, requiring premarket ap-
FDAbut must be described by the manufac- proval (PMA).
turer in an annual report [21 CFR601.12(d)].
The FDAapproves some blood-related devic-
Blood-Related Devices es as biologicsunder the PHSAct and therefore
CBER has the lead responsibility for devices requires the submission of BLAsor related sup-
marketed for transfusion and the collection and plements. These devices include reagents used
processing of blood products and hematopoietic for immunohematology testing by serologic
progenitor cells (HPCs). These devices include methods and most donor-screening infectious
apheresis machines; devices and reagents used disease assays leg, tests for human immunodefi-
for compatibility testing; blood establishment ciency virus (HN), hepatitis B virus (HBV),and
computer software; and blood and human cells, hepatitis C virus (HCV)].
tissues, and cellular and tissue-based product The FDA requires device manufacturers to
(HCTIP) screening tests for infectious diseases. register and list the products they manufacture
The medical device classificationsare based (21 CFR 807). Each device category is assigned
on the risks the device poses to the patient and a code, and all cleared or approved manufactur-
the user or on the level of controls that may be ers and products for that code are searchable in
necessary to ensure the device can be operated the Establishment Registration and Device List-
safelyand effectively": ing database on the CDRHwebsite. 16
Manufacturers and importers of medical de-
Class I medical devices represent the lowest- vices must report deaths and serious injuries re-
level risks to the patient or user. Such devices lated to medical devices to the FDA (21 CFR
are subject to a comprehensive set of regula- 803).17User facilitiesmust report deaths and se-
tory authorities called general controls. Gen- rious injuries in which a device was or may have
eral controls are applicable to all classes of been a factor. Serious injury is defined as being
devices. Examples of Class I devices include life threatening, causing permanent impairment
copper sulfate solutions for hemoglobin or damage, or needing medical or surgical inter-
screening, blood grouping view boxes, and vention. For user facilities,reports of serious in-
heat sealers. juries are sent to the device manufacturer using
Class II medical devices carry greater patient FDA MedWatch Form 3500A within 10 work-
or user risks than Class I devices. These are ing days of the event, or to the FDAif the device
devices for which general controls alone are manufacturer is unknown. Deaths must be re-
insufficient to provide reasonable assurance ported to both the manufacturer and the FDA.
of the safety and effectiveness of the device, In years when a Form 3500A report is submit-
and for which there is sufficient information ted, the user facility must send an annual user
to establish special controls to provide such facilityreport (Form 3419) to the FDAby Janu-
assurance. Most blood-related devices are in ary 1 of the following year." Users may volun-
Class II and cleared through the 51O(k)path- tarily report other device-related adverse events
way, where a device is found to show equiva- or malfunctions to the FDA (Form 3500). All
lence to a predicate. possible adverse events, whether reported or
not, must be investigated, and these records operational systems that are associated with the
must be kept on filefor a minimum of 2 years. layers of safety: quality assurance, donor eligibil-
ity, product testing, quarantine/inventory man-
fDA Inspections agement, and production and processing. With-
in each system, the investigators review
The FDA inspects regulated facilities to deter- standard operating .•procedures, personnel, and
mine compliance with regulations." These. in- training, facilities, equipment calibration .and
spections can be classified as one of the follow- maintenance, and records. Specific require-
ing: ments for individual systems and processes are
discussed in detail in their respective chapters of
Prelicense or preapproval inspection after a the CPGM20
manufacturer submits an application to the Full inspections of all systems are designated
FDA for a biologics license or to market a Level I. After two favorable inspection profiles,
new device or product. facilities with only four or five systems some-
Routine inspection of a regulated facility. times have streamlined Level II inspections of
@! "For-cause" inspection, which involves inves- three systems. Prelicense and preapproval in-
tigation of a specific problem that has come spections or for-cause investigations for com-
to the FDA's attention, such as a complaint or plaints or fatalitiesneed not follow these formats
fatality. because they are more focusedon a specificissue.
If the FDA investigator observes that signifi-
The FDA's ORA and CBER oversee inspec- cant objectionable practices, violations, or con-
tion activities related to transfusion medicine ditions are present that could result in a drug or
and blood banking. The inspection of a blood es- device being adulterated or injurious to health,
tablishment is to ensure manufacturers meet the these observations are written and presented to
standards described in applicable provisions of the facility on FDA Form 483. The FDA Form
the regulations intended to protect donors and 483 serves to notify the manufacturer of the ob-
ensure the safety, purity, and potency of the jectionable conditions and does not constitute a
products they make. These include regulations final determination of whether a violation has
for blood components in Title 21 CFRParts 600, occurred. Investigators are instructed to seek
601, 606, 607, 610, 630, and 640, as well as and record the manufacturer's intentions to
the process and production controls, equipment make corrections. The investigator documents
regulations, and quality control requirements in observations and discussions in an Establish-
21 CFR 211. (SeeTable 3-2.) The licensed man- ment Inspection Report(EIR). The FDAreviews
ufacturers must also meet any additional condi- and considers all the information provided in a
tions of licensure incorporated in their approved Form 483, EIR, and any responses from the
BLA.20 manufacturer and then determines what fur-
When a blood establishment applies for a ther action, if any, is appropriate to protect pub-
BLA,the facilityis generally inspected by a team lic health.
from CBER and ORA. Subsequent inspections The FDAcan take a number of enforcement
may be performed by ORA. actions in response to a violation." Enforcement
ORA provides and publishes policies and in- actions are categorized as advisory, administra-
structions for FDAinvestigators. There is a spe- tive, or judicial. Under advisory actions, the
cific Compliance Program Guidance Manual FDA issues a warning or an untitled letter, in-
(CPGM) for inspections of licensed and unli- forming the manufacturer of noncompliant ac-
censed blood banks. The foundations for blood tivities that could affect donor safety or result in
establishment inspections are in the general the distribution of an unsafe biological product.
FDA regulations for cGMP and drugs, and the The letters provide the facilitywith the opportu-
specific requirements for blood components. All nity for voluntary compliance. Administrative
inspections address the FDA's five layers of actions include product recalls, withdrawals of
blood safety. Investigators review the following product approvals, formal citations of violation,
and-for licensed facilities-suspension or revo- temporary adverse effects or remotely possible

cation of a license. Judicial actions range from serious problems. Class I recalls involve a rea-
seizures of products to court injunctions, civil sonable probability of serious or fatal adverse ef-
monetary penalties, and criminal prosecution. fects. All recalls are published by the FDA.25,26
Market withdrawals occur when a product
Biological Product Deviation has a minor violation that would not be subject
Reporting to FDAlegal action." The manufacturer volun-
tarily removes the product from the market or
When blood establishments discover after distri-
corrects the violation. In collection establish-
bution that a blood product was manufactured
ments, problems such as postdonation informa-
in violation of rules, standards, or specifications,
tion are often in this category. Withdrawals are
they must report the biological product devia-
not published.
tion (BPD) to the FDA [21 CFR 606.171, 21
In some blood guidance documents on infec-
CFR 1271.350(b)]. BPD events are ones in
tious diseases, the FDAhas included recommen-
which the safety, purity, or potency of a distrib-
dations on whether to notify the recipient's phy-
uted blood product may be affected, and may in-
sician about transfused units. In cases of possible
volve any event associated with manufacturing
recent infectious disease exposure in donors or
a product, including collection, testing, process-
transfusion recipients, the test-negative win-
ing, packing, labeling, or storing and distribut-
dow periods for the agent and test kits should be
ing. Licensed and unlicensed manufacturers,
consulted for scheduling prospective testing or
registered blood establishments, and transfusion
reviewing retrospective results, such as for a do-
services that are exempt from registration are re-
nor who has been retested after an exposure."
quired to report BPDs in distributed products.
"Look-back"investigations on units from do-
Bloodestablishments must report a BPD as soon
nors found after donation to have HIV or HCV
as possible, not to exceed 45 calendar days from
are discussed in Chapter 7.
the date the manufacturer became aware of the
reportable event.22 CBER publishes an annual
summary of reported BPDs.23 Most of the re- R
ports used to involve postdonation information.
The FDA no longer considers postdonation in- U
formation to require BPD reports. Blood estab- CMSregulates all US medical laboratories under
lishments should have procedures to investigate CLlA[42 USC 263(a) and 42 CFR4931and Sec-
a BPD and determine if the product should be tion 353 of the PHSAct.28,29 The law and regula-
recalled or withdrawn. tions establish the requirements and procedures
for laboratories to be certified under CLlA as
Managing Recalls and Withdrawals
both a general requirement and a prerequisite
The FDA's requirements for monitoring and infor receiving Medicare and Medicaid reimburse-
vestigating problems with drugs extend to the ment. They provide minimal standards for facili-
time after a product's release. ties, equipment, and personnel. Furthermore,
A recall is defined as the removal or correc- they require participation in a proficiencytesting
tion of a marketed product that is in violation of (PT)program.
the law (21 CFR 7.3 and 7.40). Recallsmay be To be certified, laboratories must have ade-
initiated by the manufacturers, requested by the quate facilities and equipment, supervisory and
FDA,or ordered by the FDAunder statutory au- technical personnel with training and experi-
thority. The FDA classifiesrecalls by severity." ence appropriate to the complexity of testing, a
Recalls are classified as Class I, II, or III. Most quality management system (see Chapter 1),
blood component recalls are in Class III, not and successful ongoing performance in CMS-
likely to cause adverse health consequences. approved PT.3o All laboratories must register
Class II recalls are for products that may cause with CMS, submit to inspection by CMS or one
C HAP T E R 3 Regulatory Considerationsin TransfusionMedicine and Cellular Therapies 85
of its "deemed status" partners, and obtain re- immunology.requirements. CMS has also pub-
certification every 2 years. lished.guidance on conducting surveys (inspec-
All laboratory tests are rated for complexity ttons)."
by the FDAfor CMS as waived or moderate or CMS has approved six laboratory accredita-
high complexity. Waived tests are simple and tion organizations with requirements that meet
easilyperformed with limited technicai training. CMS regulations: AABB,the American Osteo-
Examples include over-the-counter tests, urinal- pathic Association, the American Society for
ysis dipsticks, copper sulfate specific-gravityhe- Histocompatibilityand Immunogenetics (ASHI),
moglobin screens, microhematocrits, and some the. College of American Pathologists (CAP),
simple devices for measuring hemoglobin. Labo- COLA(formerlythe Commission on Office Lab-
ratories that perform only waived tests register oratory Accreditation), and The Joint Commis-
with CMS for a certificate of waiver. The Cen- sion." The Joint Commission has cooperative
ters for Disease Control and Prevention provides agreements with ASHI, CAP, and COLA to ac-
technical and advisory support to CMS for labo- cept their laboratory accreditations in facility
ratory regulation and has published good labora- surveys." CMS may perform its own follow-up
tory practice recommendations for waived- surveys to validate those of the accreditation or-
testing sites." ganizations.
Nonwaived tests are classified as being of CMS requires successfulPT for ongoing labo-
moderate or high complexity based on a scoring ratory certification of nonwaived testing. Within
system of needs for training, preparation, inter- each laboratory section, CMSregulations specify
pretive judgment, and other factors (42 CFR tests and procedures (regulated analytes) that
493.17).28The "Medical Devices" section of the must pass approved PT if the laboratory per-
FDA website provides a searchable CLIA data- forms them. The CMS website has a list of ap-
base that provides the complexity levels of spe- proved PT providers." (PTis discussed in Chap-
cific tests." Compatibility testing with manual ter 1.) CMS can remove certification or impose
reagents and infectious disease testing are gener- fines for failure to complywith its regulations.
ally considered high-complexitytesting.
Blood banks and transfusion services have
three pathways to obtain a CLIA certificate to
perform testing: 1) certificate of compliance: ap-
proval via state health department inspections
using CMS requirements; 2) certificate of
accreditation: approval via a CMS-approved ac- Facilitiesalso should be familiarwith all relevant
crediting organization; and 3) CMS-exempt state and local laws and regulations, including
status: licensure programs for nonwaived labora- professional licensure requirements for medical
tories in New York and Washington states that and laboratory personnel, as many states have
are accepted by CMS.33 regulations that apply to blood banks and trans-
The CLIA regulations delineate general re- fusion services. Furthermore, in some situa-
quirements for facilities;quality systems, includ- tions, facilitiesproviding products or services in
ing quality assurance and quality control sys- other states must comply with local regulations
tems; and management and technical personnel in the customer's location.
qualifications. High-complexity tests require CMS approves hospitals for Medicare reim-
more stringent personnel qualifications. Immu- burseIilent through state surveys or (lccredita-
nohematology laboratories have standards for tion programs from The Joint Commission, the
blood supply agreements, compatibility testing, American Osteopathic Association, and DNV
blood storage and alarms, sample retention, pes- Healthcare. These inspections cover CMS
itive identification of blood product recipients, regulations for blood administration and the
investigation of transfusion reactions, and docu- evaluation of transfusion reactions found within
mentation (42 CFR493.1103 and 493.1271).28 "Basic Hospital Functions" regulations [42 CFR
Viral and syphilis serologic tests are part of the 482.23(c)].38The Joint Commission has stan-
dards for preventing misidentification of labo- sion, or transferinto a human reciptent.f HCT/Ps
ratory specimens and transfusion recipients can be derived from deceased or living donors
(National Patient Safety Goal section- (Table 3-3). The FDAhas established a compre-
NPSG.O1.01.01, .01.03.01), checking blood hensive, tiered, risk-basedregulatory framework
products in the "Universal Protocol" preproce- applicable to HCT/Ps. These regulations, which
dure verification process ("time-out"- were published in three parts (referred to as the
UP.01.01.01), and assessing transfusion appro- "tissue rules") and contained in 21 CFR Part
priateness (MS.05.01.01).39The Joint Commis- 1271, became fully effective on May 25,2005.
sion also addresses utilization review of blood They apply to all HCT/Ps, including HPCs, that
components in the Performance Improvement were recovered on or after this date.44,45
(PI)section of the hospital accreditation require- Under this tiered, risk-based regulatory
ments. Furthermore, in the same section of stan- framework, some HCT/Ps (referred to as "361"
dards, The Joint Commission directs hospitals to HCTIPs) are regulated solely under section 361
collect data on all reported and confirmed trans- of the PHSAct (42 USC 264), which authorizes
fusion reactions, and directs that these areas the FDA to establish and enforce regulations
should "be measured regularly." The Joint Com- necessary to prevent the lntroduction, transmis-
mission includes hemolytic transfusion reactions sion, or spread of communicable diseases." For
in its Sentinel Events reporting program." an HCT/P to be regulated solely under section
Both AABBand CAP have developed stan- 361 of the PHS Act and the regulations in 21
dards for transfusion services. The AABBStan- CFRPart 1271, it must meet all of the following
dards jar Blood Banks and Trensfusion Ser- criteria in 21 CFR 1271.10(a):
vices is updated every 2 years." The CAP
transfusion medicine checklist TRM42is updated 1. The HCT/P is minimally manipulated (re-
annually. AABB-and CAP-accredited facilities lates to the extent of processing).
need to be physically surveyed every 2 years to 2. The HCT/P is intended for homologous use
receive reaccreditation. AABBand CAP can co- only as reflected by the labeling, advertising,
ordinate joint surveys of facilities seeking both or other indications of the manufacturer's
types of accreditation. objective intent. (The product performs the
same basic function or functions in the donor
HUMAN TISSUES,
g
as in the recipient.)
AND C LL RAND TISSU ~ 3. The manufacture of the HCT/P does not
involve the combination of the cells or tissues
BASED (HCT/P )
with another article, except for water, crystal-
HCT/Ps are defined as articles containing or loids, or a sterilizing, preserving, or storage
consisting of human cells or tissues that are in- agent, provided that the addition of water,
tended for implantation, transplantation, infu- crystalloids, or the sterillztng, preserving, or
TABLE 3-3. Examples of HCT/Ps
'IiSkin ,. Hematopoietic stem/progenitor cells from peripheral or cord blood

Dura mater
ill! '" Other cellular therapy products (eg, pancreatic islets, mesenchymal
• Cardiovascular tissues stem/stromal cells, fibroblasts)
It Ocular tissues @ Reproductive cells and tissues
It Musculoskeletal tissues
storage agent does not raise new clinical layed implementation. For clarification on regu-
safety concerns with respect to the HCT/P. latory expectations for specific uses, it may be
4. The HCT/Peither does not have a systemic prudent to contact the agency directly. Minimal-
effect and is not dependent on the metabolic ly manipulated, unrelated umbilical cord blood
activity of living cells for its primary function, intended for hematopoietic or immunologic re-
constitution in patients with disorders affecting
or has a systemic effect or is dependent on the
the hematopoietic system must be FDA licensed
metabolic activity of livingcells for its primary
or used under an IND protocol. Minimally ma-
function, and is for autologoususe; for alloge- nipulated marrow that is not combined with an-
neic use in a first-degree or second-degree other article (with some exceptions) and is in-
blood relative; or for reproductive use. tended for homologous use is not considered an
HCT/P.
Manufacturers of 361 HCTIPs must comply FDAregulations in 21 CFRPart 1271 require
with the requirements in 21 CFR Part 1271, HCTIP manufacturers to have a tracking and la-
which include 1) establishment registration and beling system that allows for tracking each prod-
product listing; 2) donor eligibility, including uct from the donor to the recipient and from the
screening and testing for relevant communica- recipient back to the donor. HCTIP manufactur-
ble disease agents or diseases; and 3) current ers are also required to inform the facilities that
good tissue practice (cGTP);and are not subject receive the products that they have established
to the requirements for premarket review and a tracking system. However, FDA's regulations
approval. FDA guidance documents related to for HCTIPs, including the requirements for
these requirements can be found on the agen- tracking, do not apply to facilities that receive,
cy's website." store, and administer cells or tissues but do not
If an HCT/P does not meet the criteria set perform any manufacturing steps. The Joint
out in 21 CFR l271.l0(a), and the establish- Commission has hospital standards for
ment that manufactures the HCT/P does not receiving, handling, and tracing tissues and in-
qualify for any of the exceptions in 21 CFR vestigating adverse events (TS.03.01.01 to
1271.15, the HCT/P will be regulated as a TS.03.03.01).39(See Chapter 28 for information
drug, device, and/or biological product under on these standards.)
the FD&CAct and/or section 351 of the PHS The Health Resources and Services Adminis-
Act (referred to as a "351" HCT/P) and applica- tration (HRSA)within DHHS oversees the CW
ble regulations, including 21 CFR 1271. For BillYoung Cell Transplantation Program and the
these HCT/Ps, premarket review and FDA ap- National Cord Blood Inventory for marrow and
proval will be required. During the development cord blood donations and transplant procedures
phase, an Investigational New Drug (IND) or coordinated by the National Marrow Donor Pro-
Investigational Device Exemption (IDE)applica- gram (NMDP)inthe United States.
tion must be submitted to the FDA before The Circular of Injormation for the Use of
studies involving humans are initiated. Manu- Cellular Therapy Products is jointly written by
facturers are required to comply with the regula- AABB and multiple organizations involved in
tions in 21 CFR 1271 and all the regulations for cellular therapy for users. of certain minimally
drugs, devices, or biologicalproducts, as applica- manipulated unlicensed cellular therapy prod-
ble (Table 3-4). ucts." AABB and the Foundation for the Ac-
Peripheral blood stem cells (PBSCs),or cord creditation of Cellular Therapy (FACT)set vol-
blood for autologous use or for allogeneic use in untary standards covering the collection,
a first- or second-degree blood relative, that processing, and administration of cellular thera-
meet all the other criteria in 21 CFR 1271.10(a) py products.48,49AABBand FACThave standards
are regulated as 361 HCTIPs. PBSCsfrom unre- review cycles of 2 and 3 years, respectively. (See
lated donors are regulated as 351 products; Table 3-5.) The CAP transfusion medicine
however, for some clinical indications of PBSCs, checklistf includes cellular therapy require-
these regulations remain under a period of de- ments. The World Marrow Donor Association
TABLE 3-4. US Regulations for Manufacturers of Hematopoietic Progenitor Cells
Key Regulations FDA Premarket .

Ol(ersig~fJRegulatory (21 CFRexcept as Licensure, Approval,
TYpeotHPC Product Category noted) or Clearance?
Minimally manipulated mar- Health Resources 42 US Code274(k) Not applicable

row, not combined with and ServicesAdmin-
another article (with some istration oversight
exceptions) and for homolo-
gous use
Autologous or allogeneic PHSAct Section 1271.10(a)t (must No

related-donor (first- or 361: HCT/Ps* meet all criteria);
second-degreeblood relative) 1271 Subparts A-F
HPCs
Minimally manipulated PHSAct Sections 1271 Subparts A-D Delayedimplementa-

unrelated-donor peripheral 361 and 351: HCT/Ps Applicablebiologics/ tion
blood HPCs,not combined regulated as drugs drug regulations
with another article (with and/or biological
some exceptions) and for products
homologous use
Minimally manipulated PHSSections 361 1271 Subparts A-D Yes (after October20,
unrelated-donor umbilical and 351: HCT/Ps 2011): BLA or IND
cord blood cells regulated as drugs application
and/or biological
products
HPCsthat don't meet all the PHS Sections 361 1271 Subparts A-D Yes: IND and BLA
criteria in 21 CFR1271.10(a) and 351: HCT/Ps Applicable drugs/
regulated as drugs biologics regula-
and/or biological tions
products
*As defined by 2005 tissue regulations [21 CFR1271.3(d)J.

t21 CFR1271.10(a) as applied to PHSAct Section 361 requiresthat HPCsbe: 1) minimally manipulated,2) for homologous
use only, 3) not combined with another article (exceptwater, crystalloids, or sterilizing, preserving, or storage agents with
no new safety concerns), and 4) for autologous use or for allogeneic use in a first- or second-degreeblood relative.(Seefull
rule for detalls.)
HPC = hematopoietic progenitor cell; CFR = Code of FederalRegulations; FDA = Food and Drug Administration; PHS =
Public Health Service; HCT/Ps = human cells, tissues, and cellular and tissue-based products; BLA = Biologics License
Application; IND = Investigational New Drug.
(WMDA) fosters international collaboration to nor registries that follow its global standards
facilitate the exchange of high-quality covering all aspects of unrelated hematopoietic
hematopoietic stem cells for clinical transplanta- stem cell registry operations. The NMDP stan-
tion worldwide and to promote the interests of dards set forth basic guidelines and require-
donors. WMDA also accredits and qualifies do- ments for programs working with the NMDP.
TABLE3·5. Cellular Therapy Accreditation
AABB 2 years
FACT-JACIE(Foundation for the Accreditation of 3 years

Cellular Therapy and.the Joint Accreditation
Committee of ISCT and EBMT)
National Marrow Donor Program (NMDP) 2 years
World Marrow Donor Association (WMDA) 5 years
College of American Pathologists (CAP) Not set (publishes updated checklist annually)
The standards encompass network participation FDA's risk. evaluation and mitigation strategy
criteria with requirements for transplant cen- (REMS).51Several months later, the FDA ap-
ters, recruitment centers, and product collection proved a second gene therapy, Yescarta (axi-
centers. The NMDP standards are designed to cabtagene. ciloleucel), for. the treatment of re-
ensure that donors and. patients receive high- lapsed/refractory large Bcell lymphoma in
quality care and that government standards are adults with REMS. The FDAAmendments Act
met. (FDAAA)in 2007 gave FDAthe authority to re-
The Alliance for Harmonisation of Cellular quire REMSfrom drug manufacturers to ensure
Therapy Accreditation (AHCTA),which is under that the benefits of a drug or biological product
the umbrella of the Worldwide Network for outweigh its risks. REMSis a safety strategy to
Blood and Marrow Transplantation (WBMT), manage a known or potential serious risk associ-
encompasses all the above-mentioned accredita- ated with a drug and to enable patients to have
tion organizations. AHCTAis working to harmo- continued access to such drugs by managing
nize standards that cover all aspects of the pro- their safe use."
cess, from assessment of donor eligibility to Following the approval of the first gene ther-
transplantation and clinical outcome for hema- apy, FACT published standards for immune ef-
topoietic stem cells and related cellular thera- fector cells (IECs)in March 2018. The main ob-
pies. AHCTA provides helpful documents to jective of these standards is to promote quality
navigate the different sets of participating orga- practice in IEC administration. IECs include ge-
nizations' standards. Moreover, crosswalk docu- netically engineered cells and therapeutic vac-
ments comparing the different sets of cellular cines made from dendritic cells, natural killer
therapy standards were created and are avail- cells, T cells, and B cells and are used to modu-
able on the WBMT webslte." late an immune response for therapeutic in-
tent." The IEC Standards and Accreditation
u R Program was created by a task force represent-
ing FACT, the International Society for Cellular
The FDA approved the first gene therapy in the Therapy (ISCT),the American Society for Gene
United States in 2017. Kymriah (tisagenlecleu- and Cellular Therapy (ASGCT),and the Society
cell, indicated for pediatric and young adult pa- for Immunotherapy of Cancer (SITC) leader-
tients with relapsed/refractory acute lympho- ship, as well as academicians and cellular thera-
blastic leukernla (ALL),was approved using the pists from 10 cancer centers.54
KEY POINTS
1. ;e~~~~~ ~~~~~~~:: :~~~~:ti:do~::;~:a~~~PY are highly regulated, involving multiple

2. The FDA regulates biological products, including blood and blood components, HCTIPs, and
related devices through established laws and regulations. In addition to the legally binding reg-
ulations, the FDA periodically publishes recommendations in guidance documents. The FDA
;:~~~e provides links to blood- and HCT/P-related regulations and relevant guidance docu-
3. Blood establishments and device manufacturers must register their manufacturing facilities and
list the products they manufacture. Some blood establishments (eg, transfusion services that do
not collect or process blood and blood components) are exempt from registration but must be
CLlA certified.
4. Blood establishments that manufacture or participate in the manufacture of blood and blood
components are inspected by the FDAto determine compliance with regulations. Observations
~~~.~:~::~~~~r:~~~fe~~:i~~~~c~:~;t ~~~~~ ~ a;;~~~!~;. its response and correc-
S. The FDA requires drug (and blood) manufacturers to conduct recalls or market withdrawals
;~~~:~nr~~:~~~ce is found after products are distributed, such as when postdonation infor-
6. CMS regulates all US medical laboratories under CLlA. CLlA regulations establish require-
ments for certification. This includes the use of adequate facilities, qualified personnel com-
mensurate with the complexity of testing, and ongoing successful performance in proficiency
~~~!~~
~~~~~:;r~~:~ :~~~~~~~~~~:iia~~~~o~~:eC~~~ re~~~~:;t~spections per-
7. Health-care facilities also have CMS regulations for their activities, and The Joint Commission
and other organizations accredit many hospitals for CMS compliance. CMS and The Joint
~~~~;~~~~o~~~~~~!~:e:~~ :~s~:~~!~~s~anSfusion practices, evaluating adverse trans-
8. HCTIPs are regulated by the FDAunder a tiered risk-based framework. The FDAwebsite pro-
vides links to HCT/P-related regulations and relevant guidance documents.
REFERENCES
1. Office of the Law Revision Council. United 4. Food and Drug Administration. Blood and blood
States Code. Washington, DC: US House of products. Silver Spring, MD: CBER Office of
Representatives, 2019. [Available at http://us Communication, Outreach, and Development,
code.house.gov/.] 2019. [Available at https:llwww.fda.gov/vac
2. Food and Drug Administration. Frequently cines-blood-biologics/blood-blood-products.]
asked questions about therapeutic biological 5. Foodand DrugAdministration.Officeof Regulatory
products. Silver Spring, MD: FDA, 2015. [Avail- Affairs.SilverSpring,MD: FDA,2019. [Availableat
able at https:llwww.fda.gov/drugs/therapeu https:llwww.fda.gov/about-fda/office-global-
tic-biologics-applications-bla/frequently-asked- regulatory-operations-and-policyloffice-regulato
questions-about-therapeutic-biological-prod ry-affairs.]
ucts.] 6. Electronic code of federai regulations. Washing-
3. Foodand DrugAdministration.FederalFood,Drug, ton, DC: US Government Publishing Office,
and CosmeticAct (FD&CAct). SilverSpring,MD: 2019. [Availableat https:llgov.ecfr.io/cgl-bin/
FDA,2018. [Availableat https:llwww.fda.gov/ ECFR.]
regulatory-information/laws-enforced-fda/fed 7. Food and Drug Administration. Requirements
eral-tood-drug-and-cosmenc-act-fdc-act.] for blood and blood components intended for
transfusion or for further manufacturing use; Silver Spring, MD: CDRH, 2017. [Available at
final rule. (May 22,2015) Fed Regist 2015; https://www.fda.gov/medical-devices/device-
80:29841-906. [Available at https://www.fed registration-and-listing/search-registration-and-
eralregister.gov /artic1es/20 15/05/22/20 15- listing.J
12228/requirements-for-blood-and-blood-com 17. Code of federal regulations. Title 21, CFR Part
ponents-intended-for-transfusion-or-for-further- 803. Washington, DC: US Government Publish-
manufacturing. J ing Office, 2019 (revised annually).
8. Foodand DrugAdministration.Exceptionsand alter- 18. Food and Drug Administration. Mandatory re-
nativeproceduresapprovedunder 21 CFR640.120. porting requirements: Manufacturers, import-
SilverSpring,MD: CBEROfficeof Communication, ers and device user facilities. Silver Spring, MD:
Outreach, and Development, 2019. [Availableat CDRH, 2019. [Available at https://www.fda.
https://www.fda.gov/vaccines-b1ood-biolog gov/medical-devices/postmarket-requirements-
ics/ regulation -blood-supply / exceptions-and- devices/mandatory-reporting-requirements-
alternative-procedures-approved-under-21-cfr- manufacturers-importers-and-device-user-facili
640120.J ties#1.J
9. Food and Drug Administration. Guidance, com- 19. Food and Drug Administration. What does FDA
pliance and regulatory information (biologics). inspect? Silver Spring, MD: FDA, 2018. [Avail-
Silver Spring, MD: CBEROffice of Communica- able at https://www.fda.gov/about-fda/fda-
tion, Outreach, and Development, 2019. [Avail- basics/what-does-fdadnspect.]
able at https://www.fda.gov/vaccines-blood- 20. Food and Drug Administration. Blood and blood
biologics/guidance-compliance-regulatory-infor components. Inspection of licensed and unli-
mation-biologics.J censed blood banks, brokers, reference laborato-
10. Code of federal regulations. Title 21, CFR Part ries, and contractors-7342.001. In: Compli-
607.3. Washington, DC: US Government Pub- ance Program guidance manual. Silver Spring,
lishing Office, 2019 (revised annually). MD: CBER Office of Compliance and Biologics
11. Code of federal regulations. Title 21, CFR Part Quality, 2016. [Avallable at https://www.fda.
607.65. Washington, DC: US Government Pub- gov/medial84887/download.J
lishing Office, 2019 (revised annually). 21. Food and Drug Administration. Inspections,
12. MOU 225-80-4000. Memorandum of under- compliance, enforcement, and criminal investi-
standing between the Health Care Financing gations. Regulatory procedures manual. Silver
Administration and the Food and Drug Adminis- Spring, MD: FDA, 2017. [Availableat https://
tration. (June 6, 1983) Silver Spring, MD: FDA, www.fda.gov/inspections-compliance-enforce
1983. [Available at https://www.fda.gov/ ment -and-cnminal- investigations/ compliance-
about-fdaldomestic-mous/mou-225-80-4000.J manuals/regulatory-procedures- manual#_top. J
13. Code of federal regulations. Title 21, CFR Part 22. Code of federal reguiations. Title 21, CFR Part
601.12. Changes to an approved application. 606.171. Washington, DC: US Government
Washington, DC: US Government Publishing Publishing Office, 2019 (revised annually).
Office, 2019 (revised annually). 23. Food and Drug Administration. Biologicalprod-
14. Food and Drug Administration. Guidance for in- uct deviation reports annual summaries. Silver
dustry: Changes to an approved application: Bio- Spring, MD: CBER Office of Communication,
logical products: Human blood and blood com- Outreach, and Development, 2019. [Available
ponents intended for transfusion or for further at https://www.fda.gov/vaccines-blood-biolog
manufacture. (December 2014) Silver Spring, ics/report-problem-center-biologics-evaluation-
MD: CBER Office of Communication, Out- research/biological-product-deviation-reports-
reach, and Development, 2014. [Available at annual-summaries]
https://www.fda.gov/media/86137/down 24. Food and Drug Administration. Recalls, market
load.J withdrawals, and safety alerts. Silver Spring,
15. Food and Drug Administration. Classify your MD: FDA, 2019. [Available at https://www.
medical device. Silver Spring, MD: CDRH, fda. gov/ safety / recalls-market-withdrawals-
2018. [Avallableat https://www.fda.gov/medi safety-alerts.J
cal-devices/ overview-device-regulation/ classify- 25. Food and Drug Administration. Recalls (biolog-
your-medical-device.J ics). Silver Spring, MD: CBEROffice of Commu-
16. Food and Drug Administration. Device registra- nication, Outreach, and Development, 2019.
tion and listing: Search registration and listing. [Available at https://www.fda.gov/vaccines-
blood-biologics/safety-availability-biologics/re- 36. Laboratoryservices. Facts about the cooperative

calls-biologics.] accreditation initiative. Oakbrook Terrace, IL:
26. Food and Drug Administration. Enforcement re- The Joint Commission, 2019. [Available at
ports. Silver Spring,MD: FDA,2019. [Available http://www.join tcommission, org/facts_
at https://www.fda.gov/safety/recalls-market- about_the_cooperative_accreditation_initia
withdrawals-safety-alerts/enforcement-reports.] tive/ (accessedMarch 3,2020).]
27. Food and Drug Administration. Blood guidanc- 37. CLIA approved proficiency testing programs -
es. Silver Spring, MD: CBEROffice of Commu- 2019. Baltimore,MD: Centers forMedicare and
nication, Outreach, and Development, 2018. Medicaid Services, 2019. [Availableat https://
[Available at https://www.fda.gov/vaccines- www.cms.gov/Regulations-and-Guidance/Leg
blood-biologics/biologics-guidances/blood-guid islation/ CLW downloads/ptlist. pdf.]
ances.] 38. Code of federal regulations. Condition of partici-
28. Code of federal regulations. Laboratoryrequire- pation: Nursing services. Title 42, CFR Part
ments. Title 42, CFR Part 493. Washington, 482.23(c). Washington, DC: US Government
DC: US Government Publishing Office, 2019 PublishingOffice,2019 (revisedannually).
(revisedannuaily). 39. 2019 Hospital accreditation standards. Oak-
29. United States code. Certification of laboratories. brook Terrace, IL: The Joint Commission Re-
Title 42, USCPart 263a. sources, 2018.
30. Rauch CA, NicholsJH. Laboratoryaccreditation 40. Sentinel event. Oakbrook Terrace, IL:The Joint
and inspection. Clin LabMed 2007;27:845-58. Commission, 2017. [AVailableat https://www.
31. Howerton D, Anderson N, Bosse D, et al. Good jointcommission. org/ sen tinel_ even t_policy_
laboratory practices for waived testing sites: Sur- and_procedures/ (accessedMarch 3, 2020).]
vey findings from testing sites holding a certifi- 41. Gammon R, ed. Standards for blood banks and
cate of waiver under the Clinical Laboratory transfusion services. 32nd ed. Bethesda, MD:
Improvement Amendments of 1988 and recom- MBB,2020.
mendations for promoting quality testing. 42. College of American Pathologists, Commission
MMWR Recomm Rep 2005;54(RR-13):1-25. on LaboratoryAccreditation. Transfusion medi-
32. Food and Drug Administration. Medical device cine checklist. 2019 ed. Northfield, IL: CAP,
databases. Silver Spring, MD: CDRH, 2019. 2019.
[Available at https://www.fda.gov/medical- 43. Code of federal regulations. Title 21, CFR Part
devices/ device-advtce-comprehenstve-regulato 1271.3(d). Washington, DC: US Government
ry-assistance/medical-device-databases.] PublishingOffice,2019 (revisedannually).
33. Clinical Laboratory Improvement Amendments 44. Code of federal regulations. Title 21, CFR Part
(CLlA): How to obtain a CLlA certificate. 1271. Washington, DC: US Government Pub-
(March 2006) Baltimore, MD: Centers for lishingOffice,2019 (revisedannually).
Medicare and Medicaid Services, 2006. [Avail- 45. Food and Drug Administration. Tissue guidanc-
able at https://www.cms.gov/Regulations-and- es. Silver Spring, MD: CBEROffice of Commu-
Guidance/Legislation/CLlA/downloads/how nication, Outreach, and Development, 2019.
obtalncliacertificate.pdf.] [Available at https://www.fda.gov/vaccines-
34. Interpretive guidelines for laboratories. Appen- blood-biologics/biologics-guidances/tissue-guid
dix C. Survey procedures and interpretive ances.]
guidelines for laboratories and laboratory ser- 46. United States code. Regulationsto control com-
vices. Baltimore,MD: Centers for Medicare and municable diseases.Title 42, USC Part 264.
Medicaid Services, 2018. [Availableat https:// 47. MBB, America's Blood Centers, American Red
www.cms.gov/Regulations-and-Guidance/Leg Cross, American Society for Apheresis, Ameri-
islation/ CLlA/In terpretive _Guidelines_for _ can Society for Blood and Marrow Transplanta-
Laboratories.html.] tion, College of American Pathologists, Cord
35. List of approved accreditation organizations un- Blood Association, Foundation for the Accredi-
der the Clinical Laboratory Improvement tation of CellularTherapy, ICCBBA,Internation-
Amendments (CLIA).Baltimore, MD: Centers al Society for Cellular Therapy, Joint Accredita-
for Medicare and Medicaid Services, 2017. tion Committee of ISCT and EBMT, National
[Available at https://www.cms.gov/Regula Marrow Donor Program, World Marrow Donor
tions-and-Guidance/Legislation/CLW Accredi- Association. Circular of information for the use
tation_Organizations_and_Exempt_States.html.] of cellular therapy products. Bethesda,MD:AABB,
2018. [AVailableat http://www.aabb.org/aabb fda.gov/ news-events/ press-announcemen ts/

cctl coi/Documents/ CT-Circular-of-Informa fda-approval-brings-first-gene-therapy-united-
tion.pdf (accessedSeptember 14,2019).] states.]
48. HaspelRL,ed. Standardsfor cellulartherapy ser- 52. Food and Drug Administration. REMSintegra-
vices. 9th ed. Bethesda,MD: AABB,2019. tion initiative. Silver Spring, MD: FDA, 2018.
49. FACT-JACIEInternationaistandards for hemato- [Availableat https://www.fda.gov/industry/
poietic cellular therapy product collection, pro- prescription-drug-user-fee-amendments/rems-
cessingand administration.6th ed. Omaha, NE: integration-initiative.]
Foundation for the Accreditation of Cellular 53. Foundation for the Accreditation of Cellular
Therapy, 2018.
Therapy. FACTstandards for immune effector
50. Alliance for Harmonisation of Cellular Therapy
cells. 1st ed, version 1.1. Omaha, NE: FACT,
Accreditation. Comparison of cellular therapy
standards: Crosswalk documents. AHCTA, 2018. [Availableat http://www.factwebsite.
2016-2017. [Availableat https://www.wbmt. org/iecstandards/ (accessed September 14,
org/ahcta/documents/ (accessed September 2019).]
14,2019).] 54. Maus MV, Nikiforow S. The why, what, and
51. FDA News Release: FDA approval brings first how of the new FACTstandards for immune ef-
gene therapy to the United States. (August 30, fector cells. J Immunother Cancer 2017;5:36-
2017) Silver Spring, MD: Food and Drug Ad- 40.
ministration, 2017. [AVailableat https://www.
4
Natio.nal.Hemovigiiance: The Current
State
Kevin J. Land, MD; Barbee I. Whitaker, PhD; and Lynne Uhf, MD
HE CONCEPT OF VIGILANCE IN RELA- al levels, but also the analysisof shared data and
tion to donor and recipient safety within best practices. This is encouraging progress, for
AABBand its membership has existed for donor and recipient safety are global concerns
a number of years. 1·4 The term hemovigilence has and benefit most from lessons learned next door,
beendefinedas "[a]set of surveillance procedures acrossthe country, and around the world.
of the whole transfusion chain intended to mini- Hemovigilancestructures the data and subse-
mize adverse events or reactions in donors and quent analysis using standardized definitions
recipients and to promote safe and effective use and conventions, allowing data to be widely
of blood components." Hemovigilance, howev- shared and compared in order to identify and in-
er, will have limited utility if used only to moni- fluence best practices across institutions.
tor events rather than implement changes or if Hemovigilanceis the ultimate benchtop-to-bed-
data are shared and compared only within a sin- side collaboration where stakeholders (donors,
gie institution. Also important are incidents and recipients, researchers, policy makers, blood es-
mistakes (errors) that do not result in harm tablishments, and hospitals) share data and
(near-misses). Identification of these can en- ideas, design interventions to address gaps and
hance awareness of potential risks and lead to a failures, then implement potential solutions,
safer environment for transfusion. evaluate the results, and further refine hypothe-
As implied by the phrase "the whole transfu- ses based on real-worlddata.
sion chain," hemovigilanceis considered to be a Within the United States, public and private
national-level activity and has been since incep- organizations have been encouraged to collabo-
tion, when countries were strugglingwith the rate to improve donor and patient outcomes. Al-
(primarily infectious) complications of blood though much work remains toward a compre-
component therapy. Today, hemovigilance in- hensive hemovigilance system, the combined
creasingly includes not merely the collection of efforts of governmental, public, and private or-
data acrossrnanyinstitutions and geopoliticalen- ganizations have helped close the gap between
tities at state, regional, national, and internation- US hemovigilance and that of other nations.
Kevin J. Land, MD, Adjunct Professor of Pathology, UT Health Science Center at San Antonio, San Antonio,
Texas, and Vice President of Clinical Services, Vitalant, Tempe, Arizona; Barbee 1. Whitaker, PhD, Lead General
Health Scientist, Office ofBiostatistics and Epidemiology, Center for BiologicsEvaluation and Research, US Food
and Drug Administration, Silver Spring, Maryland; and Lynne Uhl, MD, Vice Chair for Laboratory and Transfu-
sion Medicine, Beth Israel Deaconess Medical Center, Division Director, Laboratory and Transfusion Medicine,
Beth Israel Deaconess Medical Center, and Associate Professor of Pathology, Harvard Medical School, Boston,
Massachusetts
The authors have disclosed no conflicts of interest. This chapter reflects the views of the authors and should not
be construed to represent their employers' views or policles,
9S
This chapter focuses on what is being done in diseasesat a national level.8 In 1994, France be-
other countries and what is being done with the came the first European country to develop a
increased transparency and collaboration across formal national hemovigilance system in re-
the United States to improve donor and recipi- sponse to HN transfusion transmissions in that
ent outcomes. country. Since then, many other hemovigilance
systems have been developed and regularly pro-
vide annual reports. Referencesat the end of the
I NAL chapter may be consulted for a more thorough
ILAN review of international hemovigilance efforts or
a treatise on how to set up a national hemovigl-
Hemovigilance has more than two decades of lance system.Y
history in the international arena, with so many The Serious Hazards of Transfusion (SHOT)
programs being established as a consequence of reporting system in the United Kingdomis con-
concern over transfusion-transmittedviral infec- sidered a very successfulhemovigilance system.
tions and their sequelae leg, transfusion-trans- ABshown in Fig 4-1, the increased involvement
mitted human immunodeficiency virus (HN), by stakeholders and the maturity of the program
hepatitis B, and hepatitis C]. The architecture resulted in more reports but fewer fatalities.The
and oversight of these hemovigilance programs most notable achievement of the SHOT pro-
vary widely, with management and control by gram is reporting of the association of plasma
various organizations, including blood establish- and platelet transfusions derived from female
ments, governmental regulators, national medi- donors and the incidence of transfusion-related
cal societies, and public health agencles." acute lung injury (TRALI)in the early 2000s9
Hemovigilance has been implemented on a that led to changes in the management of donor
country-by-country basis, with early adopters collections and product manufacturing around
now having the more robust systems, as expect- the world.'? Equally important are the SHOT
ed with experience. In January of 1993, the Jap- data on lapses in safe transfusion practice, in-
anese Red Cross Society began aggregating including inadequate patient identification at the
formation on adverse reactions and infectious time of specimen acquisition and blood product
2000 _Total no. of reports analysed 14
1800 12
_Death definitely attributed to
1600 transfusion
~0 1400 10
0
~ 1200
....0 1000
8 ~
-e
'0
...
(1) 6 _g
.c 800 E
E z=
::::I 600 -
z
400
FIGURE 4-1. Total reports submitted to the UK's Serious Hazards of Transfusion program between
1996 and 2018, and total deaths determined as definitely caused by transfusion. (Graphic provided
courtesy of Oebbi Poles and Shruthi Narayan from SHOT.)
C HAP T E R 4 Hemovigilance 97
administration. 11 These data have prompted surveillance from clinical data is important. Al-
public campaigns engaging patients to actively though the two types of data are very similar,
participate in their care as a means to mitigate clinical decislon-maklngasks and answers funda-
risks of misidentification and transfusion errors. mentally different questions about an individu-
More resources can be found on the Joint UK al's health and subsequent care. Surveillancedata
Blood Transfusion and Tissue Transplantation focus more on trending summary data that may
Services Professional Advisory Committee web- or may not be ultimately classifiedin a manner
site." Increasingly, other countries such as Aus- suited for the clinician-patientrelationship.
tralia, are providing free and comprehensive Eventually, as hemovigilance programs have
resources to the international transfusion medi- evolved, it has become increasingly clear that
cine community on their websites. proper analysisof hemovigllance events requires
In 1998, those practicing hemovigilance in standardization of data structures and harmoni-
Europe established the European Haemovlgl- zation of terminology with simple and objective
lance Network (EHN) to bring practitioners to- definitions that lend themselves to comparison
gether to exchange ideas for improving patient (eg, expressing adverse events per transfusion or
safety and hemovigilance reporting. Eventually, per 1000 donations, etc), as well as extensive
national hemovigilance programs were required use of subject matter experts to aid in interpreta-
by the 2002 and 2005 European Blood Direc- tion and validation of the data.F:"
tives (2002/98/EC, 2005/611EC)/3,14 which
mandated implementation of hemovigilance
systems with a set of minimum common donor us HEMO\HGI NeE
and recipient data elements in all European
Union (EU)member states, and subsequent re- In 2009, US hemovigilance was described as a
porting of results to EU authorities. The EHN "patchwork of reporting processes"19with signif-
was reinvented as the International Haemovlgi- icant but limited programs to collect specific do-
lance Network (IHN) in 2009, when the inter- nor and recipient data and where only fatal
est and membership of non-Ell countries made transfusion reactions, donation-related deaths,
it clear that the desire to develop robust hemo- and product deviations were reported at a na-
vigilance systems and share critical experiences tionallevel to the US Food and Drug Administra-
was truly global. Information on hemovigilance tion (FDA).These programs capture a subset of
systems participating in the IHN is available hemovigilance data, but do not comprise a com-
(https://www.ihn-org.com/membership/cur plete system meeting all the criteria for "mature"
rent-members/). International efforts for hemovigilance. In 2003, the FDA took initial
hemovigilance data- and knowledge-sharing in- steps toward mandating more comprehensive
clude, but are not limited to: the International US hemovigilance through the proposed rule
Surveillance of Transfusion-Associated Reac- "Safety Reporting Requirements for Human
tions and Events (ISTARE)database from IHN15 Drug and BiologicalProducts"; however, at the
and Project Notify Library (http://www.noti time of this writing, this remains in draft status."
fylibrary.org) from the World Health Organiza- Hospitals and transfusion services are re-
tion (WHO).16 quired to perform an investigation of all serious
The historical focus of hemovigilancesystems adverse reactions associated with blood transfu-
has been on hospital surveillanceand the report- sion and to report complications that may be re-
ing of transfusion-associatedadverse events into a lated to the blood donor or to the manufacture
central data repository. Interest in donor hemo- of the blood components to the collection facili-
vigilance is more recent, enhanced with studies ty [Title21, Code of Federal Regulations (CFR),
showing higher reaction rates among young do- Part 606.170]. Transfusion- and donation-relat-
nors.":" The goal of hemovigilance systems ed fatalitiesmust also be reported directly to the
should be acquiring and trending high-quality, FDAby the transfusing or collecting facility.The
representative, and validated surveillance data, FDApublishes an annual report of the reported
not on capturing all potential data. Separating fatalities (https:l/www.fda.gov/vaccines-blood-
o
N
00
rl
N
U)
:::;)
c...:>
co
6
>-
Q)
2:
::::I
o:
c::
0
~
,!:::!
o :;::::;
rl :::;)
'0
c::
00
'"
c::
0
~
~
'0
c...:>
'0
0
0
05
(ij
c::
0
~
2:
c::
'0
"I', Q)
0 t:::
0
I:: V) I:: U I:: I:: 1/) co I:: (lJ 0..
a
'.j:;
I::
a
a
'.j:;
co
a
a a
'.;::;
I::
a
'-
:::l a
'.j:;
I::
a
'.j:;
til
ro
e
(J)
+-' '';:; Q. (lJ
u '.j:; u I:: u u '- u u c::
co u co (J) co co u
ro :::::; (lJ (lJ .~ 0
,_
(J) co ,_
(J)
> (lJ (lJ
(J) Q. I:: 4- 4- <:J :;::::;
(lJ a L '- ,_ I:: I::
,_ a +-' U
I::
a u
'50 ,2
.._ tfl
I:: I::
.2
I::
a
U
(J)
+-'
u
on .,_,
(lJ
I::
a
,_
Vl I::
'.0
u
rc
rc
'-
ro
';::
til
a
_c
'"
Q)
'-
V)
:::::; :::::;
V)
:::::;
tfl
::l
ro '-
(lJ :::::; :::l a (lJ ':;: (lJ
I
Vl
Q)
(J)
(lJ '.j:; l- '-
4-
Vl
4-
Vl
4-
Vl
'l-
Vl
U
a U Vl U U +-'
U >I Q)
rc I:: co co I:: m I:: (lJ ro >
I::
co
,_ (lJ co
I::
co
I::
co
V)
Vl (lJ U co (lJ ,2 +-' _Q .t:
rn
'0
+-' +-'
co
.,_.
'- l-
+-'
I-
+-' cp
I::
'-
(lJ
(lJ
+-'
....
I-
.J-'
'-
I::
Vl
:::l
,t:
E u
(lJ
,_
on '"
'0
'- U U > ro Vl 4-
Vl
Q)
(J) '50 (lJ a (J) (lJ a .2 Vl
I:: I::
+-'
,t:
Ll ~
u a -l-' V) V1 o, Vl
ro (lJ
(j)
a 0 rn ::l
'-
I :::l ro >- E +-'
'-
'u 4- I:: 4- ....,
'- +-' Vl
ro I
EI '- a V)
a V)
I::
I
I:: I:: 'u c::
a (lJ II) I::
til a ro a 0
+-'
til til
coI
ro
'- :::l
(1J
l- I-
+->
Vl
til
'en
"0
I
I::
I- 4-
V)
+J Vl
::l
I
I::
roI .....
::::I
(J)
I:: 4-
a co VI a c: c::
:2: 'Vi
,_ I:: V) ,2 ~
:::::; I- ro ::l V) f-
4-
'- 4- :::l
V) .._.
(lJ l- V)
4- co-!
I:: V)
I::
co :::l ro C ~
u I-
ro w
l-
I- (lJ
+-'
« I- l-
I-
ex:
;::)
:::::; (::J
u
« u::
biologics/report-problem-center-biologics-evalua for a broader, more cohesive and coordinated

tion -research/transfusiondona tion -fatali ties). approach, termed btovigilsnce, that encom-
Mandatory reporting of product deviations by li- passed not only blood recipient and blood donor
censed manufacturers, unlicensed registered hemovigilance but also tissues and organs, and
blood establishments, and transfusion services cellular therapy (CT) components" MBB es-
has also been defined by the FDA (21 CFR tablished an Interorganizational Task Force on
606.171). These requirements will likely con- Biovigilance,inviting key governmental thought
tinue even if a more formal US national hemo- leaders and representatives from the private sec-
vigilance system is eventually required, because tor to initiate the development of a national bio-
hemovigilance reports are retrospective, where- vigilance program. Amid concerns of additional
as reporting to the FDA following an adverse oversight by government agencies, including a
event or for product deviations must occur in perception that such oversight could result in
near-real time, as outlined in the CFR. punitive actions that might interfere with pa-
There is no mandatory nationwide, systemat- tient care, hemovigilance was established in the
ic assessment of transfusion reactions that were United States within a framework of a public-
not fatal, nor of nonfatal donation-related reac- private partnership." Through the efforts of the
tions, in the United States. However, a few task force and an international working group of
investigator-led or collaborative research lnlta- advisors representing established hemovigi-
tives [American Red Cross (ARC),Vitalant Re- lance systems, recipient and donor hemovigi-
search Institute, and collaborations sponsored lance programs were designed based on the
by the National Heart, Lung, and Blood Institute important tenets of voluntary reporting: confi-
(NHLBI) such as the Retrovirus Epidemiology dentiality, a just culture (nonpunitive data analy-
Donor Study (REDS),REDS-II,and the Recipi- sis), and efficient data reporting with a focus on
ent Epidemiology and Donor Evaluation Study- improvements to patient and donor safety. The
III (REDS-III)]have generated extensive results task force was also critical in the development of
in the areas of emerging infectious diseases, case definitions and data requirements for the
noninfectious complications of transfusion, and national recipient hemovigilance system, de-
donation-related reactions. scribed below.
In parallel, Drs. Harold Kaplan and James
Battles developed the impactful medical event Elements of Biovigilance
(incident) reporting system for transfusion medi-
cine (MERS-TM) in hospitals in the mid- Efficient reporting for recipient hemovigilance
was achieved through leveraging an existing
1990s.22In addition, National Blood Collection
hospital event reporting system, the CDC's Na-
and Utilization Survey Reports have included
summary reports of transfusion reactions and se- tional Healthcare Safety Network (NHSN)
(see Recipient Hemovigilance, below). Donor
rious donation-related adverse events since
2007 (Fig4_2).21The launch of a voluntary sys- hemovigilance required building a new infra-
structure to capture national reporting by US
tem of hemovigilance in the United States has
blood establishments (see Blood Donor Hemo-
required collaboration among all stakeholders
and many organizations. Specific acknowledge- vigilance, below). Other areas essential to com-
ment is given to the strategic partnership, re- prehensive biovigilance in the United States in-
clude tissue and organ surveillance and cellular
sources, financial support, and guidance provid-
therapies (CT) biovigilance. CT adverse event
ed by the US Department of Health and Human
and outcome monitoring occurs through nation-
Services (DHHS), including the FDA and Cen-
al and international registries, including the
ters for Disease Control and Prevention (CDC).
Center for International Blood and Marrow
Transplant Research (CIBMTR), the National
Task oraBiovigihmce
Marrow Donor Program (NMDP), and the
In early 2006, a number of prominent individu- World Marrow Donor Association (WMDA).
als in transfusion medicine recognized the need In ongoing efforts to promote hemovigilance
ideals, AABB Standards for Blood Banks and ed with patient harm. As of 2018,359 transfu-
Transfusion Services (Standards) requires ac- sion services have enrolled in the NHSN hemo-
credited blood banks, transfusion services, and vigilance module, of which 133 are actively
blood centers to adopt nationally recognized reporting data to NHSN (KracalikI, CDC per"
classifications [or internationally recognized, sonal communication, April 8, 2020). At this
such as defined by the International Society of time, hemovigi1ance reporting into the CDC
Blood Transfusion (ISBT),for countries lacking NHSN requires manual data entry, although a
their own such classifications]for adverse reac- mechanism exists for reporting monthly denom-
tions associated with blood donation and the inators through clinical documentation archttec-
transfusion of blood and blood compo- ture (CDA).
nents"25(p89) Tools to integrate hospital reporting of prod"
uct-related adverse events, such as transfusion"
associated circulatory overload (TACO),TRALI,
R IPH N or suspected transfusion-transmitted infection
H (TTl),with blood center investigations can facili-
tate comprehensive understanding of adverse
CDC's NHSN is a secure, web-based surveil" reactions. An example of this is the recently
lance system" Originally developed to capture completed harmonized form for reporting ad"
data on hospital-acquired infections (HAI),it is verse transfusion reactions to blood suppliers,
now used by over 12,000 US health-care facili- developed by members of the AABBHemovlgl-
ties to report on a variety of infection-related lance Committee, Hemovigilance Education
and other patient safety issues." The hemovigi- and Ana1yticsCommittee, and Donor Hemovigi-
lance module of the NHSN, which began na- lance Working Group (see Appendix 4"1).28
tional data collection in 2010, allows hospitals
Complex case definitions for entities such as
to monitor their own transfusion activity as well
TACO and TRALIneed periodic reevaluation as
as to share data with external groups (govern"
new information and technology are used to dl-
mental and nongovernmental; described below)
agnose and treat transfusion complications.
at the individual hospital's discretion, for man"
datory or voluntary reporting actlvities." Such exercises need to reflect the potentially dlf
The NHSN modules were established as data ferent needs of the clinical, research, and sur"
repositories for surveillance of adverse events veillance communities. Ideally, changes to these
and reactions of interest. To be comparable be" definitions should be validated by each commu-
nity.29,30
tween hospitals, data must be reported using
standard definitions for all aspects of data ele- Collaborative efforts between the CDC and
ment reporting. Ideally, reporting thresholds state departments of public health continue to
would be as similar as possible between differ" promote statewide reporting into the NHSN
ent hospitals. The components of the NHSN hemovigilance network. For example, in 2014,
hemovigi1ancemodule and the terms and deflnl- the state of Massachusetts implemented manda-
tions employed were developed and thoroughly tory reporting into the NHSN hemovigilance
vetted by subject matter experts. The major module as a means to comply with state regula"
components of the module include: 1) demo" tory requirements on transfusion activity and
graphic and utilization data, which permit classi- transfusion-related adverse events.27,31State"
fication of facilitiesfor comparisons in aggregate wide participation provided insight into varying
data analyses; 2) reports on transfusion-related transfusion practices and utilization, and a more
adverse reactions in accordance with case deft- informed picture of transfusion-related adverse
nition criteria, as well as severity and imputablll- events has emerged, including events related to
ty grading (Table 4"1); and 3) incident reporting TTl and TACO.32,33 Additionally, statewide par"
(ie, mistakes, errors, or adverse events associat- ticipation affordsthe opportunity to track trends
ed with transfusion). Reporting of incidents is in practice changes (eg, mitigation strategies
not mandatory for participation unless asscciat- for TTl, adoption of new blood component
CHAPTER 4 Hemovigilance 101
,
,
en
::::I t::
't; e
--
t::_
~
Q)
s:
e.:b <U <=

<U :::l
..c:: 0 ..c:: 0
1-"0 I- .17)
offerings such as whole blood) by transfusion Transfusion Incident Reporting

services."
Hr. ipSi.qents.associated.:witil.trapsfu-
lVlistal<~.r
p~x~p()rted·ona.<ieta.iledlIlciq~I)tform
sioIlSCl.l:l,
or summarized on a monthly incidentsummary
CDC's NHSN Hemovigilance Module website report. The CDC requires detailed incident form
provides resources and training materials for fa- use for those incidents that result in a transfu-
cilities to get access and start reporting into the sion reaction. However, nonharmful incident
system (https://www.cdc.gov/nhsn/acute-care- review and more detailed reporting can be a
hospitallbio-hemo/setup.html). Once access to valuable activity if conducted regularly and
the module has been attained, the hospital must compared with other transfusion services that
complete the Annual Hospital Survey before any use the CDC standardized event codes that are
data submission may begin. This survey includes based on the adverse event coding system devel-
a significant number of data elements that de- oped in MERS-TM(Fig4-3).34
scribe the demographics of the hospital. Follow-
ing its completion, the hospital can submit data Using Recipient Data via NHSNGroup
using web forms to complete monthly reporting Function
plans and then report monthly denominators,
Hospitals participating in the NHSN Hemovigi-
adverse reactions, and incidents. Monthly de-
lance Module can share their data with external
nominator reports include the components
partners (groups) through the group function
transfused each month and are provided in the
built into the NHSN system. This eliminates the
month following the reporting month. Adverse
need to enter data into multiple systems. Al-
events (reactions and incidents) should be re-
though group users (managers) can access data
ported upon completion of their investigation.
from all participating hospitals, hospitals cannot
access any data from other hospitals in their
Transfusion-Related Adverse Reaction
group. Hospitalsand blood centers in Massachu-
Reporting
setts comply with the state's regulatory require-
Adverse reactions should be investigated and ment by sharing their data with the NHSN
categorized according to the CDC protocol" group maintained by the state (Fig 4-4). Data
into 12 different reaction types and coded ac- can also be shared with a NHSN group main-
cording to the degree to which each reaction tained by a Patient Safety Organization (PSO)
conforms to the surveillance definition (criteria), developed under the Patient Safety and Quality
the severity or grade of the reaction, and the re- Improvement Act of 2005 (PSQIA).35,36
action's imputability to the transfusion (Table 4-
1). Imputability is an important concept for
hemovigilance; it is the degree to which the re- N
action was caused by the transfusion. With the OVIGiLAN THE
January 2017 release of the NHSN Hemovigi- UNITED STATES
lance Module, CDC instituted a function to col-
lect specific data related to case imputability and Part of maintaining a safe and adequate blood
propose an imputability score. More recent de- supply involves vigilance over donor safety and
velopments [Ianuary 2018) allow the reporting well-being. Blood establishments must be the
organization the ability to indicate its agreement stewards of safety that donors expect them to
with that computer-based assessment. As the be. This commitment to the donor requires at-
formal definitions of imputability and severity tention to continuous improvement in donor re-
continue to mature, hemovigilance systems cruitment, donor management, the process of
around the world are beginning to incorporate blood collection, and donor safety; hence, the
them into their reports. need for a donor hemovigilancesystem.
Incident Codes
Note: lncldent codes are based on MERS TM (US) and TESS (Canada) incident classification schemes.
Product Check-In ProductlTest Request

(Transfusion Service) (Clinical Service)
Events that occur during the shipment and receipt of Events tnet occur when the clinical service orders
products into the transfusion service from the patientlesls or blood products for transfusion.
supplier, another hospital site, satellite storage, or PR 00 Detail not specified
Clinical area. PR 01 Order for wrong patient
PC 00 Detail not speolfled PR 02 Order Incompletely/incorrectly ordered (online
PC 01 Data entry incomplete/Incorrect/not performed order entry)
PC 02 Shipmenllncomplelelincorrecl PR 03 Special proces.slng needs not Indicated (e.q.,
PC 03 Products and paperwork do not match eMV negative, autologous)
PC 04 Shipped/transported under inappropriate PR 04 Order not done
conditions PR 05 Inappropriate/unnecessary (intended) test
PC 05 Inappropriate return to inventory ordered
PC 06 Product confirmation incorrect/not performed PR OS lnappropnate/unnecessary (intended) blood
PC 01 Administrative check not Incorrect/not product ordered
performed (record review/audit) PH 07 Incorrect (unintended) test ordered
PC 08 Product label incorrect/rnlsslnq PR 08 Incorrect (unintended) blood product ordered
Product Storage ProductlTest Order Entry

(Transfusion Service) (Transfusion Service)
Events thai occur during product storage by the Events that occur when the transfusion service
transfusion service. receives a patient order. This process may be
US 00 Delail not specified excluded if clinical service uses online ordering.
US 01 Incorrect storaqe condittons OE 00 Detail not specified
US 03 Inappropriate monitoring of storage device OE 01 Order entered for wrong patient
US 04 Unit stored on Incorrect shelf (e.g., OE 02 Order Incompletely/Incorrectly entered online
ABO/autologoJJs sldirecled) OE 03 Special processing needs not entered (e.g.,
US 05 Incorrect storage location CMV·, autologous)
OE 04 Order entry not done
Inventory Management OE 05 Inappropriate/unnecessary (intended) tesl
(Transfusion Service) order entered
Events that involve quality management of the blood OE 06 Inappropriate/unnecessary (Intended) blood
product inventory. product order entered
1M 00 Detail not specified OE 071noorrecl (unintended) test ordered
1M 01 Inventory audit mcorrect/not performed OE 08 Incorrect (unintended) blood product ordered
1M 02 Product status lncorrecUylnot updated online
(e.g., available/discarded) Sample Collection
1M 03 Supplier recall/traceback no! appropriately (Service collecling the samples)
addressed/not performed Events that occur during patient sample collection.
1M 04 Product order mcorrecuyrnot submitted to se 00 Detail not specified
supplier SC 01 Sample labeled with locorrect patient name
1M 05 Outdated product in available inventory se 02 Not labeled
1M 06 Recalled/quarantined product in available se 03 Wrong patient collected
Inventory SC 04 Collected in wrong lube type
se 05 Sample eNS
se 06 Sample hemolyzed
se 07 labellncomplete/Hlegiblefln,correct (other than
patient name)
se 08 Sample collected In error
SC 09 Rsqulsnton arrived without samples
se 10 Wristband incorrect/no! available
se 11 Sample contaminated
FIGURE 4-3. NHSN Hemovigilance Module incident codes. [Reprinted from NHSN Manual: Biovigilance
Component Protocol April 2018 (page 1 of 3).26]
FIGURE 4·4. Hemovigilance in the United States.

*Regulated by HHS Agency for Healthcare Research and Quality (AHRQ), authorized by the Patient
Safety and Quality Improvement Act of 2005 (PSQIA).
FDA = Food and Drug Administration; CDC = Centers for Disease Control and Prevention; NHSN =
National Healthcare Safety Network; PSO = patient safety organization; ARC = American Red Cross;
BloodSCAN = Blood Safety Continuous Active Surveillance Network.
Blood establishments collect a wealth of data staff on donor hydration and salt loading of do-
every day centered on donor-related activities. nors, improving the ability to predict donors at
They are driven to maximize the efficiency of risk for vasovagal reactions with loss of con-
collecting each donation type while also reduc- sciousness (especially off-site), and developing
ing the risk of donor harm. Blood-center-based strategies to decrease deferrals (eg, monitoring
donor hemovigilance systems and collaborative iron stores.and iron replacement) and increase
research have evolved primarily to capture ad- donor return rates. Blood establishments,
verse reaction rates and have been used in through a mature donor hemovigilance system,
donor safety studies to mitigate adverse use collected data not only to improve donor
events.34,37-39 Donor hemovigilance programs safety and satisfaction but also to inform deci-
within large blood systems have demonstrated sions about complex business and.operational is-
their utility in identifying and implementing sues.
measures that improve donation safety for
young donors.r':" Blood center data have also Background
been analyzed to help identify donation deter-
rents and motivators and to increase overall do- As with the recipient hemovigilance system in
nor satisfaction.41-42 Improvements in donor the United States, US donor hemovigilance was
safety can occur through a variety of methods, developed through the AABBInterorganization-
such as avoiding donors at high risk for vasova- al Task Force on Biovigilance.A donor hemovlg-
gal reactions (eg, potential donors with low esti- Hance working group was established with
mated blood volume), educating donors and representation from US blood establishments,
hospital-based blood collection programs, the others) have worked together to make well-
US Armed Services Blood Program, Canadian accepted and universal definitions, terms, and
Blood Services, the Plasma Protein Therapeutics ideas for donor hemovigilance. A formal revi-
Association (PPTA), and international liaisons sion group made up of representatives from the
from the ISBT and the International Haemogi- ISBT Hemovigilance Working Party, the IHN,
lance Network (IHN),in partnership with repre- and the AABB Donor Hemovigilance Working
sentatives from the US DHHS. The working Group collaborated to revise the 2008 ISBT
group was charged to develop and implement a standard for surveillance of complications relat-
national monitoring program on donor safety is- ed to blood donation. The group reconciled the
sues. Tasksincluded developinga set of common differencesbetween the AABBand ISBTsurveil-
definitions for US donor hemovigilance based lance terms and, in December 2014, published
on existing models and providing subject matter the first internationally harmonized AABB-ISBT
expertise for the development of software, fund- standard definitions for complications related to
ed by DHHS, that would gather donor data and blood donation. The Alliance of Blood Opera-
provide a systematic and standard mechanism tors, the European BloodAlliance, and the IHN
to calculate and compare rates. This yielded the have formally endorsed these definitions.40,45,48
Donor Hemovigilance Analysis and Reporting Additional validation of this standard has result-
Tool (DonorHART),which offereda robust set of ed in confirmation of its applicabilityand recom-
reporting and graphing tools designed to simplify mendations for the additional development of a
initial data reporting and aggregatedata analysis. harmonized structure for the severity and im-
Observed variation in reaction rates, reported putability assessments of donor reactions. Be-
to DonorHART,led to changes in practice in one tween January 2018 and June 2019, an interna-
blood center that likely reduced reaction rates tional hemovigilance working group involving
among donors younger than 30 years 01d.43,44 AABB,ISBT, and PPTA representatives devel-
Donor hemovigilance reports have been pub- oped and validated a Donor Adverse Reaction
lished based on data from the small number of Severity Grading Tool,49,50 which is designed to
participating blood collectors (Fig4-5 and Table be used along with the 2014 AABB-ISBTstan-
4_2).45-47These generallyconfirm findingsreport- dard definitions for complications related to
ed in the literature by large proprietary donor blood donation (see Appendix 4-2). The tool
hemovigilance programs (ARC,Vitalant). US do- provides a standardized, objective framework
nor hemovigilance is still active, with organiza- for grading severity and addresses the lack of
tions such as ARC and Vitalant Research Insti- consistency in severity grading seen during vali-
tute contributing significantly. However, use of dation of the 2014 definitions. This harmoniza-
the DonorHARTprogram ended September 25, tion of definitions facilitates easy national and
2018, as a result of system maintenance costs. international comparison between blood centers
Ongoing efforts to standardize donor hemovigi- and systems. The tool is patterned after Com-
lance continue, as evidenced by the recent re- mon Terminology Criteria for Adverse Events
quirement in AABB Standards to use standard- (CTCAE1),a well-known clinical severity scale.
ized classifications. In addition, there was an CTCAE1 rates severity from Grades 1 through
international project in which members from 5, roughly associated with mild, moderate, se-
the AABB Donor Hemovigilance Working vere, life-threatening, and fatal levels, respec-
Group along with ISBT and PPTA representa- tively."
tives collaborated to standardize the severity
grading of donor adverse reactions (see below
and Appendix 4-2).
A Global Standard
In a 2009 DHHS report on the critical gaps in
Over the years, many organizations (ISBT,IHN, US biovigilance," Sixteen gaps were identified
ARC, America's Blood Centers, Vitalant, and for blood, tissue, and organ national vigilance
suoueuoq ODD', rad ale\j uo!pea\j

o o o o
<D II) N1 o
'"'
~
o
N
...
r--
1il
c cr:
QJ
en
......
0 ..r::::
c
~
c
0 .2'
0 t ..r::::
c QJro .2'
~ cr: ::r::
0
N '*>= >=
<J.)
c.:>
+ +
0 0 c::
CI:l
'" '"
1 c::n
.s:
~ I 0
E
<J.)
::r::
'-
0
c::
0
~ 0
0
N co
QJ co
1il
c cr:
0
c
~
~c 0 I'-
o"B
c roQJ
T-
0
cc N
'*>=>=
N
I
T-
en en 0
'\' '\' ~
'" '"'"
rv <J.)
c::n
:; CI:l
0
N U .._
0
c::
0
-0
»
.Cl
......
<J.)
e
c::
~ 0
.,cro
0 cr:
0
c
:p
c.:>
CI:l
m 0
c
tro ,_
<J.)
'"'
0 Cl
QJ
,_
N 0
cc
'*~ >= 0
-0
c::
'"'" '" N
-0
c::
<D <D
M M CI:l
c::
N
<D
f 0
~c::
r-, 0
-0
<J.)
~ c::n
0
N ......
CI:l
c::
<J.)
f::
<J.)
CL.
Lt:i
..;.
'J!, 'J!, 'J!, 'J!, 'J!, 'J!, 'J!, 'J!, 'J!,
0
0
M
0
en
0
00
0
.... 0
<D
0
II)
0
er
0
N1 *'
0
N
0
'"' *'
0
LI.I
I:C
:::I
~
UO!leUOaa3elUaJJad
u::
7.
-=....
00 "! OC! L() CJ:> N -e-
(V) N 0 '<j"
5 5
=
,...
ci
N
(V)
..- 0 C'J ci cv:i
0
ci ci ci ci
0
ci
0
ci
0
ci
,-...
,... N '<j" ..- L() N (V)
CJ:> (V) (V)
II '<j" L() N 0 0 N 0 0 0 0 0
C'J " II "<i C'J
<.0
C'J
N
ci
-r-'
cv:i
0 N
ci
5 ,.-
ci
0
c: S N ci ci ci
~ It)
'5 6
.s; N
0
E
CD
:::c c;-
"-
0
,... 00 (V) 0 (V) ,.- N 00
(V) ,.- ,.- ,.-
c: II C'J cO "<i 0 0
0 S N ,.- C'J cv:i ci C'J ci ci ci 0
Cl
,...
'<:I'
co
co
« =
N
-c
,...
I'-
0
C'\J iii' (V) '<j" 0)
L()
"! 0
(V)
..- ,.- ,.-
I
,...
C'\J
II " II C'J to cv:i 0 0
m S N C'J cv:i ci C'J ci ci ci ci
CD
"- ,...
N
"-
0
c:
0
=
N
'0
»
..c
en
c:
0
~c:
0 '2
Cl 0
c::
0
0 'E
,_
,...
0" 0
'0
"- <1> CI)
CD
0.. ttl
.9
=s
CI)
en
=
CI)
CD Ci
t;:; E i::::-
a: 0 .s ::::J
'2
-
-=
(.j
c: c:: CI)
0 ::::J <1> Ci3
:;:;
c:.:>
m
=...
U>
<=
:;::
(.)
0
«I
a;
a: c::
en
.:::;
05
f':
::::J
0>
c::
:0
en
en
<1>
>
CD
a: CI) ....I 0
.~ -.... ts <1>
z-
=. «I ~ c:: '0 i::::-
....>- CI)
a: "ii
0
c::
= :g ro ::::J E <1>
zs 0 ::::J ::::J
-"e E ~ 0
= S 0.. c:: '2 '2
~ =
<= ....
«I !: c::
...<= e <1>
0
ttl ~<1> S
'0
<1>
0
~<1>
iD
0 0 0
"'"
.....
W ~I'CI CI)
>
<=
U>
«I
:§ (.)
0
"ii
..... <1>
E
<1> t::
'+-
c::
'ca
~
Ci3 '+- ·til ·til c::
~
CCI CI) Q > 0.. ....I Z :r: « 0.. Cl E 2 2
~ a:
M
c:i
M
o """
o
o
c:i c:i
c.o c; M -e- ,.-

N o o o o
c:i o o o o
c:i c:i c:i c:i c:i
I.!)
M ,.-
N o o o
c:i o
c:i c:i c:i
'er-
o c; ,.-
o
C\J
c:i o o
N
c:i """
o
c:i
0.
c:i c:i c:i
,.-
N
I.!)
o o N "<:t"
o c;
c:i o o c:i o
c:i c:i c:i c:i
,.-
C\J c; N
"""
o o
c:i o c:i 0.
c:i c:i o
-0
~ Cl)
"t;:;
E 0.
ts
e 'E
::l
0 '"
0.
e
::l
..c: "t;:;
:E
.=..c:"""
0 C\J
C/l
I-
..c:
-e-
'"
'"
;g
::l C\J :1:: 'u
0
..c: c:: :s: ..c:>!
....... :E .:.:: I:
"""c::
C\J c:: :t:
:s:
(.) 0
n~
~'"
Q)
"C
:E '(3 c::
(..) 0 (.) "0
~ '" ~ 'E <.:>
-0
C/l
";:: .......
l§ ~c:: Q) 0
s: 0
0 C/l ::l (.) zs
C/l
C/l
::l
c::
C/l
"X U e..... (..)
C/l
Cil .!!1 "0
"C:;;
>.
0
..0
0 (.)
"E '"
;;,
Q)
a.
'" (..) 0 '"
>
,_
:c
@
.......
c::
Q)
~
~Cl)
Q)
"§ 0
E
E
UJ
,_
~
±: Cil
Q)
"Ii)
sz:
a. '"
c::
'0, '"
~
..0
Q)
(.)
0
"U:;
c:: E
::>
:t: Q) <;:: (.)
>. '"
c:: c::
I-
>.
'"
..... I:
'"
0 Q)
c..:> :::c < c:: _J C/) « « c..:> c..:> :2 l- II
I:
programs, eight of which were focused on data on donor adverse events (Gap 5), beyond
blood-relatedactivities(Table4-3). Ten years lat- the FDA-required fatality reporting. Unfortu-
er, many of the gapshave been addressed, while nately, tissue and organ biovigilance efforts
others persist. Gaps 4 and 6 are largely closed, worldwide (not just the United States) lag be-
with precise recipient and donor definitions and hind blood donor and recipient hemovigilance
identified denominator data; however, the sys-
tem is still very fragmented (Gap 1) and lacking efforts,meaning that Gaps 9 through 16 are still
in comprehensive national reporting, with only relevant today. Project Notify,however, is an ex-
a small number of hosptials participating in the ample of how international experts are working
CDC NHSN HemovigilanceModule and a few together to address tissue and organ biovigi·
blood collectors voluntarily providing detailed lance.
TABLE 4-3. Gaps Identified in US Biovigilance, 200918

Gap 1 Patchwork and sometimes fragmented system of various adverseevent reporting.
Gap 2 Likely underreporting of transfusion adverseevents.
Gap 3 Challengeswith FDA-requiredreporting.
Gap4 Needfor accurate recipient denominator data, precise definitions, and training.
Gap 5 No national surveillance of donor serious adverseevents other than fatalities.
Gap 6 Needfor accurate donor denominator data, precise definitions, and training.
Gap 7 Needfor accuratetracking of all donor infectious diseasetest data.
Gap 8 Needfor timely analysis of reported data.
Gap 9 Limited information on the potential for HCT/Psto transmit
infectious disease.
Gap 10 Ability to ascertain that reported infections in HCT/Precipients
can be attributed to the tissue is limited.
Gap 11 Regulationsconcerning HCT/Padverse reaction reporting do not
extendto the level of the health-carefacility or health-care provider.
Gap 12 Current mechanisms for tracking HCT/Pgrafts to the level of the
recipient are limited.
Gap 13 Adverse reaction reporting for HCT/Psregulated solely under
Section 361 of the PHSAct is limited to infectious diseases.
Gap 14 Information about adverse reactions in other recipients of
HCT/Psfrom an implicated donor may not be readily available.
Gap 15 Lack of nationwide common organ/tissue donor network system
for real-time reporting, data collection, communication, and
analysis of donor transmitted diseasesin organ and tissue
transplant recipients, including a common donor identifier
necessaryfor linkage backto implicated donors of both organs and tissues.
Gap 16 No requirementto retain donor and recipient samples.
FDA= Food and Drug Administration; HCT/Ps = human cells, tissues, and cellular and tissue-based products; PHS = Public
Health Service.
Although reporting institutions technically ply active surveillance and advanced technical
should have always had access to. their own da- tools, including artificial intelligence and natural
ta, the complexities of hospital information sys- language processing to hemovigilance questions.
tems and blood establishment computer systems Such nonstructured data investigations show
(BECS) have not lent themselves to easy accessi- promise for linking textual data from the elec-
bility to and analysis of data, Fortunately, this is tronic heaith record (EHR)with transfusion ex-
beginning to be addressed in broad terms with posure. These projects go a long way toward ad-
the successful completion of national repository dressing several of the key hemovigilance
databases leg, the Transfusion-Transmissible In- deficiencies, including real-time data availability.
fections Monitoring System (TTIMS)], the pub- Although such data mining may speed up dis-
lishing of annual reports, and (limited) access to covery of emerging events, there is also the in-
composite data for researchers and analysts. herent risk of inaccurate and/or incomplete da-
Short of a formal revamping of the current re- ta, resulting in false signals.
porting system and increasing what is required The Observational Health Data Sciences and
to be reported, Gaps 2, 3, and 5 will likely re- Informatics (OHDSI, pronounced "Odyssey";
main, at least at the national level. It is encour- ohdsi.org) program is another example of how
aging that some progress is being seen in ad- people are collecting and analyzing health-care
dressing these gaps at the state level (ie, data in new and powerful ways. ODHSI is a
Massachusetts), which eventually may be seen multi-stakeolder, international, interdisciplinary
as the necessary next step to integrate reporting collaborative creating open-source solutions to
at the national level. Real-time data availability large-scale analytics or observational health da-
(Gap 8) is also a difficult issue, but steps are beta. The mission of OHDSI is "to improve health
ing made to shorten the reporting window. by empowering a community to collaboratively
A5 the blood banking industry and threats to generate the evidence that promotes better
blood and transfusion safety continue to evolve, health decisions and better care."
so do the tools needed to evaluate the risk and With the constant threat of existing (eg, HIV)
the impact of professionals' decisions. Because and emerging infectious diseases (EIDs; eg,
of ongoing concerns about the need for real- West Nile virus, Zikavirus, and Chagas disease),
time assessment of postmarket recipient risk and improved infectious disease surveillance and
as a result of a congressional mandate (Food and rapid alert systems are still considered primary
Drug . Administration Amendments Act of parts of any national hemovigilance program
2007), the FDA developed the Sentinel Blood (Gap 7). MBB maintains a stand-alone online
Safety Continuous Active Surveillance Network network specifically designed for reporting of
(BloodSCAN) Program in 2011 as an active sur- West Nile virus and Zika virus testing results
veillance system to monitor the safety of FDA- (see http://www.aabb.org/research/hemovigi-
regulated medical products,' including blood lance). This system tracks donor screening
components (see www.sentinelinitiative.org for data-including confirmatory testing data to dif-
more information).51,52It is a distributed data- ferentiate true from false positives-on a real-
base composed of data from 18 data partners, time basis for each EID and includes the ability
covering over 300 million person-years of health to produce reports by geographic location and
data. The project's aim is to help the FDAwork time period.
with public, .academic, and private entities to In early 2015, the FDA's Center for Biologics
collect quality data, provide a rapid response Evaluation and Research (CBER)and NHLBIan-
(days to weeks) to safety questions, and make nounced a joint program to develop TTIMS, by
the results accessible to the public domain. leveraging •.existing .and ••successful programs
More recently, in 2017 the Biologics Effec- (REDS-II)to establish a framework for TTl sur-
tiveness and Safety (BEST)initiative expanded veillance to include known pathogens and EIDs.
use of the common data models (a way of orga- Such programs are invaluable to objectively
nizing data from different primary sources into a assess the value of new blood safety initiatives,
standard secondary structure for analysis) to ap- such as the FDA's changes in policy for male-to-
male (MSM) sexual contact from lifetime defer- Instead of an ineffective "patchwork," as
ral to a deferral of 1 year, and further to 3 once described, perhaps it is better to describe
months, since the last MSM sexual contact. Pro- the current US biovigilance system as a true
grams such as Sentinel BloodSCAN,BEST,and "network" of independent and interdependent
TTIMS are needed for the FDA to have the ob- solutions, each built by committed stakeholders
jective national data necessary for other poten- from a variety of disciplines and interests, na-
tial policy changes, such as discontinuing tionally and increasingly internationally, with an
certain required tests or changing other identifi-
overarching common goal of improving donor
cation testing requirements for components
and patient safety. Countries like the United
modified with pathogen reduction technology.
Given the privatized, competitive nature of States no longer have to rely soley on their own
the US health-care system, the United States resources. If a country's hemovigilance system
may never have a hemovigilance program like can be considered a network, then globally,
SHOT or other programs in countries with sin- hemovigilance is best imagined as a network of
gle-payer health-care systems. Although the networks currently being woven together along-
availability of long-term funding for private-sec- side a variety of research, public, and private
tor programs may be concerning (DonorHART hemovigilance programs whose net goal is the
and the AABB Center for Patient Safety pro- optimal care of blood donors and patients re-
grams are no longer active), most of the other ceiving transfusions worldwide.
national programs described above have more
stable funding commitments in place. To be suc-
cessful, systems must continue to provide stake- IN L
holders with quality data to drive new safety hy-
potheses and objective analytics to demonstrate The authors thank Srijana Rajbhandary for her
outcomes. assistance in updating the chapter.
KEY POINTS
1. Hemovigilance is surveillance of the whole transfusion chain to minimize adverse events or re-
actions in donors and recipients and promote safe and effectiveuse of blood components. It is a
collaborationwhere stakeholders(researchers,policymakers,bloodestablishments,and hospitals)share
data andideas,implementpotentialsolutions,evaluatethe results,andfurtherrefinethembasedon real-
worlddata.
2. The earliest hemovigilance efforts arose out of concerns over transfusion-transmitted viral in-
fections and their sequelae for blood recipients.
3. In the United States, national hemovigilance has lagged. Currently, it consists of independent
and interdependent programs functiOningas a loose network collecting specific, high-level do-
nor and recipient data at national or near-national levels. Some individual programs, however,
although not national in scale, are very robust in their abilityto capture and validate their data.
4. Surveillance and clinical deciSion-making are overlapping but separate concepts. Clinical
decision-making asks and answers questions important to the health and care of an individual
patient, whereas surveillance focuses more on trending summary data that mayor may not ul-
timately be classifiedsimilarlyto the clinical findings.
5. Patient safety organizations (PSOs)were formed to improve health safety by providing a confi-
dential environment for voluntary reporting and analysis of patient safety events.
6. Blood establishments have an obligation to minimize the risks of blood collection. The rise in
formal donor hemovigilance efforts over the past decade resulted in internationally harmo-
nized AABB-ISBTstandard definitions for complications related to blood donation.
C H A PT E R 4 Hemovigilance 113
REFERENCES
l. Busch MP, Lee LL, Satten GA, et al. Time 11. PHBBolten-Maggs,Poles D, et al for the Serious
course of detection of viral and serological Hazards of Transfusion (SHOT)Steering Group.
markers preceding human immunodeficiency The 2017 Annual SHOT Report. Manchester,
virus type 1 seroconversion: Implications for UK: SHOT, 2018. [Availableat https://www.
screening of blood and tissue donors. Transfu- shotuk.org/shot-reports/ (accessed April 8,
sion 1995;35(2):91-7. 2020).]
2. Jennings ER, Hindman WM, Zak B, et al. The 12. Joint United Kingdom (UK)Blood Transfusion
thymol turbidity test in screening of blood do- and Tissue Transplantation ServicesProfessional
nors. AmJ Clin Pathol1957;27(5):489-502. Advisory Committee. Welcome to the National
3. Keating LJ, Gorman R, Moore R. Hemoglobin Blood Transfusion Committee. [Available at
and hematocrit values of blood donors. Transfu- https://www.transfusionguidelines.org/uk-
sion 1967;7(6):420-4. transfusion-committees/national-bloodtransfu
4. AuBuchon JP, Whitalcer BI. America finds sion-committee (accessedApril8, 2020).]
hemovigilance!Transfusion2007;47:1937-42. 13. European Union. Directive 2002/98/EC of the
5. De VriesRRP,Faver }C, eds. Hemovigilance:An European Parliament and of the Council of 27
effective tool for improving transfusion safety. January 2003 setting standards of quality and
West Sussex,UK:Wiley-Blackwell,2012. safety for the collection, testing, processing,
6. Strengers PFW. Haemovigilance- Why? [Avail- storage and distribution of human blood and
blood components and amending Directive
able at http://www.ztm.si/uploads/publica
2001l83/EC. OfficialJournal of the European
tion/990/l 009 .pdf (accessed September 16,
Union 2003;L33:30AO. [Available at http://
2019).]
cur-lex. europa. eu/Iegal-con tentiEN/TXT /
7. DHHS Advisory Committee on Blood and Tis-
?uri=CELEX%3A32002L0098.]
sue Safety and Availability meeting minutes, 14. European Commission. Commission Directive
April 30-May 1 and November 19-20, 2009 2005/611EC of 30 September 2005 imple-
transcripts. Silver Spring, MD: Department of menting Directive 2002/98/EC of the Europe-
Health and Human Services,2009. an Parliament and of the Council as regards
8. Japanese Red Cross Society.Haemovigilancean- traceability requirements and notification of se-
nual report 1993-2001. Tokyo: Japanese Red rious adverse reactions and events (text with
Cross Central Blood Center, 2003. [Availableat EEArelevance). OfficialJournal of the European
http://www.jrc.or.jp/mr/english/ (accessed Union 2005;L256:350-8. [Availableat http://
September 6,2018).] eur-lex.europa.eu/legal-contentiEN/ ALL!
9. Serious Hazards of Transfusion. TRALItables. ?uri=CELEX%3A32005L0061.]
Manchester, UK: SHOT, 2010. [Available at 15. Politis C, WiersumJC, Richardson C, et al. The
www.shotuk.org/shot-reports/trali-tables/ (ac- International Haemovigilance Network Data-
cessed April8, 2020).] base for the Surveillance of Adverse Reactions
10. Bolten-MaggsPHB,Cohen H. SHOThaemovigi- and Events in Donors and Recipients of Blood
lance and progress in improving transfusion Components: Technical issues and results. Vox
safety. BrJ Haematol2013; 163(3):303-14. Sang2016 Nov;lll (4):409-17.
16. World Health Organization. Blood transfusion conducted in the United States? Transfusion
safety. Haemovigilance. Geneva: WHO, 2019. 2015;55(4):703-7.
[Availableat http://www.who.intlb1oodsafety/ 28. AABB.Common transfusion reaction reporting
haemovigilancel enl (accessed September 6, form (report of adverse transfusion reaction to
2018).J blood suppliers). Bethesda, MD: AABB,2019.
17. Zins C. Conceptual approaches for defining da- [AVailableat http://www.aabb.org/re searchl
ta, information, and knowledge. J Am Soc In- hemovigilance/Documentsl AABB-Transfusion-
form SciTech 2007;58(4):479-93. Adverse-Reaction-Form.pdf (accessed Septem-
18. Strehlow RA.Content analysisof definitions. In: ber 16, 2019).J
Wright SE,Strehlow RA,eds. Standardizing and 29. Wiersum-OsseltonJC, Whitaker B, Grey S, et al.
harmonizing terminology: Theory and practice. Revised international surveillance case defini-
ASTM STP 1223. Philadelphia: American Soci- tion of transfusion-associated circulatory over-
ety for Testing and Materials, 1994:53-62. load: A classification agreement validation
19. Public Health Service BiovigilanceTask Group. study. Lancet HaematoI2019;6(7):e350-8.
Biovigilance in the United States: Efforts to 30. VlaarAPJ,Toy P, Fung M, et al. A consensus re-
bridge a critical gap in patient safety and donor definition of transfusion-related acute lung inju-
health. Washington, DC: DHHS, 2009. [Avail- ry. Transfusion2019;59(7):2465-76.
able at https:llwayback.archive-it.org/3922/ 31. Cumming M, Osinski A, O'Hearn L, et al.
20140403203201 Ihttp://www.hhs.gov I ashl Hemovigilance in Massachusetts and the adop-
bloodsafety Ibiovigilancel ash_to_acbsa_ tion of statewide hospital blood bank reporting
oct_2009.pdf.J using the National Healthcare Safety Network.
20. Food and Drug Administration. Safetyreporting Transfusion2017;57(2):478-83.
requirements for human drug and biological 32. Haass KA, Sapiano MRP, Savinkina A, et al.
products; proposed rille. (March 14, 2003) Fed Transfusion-transmitted infections reported to
Regist2003;68: 12405-97. [AVailableat https:11 the National Healthcare Safety Network Hemo-
www.federalregister.gov/documents/2003/ vigilance Module. Transfus Med Rev 2019;
03/14/03-5204/safety-reporting-requirements- 33(2):84-91.
for-human-drug-and-biological-products.] 33. Massachusetts Department of Public Health.
21. US Department of Health and Human Services. Hemovigilanceprogram data summaryJanuary _
National Blood Collection and Utilization Sur- December 31,2017. Jamaica Plain,MA: Bureau
vey (website). Washington, DC: Officeof Infec- of Infectious Disease and Laboratory SCiences,
tious Disease and HIVIAIDS Policy, 2020. 2018. [Availableat https:llwww.mass. gov/ser
[Available at https:llwww.hhs.gov/oidp/top vice-details/reporting-requirements-for-blood-
ics/blood -tissue-safetyI surveysl national-bloodbanks-and-hemovigilance-in-massachusetts.J
collection-and-utilization-survey/index.html.J 34. Eder AF, Dy BA, Kennedy JM, et al. Improved
22. KaplanHS, BattlesJB,Van der SchaafTW, et al. safety for young whole blood donors with new
Identification and classificationof the causes of selection criteria for total estimated blood vol-
events in transfusion medicine. Transfusion ume. Transfusion2011;51:1522-31.
1998;38(11-12):1071-81. 35. Patient Safety and Quality Improvement Act of
23. Strong DM, AuBuchonJ, Whitaker B, Kuehnert 2005. Pub. L. No.109·41, §§ 921-26,119
MJ. Biovigilanceinitiatives. ISBTScience Series Stat.424 (2005). Rockville, MD: Agency for
2008;3:77-84. Healthcare Research and Quality, 2005. [Avail-
24. AuBuchon JP, Whitaker BI. America finds able at http://www.gpo.gov/fdsys/pkglPLAW-
hemovigilance!Transfusion2007;47:1937-42. 109publ41/pdf/PLA W-109pubI41.pdf.J
25. Gammon R, ed. Standards for blood banks and 36. PSO Privacy Protection Center. Common for-
transfusion services. 32nd ed. Bethesda, MD: mats background. Rockville, MD: Agency for
AABB,2020. Healthcare Research and Quality, 2005. [Avail-
26. National Healthcare SafetyNetwork. Bloodsafe- able at https:llwww.psoppc.orglpsoppc_web/
ty surveillance.Atlanta, GA: Centers for Disease publicpagesl commonFormatsOverview (ac-
Control and Prevention, 2018. [Available at cessedApril12,2020).]
http://www.cdc.gov/nhsn/acute-care-hospi 37. Custer B, BravoM, Bruhn R, et al. Predictors of
tal/bio-hemol.J hemoglobin recovery or deferral in blood donors
27. Chung KW,HarveyA, BasavarajuSV,Kuehnert, with an initial successful donation. Transfusion
MJ. How is national recipient hemovigilance 2014;54(9):2267-75.
38. Wieling W, France CR, van Dijk N, et al. Physio- hemovigilance report. Bethesda, MD: AABB,
logic strategies to prevent fainting responses 2016 ..[Available at http://www.aabb.org/
during or after whole blood donation. Transfu- research/hemovigilance/Pages/ donor-hemovig
sion 2011 ;51 (12);2727-38. ilance.aspx (accessed September 6, 2018).J
39. Eder AF. Current efforts to reduce the risk of 47. AABB. 2012-2017 Donor hemovigilance high-
syncope among young blood donors. Curr Opin lights. Bethesda, MD: AABB, 2019. [Available
HernatoI2012;19(6):480-5. at..http://www.aabb.org/research/hemovigi
40. Tomasulo P, Kamel H, Bravo M, et al. Interven- lance/Documents/20 12-20 17-AABB-Donor-
tions to reduce the vasovagal reaction rate in Hemovigilance-Highlights.pd£(accessedApril 8,
young whole blood donors. Transfusion 20 11; 2020).J
51:1511-21. 48. Goldman M, Land K, Wiersum-Osselton J. De-
41. Kamel HT, Bassett MB, Custer B, et al. Safety velopment of standard definitions for surveil-
and donor acceptance of an abbreviated donor lance of complications related to blood dona-
history questionnalre. Transfusion 2006;46(10): tion. Vox Sang 2016; 110:185-8.
1745-53. 49. Donor hemovigilance. Bethesda, MD: AABB,
42. Bednall TC, Bove LL. Donating blood: A meta- 2019. [Available at http://www.aabb.org/
analytic review of self-reported motivators and
research/hemovigilance/Pages/ donor-hemovig
deterrents. Transfus Med Rev 2011 ;25(4):317-
ilance.aspx (accessed September 17, 2019).J
34.
50. Townsend M, Kamel HT, Van Buren N, et al.
43. Townsend M, Land KJ,Whitaker B, et al. US do-
nor hemovigilance system: Mechanism for na- Development and validation of donor adverse
tionwide reporting (abstract 3A-SI-02). Vox reaction severity grading tool: Enhanciing objec-
Sang 2001;101 (Suppll): 11-12. tive grade assignment to donor adverse events.
44. Land KJ. Update on donor hemovigilance. Pre- Transfusion 2020 (in press).
sented at DHHS Advisory Committee on Blood 51. Menis M, Izurieta HS, Anderson SA, et al. Out-
Safetyand AVailability,November 4-5,2010. patient transfusions and occurrence of serious
45. Land KJ, Whitaker BI, for the AABBUS Donor noninfectious transfusion-related complications
Hemovigilance Working Group. The 2012 among US elderly, 2007-2008: Utility of large
AABBdonor hemovigilance report. Bethesda, administrative databases in blood safety re-
MD: AABB, 2013. [Available at http://www. search. Transfusion 2012;52: 1968-76.
aabb.org/research/hemovigilance/Pages/ 52. Menls M, Anderson SA, Forshee RA, et al.
donor-hemovigilance.aspx (accessed September Transfuslon-relatedacute lung injury and poten-
6,2018).J tial risk factors among the inpatient US elderly
46. Rajbhandary S, Stubbs J, Land KJ, Whitaker BI, as recorded in Medicare claims data, during
for the AABB US Donor Hemovigilance 2007 through 2011. Transfusion 2014;54:
Working Group. The 2012-2014 AABBdonor 2182-93.
APPENDIX 4-1.
Protocol for Hospital Reporting Adverse Events to Blood Suppliers
REPORT OF ADVERSE TRANSFUSION REACTION TO BLOOD SUPPLIERS
'.IIISTRLiCTlClNS: Send the form to ALLblood suppllers that provided blood components to thls patient. Ilmely reportlnq is
important, so that,. ifappropriat~, the blood supplier may prevent the transfusion of other products from the same donor(s).
[Complete areas which are not included in your internal hospital work-up and attach work-up.]
Do you suspect this reaction is the result of an attribute specific to the donor or the blood product?
o Yes or suspected:
Reaction did not result in fatality: Complete this form and forward to the blood supplier(s).
Reaction resulted in fatality: Complete this form, forward to the blood supplier(s), ANDreport fatality to FDA.
o No: Stop, do not report to the blood supplier.
o Other: Consult with the blood supplier physician.
Additional Blood Supplier Instructions for the Hospital Transfusion Service, as applicable:
GENERALINSTRUCTIONS
Please attach the following:
Copy of completed hospital internal Transfusion Reaction Work-up Form
Copy of Pre- and Post- transfusion chest x-ray reports for suspected TRALI reactions
Copy of Culture and pending tests (when available) for suspected sepsis cases
Copy of applicable Admission Note, Physician notes regarding reaction, Discharge Note
Copy of allergy and medication list for suspected allergy reactions
-"!0~- ~ -=~
" rlotlilfooB SURIlIil!rlIrS,h:)lil5i= .~ Case Identification # Date Received (mm/dd/yy)
'"'~= ~"" w 4"" = =;;;
="'~R~~"';'
iJil
APPENDIX 4-1.
Protocol for Hospital Reporting Adverse Events to Blood Suppliers (Continued)
Name of Person Filling Out Form
Facility Addrress
Telephone Number Fax # Email
Transfusion Services Medical Director
Transfusion Services Medical Director Email Phone #
Blood Bank Medical Director
Blood Bank Medical Director Email Phone #
PATIENT/RECIPIENT INFORMATION
Medical Record # Name (optiona/)
Age Date of Birth (mm/ddlyy)(optional)
Weight Sex
Attending Physician (optional) Attending's Phone # (optional)
Admitting or Primary Diagnosis
Indication forTransfusion
Relevant Severe Co-morbidities (If applicable)
Pertinent Medications
List transfusion history within 24 Hours PRIOR to reaction (Attach additional sheets ifnecessary.)
List transfusion history within 24 hours AFTER reaction
o Returned to pre-transfusion status.
o Still requires support related to transfusion reaction.

o Expired (Not transfusion related)
/ / (mm/ddlyy)(lfavailable)
o Other/Unknown, Specify:
* Report to FDA within 24 hours.
ilil 2
(Continued)
J~ f
a.
V>
f.;r t t t
V> V> V>
f
a.
V>
~
2i
V>
t
V>
~'" ~'" ~'" ~'" ~'" ~'"

vi' vi'
~ ~
D D D D D D D D D
o o o o o o o o o
z z z z z z z z z
D D D D D D D D D
'>:E
""
'>: E
... :a-'" ~g_ ~g_
:g 0
~ 0 ~ 0 ~ 0
E
.s E.. .s ~
o o - o..
e E - o..
e E
..-
"0
Cl)
"I J
e I I
e
:::J
C
:;::;
C
0
8
en
._
.!:!2
0.
0.
:::J
CfJ
"0
0
.Q
co
.....0
en
.....
c
Cl)
>
u.J
Cl)
en
._
Cl)
>
"0
<C
OJ
c
:e
0
0.
Cl)
c:c:
ro
~
0.
,..., :::cen
0
oo::t ._
=:s
e
0
"l-
0
z
w
(..)
e, .....
0
._
0
a..
< o,
APPENDIX 4-1.
REACTION INFORMATION
Dat~/Time Transfusion
Date/Time Reaction Started: (mm/ddlyy) (hh:mm) 0 am 0 pm
Date/Time Transfusion Stopped: (mm/ddlyy) (hh:mm) 0 am 0 pm
(mm/ddlyy) (mm/dd/yy) (mm/ddlyy)

Date/Time
(hh:mm) 0 am 0 pm (hh:mm) 0 am 0 pm (hh:mm) 0 am 0 pm
Temperature "(I"F "(I"F "(N
Blood Pressure (Systolic) mmHg mmHg mmHg
Blood Pressure (Diastolic) mmHg mmHg mmHg
Pulse bpm bpm bpm
Respiratory Rate rpm rpm rpm
O,Sat % % %
o Abdominal pain/cramps [1AJ o Dyspnea [1,2,3, 4J o Loss of consciousness [1]

o Angioedema [1J OEdema - pulmonary [2,3J o NauseaNomiting [1, 4J
o Anxiety [1 J OEdema - Pedal [3J o Oliguria [4J
o Arrythmia [1J o Erythema [1J o Orthopnea [3J
o Back pain [4J o Fever [2, 4J o Pain at infusion site [4J
o Cardiac arrest [1J o Flushing [1J o Prurltls [1J
o Chest pain [4J o Headache [3, 4] o Shock [1,4]
o Chest tightness [1, 3J o Hoarseness/Stridor [1J o Substernal pain [1]
o Chills/Rigors [4J o Hypertension [2, 3] o Tachycardia [1,2,3, 4J
o Cough [3,4J o Hypotension [1,2, 4J nTachypnea [2,3J
o Cyanosis [1,2, 3J o Hypoxemia [2, 3J o Urticaria [1J
o Diarrhea [1J o Impending doom [1J o Wheezing [1, 4J
ODIC [4J o Jugular venous distension [3J o Widened pulse pressure [3J
o Other, specify:
Additional Information:
(If more than one possibility,
assign priority.)
-...
-..
~-- ....
-~....
-....•
-.•...
~~.•.•.
-~--.~.---.-~'-'-~"-.~'--'~-~".-'~';l
* Pleaserefertothe[!l~!~~~~IH.!~!~~!~~a.f~trN'!~_IlI'!'.~~v.i9II~_I1,c!£~l11pon~t_II!!'1.«>!il!il~~!I~.d.~I.!~~!",!!~(lI!C.I!.f!~~~~~I.I
forcomplete
definitions.
Alta(h atleryyand medication list
00. t
, Atta(hChestx·ray
, Pleaseforward results ofrulture and pending tests when available
II
(Continued)
120 A A B B TEe H N I CAL MAN UA L
APPENDIX 4-1.
PULMONARY-ALLERGIC-ANAPHYLACTIC REACTION INFORMATION
o Acute Respiratory Distress o Severe sepsis o Upper airway obstruction

Syndrome (ARDS) o Shock o Diffuse alveolar damage
o Aspiration o Multiple trauma o Chemotherapy
o Pneumonia o Burn o Amiodarone
o Toxic inhalation o Acute pancreatitis o Disseminated intravascular
o Lung contusion o Cardiopulmonary bypass coagulation
o Near drowning o Drug overdose o Radiation to thorax
o Pulmonary hemorrhage o Volume overload o Massive blood transfusion
o Renalfailure
Additional comments (Other risk factors)
0, sat s 90% on room air
Pa0,tFiO, :;; 300 mm Hg
ChestX-ray: Bilateralinfiltrates
(Attach chest x-ray If available)
Chest X-ray: Widened cardiac

silhouette (cardiomegaly)
Elevated BNP
(Provide value in pglmL.)
o BNP 0 NT-proBNP
Elevated Central Venous
Pressure greater than 12
mm Hg (Provide values.)
Positive Fluid Balance

(in mL) (Attach patient 110
report if available.)
Transientdecrease
White Blood Cell Count
ilil. 5
APPENDIX 4-1.
Acetaminophen DYes DYes
Antihistamines DYes DYes
Bronchodilators DYes DYes
Diuretics DYes DYes
Epinephrine DYes DYes
IntubationlVentiiatory support DYes DYes
Oxygen supplementation DYes DYes
Steroids DYes DYes
Vasopressors DYes DYes
Other (specify): DYes DYes
Additional comments (Attach additional C/in/callnformation if available.)
Recipient HLA type:
Recipient HNA type:
Recipient HLA/HNA antibody status:
Donor HLA/HNA antibody result (if performed on unit):
Donor HLA type (If available):
iJiJ.
(Continued)
APPENDIX 4·1.
SUSPECTED BACTERIAL CONTAMINATION

Were the suspect components returned to the blood bank? D No DYes
On repeat visual inspection, does the component reveal any abnormalities (e.g. clumps,discoloration, hemolysis)?
D No D Yes:Describe: D Unevaluabie
Suspect component - Source used: D Bag D Segment D Not done
Gram stain performed:
Result (organism identified, If positive):
D Negative D Positive D Not done
Culture performed:
Result (organism identified, if positive):
D Negative D Positive D Pending D Not done
Was a secondary test performed by the hospital for this component (PGD or equivalent)?
D No D Yes,Specify:
Patient's pre-transfusion blood culture: D Negative D Positive D Pending D Not done
DatelTime: (mm/ddlyy)
Result (organism identified, If positive):
(hh:mm) D am D pm
Patient's post-transfusion blood culture result: D Negative D Positive D Pending D Not done
DatelTime: (mm/ddlyy)
Result (organism identified, if positive):
(hh:mm) D am D pm
Does the patient have history offever or other infection related to his/her underlying medical condition? D No DYes
Was the patient on antibiotics at the time of transfusion? D No D Yes,Name:
Is the patient currently being treated with antibiotics? D No D Yes,Name:
Did the patient have an absolute neutropenia (neutrophil count less than 500 per Ill) prior to transfusion? D No DYes
Comments:
FOR TRANSFUSION MEDICAL DIRECTOR REVIEW
Reaction D Allergic/Anaphylactic DTRALI DTACO D Septic Transfusion Reaction D Other:
Case definition
D Definitive D Probable D Possible
criteria
Severity D Non-severe D Severe D LifeThreatening D Death
Imputability D Definite D Probable D Possible D Doubtful D Ruled out D Not Determined
Notes
Tranfusion Medical Director contact/phone/email
Tranfusion Medical Director (or designee) signature

I-----------.-- ...-..-.-----~-·--··---~-·--·-----·-··----~-----"·---·---"---"-~"-I
* Pleaserefertothe LNatio~~1Healthcarl!Saf~~~_~t\li0rkB~~g!lan(I!_C~I_'IIl~.'I!H.~~1I~1~~_(eMo~~:~_ur1l:iI~~~e~:o!~(o.'.,forcompletedefinitions.
jJil 7
APPENDIX 4-1.
Case definition
criteria
o Definitive 0 Probable 0 Possible
Severity o Non-severe 0 Severe 0 life Threatening 0 Death
Imputability o Definite 0 Probable 0 Possible 0 Doubtful 0 Ruledout 0 Not Determined
Notes (Attach additional reports, Ifavailable.)
Blood Supplier contact/phone/email

<~.-.-.-.-. .---~~------~-~. ~"--'''~-~' ---.-----.~.~~~~~.
* Please refer to the [~,~~~-"--~~~-."---~~----
National Healthcare Safety Network Biovigilance Component Hemovigilance Module Surveillance Protocol
... .~-~-"~---,---
---,,--,,-"-"~,~---,-,-,,~-,~---,-"'~~'",~~~~~~---'
for complete definitions.
jJj]. II
....0 ~
1::
§
0 ::r:
0- 'd
</J
~ ~
ro
til .,_,
>-<
~ {§ ~
U)
::r:
l
<J) <J)
.~ t;
;::l
0<
&1 1:i
~ <J)
>-<
<J)
</J
>-< '"
r."1 +-'
ro r::: ~
{3.9 ~.... fr
Q
~ t:~
'-' 1':'..f3 0
r::: f3
~E 0
Cl
~
0
~l::; '1j
'"~
~~ 0
ot.8
~
0
...
"S'"
gj~ 4-<
:>~ 0
'0 !Ii
0
f-<
~
'd
eo
r::: ~'"
.aro ""
-g",
(3 ~:a
O·oP
.€
<J)
~.s
.~~
g,
U)
<J)
~8
cS -E]
.S ~·s
~ "'~
~"
0
il
.,_, ~~
0
"'::r:....
.8
r::: 5:iJ0
°
<J)
~
l{")
<J)
'1j
])
.~
:rJ
oj ~'"iil
'"'
c.?
'1j
'"~.~'"
00
r::: oZ
~t:: ro uul'
"T .. 1l
Q) <J)
'1j
~·E
dJ ro
(3 ~Jl
Q)
e tl
Q)
>
"'C ~
«
....
0
o '"
\,!)
t:: M
o :::
Cl ~
"'C
o
'"
0
;?;
o
co
....o '"
OJ
..... "
<Il
rEb
o
.~
0
o S
I- <J)
Ol ::r:
Nt:: '"
0
~-g "
0
>< ....
_C!:i
c:l
!XI
CIa
z.-
w"'"
c..Q) ~
c..ii;
<CCf.J
:c-
Cl)
:::>
c::
0';:::
c::
o
S
en
-I-'
c::
Q.)
>
UJ
Q.)
e
Q.)
>
""0
«
.._
o
c::
o
t.::::l
""0
o
o
CQ
.._
o
..,_
.-
"0
CD
::I
r:::
:;::::;
t::
o
c..:>
'-"'
~t::
CD
>
L.I.J
CD
e
CD
"0
>
«
'--
o
r:::
o
Cl
"0
o
o
en
'--
o
......
o
o
I-
0)
Nt::
~~
><'--
_C!'
c~
z.-
w'--
c..CD
c..6:;
CC(J)
Allogeneic and Autologous Blood
Donor Selection
Debra A. Kessler, RN, MS, and Susan N. Rossmann, MD, PhD
HE FOREMOST RESPONSIBILITYOF BLOOD

collection facilities is to maintain a safe
and adequate blood supply. The selection
of appropriate blood donors is essential to pro- The blood collection facilitymust determine do-
tect donors' health during and following dona- nor eligibility in accordance with federal and
tion and to ensure the safety, quality, identity, state regulations and MBB's voluntary accredi-
purity, and potency of the donated blood com- tation standards. Specific criteria used to select
ponents to protect the transfusion recipient. Key donors are established by FDA regulations and
elements of the selection process, as part of the recommendations in guidance and memoranda.
overall approach to blood safety, are donor edu- In addition, MBB has developed professional
cation, use of a Food and Drug Administration standards for donor selection with which ac-
(FDA)·accepteddonor history questionnaire and credited blood collection facilitiesmust comply.1
materials, a focused physical examination, and Blood collection facilities provide prospec-
the results of infectious disease testing (see tive blood donors with information on the dona-
Chapter 7); management of all information tion process and potential donation-related ad-
about the donation, including subsequent post· verse effects and instruct them not to donate if
donation information, and the donation process they have risk factors associated with an RTTI.
itself must be completed in accordance with The donor screening process includes a focused
current good manufacturing practice. physical examination and direct questioning
This chapter describes the current federal reg- about specific risk behaviors, medications, trav-
ulations, accreditation requirements, and medi- el, and other factors that potentially affect trans-
cal considerations related to both screening fusion recipient or donor safety. The donor
blood donors before their blood is collected and screening questions. address risks related to
donor testing for relevant transfusion-transmitted RITIs for which sensitive tests are currently per-
infections (RTTIs),as defined by the FDAin the formed leg, human immunodeficiency virus
Code of Federal Regulations (CFR), Title 21, (HIV)],for which tests are not universally used
Part 630.3(h).1'3 (eg, Babesia), and for which licensed screening
Debra A. Kessler, RN, MS, Vice President, Medical Programs and Services, New York Blood Center, New York,
New York; and Susan N. Rossmann, MD, PhD, Chief Medical Officer, Gulf Coast Regional Blood Center, Hous-
ton, Texas
127
tests are not yet available (eg, sporadic and varied without prior public comment. The FDA in-
ant Creutzfeldt-]akob diseases (C]D) and malar- tends to revise and replace the HIV and malaria
ia). In addition, centers may ask donors questions guidances after considering feedback from the
about a history of pregnancies to determine comment period. However, the agency expects
whether testing for HLA antibodies is necessary, that the recommendations set forth in the guid-
as part of a risk reduction strategy for transfusion- ances will continue to apply.
related acute lung injury (TRALI).Facilities must A precautionary approach attempts to reduce
establish donor eligibility on the day of donation the risk of known or potential RTTIs but also re-
and before collection. If a donor's responses to sults in the deferral of many healthy donors.
the screening questions are incomplete or un- Medical directors of blood collection facilities
clear, the blood collection establishment may are responsible for determining donor eligibility
clarify the responses within 24 hours of dona- policies on issues that are not covered by regula-
tion and remain compliant with the "day of col- tions or standards.v"" Consequently, medical
lection" requirement [21 CFR 630.10(c)]. decisions regarding the same issue may differ
If individuals are instructed not to donate among facilities or even among physicians at the
blood for others because of their health history, same facility. Considerable variability exists in
reactive test results, behavioral risks, or medical national and international practices for deter-
reasons, they must be added to a confidential mining donor eligibility, which reveals the in-
deferral list at the blood center to prevent future herent uncertainty in risk assessment." The fa-
donations [21 CFR 606.160(e)(1) and (2), and cility's collection staff should be able to explain
630.10(d)(1)]. Depending on the causative rea- to donors the intended purpose of AABB and
son, deferrals may be for a defined interval of FDA requirements, as well as their center-specif-
time, for an indefinite period (for which there ic eligibility screening practices.
may be the possibility of reinstatement to the Answers to the most frequently asked ques-
donor pool), or permanent with no potential for tions about federal regulations defining blood
reinstatement as a blood donor in the future. 1 In donor eligibility are available to the public in the
addition, collection facilities are required to "Questions about Blood" section of the FDA
manage postdonation information that could af- website. IS Frequently asked questions about the
fect the safety, purity, and potency of the blood interpretation or underlying rationale of AABB
components from the current donation and any Standardsjor Blood Banks and Transfusion Ser-
previous donations, and affect the future eligibil- vices (Standards) can be found through the
ity of the donor.' AABBStandards Portal," or questions may be di-
The criteria to evaluate individuals who are rected to the AABB Standards Department (stan-
donating blood for their own use (autologous do- dards@aabb.org); responses and discussion of se-
nation) may be less stringent than those for peo- lected issues are posted on the AABBwebsite.
ple who donate for use by others (allogeneic do-
nation). However, the focus remalns on providing
the safest possible blood for transfusion to the do- I
nor-patient and on evaluating the risks that the
D
collection procedure poses to his or her health.'
The AABB Donor History Questionnaire Registration and Donor Identification
(DHQ) is currentiy used by most blood collec-
tion facilities in the United States."? The refer- In the United States, blood components for allo-
ences in this chapter are to DHQ version 2.1, geneic transfusion are typically collected from
which at the time of writing was still under re- volunteer, nonremunerated donors; otherwise,
view by the FDA. The v2.1 DHQ incorporates components must be labeled as being from paid
the recommendations of the April 2020 FDA donors [21 CFR 606.121 (c)(8)(v)].
guidances for C]D, HN, and malaria+" The Prospective blood donors should provide an
April 2020 C]D guidance was issued as final, acceptable form of identification. Acceptable
but those for HN and malaria were implement- forms of identification include government-
C HAP T E R 5 Allogeneic and Autologous Blood Donor Selection 129
issued documents, such as a driver's license or AABBDHQ Blood Donor Educational Material
passport, or a blood-center-issued donor card incorporates all necessary elements required by
with a unique numeric or alphanumeric code, federal regulations and AABBStandards, and fa-
or other forms of identification as determined by cilities may add elements to protect both blood
the blood collection facility. Most facilities no donors and transfusion recipients.P" At each
longer use Social Security numbers for donor encounter, the donor must be informed about
identification because of donor privacy con- the collection procedure in terms that the donor
cerns. understands, and the donor's consent and/or
Accurate records are essential to identify all acknowledgment must be documented. It must
prior donations from any given donor, including also be documented that the donor has read the
whether the donor has ever given blood under a educational material, has had an opportunity to
different name, so that the link with all prior do- ask questions, agrees pot to donate lithe dona-
nations within the blood system is maintained. tion. cOMldresult lna risk to recipients as out-
Accurate records are also essential to ensure lined in the educationallllateriCll.,.and may with-
that the donor can be contacted following the draw from the donation procedure. 1The setting
donation and informed of test results or other used for the donor screening process should pro-
relevant information from the current donation, vide adequate privacy for donors to be comfort-
if necessary. Facilities must make reasonable at- able discussing confidential information. The do-
tempts to notify the donor within 8 weeks of the nor should be informed about possible adverse
donation if any test results disqualifythe individ- reactions to the collection procedure and the
ual from continued donation." FDA regulations tests that will be performed for RTTIson his or
published in the CFR require facilities to ask her donated blood. The donor must also be in-
donors to provide a postal address where they formed of the notification process for positive
can be reached during this interval for counsel- test results, any reporting requirements to pub-
ing or other follow-up, if necessary [21 CFR lic health authorities, and the possibilityof inclu-
630.10(g)(1)]. Accurate donation records must sion in the facility's deferral registry and subse-
be kept by the facility for the requisite amount quent deferral from future donation. Donors
of time, according to current regulations and should also be informed if investigational tests
standards [21 CFR 606.160; AABB Reference or other research may be performed on samples
Standard 6.2N(pp73'75)]. or information collected during the blood dona-
Individual blood collection facilities or sys- tion. Finally,the limitations of the tests to detect
tems must maintain a list of deferred donors, early infections and the possibility that a test
but, notably, there is currently no national regis- may not be performed if samples are not ade-
try of deferred blood donors in the United quate should be explained to the donor. The
States. Individuals who are deferred by one Blood Donor Educational Material instructs the
blood center may be eligible in another blood donor not to donate blood for the purpose of re-
system. The available evidence suggests that na- ceiving free infectious disease testing. Blood col-
tional deferral registries are not necessary or lection facilities must comply with applicable
useful because they do not contribute to blood state laws to obtain permission from parents or
safety and do not prevent the release of unsuit- guardians for minors (ie, 16- or 17-year-olds).1
able components." In addition, national regis- Moreover, AABBStandard 5.2.2 requires that
tries have raised privacy concerns. blood collection facilities have a process to pro-
vide information about the donation process to
Educational Materials and Donor parents or legally authorized representatives of
donors when parental permissionis required.1(PI6)
Acknowledgment and Consent
Blood collection facilities should also estab-
US blood collection facilitiesprovide all prospec- lish policies on accommodating individuals who
tive blood donors with information about blood are not fluent in English or are illiterate, are vi-
donation via educational materials, donor ac- sion or hearing impaired, or have other physical
knowledgment, and informed consent. The disabilities. Many facilities try to make reason-
able accommodations for donors' special needs. In general, neither the FDAnor MBB speci-
However, facilitiesmust also ensure that the col- fies the test method, specimen type [capillary
lection procedure does not pose undue risk to (finger stick) or venous blood], or acceptable
donors or staffmembers, that an accurate health performance characteristics for tests used for he-
history can be obtained, and that the acknowl- moglobin/hematocrit screening. One exception
edgment/ consent process is not compromised. is that a capillary sample collected from an ear-
The final authority for decisions on such issues lobe puncture is not an acceptable specimen for
rests with the facility's medical director who is hemoglobin/hematocrit screening for allogeneic
responsible for all aspects of donor qualification or autologous donors because of its poor accura-
and phlebotomy. 1 cy.l Most US blood collectionfacilitiesuse finger
stick samples for hemoglobin/hematocrit deter-
Donor Qualification by Focused mination. These samples tend to give slightly
Physical Examination and Hemoglobin
higher values than venous samples."
The methods to measure hemoglobin or he-
or Hematocrit Measurement
matocrit are generally selected for their ease of
Qualification screening procedures for blood do- use in the mobile blood collection setting. The
nation include a focused physical examination copper sulfate density method (Method 6-1) is
and a hemoglobin or hematocrit measure- still an acceptable screening tool in blood cen-
ment. 1,2The donor eligibilityregulations define ters in the United States but has been largely re-
not only the specific physical requirements but placed by methods such as spectrophotometric
also the level of medical supervision required for measurement of hemoglobin with portable de-
the assessment (21 CFR630.5; 21 CFR630.10). vices or hematology analyzers to measure hema-
This evaluation has potential implications for tocrit. The point-of-caremethods that use porta-
the potency or safety of the collected compo- ble devices yield quantitative hemoglobin
nent and/or the well-being of the donor. Addi- results, with a coefficient of variation (CV) of
tional requirements apply to apheresis donors, 1.3%.19A typical automated analyzer measures
who must meet the height, weight, and hemo- hemoglobin levels in a venous sample with a CV
globin or hematocrit requirements approved by :0;1.2%.19 Most quantitative methods currently
the FDA for the automated collection device. in use reliably measure hemoglobin levels with-
The collector must weigh the donor and not rely in approximately 0.2 to 0.5 g/dL, and the vast
on self-reported weight for donation of any plas- majority of donors deferred for hemoglobin or
hematocrit have values near the cutoff. For
ma product collected byapheresis.
capillary-sample-basedmethods, the most likely
The donor must have his or her blood pres-
source of preanalytical error is the sampling
sure measured and is eligible to donate only if it
technique, and testing must be performed in
falls within the range of 90 to 180 mm Hg for compliance with the manufacturer's instruc-
systolic and 50 to 100 mm Hg for diastolic pres- tions. A noninvasive measurement, not involv-
sures. If the measurement falls outside of these ing a blood sample but using measurements
ranges, the donor must be deferred unless seen through the skin, has recently been approved.
in person by a physician to assess the safety of Donor hemoglobin screening may help en-
performing a collection. For pulse, the rate must sure a minimum content of hemoglobin in a
be between 50 and 100 beats per minute with- unit of Red Blood Cells (RBCs),but currently
out irregularities in rhythm. A physician may ap- neither the FDAnor AABBdefine potency stan-
prove donation for rates outside this range or ir- dards for RBCunits prepared from whole blood
regularities in rhythm, using his or her collection. If a donor's hemoglobin level is
judgment. This approval may be in person or by 12.5 g/dL, a 500-mL whole blood collection is
telephone. Approval for blood pressure or pulse expected to yield about 62.5 g of hemoglobin
outside the stated ranges cannot be delegated to per unit of RBCs,but determining the final con-
a nonphysician nor defined by a standard operat- tent of hemoglobin in an RBC unit prepared
ing procedure (SOP). from whole blood is not required. MBB Stan-
dards requires apheresis RBC units to be pre- deficient erythropoiesis or advance to frank ab-
pared using a method known to ensure a final sence of iron stores." Without iron supplementa-
component containing a mean hemoglobin level tion, two-thirds of donors may not recover iron
of 60 g, with 95% of the units sampled contain- stores even after 168 days (24 weeks).27Recent
ing >50 g of hemoglobin. I(p27) studies have shown the benefit of iron supple-
As of May 2016, FDAregulations define the mentation in improvingiron stores and hemoglo-
minimum acceptable hemoglobin concentra- bin in these donors. The amount of elemental
tion for male donors as 13.0 g/dL or the essen- iron found in an over-the-counter dallymultivita-
tially equivalent hematocrit of 39% [21 CFR min is typically19 mg; iron tablets availableover
630.10(f)(3)(i)(B)].For females, the acceptable the counter may contain 38 mg of elemental
hemoglobin concentration is 12.5 g/dl. or 38% iron. Either can be an effective supplement for
hematocrit [21 CFR 630. 1O(f)(3)(i)(A)].Hemo- blood donors with adverse effects indistinguish-
globin or hematocrit screening may help pre- able from placebo." Another successful ap-
vent collection of blood from a donor with sig- proach is to perform ferritin testing and to notify
nificant anemia, but it is clear that many donors donors who have low levels, offering them the
do not have adequate iron stores even though option of iron supplementation or delaying fur-
they meet donor hemoglobin requirements." ther donation.23-2B Simple notification of donors
This could have implications for the health of of their low iron status was shown to be essen-
the donor as well as the potency of the collected tially as effective as providing iron supple-
component. If tile organization wishes to collect ments. 2B Variousmethods can be used to encour-
blood from female donors with hemoglobin lev-
age iron replacement with blood donors,
els of 12.0 to 12.5 g/dl, or 36% to 38% hema-
including providing iron tablets to the donor at
tocrit, they may do so if additional steps are fol-
the donation site, providing coupons for iron
lowed to ensure that the health of the donor
products, and/or informing donors of the need
will not be adversely affected by the donation,
in accordance with a procedure that has been for supplemental iron. Low ferritin levels can
found acceptable for this purpose by the FDA also serve as the basis for an extended deferral
[21 CFR 630.10(fj(3)(i)(A)].At the time of this from donations that include red cell compo-
writing, the procedures the FDAwill approve to nents. Extended deferrals without providing in-
accept female donors with hemoglobin values of formation about ferritin or iron supplementation
12.0 to 12.5 g/dL are not clear. The strategies will not restore iron levels in a reasonable time.
might involve extended intervals between dona- Donors with an abnormally low or high ferritin
tions, iron supplementation, and/or ferritin test- level should be referred to their health-care pro-
ing (usinga predonation sample). Unfortunately, vider for evaluation, as appropriate.
no point-of-caretests for assessingiron stores are Finally, before phlebotomy, the collection
availableat this time. staff inspect the donor's antecubital skin to de-
Low hemoglobin/hematocrit is the most termine that it is free of lesions and evidence of
common reason for blood donor deferral at most injection drug use, such as multiple needle
donor centers. The various strategies to mitigate punctures (eg, small scars lined up in "tracks")
nonanemic iron deficiency in blood donors ad- and that the veins are adequate for donation.
dress the concern that has arisen about possible Scars or pitting on the forearm associated with
health effects of low iron, particularly among frequent blood donation should not be mistaken
teens, females of childbearing potential, and fre- for evidence of injection drug use. Common and
quent donors of either gender.2l-25Both physical mild skin disorders (eg, poison ivy rash) are not
issues and impaired cognitive functions have a cause for deferral unless there are signs of lo-
been suggested in individuals with low iron calized bacterial superinfection or the condition
stores, even in the absence of anemia." Donors, interferes with proper skin disinfection in the
particularly frequent donors, may develop iron- antecubital site before phlebotomy.
Health History Assessment-AABB ing or a combination of both methods to admin-

DHQ ister the DHO.
If AABBstandards or FDAregulations do not
The AABBDHO is now used by most blood cen- address specificmedical conditions that a blood
ters in the United States. The DHO includes the collection facility has chosen to include in the
information required for compliance with both DHO, the facility must develop SOPs for deter-
AABB Standards and the FDA. Its use is not mining the criteria for acceptance or deferral of
mandated by the FDA/,7 but alternative proce- a donor. A rational approach to donor health his-
dures for collecting required information from tory assessment should attempt to balance the
blood donors must be submitted for FDA ap- need to take appropriate precautions to protect
proval in a Prior Approval Supplement under 21 the blood supplywith avoidingunnecessarily re-
CFR 60 1.12(b) before implementation. In addi- strictive policies that disqualify large segments
tion, the FDAacknowledges that the DHO doc- of the population without contributing to either
uments contain questions related to the follow- recipient or donor safety.' Decisions about do-
ing issues not addressed by any FDAregulations nor eligibility should be based on available evi-
or recommendations: cancer; certain organ, tis- dence regarding the risk that the medical condi-
sue, or marrow transplants; and bone or skin tion or history poses to the blood donor and the
grafts. However, AABBrecommends that blood transfusion recipient.
collection facilities implement the DHO materi- If a potential risk exists for the transfusion re-
als, including the following documents, as ac- cipient or donor, the effectiveness and incre-
cepted by the FDAand in their entirety: mental benefit of screening donors by question-
ing should be evaluated, especially in light of
Blood Donor Educational Material. other safeguards that protect the donor or other
Full-LengthDHO. transfusion practices that mitigate potential risks
'Ii User Brochure, including glossary and refer- to the recipient. If the facilityreceives postdona-
ences. tion information that should have been cause for
Medication Deferral List. deferral had it been reported at the time of do-
nation, then any subsequent actions, such as
The use of the DHO flowcharts as a resource product quarantine, retrieval, market withdraw-
is optional, and facilities may implement an al, or consignee notification, should be com-
equivalent method to evaluate responses to the mensurate with the potential hazard and likeli-
DHO. The current FDA-recognized DHO and hood of possible harm to the recipient. The
accompanying materials, Version 2.0, materials facility's approach to developing donor deferral
are available to the public on the AABBweb- criteria should take into account evidence as it
site." becomes available to modify those decisions. Is-
The wording, order, and text of the DHO sues that allow for medical judgment and for
questions must not be changed because FDA which questions exist in the DHO can be ex-
has accepted the current DHO as presented. plored further with the donor, but each donor
The User Brochure for the DHO details the pur- center must develop and follow its own proce-
pose and limitations for the use of the DHO and cures.'
related materials. Changes to the DHO ques-
tions, even minor, are not permitted. Once re-
vised, the documents are no longer recognized
by FDAas the accepted AABBdocuments. Facil-
ities may choose to include additional questions
as long as they are 1) in the designated area for Blood collection facilities have recognized for
additional questions at the end of the DHO, and years that frequent and repeat donors, notably
2) more restrictive. The DHO documents are in- platelet and plasma donors, must answer the
tended to be self-administeredby the donor, but same questions at every donation about remote
facilitiesmay choose to use direct oral question- risk factors that are not likely to change-a situ-
ation that leaves many dedicated donors dissat- screening questions may become a permanent
isfiedwith the donation experience. An abbrevi- part of donor screening or may be discontinued
ated questionnaire for frequent donors may as the risk recedes.
improve their experience. The FDA ailows the
administration of AABB's abbreviated DHO
(aDHO)for frequent donors who qualifyby suc- D IN
cessfully completing the full-length DHO on at I
least two separate occasions, with one or more
donations within the past 6 months. The User
Brochure for the aDHO details the purpose and Unlike questions about potential risks to transfu-
limitations on the use of the aDHO. The AABB sion recipients, most selection criteria directed
aDHO, which was developed and validated by primarily at protecting donor safety are left to
the Donor History Task Force (DHTF) along the discretion of the blood center's medical di-
with the full-length DHO, has been officiallyrec- rector. Consequently, practice varies at different
ognized in FDA guidance as "acceptable" and blood centers+" AABBStandards requires that
can be implemented by blood collection facili- prospective donors appear to be in good health
ties using the corresponding full-length DHO.7 and be free of major organ disease (eg, diseases
In the aDHO, two "capture questions" about of the heart, liver, or lungs), cancer, or abnormal
new diagnoses or treatments since the last dona- bleeding tendency, unless determined eligibleby
tion replace 14 previous questions about remote the medical director.1(p64) The rationale for each
risks (eg, blood transfusion and babesiosis). deferral for medical conditions should be care-
In some cases, measures must be immediate- fully considered because even temporary defer-
ly taken to reduce risk from an emerging or re- rals adversely affect the likelihood that individu-
emerging RTTI.Donor screening may playa par- als will return to donate blood."
ticularly important role if testing is not available
or used, or if the agent has not been shown to Cancer
be reduced by available pathogen-inactivation
measures. Donor information and/or education Eachyear in the United States, blood collection
materials requesting self-deferral can be imple- facilitiesreceive hundreds of reports of cancer in
mented quickly, asking prospective donors not individuals who had donated blood. Direct
to donate if, for example, they have traveled to transmission of (:ancer.through blood transfu-
certain regions where the RTTI is common or sion__although biologically plauslble=has not
there is an outbreak. Donor screening questions yet been documented to occur even though peo-
may also be added.t9 the endof the PHO, as- ple with cancer frequently donate blood before
sessing travel risks or exposure to others who discovering .their.di~gnosis. 30 .A retrospecMve
may have been infected. The interval before study examined the incidence of cancer among
such screening questions are implemented will patients in Denmark and Sweden who received
vary with the length of time it takes to modify a blood from donors with subclinicalcancer at the
blood collection facility's operational methods time of donation. Of the 354,094 transfusion re-
involved in donor screening. This may include cipients, 12,012 (3%) were exposed to blood
its blood establishment computer system components from precancerous donors, yet
(BECS),SOPs, and staff training. It may be nec- there was no excess risk of cancer among these
essary to temporarily stop collections entirely in recipients compared with recipients of blood
locations where the risk otherwise cannot be ef- from donors without cancer." A similar study
fectively managed. The FDA, other authorities, indicated no risk for recipients of blood from do-
and/or AABBwill provide guidance in emergen- nors who were later demonstrated to have
cy situations, to which blood collection facilities chronic lymphocytic leukemia.F These data in-
should remain alert and flexible to meet such dicate that cancer transmission by blood collect-
emergencies. Depending on the situation, mea- ed from blood donors with incident cancer, if it
sures such as donor education information and occurs at all, is so rare that it could not be
detected in a large cohort of transfusion recipi- should not contain significant inhibitory or pro-
ents that included the total blood experience of thrombotic factors. Similarly, platelet compo-
two countries over several years. nents intended as the sole source for patients
In considering the future eligibilityof donors should contain platelets that have adequate
with cancer, some degree of caution is warrant- function and are not irreversiblyimpaired by the
ed to allow sufficient time for donors to recover presence of inhibitors.
after chemotherapy or other treatment. There Individuals with a history of a significant
are currently no USfederal regulations or profes- bleeding diathesis are usually counseled to avoid
sional standards regarding the criteria that blood donation. However, screening donors for
should be used to evaluate donors with a history such a history does not prevent the rare but seri-
of cancer. For this reason, a blood center's medi- ous thrombotic or hemorrhagic complicationsin
cal director has considerable flexibilityin deter- otherwise healthy blood donors. Individuals
mining donor eligibilitypolicies. with hemophilia, clotting factor deficiencies, or
Almost all licensed blood collection facilities clinicallysignificantinhibitors-all of which are
currently accept donors who report localized manifested by variable bleeding tendencies-
cancers after treatment, with no deferral period. require deferral for both donor safety and prod-
These cancers include skin cancer (eg, basal cell uct potency considerations. The exception is
or superficialsquamous cell carcinoma) and car- Factor XII deficiency, which is not associated
cinoma in situ (eg, cervical)that have been fully with either bleeding or thrombosis.
excised and are considered cured. Most facilities Carriers of autosomal-recessive or sex-linked
defer individuals with a history of a solid organ recessive mutations in clotting factors usually
or nonhematologic malignancy for a defined pe- are not at risk of bleeding. They typically have
riod after completion of treatment, provided decreased factor levels but are accepted by most
that the donor remains symptom free without facilitiesbecause of the normal, wide variability
recurrence. The deferral period following com- in clotting factor activity levels (50% to 150%)
pletion of treatment for cancer ranges from 1 to compared to the much lower relative activity
5 years." Centers vary in approach to deferrals that is necessary to maintain hemostasis (5% to
for donors with hematologic malignancies and 30%V Individuals with von Willebrand disease
invasive melanoma. These various deferral poli- are typicallydeferred by most facilities,although
cies are currently defensible but should be re- some may allow individuals with mild disease
evaluated if new information becomes available and no history of bleeding to donate red cells.
about the potential for cancer transmission Antithromboticmedicationsare discussedbelow.
through blood transfusion.
Heart and Lung Conditions
Bleeding Conditions or Blood Diseases
Cardiovasculardisease is common in the United
Bleeding conditions and blood diseaseshave the States, affecting an estimated 86 million (more
potential to affect donor safety, as well as prod- than 1 in 3) adults." Prospective blood donors
uct potency, and blood collection facilitiesmust are asked if they have ever had problems with
define SOPs for handling donors with hemato- their heart or lungs as a donor safety measure,
logic disorders. In general, prospective donors but the criteria for accepting donors with a his-
should be evaluated for bleeding conditions or tory of heart or lung disease are defined by each
blood diseases that 1) place the donors at risk of blood center.
bleeding or thrombosis as a result of the collec- The collective, published experience with
tion procedure or 2) may affect the hemostatic autologous donation by patients scheduled for
efficacy of their blood and its suitability for cardiac procedures has demonstrated that ad-
transfusion to others.' verse effects are not more frequent than in do-
Plasma components and Cryoprecipitated nors without a history of cardiac disease.34-38 De-
Antihemophilic Factor should contain adequate spite the relative frequency of cardiovascular
amounts of functional coagulation factors and disease in the adult population, vasovagal reac-
tions occur in only about 2% to 5% of whole as developed by AABB and reviewed by the
blood donations by healthy donors and are actu- FDAor have added only a few drugs. The DHTF
ally more likely to occur among young, healthy has encouraged facilities to fully consider the
adolescents than older adults at greater risk for reasons behind each local deferral and avoid un-
cardiac conditions.Y" necessary deferral practices.' The recent explo-
A rational approach to screening donors with sion in the number of drugs affecting platelet
a history of cardiac disease allows the accep- function or clotting, such as the direct Factor X
tance of donors who are asymptomatic on the inhibitors, which are increasingly used instead
day of donation, have been medically evaluated, of warfarin, are often encountered as a cause for
and report no functional impalrment or limita- deferral [21 CFR 630.10(e)j 21 CFR 640.21(b)
tions on dally activity after being diagnosed or and (c)l.
treated for cardiac disease. Some donor centers FDA's older pregnancy-riskcategories, which
advise individuals to walt at least 6 months after are designed to assess the benefit-vs-riskratio if
a cardiac event, procedure, or diagnosis. These drugs are taken during pregnancy, are often in-
centers then allow these individuals to donate if appropriately used for blood donor selection.
they have been asymptomatic and able to per- For example, categories D and X include some
form their usual dally activities during this inter- commonly used drugs (eg, oral contraceptives
val. Indications for deferral may include recent and anticholesterol agents) that may be contra-
symptoms, limitations on activity or functional indicated in pregnancy but pose negligible, if
impairment resulting from unstable angina, re- any, risk to any transfusion recipient. In 2016,
cent myocardial infarction, left maln coronary FDA eliminated the risk categories and intro-
disease, ongoing congestive heart failure, or se- duced a new descriptive approach to prescrip-
vere aortic stenosis.3 tion drug labeling for risk to pregnant or breast-
feeding women, which may also make the
Medications inappropriate application to blood donor eligibil-
ity less of a problem."
The DHQ and Medication Deferral list contain Localmedication deferrals are often based on
the requirements for deferrals for specific medi- concerns about the reason for the potential do-
cations as stipulated by the FDA and AABB. nor's use of the medication and his or her under-
These requisite medication deferralsfallinto five lying medical condition rather than on any in-
broad categories: herent threat posed by residual medication in
the collected blood component. Most drugs
@ Potent teratogens that pose potential harm to used by donors pose no harm to recipients, and
unborn children (although there have been many factors should be considered when evalu-
no documented cases of adverse fetal out- ating the potential risk of a drug's use by a donor
comes related to transfusions from donors (eg, the medication's half-life, mean and peak
taking these medications). plasma concentration, residual concentration in
@ Antibiotics or antimicrobials to treat an infec- a blood component, and dilution when trans-
tion that could be transmitted through blood fused to a recipient).
transfusion (excluding preventive antibiotics
for acne, rosacea, and other chronic condi-
tions with a low risk of bacteremia).
Anticoagulants and antiplatelet agents that
affect component potency (plasmaor platelet
components only).
Although blood collection facilities may add

Exceptional Medical Need
medications whose use requires local donor de-
ferral to the Medication Deferral List, many In certaln limited clinical circumstances, a recip-
have chosen to use the Medication Deferral List ient may benefit from blood components collect-
ed from a specific donor. Such a recipient might with positive test results may be difficult to
be a patient with multiple antibodies or with an- maintain. Nevertheless, directed donations are
tibodies to high-prevalence antigens who re- sometimes sought by patients and their fami-
quires units from donors whose red cells are lies, particularly for neonatal and other pediat-
negative for the corresponding antigens. Fre- ric patients.
quent donation by a specific donor for a specific Directed donors must meet the same criteria
patient with a medical need requires that the as voluntary donors, and their blood can be used
blood collection facility have a procedure that for other patients if not needed by the individual
typically calls for both a request from the pa- for whom the donations were initially intended.
tient's physician and approval by the donor cen- The facility should clearly communicate its
ter's physician. The donor must meet all alloge- directed-donation procedures so that the expec-
neic donor selection requirements, with the tations regarding availability of directed-donor
exception of donation frequency, provided that units are known to the hospital, ordering physi-
he or she is examined and certified by a physician, and patient. The communication required
cian [21 CFR 630. 15(a)(1)(ii)(B)].In emergency includes defining the mandated interval be-
medical situations, blood components can be re- tween collection of the blood and its availability
leased before test results for RTTIsare available to the patient, mentioning the possibility that
provided that the units are labeled and managed the patient will identify donors who are not
in accordance with the CFR. Granulocyte com- ABO compatible or not otherwise acceptable
ponents are generally released this way, as they blood donors, and defining the policy for release
expire in 24 hours. Testing on the units must be of donor-directed units for transfusion to other
completed as soon as possible after release or patients.
shipment, and results must be promptly report-
ed to the hospital or transfusion service.' Autologous Blood Donations
Autologous donations have declined dramatical-
Directed Blood Donations
ly in the United States since the 1990s. Waning
The use of directed donors, that is, when pa- interest in autologous donations may reflect the
tients ask the blood center if they can designate decline in viral risk associated with allogeneic
their own blood donors (usually relatives or blood transfusion, the lower rate of surgical
friends) for their anticipated transfusion needs, transfusion generally, and, consequently, the
has decreased in recent years, but there is still minimal medical benefit and increased cost of
ongoing demand. The concern likely reflects an autologous blood.43.45 The most appropriate can-
inaccurate perception continuing among the didates are alloimmunized donors for whom
general public of the risk for RTTIs associated compatible blood is hard to collect, who are un-
with blood transfusion from the general invento- dergoing elective surgery for which transfusion
ry. Most facilities and hospitals accommodate will likely be required, and who have adequate
the associated collection, storage, tracking, pay- time before the procedure to replace the hemo-
ment, and logistical difficulties to provide a di- globin lost via phlebotomy.
rected donation service. In general, the use of preoperative autolo-
Directed donations have higher viral marker gous blood donation alone provides only a rela-
rates than volunteer donations, mostly but not tively small benefit in reducing the probabilityof
entirely reflecting the higher prevalence of first- allogeneic transfusion and may actually in-
time donors among the former group." There is crease the risk of lower postoperative hemato-
no evidence that directed donations are safer to crits. Preoperative autologous donations may
use than donations from volunteer community still be used in conjunction with other blood
donors. On the contrary, some concerns persist conservation methods, such as acute nor-
that directed donors may feel unduly pressured movolemic hemodilution, perioperative blood
to give blood, which could compromise blood recovery, and pharmacologic strategies (see fur-
safety. The confidentiality of directed donors ther discussion in Chapter 20).
Patients identified as candidates for autolo- Contraindications to autologous blood dona-

gous donation are evaluated by the donor center tion should be defined by the blood center and
as well as the referring physician. The following may include medical conditions associated with
criteria for autologous donations are specifiedby the greatest risk from blood donation, such as I)
unstable angina, 2) recent myocardial infarction
the FDA, MBB, or both:
or cerebrovascular accident, 3) significant cardi-
ac or pulmonary disease with ongoing symp-
A prescription or order from the patient's
toms but without an evaluation by the treating
physician. physician, or 4) untreated aortic stenosis." Both
Minimum hemoglobin concentration of the ordering physician and the donor center
11 gldL or hematocrit of 33%. physician need to carefully balance the risks of
Collection at least 72 hours before the antici- the collection procedure against any perceived
pated surgery or transfusion. benefit to the patient-donor. The FDA has issued
Absence of conditions presenting a risk of guidance on the process by which autologous
bacteremia. donations may be collected, making it clear that
Use only for the donor-patient if labeled "au- rules for allogeneic donors may not necessarily
tologous use only." be applied."
REFERENCES
1. Gammon R, ed. Standards for blood banks and 3. Eder AF, Goldman M, eds, Screening blood do-
transfusion services. 32nd ed. Bethesda, MD: nors with the donor history questionnaire.
MBB,2020. Bethesda, MD: MBB Press, 2019.
2. Code of federal regulations. Title 21, CFR Parts 4. lou S, Eder AF, Musavi F, et al. ARCNETStudy
600 to 799. Washington, DC: US Government Group. Implementation of the uniform donor
Publishing Office, 2019 (revisedannually). history questionnaire across the American
Red Cross Blood Services: Increased deferral creu tzfeld t -jako b -dis eas e -and -varian t-
among repeat presenters but no measurable im- creutzfeldt.]
pact on blood safety. Transfusion 2007;47: 1990- 11. Eder AF. Evidence-based selection criteria to
8. protect the blood donor. J Clin Apher 2010;25:
5. Fridey JL, Townsend M, Kessler D, Gregory K. 331-7.
A question of clarity: Redesigning the AABB 12. Eder AF, Goldman M, Rossmann S, et al. Selec-
blood donor history questionnaire-a chronolo- tion criteria to protect the blood donor in North
gy and model for donor screening. Transfus America and Europe: Past (dogma), present (evi-
Med Rev 2007;21 :181-204. dence), and future (hemovigilance). Transfus
6. Blood donor history questionnaires. Version 2.0. Med Rev 2009;23:205-20.
Bethesda, MD: AABB,2016. [Availableat http:// 13. Strauss RG. Rationale for medical director ac-
www.aabb.org/tm/questionnaires/Pages/dh ceptance or rejection of allogeneic plateletphere-
qaabb.aspx (accessed August 24,2018).] sis donors with underlying medical disorders.
7. Food and Drug Administration. Guidance for in- J Clin Apher 2002; 17:111-17.
dustry: Implementation of acceptable full-length 14. Reik RA, Burch ]W, Vassallo RR, Trainor L.
and abbreviated donor history questionnaires Unique donor suitability issues. Vox Sang 2006;
and accompanying materials for use in screen- 90:255-64.
ing donors of blood and blood components. 15. Food and Drug Administration. Questions about
(May 2016) Silver Spring, MD: CBEROffice of blood. Silver Spring, MD: CBEROffice of Com-
Communication, Outreach, and Development, munication, Outreach, and Development,
2016. [Available at https://www.fda.gov/ 2016. [Available at https://www.fda.gov/vac
downloads/BiologicsBloodVaccines/Guidance cines-blood-biologics/blood-blood-products/
ComplianceRegulatoryInformation/Guidances/
questions-about-blood]
Blood/UCM273685.pdf.]
16. Standards portal. Bethesda, MD: AABB,2019.
8. Food and Drug Administration. Guidance for In-
[Available at http://www.aabb.org/sa/Pages/
dustry: Revised recommendations to reduce the
Standards-Portal.aspx (accessed September 18,
risk of transfusion-transmitted malaria. (April
2019).]
2020) Silver Spring, MD: CBEROffice of Com-
17. Cable R, Musavi F, Notari E, Zou S. ARCNET
munication, Outreach, and Development,
2020. [Availableat https://www.fda.gov/regu Research Group. Limited effectiveness of donor
latory- information/ search -fda-guidance-docu deferral registries for transfusion-transmitted dis-
ments/revised-recommendations-reduce-risk- ease markers. Transfusion 2008;48:34-42.
transfusion-transmitted-malaria.] 18. Cable RG, Steele WR, Melmed RS, et al for the
9. Food and Drug Administration. Guidance for In- NHLBIRetrovirus Epidemiology Donor Study-Il
dustry: Revised recommendations to reduce the (REDS-II).The difference between fingerstick
risk of human immunodeficiency virus transmis- and venous hemoglobin and hematocrit varies
sion by blood and blood products. (April2020) by sex and iron stores. Transfusion 2012;52:
Silver Spring, MD: CBEROffice of Communica- 1031-40.
tion, Outreach, and Development, 2020. [Avail- 19. Cable RG. Hemogiobin determination in blood
able at https://www.fda.gov/regulatory-infor donors. Transfus Med Rev 1995;9: 131-44.
ma ti on/ se arch -fda -guidanc e-documen ts/ 20. Cable RG, Glynn SA,KissJE, et al, for the NHL-
revised-recommendations-reducing-risk-human- BI Retrovirus Epidemiology Donor Study-Il
immunodeficiency-virus-transmission-blood- (REDS-II).Iron deficiency in blood donors: Anal-
and-blood.] ysis of enrollment data from the REDS-IIDonor
10. Food and Drug Administration. Guidance for In- Iron Status Evaluation (RISE)study. Transfusion
dustry: Revised recommendations to reduce the 2011;51:511-22.
risk of transmission of Creutzfeldt-Jakob disease 21. Beutler E, Waalen J. The definition of anemia:
and variant Creutzfeldt-Iakob disease by blood What is the lower limit of normal of the blood
and blood components. (April 2020) Silver hemoglobin concentration? Blood 2006; 107:
Spring, MD: CBER Office of Communication, 1747-50.
Outreach, and Development, 2020. [Available 22. Simon TL, Garry PJ, Hooper EM. Iron stores in
at https://www.fda.gov/regulatory-informa blood donors. JAMA 1981;245:2038-43.
tion/search-fda-guidance-documents/recom 23. Cable RG, Glynn SA,KissJE, et al. Iron deficien-
mendations-reduce-possible-risk-transmission- cy in blood donors: The REDS-IIDonor Iron Sta-
tus Evaluation (RISE)study. Transfusion 2012; cardiac disease are not necessarily caused by
52:702·11. blood donation. Transfusion 1998;38:669-73.
24. Updated strategies to limit or prevent iron defi- 35. Mann M, Sacks HJ, Goldfinger D. Safety of au-
ciency in blood donors. Association bulletin tologous blood donation prior to elective sur-
#17-02. Bethesda, MD: AABB,2017. [AVailable gery for a variety of potentially high risk pa-
at http://www.aabb.org/programs/publica tients. Transfusion 1983;23:229-32.
tions/bulletins/Docsl ab17-02.pdf#search=as 36. KlapperE, Pepkowitz SH, Czer L, et al. Confir-
sociation%20bulletin%20iron (accessed August mation of the safety of autologous blood
24,2018).J donation by patients awaiting heart or lung
25. Bialkowski W, Bryant BJ, Schlumpf KS, et al. transplantation: A controlled study using hemo-
The strategies to reduce iron deficiencyin blood dynamic monitoring. J Thorac Cardiovasc Surg
donors randomized trial: Design, enrollment 1995;110:1594-9.
and early retention. Vox Sang 2015; 108: 178- 37. Dzik WH, FleisherAG, CiavarellaD, et al. Safe-
85. ty and efficacyof autologous blood donation be-
26. Eder AF, KissJE. Adverse reactions and iron de- fore elective aortic valve operation. Ann Thorac
ficiencyafter blood donation. In: SimonTL, Mc- Surg 1992;54: 1177-80.
Cullough J, Snyder EL, et al, eds. Rossi'sprinci- 38. PopovskyMA, Whitaker B, Arnold NL. Severe
ples of transfusion medicine. 5th ed. Chichester, outcomes of allogeneic and autologous blood
UK:John Wiley and Sons, 2016:43-57. donation: Frequency and characterization.
27. KissJE, BrambillaD, Glynn SA,et al, for the Na- Transfusion 1995;35:734-7.
tional Heart, Lung, and BloodInstitute (NHLBI) 39. EderAF, Dy BA,KennedyJ, et al. The American
Recipient Epidemiology and Donor Evaluation Red Cross donor hemovigilanceprogram: Com-
plications of blood donation reported in 2006.
Study-III (REDS-III).Oral iron supplementation
Transfusion2008;48: 1809-19.
after blood donation: A randomized clinical tri-
40. WiltbankTB, GiordanoGF, KamelH, et al. Faint
al.JAMA2015;313:575-83.
and prefaint reactions in whole-blood donors:
28. Mast AE, BialkowskiW, Bryant BJ,et al. A ran-
An analysis of predonation measurements and
domized, blinded, placebo-controlledtrial of ed-
their predictive value. Transfusion 2008;48:
ucation and iron supplementation for mitigation 1799-808.
of iron deficiency in regular blood donors. 41. Food and Drug Administration. Content and for-
Transfusion 2016;56: 1588-97. mat of labeling for human prescription drug and
29. Custer B, Schlumpf KS,Wright D, et al. NHLBI biologicalproducts; requirements for pregnancy
Retrovirus EpidemiologyDonor Study-Il, Donor and lactation labeling; final rule. Title 21, CFR
return after temporary deferral. Transfusion Part 201. (December 4,2014) Fed Regist2014;
2011 ;51:1188-96. 79:72063-103. [Availableat https:llwww.fed
30. Eder AF. Blood donors with a history of cancer. eralregister .gov1 documents/20 141 121041
In: Eder AF, Goldman M, eds. Screening blood 2014-282411 content-and-format-of-labeling-for-
donors with the donor history questionnaire. human-prescription-drug-and-biological-prod
Bethesda, MD: AABBPress, 2019:63-78. ucts-requirements-for.J
31. Edgren G, Hjalgrim H, ReillyM, et al. Risk of 42. Dorsey KA,Moritz ED, SteeleWR, et al. A com-
cancer after blood transfusion from donors with parison of human immunodeficiency virus,
subclinicalcancer: A retrospective cohort study. hepatitis C virus, hepatitis B virus and human
Lancet 2007;369: 1724-30. 'lIymphotropic virus marker rates for directed
32. HjalgrimH, RostgaardK,Vasan SK,et al. No ev- versus volunteer blood donations to the Ameri-
idence of transmission of chronic lymphocytic can Red Cross during 2005 to 2010. Transfu-
leukemia through blood transfusion. Blood sion 2013;53:1250-6.
2015;126:2059-61. 43. Brecher ME, Goodnough LT.The rise and fallof
33. AHA Statistics Committee and Stroke Statistics preoperative autologous blood donation. Trans-
Subcommittee. Heart disease and stroke statts- fusion 2002;42: 1618-22.
tics-20 16 update: A report from the American 44. SchvedJF. Preoperative autologous blood dona-
Heart Association. Circulation 2016;133:e38- tion: A therapy that needs to be scientifically
360. evaluated. Transfus Clin Biol2005; 12:365-9.
34. KasperSM, Ellering I, Stachwitz P, et al. All ad- 45. Goodnough LT.Autologousblood donation. An-
verse events in autologous blood donors with esthesiol Clin North Am 2005;23:263-70.
46. Food and Drug Administration. Guidance for in- policy. (August 2016) Silver Spring, MD: CBER
dustry: Determining donor eligibilityfor autolo- Office of Communication, Outreach, and Devel-
gous donors of blood and blood components in- opment, 2016.
tended solely for autologous use-compliance
Whole Blood and Apheresis Collection
of Blood Components Intended for
Transfusion
Jason Acker, MBA, PhD, and Anna Razatos, PhD
LOOD HAS BEEN COLLECTEDAND N R PR N N

transfused for over 100 years. 1 This AR
chapter describes current methods avail-
able for collecting, preparing, storing, and modi- Donor Consent
fying blood components, specificallyRed Blood
Cells (RBCs),platelets, and plasma, for transfu- Potential donors must be provided with predo-
sion according to AABBand international stan- nation education, counseling about the blood
dards. Detailed descriptions of the various donation process, and an opportunity to have
blood components can be found in the Circular their questions aIlsweredbefore every blood do-
of Information for the Use of Human Blood nation. Per AABB Standards for Blood Banks
and Blood Components/ Blood components and Ttsnsfusion Services (Standards) and the
can be manually separated in the laboratory Code of Federal Regulations (CFR), blood cen-
from whole blood (WB) collected from blood ters are required to obtain donors' written ac-
donors. Recent technological advances have in- knowledgment of the following elements [AABB
troduced automated methods to separate blood Standards 5.2-5.4 and Title 21 CFR, Part
components from WB in the laboratory. Aphere- 630.10(g)j3,4(PP1S.17):
sis devices can also be used to separate and col-
lect .blood components directly from the donor The donor has been provided and has re-
and return the remaining portions. "Hemapher- viewed information regarding the risks and
esis" or "apheresis" refers to automated blood hazards of the specificdonation procedure.
component collection procedures and is derived The.donor has reviewed the educational ma-
from the Greek word "aphairos," meaning "to terial regarding relevant transfusion-transmit-
take from." Blood centers use a combination of ted infections.
WB collections and apheresis to meet transfu- A sample of the donor's blood is tested for
sion demands. specified transfusion-transmitted pathogens.
Jason Acker, MBA, PhD, Senior Research SCientist,Canadian Blood Services, and Professor, Laboratory Medicine
and Pathology, University of Alberta, Edmonton, Alberta, Canada; and Anna Razatos, PhD, Senior Medical Sci-
ence Liaison, Terumo BCT, Lakewood, Colorado
141
If the donation is determined to be unsuit- data are associated with the correct DIN and
able or if the donor is deferred from dona- blood components.
tion, the donor's record will identify the do-
nor as ineligible to donate, and the donor Vein Selection and Disinfection
will be notified of the basis for the deferral Methods for the Venipuncture Site
and the period of deferral.
The phlebotomist inspects both arms of the do-
The donor has the opportunity to ask ques-
nor to select a prominent, large, firmvein in the
tions and withdraw from the donation proce-
antecubital fossa to permit a single, readily ac-
dure.
cessible phlebotomy site that is devoid of scar-
ring or skin lesions.
In addition, the CFRrequires the blood cen-
Specificinstructions in the package insert for
ter's responsible physician or appropriate desig-
the use of approved agents should be followed
nee to obtain donors' informed consent for plas-
for phlebotomy site disinfection. These meth-
mapheresis and plateletpheresis collections. The
ods provide surgical cleanliness, but none of the
physician or appropriate designee should ex-
methods can achieve an absolutely aseptic site.
plain the risk of the procedure to the donor, pro-
Approximately 50% of donors had no bacterial
vide an opportunity for the donor to ask ques-
colonies in studies using a contact plate culture
tions, and give the donor a chance to refuse to of the venipuncture site after disinfection with
donate. The informed consent process for plate- povidone iodine or isopropylalcohol plus iodine
letpheresis must be performed before the first tincture, whereas the remaining donors had low
donation and annually thereafter [21 CFR Parts colony numbers (1 to 100).s Rarely(1%) did do-
640.21 (g) and 630.5]. Likewise, the informed nors have more than 100 colonies after arm dis-
consent process for plasmapheresis stipulates infection." Bacteria residing deep within skin
these requirements, and that the process be re- layers are not accessible to disinfectants and
peated if more than 6 months elapse between may contribute to product contamination. In
plasmapheresiscollections (21 CFR630.15). one study, pigskin epidermal cells were detect-
able in the lavage fluid in lout of 150 punc-
Donor Eligibility and Identification tures.? Diversionof at least the first 10 milliliters
Phlebotomy must be performed only after the of blood into a special diversion pouch can cap-
donor has been found to be eligiblefor blood do- ture skin debris and has been shown to reduce
nation. Identification of blood components and the proportion of platelet components contain-
maintaining test results linked to the donor are ing viable bacteria.4(p22),7-9 Blood in this pouch
critical to ensure recipient safety and to permit can be used for laboratory testing.
look-back investigations and product withdraw-
als if indicated. The blood component identifica- Donor Care after Phlebotomy
tion process uses both a bar-coded and an eye- Immediately after collection, the needle is with-
readable unique donation identification number drawn into a protective sleeve to prevent acci-
(DIN) that is assigned to the donation record, dental injuries. Localpressure is appliedby hand
donor history questionnaire, each sample tube, to the gauze placed directly over the venipunc-
and each component prepared from the dona- ture site while the donor's arm is kept elevated.
tion. Electronic records of the donation are as- Pressure is applied until hemostasis is achieved
signed the same number. The DIN should be and a bandage or tape may be applied.
verified on the donation record, collection pri- MBB Standard 5.3.3 requires blood centers
mary and secondary containers, and sample to provide donors with written instructions
tubes before the blood collection can proceed. A about care after donation.4(p17) Postphlebotomy
final check of appropriate labeling before phle- care includes observing the donor for signs or
botomy helps to ensure that the donor history symptoms of reactions. If donors tolerate a sit-
data, laboratory data, and other manufacturing ting position without problems, they may pro-
C HAP T E R 6 WBand ApheresisComponents 143
ceed to a recovery area and should be encour- than bruises, occurring in 1.7% of donors."
aged to drink fluids and have light snacks and Bruises and minor hematomas (less than 2 by 2
remain in the recovery area for about 15 min- inches) generally do not prevent donors from
utes or until they feel comfortable to leave. In donating again,"
addition, blood centers may encourage the do- Local Nerve Injury. Phlebotomy-related
nor to drink more fluid and refrain from heavy nerve injuries are relatively uncommon but still
lifting or vigorous exercise or activities that inevitably occur even with good phlebotomy
might put the donor or others at risk for several technique because of anatomic variation and the
hours after blood donation. The donor is also in- close association of nerves with veins. Donors
structed to apply local pressure to the phleboto- may complain of sensory changes away from the
my site if any bleeding recurs and to call the phlebotomy site, such as in the forearm, wrist,
blood center if the bleeding does not stop with hand, upper arm, or shouider. These injuries are
pressure. Appropriate contact information is pro- usually transient, followed by full recovery."
vided so that the donor can report if he or she However, in 7% of injured donors, recovery may
feels that the donated unit should not be used, take 3 to 9 months." In severe cases, referral to
has any reactions, or experiences any signs or a neurologist may be indicated.
symptoms of infection. Arterial Puncture. Arterial puncture is a
rare event occurring in less than 1 in 10,000 do-
Adverse Donor Reactions nations." Presence of bright red blood, rapid
Adverse reactions can occur at the time of dona- collection (within 4 minutes), and a pulsating
tion or after the donor has left the blood center. needle suggest arterial puncture, although not
In a comprehensive donor hemovigilance pro- all signs might be present." Hematomas are
gram reported by the American Red Cross, ad- more likely to occur with arterial punctures.
verse reactions were reported for WB collec- When puncture is recognized early, the needle
tions (349 in 10,000), plateletpheresis (578 in should be pulled out immediately, and local
10,000), and double RBCunit collections (538 pressure should be applied for an extended peri-
in 10,000), the vast majority of which were mi- od. Most donors recover quickly, but some
nor presyncopal reactions and small hemato- might present with waxing and waning hemato-
mas." Serious adverse reactions were slightly mas, a mass that should be evaluated, or a pseu-
more common for WB collections (7.4 in doaneurysm.
10,000) compared with plateletpheresis (5.2 in
10,000) and double RBCunit collections (3.3 in Systemic Reactions
10,000).!0 Reactions that needed medical care Vasovagal Reactions. Vasovagalreactions (also
after the donor left the donation site occurred in referred to as pre-faint or presyncope) include
roughly 3 in 10,000 donations." A population- dizziness, sweating, nausea, vomiting, weak-
based European study found the rate of compli- ness, .apprehension, pallor, hypotension, and
cations leading to long-term morbidity or dis- bradycardia." The reaction might progress to
ablement to be 0.5 in 10,000 donations and syncope (lossof consciousness);convulsions and
0.23 in 10,000, respectively." All adverse reac-
loss of bladder and bowel function might also
tions occurring during collection procedures
occur. Syncope can also result from orthostatic
must be documented along with the results of
blood pressure changes after donation. Vasova-
thorough investigations.
gal reactions are distinguished by a low pulse
rate, whereas reactions related to volume deple-
Needle-Related Injuries
tion are associated with an increased pulse rate.
Bruise or Hematoma. Bruises are the most This difference, however, has no practical value,
common adverse event after phlebotomy, occur- as both mechanisms are treated similarly. In
ring in 23% of donors based on postdonation in- case of vasovagal reaction, phlebotomy should
terviews." Hematomas, defined as the accumu- be stopped, and the donor should be placed in a
lation of blood under the skin, are less common recumbent position. Applying cold wet towels
to the donor's neck and shoulders and loosening Fatalities Due to Blood Donation
the donor's clothes can assist in symptom man- The Food and Drug Administration (FDA) re-
agement. Some donors with severe reactions or quires blood establishments to report deaths as-
with prolonged recovery times may need short- sociated with blood donation. According to the
term observation or intravenous fluid adminis- FDA, in 2015, allogeneic blood donations ac-
tration in an emergency room. Telephone counted for 12.0 million WB-derivedand apher-
follow-up for donors who have experienced se- esis RBCcomponents, 2.4 million platelet com-
vere reactions is helpful to assess the donors for ponents, and 3.7 million plasma components,
any residual symptoms. Donor reactions after whereas in 2016, there were 38.3 million
WB donation do not accurately predict the pos- source plasma donations." Fatalities associated
sibilityof recurrent syncope in returning donors, with blood donation are extremely rare. Over
although they reduce the likelihood of future the 5-year reporting period of 2013 to 2017,
donations.13 there were 47 reported donation-associated fa-
Most reactions occur at the collection site, ei- talities, with seven cases since 2014 having an
ther in the donation chair or the recovery area." imputability of definite/certain, probable/likely,
The main predictors of immediate and delayed or possible."
vasovagal and presyncopal reactions are young
age, low estimated blood volume (<3.5 L), fear, Therapeutic Phlebotomy
and first-time donation status.14-17 In donors ex- Therapeutic phlebotomy is a treatment for blood
periencing a reaction, loss of consciousness is disorders, such as hemochromatosis, polycythe-
most concerning as it can lead to injury, espe- mia vera and other conditions associated with
cially if the donor has left the donation site." erythrocytosis such as testosterone supplemen-
Blood Systems, Inc, reported a rate of syncope ration," and certain porphyrias, in which the re-
during and after phlebotomy as 27 per 10,000 moval of red cells or reduction of iron stores is
WB donations, with an associated injury rate of an effective method for managing the dis-
1.3 per 10,000 WB donations." Approximately ease.23,24The CFR [21 CFR 630. 15(a)(2)] and
10% of loss of consciousness reactions occur af- AABB Standard 5.6.7.1 specify that units col-
ter the donor leaves the donation site.14Deferral lected as therapeutic phlebotomies can be used
strategies for young donors with estimated low for allogeneic transfusion when the individual
blood volume and physiologicstrategies to mini- donor meets other allogeneic blood donor crite-
mize donor reactions are aimed at improving ria.4(p24)
Generally,units from therapeutic donors
donor safety.1S,18 Donor education, environmen- intended for allogeneic transfusion must be la-
tal controls, instructions to donors to drink fluid beled with the disease or condition requiring
before and after donation, distraction, and mus- therapeutic phlebotomy. However, allogeneic
blood donation is accepted, for example from in-
cle tension have been identified as strategies to
dividuals with hereditary hemochromatosis or
reduce reactions in young donors.P'" Deferral
individuals with erythrocytosis from testoster-
of low-blood-volume (less than 3.5 L) donors one therapy, if the FDAhas approved an "excep-
may be helpful in reducing the risk of reactions, tion or alternative procedure" under 21 CFR
especiallyin young donors. 640.120. Such variances can waive both the la-
Citrate Reactions. During apheresis, blood beling requirement and limitations on the fre-
that is anticoagulated with citrate in the aphere- quency of donation.
sis systems is returned to the donor at an accept- Treatment by phlebotomy primarily consists
able rate." In healthy individuals, citrate is rap- of donating 500 mL of WB, although donation
idly metabolized, but some donors can of RBCsby apheresis can also be considered."
experience mild citrate reactions (paresthesias Several studies have demonstrated that blood
or tingling sensattons)." Oral calcium supple- collected from stable or uncomplicated hemo-
mentation is advised for symptomatic manage- chromatosis patients is safe for transfusion in
ment of hypocalcemia." terms of transfusion-transmitted infections.26-28
Hemochromatosis.is a genetic disorder char- the manufacturer's specifications for allogeneic

acterized by absorption of excess iron that can collections should be discarded. Low-volume al-
accumulate in tissues and organs, potentially re- logeneic collections should be relabeled as
sulting in toxicity and organ damage.24,29Treat- "RBCsLow Volume." WB-derived RBCsthat are
ment of hemochromatosis by regular phleboto- labeled as low-volume units are made available
my involves an iron depletion phase to lower for transfusion when 300 to 404 mL ofWB is
serum ferritin concentration to an acceptable collected into an anticoagulant volume calculat-
level, followed by a maintenance phase to keep ed for 450 ± 45 mL, or when 333 to 449 mL of
serum ferritin levels low.29,3oIn polycythemia WB is collected into an anticoagulant volume
vera and other conditions associated with eryth- calculated for 500 ± 50 mL (AABBStandard
rocytosis, phlebotomy is used to reduce the risk 5.7.4.7).4(P27)Evidence indicates that the vol-
of venous thromboembolism that is associated ume in undercollected and overcollected units
with hyperviscosity. The frequency required is (275-600 g) does not affect in-vivo red cell re-
titrated to the hematocrit of the patient, at- covery even after 21 to 35 days of storage.33,34
tempting to maintain levels ::;52%and ::;48%in Plasma and platelets from low-volume units
males and females, respectively." should be discarded.
The average time to collect 500 mL of WB is
less than 10 minutes. A draw time longer than
BLO D LECTiON 15 to 20 minutes may not be suitable for collect-
ing platelets or plasma for transfusions, as deter-
Whole Blood Collection mined by blood center policy. The collection bag
MBB Reference Standard 5.4.1A permits col- should be periodically mixed during the collec-
lection of 10.5 mL of blood per kilogram of the tion to ensure uniform distribution of anticoagu-
donor's weight for each donation, including the lant.
blood unit and all samples for testing.4(PP62.69)
In WB is collected into sterile, plastic bags
North America and Europe, the volume of WB containing anticoagulant. Bags are typically
collected during routine phlebotomy is typically plasticized polyvinyl chloride (PVC).35Typical
either 45Q ± 10% (405-495 mL) or 500 mL ± anticoagulants would include citrate-phosphate-
10% (450-550 mL). The volume may be differ- dextrose (CPD), citrate-phosphate-dextrose-
entin other.regions and may be as low as 200 to dextrose (CP2D), or citrate-phosphate-dextrose-
250 mL. For general blood banking applications, adenine (CPDA-l). WB in acid-citrate-dextrose
the volume of WB collected can be determined (ACD), CPD, or CP2D has an expiration date of
from the net weight in grams collected divided 21 days when stored at 1 to 6 C; the maximum
by the density ofWB (1.053 g/ml.]." Donors in storage time for units in CPDAI is 35 days
the United States must weigh a minimum of (MBB Reference Standard 5.1.8A).4(pP52'61)
50 kg (110 lb) with a minimum hemoglobin (or Blood bags should list an expiry date and in-
hematocrit) of 12.5 g/dl, (38%) for women and clude label data required by regulatory authori-
13.0 gldL (39%) for .men (MBB Standard ties. Sterile collection systems can contain inte-
5.4.1A).4IPp6269)The CFRallows blood collection grally attached tubing to allow aseptic 'fluid
from females with hemoglobin (or hematocrit) transfer to satellite containers for component
of 12.0 to 12.5 g/dl, (36-38%), provided blood preparation, as well as integral access ports for
centers take additional steps to ensure that the open connection ofinfusion sets or other spiked
health of the donor is not adversely affected [21 entry. Open spiked access reduces the expira-
CFR 630.1 0(f)(3)(i)(A)].Donor qualification cri- tion from the time of entry to reduce the risk of
teria vary around the world depending on local bacterial sepsis. Innovations in WB collection in-
regulatory requirements. clude scales for monitoring the collection vol-
Collection volumes must be within the man- ume with automatic mixing and devices to add
ufacturer's specified range to ensure the correct anticoagulant at a fixed ratio as the blood is
anticoagulant-to-WB ratio. Volumes exceeding withdrawn from the vein.
The acceptable temperature during short- low component manufacturing to take place in a
term storage, transport, and handling of WB im- closed system. The blood container should not
mediately after collection is determined by probe entered before issue except for the purposes
cessing requirements for component prepara- of sample collection, postcollection processing,
tion. In some cases, this may mean that or transfer of components to a different contain-
collections at mobile blood drives or fixed col- er. Components prepared with an open system
lection sites should be transported as soon as require a reduction in their expiration time from
possible to the central component preparation the time that the system was opened to reduce
laboratory. With other processing methods, the risk of bacterial contamination. The expira-
transportation may not be as urgent. Require- tion period of an open system is 24 hours for
ments for cooling and transportation methods RBCsstored at 1 to 6 C; 4 hours for WB, RBCs,
are quite variable, and the specifications of the and platelets held at room temperature; or 4
appropriate device manufacturer should be care- hours for thawed cryoprecipitate and plasma
fully followed. WB destined for platelet prepara- (AABBReference Standard 5.1.8A).4(ppS2-61) The
tion should not be cooled to less than 20 C. In use of approved sterile connection devices main-
the United States, platelets must be separated tains a functionally closed system when various
from WB within 8 hours of collection (AABB connections are performed, such as pooling or
Reference Standard 5.1.8A).4(pPS2-61)It can take sampling, thereby maintaining the component's
10-16 hours from time of collection for a unit of original expiry date (AABBStandard 5.7.2).4(p24)
blood to reach 20 C when packed in a crate at All current methods for separation and prepa-
20-24 C.36 To cool blood more rapidly, units are ration of the three major blood components-
typically placed in specific storage environ- RBCs, platelets, and plasma-rely on one or
ments, or some centers use cooling plates that more centrifugation and expression steps. Cen-
provide rate-controlled cooling toward 20 C. trifuges and both manual and automated expres-
These cooling plates contain 1,4-butanediol, sion devices should be properly validated,
which has a melting temperature of 20 C and maintained, and calibrated, or checked in a sys-
serves as a heat absorber. With the cooling tematic manner to verify the processing condi-
plates, about 2 hours are needed for the collect- tions (AABBStandard 3.5.1).4(pS)
ed blood to reach 20 C.36 WB that will not be Following separation by centrifugation, com-
used to prepare platelets should be cooled to re- ponents must be carefully divided into separate
frigerator temperature as soon as possible;this is containers for further processing. Many labora-
often accomplished by placing the unit on wet tories use manual extractors for this purpose, in
ice or other appropriate cooling media. which case the component laboratory staff iden-
Tests for ABOgroup, Rh type, unexpected al- tifieswhen the red cell interface approaches the
loantibodies, and transfusion-transmitted infec- tubing and stops expression using a hemostat.
tions are performed. Each collection must be Bloodcomponent extractors are availablefor au-
tested unless the donor is undergoing repeated tomating the separation of WB components by
procedures to support a specificpatient-for ex- detecting the component interfaces and auto-
ample, for some apheresis procedures, testing matically stopping the extraction process. After
for infectious disease markers may be required primary centrifugation, WB is placed in the ex-
only at 30-dayintervals [21 CFR 610AO(c)]. tractor, and a pressure plate creates an outflow
of components from the container. Outflow can
Component Preparation Methods from occur from the top and/or bottom of the con-
tainer, depending on the method of manufactur-
Whole Blood
ing that is used at the blood center.
Blood collection and component separation sys- Component processing methods from WB
tems may be designed in different configura- are typically defined based on the method used
tions depending on the needs of the blood man- to separate the platelets from the WB.37,38For
ufacturer (Fig6-1).37Satellite bags and integrally preparation of platelets from WB, two primary
attached tubing that are hermetically sealed al- methods are available:preparation from platelet-
"0
o
o
:0
..!E
o
.s:
:s:
E
-e
B
c::
a.>
c::
o
c.
E
o
(..)
C<j
E
~
C5..
"0
c::
C<j
"li)"
r::: ~
0 C5..
+i-
III~ Q)tI Ci5
.l:i'£: '08
(..)
:::ti
I.L.J!!
1:._
~al
Q)
>
,_
+=i+=
C
0
~
"0
e
tI_ a.>
"OE
010
tI 0
~(/)
(..)
::l
o "0
0:11
_0. (\ CIl ec.
-
er::
al~ o
ale.
-0
~
()) C ,Q
C "0
a.>
ot. )',Q 1il CIl
::l
.r::: Ut).l:l
o~::: CIl
"0
~ 'StiLL
(})(});"
o
.JO::co
~
E
CIl
()) C C ::l
)',Q ,Q
U~ ... o
o U III
~:J'p
.~
~tI:::
w(})LL
-
C. a.>
«I .JO::;" er:: .s:
......
E il co c.
o
1/1 III
! '0
I:.
...... ~
(..)
Il. ;:: E
a.>
.r::: ..c::
(..)
(J CIl
i2 o ~
a.>
>
....I III
er:: - 0
me
())
~c::
! (}) C
,5 0
(/)0 a.>
....Q) >
:as 0 CIl
ec.
! '0_
'0 0 a.>
Il. ~(/)
CC
rich plasma (PRP) or preparation from a buffy pooled with 1 unit of plasma and then centri-
coat. Both methods involve centrifugation, and fuged under a low gforce (ie, soft or light spin)
parameters must be tailored for the blood bag to separate the platelet concentrate for additional
system and the method of platelet preparation to processing, such as leukocyte reduction. These
ensure that safe, high-quality products are pro- processes are often associated with extended
duced. holding of the WB and buffy coats for 8 to 24
Platelet-Rich Plasma Manufacturing hours at 20 to 24 C.42,43 Hold times are generally
Method. Preparation of components from PRP adjusted to ease the operational logisticsfor the
begins with a "soft spin" of the WB to separate blood center. Compared to the PRPpreparation
red cells from the PRP and is followed by ex- method, the buffy-coatmethod yieldsmore plas-
pression and hard spin of the PRPto concentrate ma, greater red cell loss, better initial white
platelets. This is done either manually, using blood cell reduction before filtration,and moder-
semiautomated extractors, or with fully auto- ate reduction in viable bacteria in the platelets
mated systems. Plasma for further processing is that interact with leukocytes.Y"
removed from the platelet pellet, which is held Whole-Blood Filtration Manufacturing
undisturbed for 0.5 to 2 hours before being re- Method. When platelet components cannot be
suspended in the residual plasma.":" The red produced from a WB unit due to logisticalrea-
cell concentrate from the PRP method can be sons or product demand, the WB unit may be
further leukocyte-reducedby inline filtration and cooled to 1 to 6 C within 8 hours of collection
stored in an approvedred cell additivesolution. and subsequently processed into red cell and
Singlewhole-blood-derivedplatelet units pre- plasma components. Leukocyte reduction by fil-
pared using the PRP must contain ::::5.5x 1010 tration of the anticoagulated WB can be per-
platelets in 90% of units sampled, must have a formed before centrifugation, followed by man-
pH ::::6.2 (AABB Standard 5.7.4.20) at out- ual or semiautomated separation of the red cell
date,4(p29)and are usually suspended in 40 to 70 and plasma components. Methods to separate
mL of plasma. Studies have shown acceptable the red cells from the plasma should result in an
recovery and survival rates when platelets are RBC component with a hematocrit of ::;80%
stored in plasma volumes of 35 to 40 mL.40,41 (AABBStandard 5.7.4.1.1 ).4(P26) Additive solu-
Four to 6 units of platelets are typicallypooled tions may be used to extend the storage of the
to create one therapeutic dose, which is labeled RBCs. Prestorage leukocyte reduction by filtra-
and stored in plasma in an approved container. tion of WB using platelet-sparingfiltershas been
Pools must be transfused within 4 hours (AABB used to produce leukocyte-reduced RBC,plate"
ReferenceStandard 5.1.8A).4(pp52'61)Platelets pre- let, and plasma components that meet FDAcri-
pared by the PRP method can be filtered to re- teria for product quality."
duce leukocytes as single PRP units or as a
pooled platelet concentrate using a leukocyte re- Automated Production of Components from
duction filter. Leukocyte-reduced pooled plate- Whole Blood
lets must have a residual leukocyte count of <5
x 106per transfusable dose in the United States Automated production of blood components can
or < 1 x 106 in Canada and Europe, as well as a improve the standardization of components. Au-
pH of ::::6.2at the end of storage (AABBStan- tomated devices control the rate of extraction,
dard 5.7.4.20).4(p30) detect an interface with an optical sensing de"
Buffy-Coat Manufacturing Method. The vice, provide clamping and sealing of tubing,
buffy-coatmethod is employed in many coun- monitor component weights, add storage solu-
tries but is not approved for use in the United tions, and perform other useful functions that
States. In short, non-leukocyte-reduced WB assist in the consistent preparation of blood
units are first centrifuged under a high g-force components. Some systems combine all of these
(ie, hard or heavy spin), and plasma, red cells, functions, including centrifugation, without reo
and buffycoats are separated for further process- lying on operator interventions. These fully au"
ing. Individual buffycoats from 4 or 5 units are tomated systems for the preparation of RBCs
and plasma and platelet concentrates from WB Similar to WB, RBCs collected by apheresis
are used in Europe and other international sites can be stored in additive solutions to extend the
and was recently approved for use in the United shelf life of the components. Storage solutions
States. The quality of RBC,platelet, and plasma can be added manually or by the apheresis de-
components produced using automated produc- vice after RBCcollection.
tion systems have been shown to meet quality
spedflcations.P'" Platelet Apheresis
In the United States, the use of apheresis plate-
Apheresis Colledion of Blood
lets has been steadilyincreasing over the past 25
Components
years. It is estimated that 92% of platelets trans-
Apheresis devices are continuous systems that fused in the United States are apheresis plate-
remove blood from a donor, separate the blood lets." Apheresis devices are designed to collect
into the desired components, and return the re- single, double, or triple units of platelets from in-
maining blood back to the donor. Apheresis al- dividual donors depending on donor characteris-
lows for the concurrent collection of RBCs, tics. Apheresis platelets can be collected concur-
platelets, and plasma from a. single donor, de- rentiy with single RBCunits and/or plasma.
pending on the regulatory approval of each indi- AABBStandard 5.7.4.21 requires that plate-
vidual device. The type or combination of com- lets collected by apheresis contain at least 3 x
ponents that can be collected by an apheresis 1011 platelets in 90% of units tested.4(p30) Units
device from a single donor is dependent on do- containing less than 3.0 x lOll platelets should
nor characteristics such as height, weight, sex, be labeled with the actual platelet count."
platelet count, and hematocrit or hemoglobin. Apheresis platelet donors may donate more
Apheresis devices employ software algorithms frequently than WB donors but must meet all
that determine donor eligibilitywhile maintain- other criteria for WB donation. The interval be-
ing donor safety. tween donations should be at least 2 days, and
The collection of components by apheresis donors should not undergo plateletpheresis
follows many of the same rules and standards more than twice in a week or 24 times in a
that apply to WB donation. Anticoagulants ap- rolling 12-month period (AABB Standard
proved for use with apheresis devices include 5.5.3.1 )_4(P20)
If a unit of WB is collected or if it
ACD formula A (ACD-A)or ACD formula B becomes impossible to return the donor's red
(ACD-B).Although the apheresis collection and cells during plateletpheresis, at least 8 weeks
preparation processes differ from WB-derived should elapse before a subsequent plateletphere-
components, the storage and transportation resis procedure unless the red cell extracorporeal
quirements and several quality-control steps are volume is less than 100 mL (AABBStandard
the same for both processes. 5.5.3.2).4(p20)Platelets may be collected more
frequently from donors if there is exceptional
Red CellApheresis medical need for a specific recipient and the
Roughly 15% of RBCsin the United States are blood center's responsible physician determines
collected by apheresis vs WB.49RBCs collected that the health of the donor will not be adverse-
by apheresis contain at least 60 g of hemoglobin ly affected by the collection. Donors who have
or 180 mL red cell volume per unit (AABBStan- taken antiplatelet medications that irreversibly
dard 5.7.4.8).4(p27)
Apheresisdevicesare approved inhibit platelet function are deferred for specific
to collectRBCsin the followingcomblnations'P: intervals before donation (48 hours for aspirin/
aspirin-containing medications and piroxlcarn;
• Single RBCunit. 14 days for clopidogreland tic1opidine)because
* Single RBC unit in combination with plate- apheresis platelets are often the sole source of
lets and/or plasma. platelets given to a patient (AABB Standard
@ Double RBCunits only. 5.4.1A).4(Pp6269)
AABBStandard 5.5.3.4 permits qualification rate blood into components while the donor is
of a donor with a platelet count from a sample connected to the device. The Trima Accel system
collected immediately before the procedure or consists of the device, software, and sterile,
one obtained either before or after the previous single-use, disposable tubing sets with integrated
procedure.4(P21) Triple collections from first-time blood storage bags. WB is drawn from the donor
donors require a qualifyingplatelet count from a and mixed with anticoagulant in the disposable
sample collected before the donation (MBB tubing set. The Trima Accel is cleared for use
Standard 5.5.3.4.1 ).4(p21) Exceptions to these with the anticoagulant ACD-A. The blood and
laboratory criteria should be approved in writing anticoagulant are pumped into the separation
by the apheresis program physician based on channel, which is spun in the centrifuge to sepa-
documented medical need. For apheresis collec- rate the blood into its components. Blood that is
tions, FDA guidelines require a periodic review not collected is returned to the donor during the
of donor records to monitor platelet counts." procedure. Plasma is leukocyte reduced in the
Apheresis devices are capable of collecting separation channel. Platelets and WBCsflow out
platelets in less plasma volume, in a volume- of the separation channel and into the leukocyte
reduced or hyperconcentrated state. Hypercon- reduction system chamber, which separates the
centrated platelets must be diluted with platelet platelets from the white cells based on size;
additive solution (PAS)to support storage up to leukocyte-reduced platelets flow out of the cham-
either 5 or 7 days depending on local regulatory ber into the final platelet storage bag. The Trima
approvals.52·54 Accel system can leukocyte-reduce red cells via
in-line filtration as well as automate the addition
Plasma Apheresis of the appropriate volume of RBC storage solu-
tion, or these steps can be completed manually
Plasma collected by apheresis represents 12% of after the collection.The TrimaAccelis cleared for
plasma intended for transfusion in the United use with the red cell storage solution AS-3in the
States." Apheresis devices can collect plasma United States, and SAGM (saline, adenine, glu-
alone or in combination with single RBC units cose, and mannitol) in Europe. It can also collect
and/ or platelets. The total volume of plasma plasma-reduced platelet concentrates with auto-
collected is based on the physical characteristics mated addition of the appropriate volume of PAS.
of donors and limited by the labeling of the The Trima Accel system can be used to col-
apheresis devices. Plasma collected by apheresis lect the following components alone or in com-
devices is typicallyleukocyte reduced. bination depending on donor size, sex, platelet
A distinction is made between infrequent count, and hematocrit:
plasmapheresis, in which the donor undergoes
plasmapheresis no more frequently than once Single, double, or triple units of platelets
every 4 weeks, and serial plasmapheresis or stored in plasma or PAS.
source plasma collection. The latter is the pro- Plasma that can be prepared into Fresh Fro-
cess to collect plasma for fractionation into plas- zen Plasma (FFP), Plasma Frozen Within 24
ma derivatives, in which the donation is more Hours After Phlebotomy (PF24), and Plasma
frequent than once every 4 weeks. For donors Frozen Within 24 Hours After Phlebotomy
in infrequent plasmapheresis programs, donor Held At Room Temperature Up To 24 Hours
selection and monitoring requirements are the After Phlebotomy (PF24RT24).
same as those for WB donation. <II Single or double units of leukocyte-reduced
RBCsstored in RBCstorage solution.
Apheresis Devices Single or double units of RBCsstored in RBC
The Trima Accel System (Terumo BeT) storage solution.
The Trima Accel Automated Blood Collection In the United States, the Trima Accel Extend·
System (TrimaAccel)is a continuous-flow, single- ed We Platelet (ELP)storage bag is cleared to
stage system that uses centrifugal force to sepa- store leukocyte-reduced platelets in 100% plas-
ma for up to 7 days and in PAS up to 5 days. Plasma that can be prepared into FFP, PF24,
Platelet storage duration outside of the United and PF24RT24.
States is dependent on local regulations and/or Single units of leukocyte-reduced RBCs
regulatory approvals. stored in RBCstorage solution.
The Amicus Separator System (Fresenius Kabi) The Amicus device is also capable of per-
forming mononuclear cell collections and thera-
The Amicus Separator System is an automated
peutic apheresis procedures such as therapeutic
blood cell separator intended for the collection
plasma exchange.
of blood components and mononuclear cells.
The separator can be configured to collect red
AlyxComponentCol/ection System (Fresenius
cells and plasma concurrently with platelets.
The Amicus Separator System is composed of KabO
the Amicus separator instrument and a dispos- The Alyx Component Collection System (Alyx)
able apheresis kit specificto the procedure being is an automated device designed to collect and
performed. The instrument is a continuous-flow, separate WB from donors. The WB is centrifu-
centrifugal device that draws WB from a donor, gallyseparated into its cellular and plasma com-
separates the blood into its components, collects ponents. Cellular and/or plasma components
one or more of the blood components, and re- are retained in collection containers or returned
turns the remainder of the blood components to to the donor according to the predetermined
the donor along with saline for fluid replace- collectionprocedures. Alyxuses a rigid, cylinder-
ment. Amicus is cleared for use with the antico- shaped chamber in the centrifuge to separate
agulant ACD-A. The first stage employs a soft the plasma from the cells. During reinfusion, the
spin to separate the heaviest cells-red cells and plasma and saline are returned to the donor.
white cells-from the platelets and plasma, re- When the collection is complete, Alyx automati-
sulting in leukocyte reduction of the platelets. In cally adds the preservative solution and pumps
the second stage, the PRP.is pumped into the the red cells through an in-line leukocyte reduc-
collection chamber, where the platelets are con- tion filter into the finalstorage bags.
centrated. Alyx has a closed, disposable kit which has
The Amicus separator allows for the use of a all solutions and containers preattached. It is
single- or double-needle configuration. Single- cleared for use with the anticoagulant ACD-A
needle platelet collection procedures provide an and AS-! in the United States, or SAGM (in Eu-
optional concurrent red cell collection. Both rope) as the red cell preservation solution.
single-needle and double-needle procedures pro- Alyx can be used to collect the following
vide for optional concurrent plasma collection. components depending on donor weight and
The Amicus platelet storage container is hematocrit/hemoglobin:
cleared to store platelets in J 00% plasma for up
to 7·days and to store platelets in a mixture of Double units of leukocyte-reduced RBCs.
35% plasma, 65% PAS-3up to 5 days. The Am- Single units of leukocyte-reduced RBCs and
icus is cleared for use with the RBCstorage solu- 2 or 3 units of plasma.
tion (additive solution) AS-! in the US market, Up to 4 units of plasma.
and SAGMin the European market.
Amicus can be used to collect.the following Plasma can be prepared into FFP, PF24, and
components alone or in combination depending PF24RT24.
on donor weight, height, platelet count, and he-
matocrit: Aurora Plasmapheresis System (Fresenius KabO
Single, double, or triple units of platelets The Aurora Plasmapheresis System, consisting
stored in plasma or a mixture of 35% plasma, of the instrument (hardware and software) and
65% PAS-3. a single-usedisposable set, is an automated plas-
mapheresis system intended for routine collec- mixes it with anticoagulant solution. WB is sep-
tion of source plasma. The Aurora system uses arated into its components using the Latham
4% sodium-citrate anticoagulant and allows for a bowl separation technology. The buffy-coatlayer
saline replacement option. The collection of containing platelets and WBCs is formed within
plasma by the Aurora Xi System is a fully auto- the bowl, supported by plasma-controlled man-
mated procedure with the donor connected to agement of the hematocrit (critical flow). Plate-
the PLASMACELLXi disposableset. The Aurora lets are elutriated from the bowl and collected
Xi System is based on a rapidly rotating separa- by using rapid plasma flow through the cellular
tor (membrane filter) to separate plasma from layers (surge technique). Bloodcomponents that
WB. The collection procedure requires a single are not collected are returned to the donor, op-
venipuncture, which means that one access site tionally with a configurab1evolume of saline to
is used to draw WB and return concentrated cel- compensate for the volume loss.
lular components. The procedure involves alter- Depending on donor size and platelet count,
nating cycles, in which blood is drawn and plas- the MCS+ LN9000 is capable of collecting sin-
ma is separated and collected, followed by gle, double, or triple units of apheresis platelets
return of residual cellular components. Venous with or without concurrent plasma collection.
pressure is continuously monitored to avoid ex- Disposable sets are offered with an integrated
ceeding the flow capacity of the donor's vein. leukocyte reduction filter. The filtration of plate-
let units by gravity is automatically performed
The MCS+ LN8 750 Device (Haemonetics) during the collection procedure, making final
The Haemonetics MCS+ LN8150 system con- leukocyte-reduced platelet components avail-
sists of a device, protocol software, and single- able at the end of the procedure.
use, disposable set components. Using a single-
access, functionally closed kit, the device draws The PCS2Device (Haemonetics)
WB from the donor's vein and proportionally
The Haemonetics PCS2 system consists of a col-
mixes it with anticoagulant solution. Using the
lection device, protocol software, and a compati-
blow-molded bowl technology, it separates do-
ble single-use, disposable set optimized for the
nor WB into its components.
Depending on donor size and hematocrit, collection of source plasma and plasma for trans-
the MCS+ LN8150 is capable of collecting sin- fusion. With the Haemonetics single-use, single
gle or double units of RBCswith or without con- venous access component set, the PCS2 device
current plasma collection, depending on donor draws WB from the donor's vein and propor-
size and hematocrit. Blood components that are tionallymixes it with anticoagulant solution. Us-
not collected are returned to the donor, option- ing blow-molded bowl technology, the PCS2
allywith a configurab1evolume of saline to com- separates WB from a donor into its components
pensate for the volume loss. The device auto- and collects a user-configurable volume of plas-
matically administers the appropriate volume of ma based on the donor profile into a collection
red cell preservative solution to the collected container. Blood components that are not col-
unit(s). Disposable sets are offered with an inte- lected are returned to the donor, optionallywith
grated leukocyte reduction filter. The filtration a configurable volume of saline to compensate
of RBCunits by gravity is performed off-line af- for the volume loss.
ter the collection is complete.
The NexSysPCSDevice with YESTechnology
The MCS+ LN9000 Device (Haemonetics) (Haem on etics)
The Haemonetics MCS+ LN9000 system con- NexSys PCS with embedded yield-enhancing
sists of a device, protocol software, and single- solution (YES)technology is a collection device
use, disposable set components. Using a single- that uses a disposable, single-use, single venous
access, functionally closed kit, the device draws access component set to draw WB from the do-
WB from the donor's vein and proportionally nor's vein and proportionally mix it with antico-
agulant solution. Using blow-molded bowl tech- such as butyryl-trihexyl-citrate (BTHC), 1,2-
nology, the NexSys PCS system separates WB cyclohexane dicarboxylic acid diisononyl ester
from a donor into its components and collects (DINCH), and di(2-ethylhexyl) terephthalate
an operator-configurable volume of source plas- (DEHT) have been explored as alternatives to
ma and deposits it into a collection container. DEHP.59 The protective effect of DEHP on red
Blood components that are not collected are re- cells has made finding an equally effective sub,
turned to the donor, optionally with a configu- stitute for DEHP challenging. In the absence of
rable volume of saline to compensate for the vol- DEHP to stabilize the red cell membrane, it has
ume loss. been suggested that next-generation additive
solutions might be able to compensate for this
lack of stabilization when alternate plasticizers
P IN are employed." PAGGSM (phosphate-adenine-
glucose-guanosine-saline-mannitol), AS-3, or
PAGGGM (phosphate-adenine-glucose-guano-
Cold-Stored Whole Blood for sine-gluconate-mannitol) in DINCH- or DEHT-
Transfusion plasticized collection and storage systems per-
form similarly to DEHP-PVC containers with
WB is most often separated into components; AS-lor SAGM.60,61
however, it can be stored as WB for transfusion RBCsfrom WB in anticoagulant-preservative
for up to 35 days in approved anticoagulant! CPO or CP2D have a shelf life of 21 days at 1 to
storage solutions (MBB Reference Standard 6 C with a hematocrit of 65% to 85%, or 35
5.1.8A).4(PP52-61)
WB can be stored non-leuko- days in CPDAI with a hematocrit of <80%
cyte-reduced or can be leukocyte-reduced using (MBB Reference Standard 5.1.8A).4(PP52-61) The
a platelet-sparing filter. Recently, successful use use of additive solutions enables extension of
of low-titer anti-Azanti-B, group 0 WB in mili- RBCshelf life up to 42 days in the United States
tary trauma resuscitation has renewed interest and to 56 daysin some other jurisdictions. Addi-
in use of stored WB by the civiliantrauma medi- tive solutions reduce the hematocrit to approxi-
cine community.I,55-57 WB and RBCshave simi- mately 55% to 65%. RBCs stored for less than
lar volume and identical storage and transporta- 7 to 10 days are often issued for neonatal or pe-
tion temperature requirements. WB offers diatric transfusions, although practice varies
operational simplicity compared to balanced both with regard to length of storage and pre-
component therapy (deliveryof a balanced ratio ferred anticoagulant or additive solution at dif-
of RBCs, plasma, and platelets) for massively ferent institutions, with sparse evidence to sug-
bleeding panents." WB contains cold-stored gest .an optimal approach. For RBCs collected
platelets that appear to have equivalent or better either from WB or apheresis, hemolysis at the
hemostatic effect than platelets stored at room end. of storage should be less than 1% in the
temperature.w" Transfusion of WB has the ad- United States or less than 0.8% in the European
vantage of providing a balanced resuscitation Union (EU) and other sites outside the United
fluid in one bag rather than up to four bags of States.62,63
RBCs,platelets, and plasma." Visual inspection of RBCunits can detect ab-
normal color or appearance caused I:>Y hemoly-
Red Blood Cell Component Storage sis, or clots. Abnormal color in the bag can also
be caused by bacterial contamination." An ab-
Containers for WB and RBC storage often are normal-appearing unit can be centrifuged to
composed. of PVC plasticized with di(2-ethyl- facilitate inspection of the supernatant for hemol-
hexyl) phthalate (DEHP). DEHP not only im- ysis. In the case of suspected bacterial contami-
parts flexibility to the PVC but also has been nation, visual inspection of the supernatant may
shown to protect red cells against hemolysis reveal murky, brown, or red fluid." However,
during storage. Because of concerns over the visual inspection will not detect all contaminat-
possible toxicity of DEHP, alternative plasticizers ed units. Blood clots in RBCunits are often too
small to be detected visually and are revealed plasticizedwith either BTHCor tri(2-ethylhexyl)
during transfusion when they clog the transfu- trimellitate or, alternatively, are composed of
sion filter, or in the component laboratory when ethylvinyl acetate, polyolefins (polyethylene or
the units fail to pass through a leukocyte reduc- polypropylene), or tluoropolymers." Agitation
tion filter. Units that fail visual inspection or are supports platelet metabolism by ensuring effec-
otherwise found to contain clots should not be tive exchange of oxygen, carbon dioxide, and
released for transfusion. lactic acid between the platelets and the sus-
Hemoglobincontent and hematocrit per RBC pending media. Long periods of static storage of
unit vary because of differencesbetween donors platelets disrupt oxidative metabolism and en-
and between blood component manufacturing hance glycolysis, resulting in increased lactic
methods.44,65,66 For example, more hemoglobin acid production and consequently a decline in
is generally lost and there is a resulting lower pH.72If pH levels decline to less than 6.2, plate-
hematocrit in the RBC unit produced when lets will have unacceptably low in-vivorecover-
platelets are manufactured by buffy-coat prepa- ies." During transport from blood centers to
ration methods than with PRP-type methods. hospitals or to other blood centers or during
Hemoglobin content per unit and the final he- long-distance air transport, platelets are not agi-
matocrit may be more precisely controlled with tated. In-vitro studies have shown minimal
apheresis collections. Total hemoglobin content platelet damage when they are stored without
is not directly regulated but has a lower limit of agitation for up to 24 hours.72,74,75
However, lon-
45 g per unit in the United States or 40 g per ger periods without agitation can lead to unac-
unit for leukocyte-reduced RBCs in Europe.f ceptable pH decline."
Some experts have advocated for tighter stan- With 20 to 24 C storage, contaminating bac-
dardizing of hemoglobin and hematocrit in RBC teria can proliferate during storage and result in
units.67,68 septic transfusion reactions, some of which are
Segments are made from tubing from the fatal. Platelet storage containers can support
RBC container, marked with repeating serial shelf life up to 5 or 7 days; however, platelet
numbers, and may be sealed at several locations storage is limited by local regulations to mini-
with either a dielectric (heat) sealer or a metal mize the risk of septic transfusion reactions.
clip (grommet) to prepare approximately 13 to These limits are determined by the different mit-
15 segments with the unique number. These igation strategies implemented in the various
segments may be used later for ABO/Rh typing, world regions. Mitigation strategies include do-
crossmatching, antigen typing, investigation of nor screening, skin disinfection, blood diversion,
adverse transfusion reactions (with the excep- component testing by culture, point-of-issue
tion of bacterial contamination), or other labora- testing, and pathogen inactivation. In the Unit-
tory tests. However, care should be taken in used States, according to the September 2019
ing segments to evaluate the quality of the RBC guidance for industry from the FDA/6 platelet
component, as significant differences between shelf life can be extended to 5 days if pathogen
measurements of hemoglobin, hemolysis, and inactivation is used, or up to 7 days with sec-
hematocrit have been shown to exist.69,7o ondary bacteria testing (reculture or rapid test-
ing) or large-volume, delayed samplingusing ap-
proved devices. Platelet shelf life will be limited
Platelet Component Storage
to either 5 days, with primary culture plus sec-
Platelets are stored and shipped at 20 to 24 C in ondary testing cleared as a safety measure or
plastic containers that have greater gas permea- pathogen inactivation, or 7 days, in an approved
bility than those for RBCs or plasma and must container, with the addition of secondary testing
be continuously agitated during storage to sup- by culture or point-of-issuetesting. Other coun-
port platelet metabolism and ensure adequate tries already allow up to 7-day platelet shelf life
in-vivo recovery (MBB Reference Standard with pathogen inactivation (eg, Switzerland) or
5.1.8A).4(PP52-61),51,71
Modern platelet containers large-volume, delayed sampling (eg, Canada and
with high gas permeability are composed of PVC the United Kingdom]."
Visual inspections of platelets after they are mercially available PAS formulations support
prepared show an absence of visible red cells in platelet storage with 30% to 40% plasma carry-
the vast majority of units, which implies that the over." Lower plasma carryover requires investi-
units contain fewer than 0.4 x 109 red cells. gational PASformulations that contain bicarbon-
Generally, the number of red cells in a standard ate and glucose to maintain platelet metabolism
transfusable unit of platelets does not exceed and viability. Benefits of PAS-storedplatelets in-
1.0 x 109, although occasionally WB-derived clude fewer adverse transfusion reactions, lower
platelets contain more red cells." If red cells are titers of anti-A and B antibodies, as well as in-
visible in a component, the hematocrit should creased availability of plasma for further manu-
be measured. AABBStandard 5.15.5 states that facturing.P The negative effect of platelets
if the component contains greater than 2 mL of stored in PASis lower posttransfusion count in-
red cells, the red cells must be ABO compatible crements.F In the United States, current PAS
with the recipient's plasma and be cross- formulations are approved for storage of aphere-
matched. In such cases, a sample of donor blood sis platelets only and cannot be used for the stor-
is attached to the container for compatibility age of platelets prepared from WB.
testing.4(P38}
On the day of preparation, some WB-derived Plasma Component Storage
platelet units, as well as apheresis platelet units,
Plasma preparations are defined and regulated
may contain clumps composed of platelet aggre-
through extensive. combinations of differences
gates.78,79 In routine practice, visual inspection is
in collection methods,. storage temperatures,
adequate to determine the degree of clumping
freezing methods, secondary processing, timing,
subjectively and ensure that units with exces-
and storage after thawing. These specifications
sive clumping are not labeled for distribution
are covered in an array of standards, rules, and
and transfusion. Most of the clumps seen on day
guidelines overlaid with various requirements of
o dissipate on day 1 of storage With continuous the country where the plasma is prepared and/
agitation, particularly those units showing light
or used. Major, although not exclusive, sources
to moderate clumping." The temperature, cen-
of this information include the US CFR, FDA
trifugation speed used inthe production of WB
guidance documents, the US Circular of Infor-
platelets, and leukocyte reduction by filtration
mation, MBB Standards, and EU directives.
may influence the presence of platelet aggre-
The definitions and requirements of the country
gates.79,80 Some blood centers have also reported
where the plasma is prepared should always be
donor dependence, where platelet aggregates
consulted.
are observed in multiple donations from the
Plasma from WB and apheresis collections is
same donor."
generally frozen to maintain factor activity and
Occasionally, because of adverse shipping
provide an extended shelf life. Delayedfreezing
conditions, temporary equipment failures, or
of fresh plasma can result in lower levels of Fac-
power outages, platelets cannot be maintained tors V and VIII.85,86 Frozen plasma is thawed for
at 20 to 24 C. One study indicated that platelets
clinical use and may be maintained at. 1 to 6 C
can maintain their in-vitro properties after expo"
for some time before use. Frozen plasma is also
sure to 37 C for 6 hours, followed by room-
the source. of cryoprecipitate and cryoprecipitate-
temperature storage without agitation for an ad-
reduced plasma. Plasma may be used for trCl!l~-
ditional 18 hours." However, negative effects fusion or for the preparation of specific plasma
on in-vivo platelet recovery and survival when protein products through fractionation process-
platelets are stored at temperatures less than
es.
20 C have been reported.83,84 Thus, proper steps
should be taken to maintain the required range
Fresh Frozen Plasma
of temperatures during storage at the blood cen-
ter and during transport. In the United States, FFP can be prepared from
In Europe, apheresis and WB-derived plate- plasma collected either from WB or by aphere-
lets can be stored in either plasma or PAS. Com- sis. A plasma unit derived from WB contains, on
'56 A A B B TEe H N I CAL MAN UA L
average, 300 mL, but apheresis units may con- -20 C and -35 C.87At and below these tem-
tain as much as 400 to 600 mL of plasma. FFP peratures, the container is brittle and is fragile
contains acceptable levels of all coagulation fac- enough to break during transport and handling.
tors, antithrombin, and ADAMTS13.86 FFP Leaky containers should not be used for transfu-
must be placed in the freezer within 8 hours of sion and should be discarded.
collection (MBB Reference Standard 5.1.8A) or MBB requires interventions to reduce the
as directed by the manufacturer's instructions risk of transfusion-related acute lung injury
for use of the blood collection, processing, and (TRALI)from plasma-containing components.
storage system (AABB Standards 5.1.8A and Current approaches for apheresis platelets, plas-
5.7.4.9).4(Pp28,52-61)
FFP has a shelf life of 12 ma, and WB for transfusion include collection
months when stored at -18 C or colder and, from males, never-pregnant females, or parous
with FDA approval, may be stored for up to 7 femaleswho test negative for HLAantibodies, to
years from collection at -65 C (MBB Reference minimize the risk of exposing patients to HLA
Standard 5.1.8A).4(pP52-61) alloantibodies that could cause TRALl (AABB
The Council of Europe, somewhat more Standard 5.4.1.3).4(p18)
stringently, defines "plasma, fresh frozen" as Quarantined plasma is held in storage until
prepared from either WB or plasma collected by the donor returns for a subsequent donation. It
apheresis. Plasma freezing must be initiated was introduced to increase the viral safety of
within 6 hours, or within 24 hours if WB is rap- plasma. The Council of Europe permits the re-
idly cooled to 20 to 24 C following collection.f lease of FFP from quarantine after the donor re-
Freezing must be completed within 1 hour and turns to the blood center after a minimum quar-
achieve a temperature of less than -25 C. Rapid antine period that is greater than the diagnostic
freezing of plasma can be accomplished using a window period for viral infection (typically 6
blast freezer, dry ice, or a mixture of dry ice months). Donors must have negative test results
with either ethanol or antifreeze. Plasma, fresh for at least hepatitis B surface antigen, antibod-
frozen has an expiry time of 36 months if held ies to human immunodeficiency virus, and anti-
at less than -25 C, or 3 months if held at -18 C bodies to hepatitis C virus. With the use of nu-
to -25 C.62The Council of Europe does not stip- cleic acid tests for viral screening, this window
ulate specific thawing methods-only that period for quarantined FFPmay be reduced."
thawing be performed in a properly controlled
environment with no insoluble cryoprecipitate Plasma Frozen Within 24 Hours After
visible upon completion of the thaw proce- Phlebotomy
dure.f
Plasma should be thawed immediately after The FDA defines plasma that is frozen to less
removal from storage at 30 to 37 C in a water- than -18 C within 24 hours of collection as
bath or by using an FDA-cleareddry-thawing de- PF24. PF24 can be prepared from plasma col-
vice. When a waterbath is used, the component lected either from WB or by apheresis. PF24 is
must be placed in a protective plastic overwrap prepared when WB or plasma cannot be trans-
(MBB Standard 5.7.4.9.1).4(p28)Thawing of ported, processed, and frozen within 8 hours af-
larger units of FFP collected by apheresis may ter phlebotomy due to geographic or logistical
require more time. FFP, once thawed, has a constraints. PF24 is equivalent to FFP with the
shelf life of 24 hours at 1 to 6 C. Thawed plasma exception of Factor V, Factor VIII,and protein C
held longer than 24 hours must be relabeled as levels.' The component prepared from apheresis
Thawed Plasma and can be stored for an addi- collections must be stored at 1 to 6 C within 8
tional 4 days at 1 to 6 C if prepared in a closed hours of collection and frozen within 24 hours
system (AABB Standards 5.7.4.13 and (MBB Standard 5.7.4. 10).4(P28) Once thawed,
5.1.8A).4(pP29,52-61) PF24 has a shelf life of 24 hours at 1 to 6 C
The glass-transition temperature of plasma (MBB Reference Standard 5.1.8A).4(pP52-61)
storage bags is dependent on the material com- Thawed plasma held longer than 24 hours must
position, but for PVC containers it is between be relabeled as Thawed Plasma, which can be
stored for an additional 4 days at 1 to 6 C (AABB Liquid Plasma

Standards 5.7.4.13 and 5.1.8A).4(PP29,5261)
In the United States, liquid plasma for transfu-
sion can be separated from WB at any time
Plasma Frozen Within 24 Hours After during the WB storage period and stored at 1 to
Phlebotomy Held At Room Temperature Up 6 C for up to 5 days after the WB expiration
To24Hours After Phlebotomy date [AABBReference Standard 5.1.8A and 21
Plasma held for up to 24 hours after collection CFR 610.53(b)].4(pP52-61)For WB stored in ACDI
at room temperature and then stored at less CPD or CP2D, the expiration for Liquid Plasma
than -18 C is PF24RT24 (AABB Standard is 26 days. If WB is stored in CPDAl, the Liq-
5.7.4.11).4(p28)PF24RT24 can be prepared from uid. Plasma expiration date is 40 days following
collection. Liquid Plasma has acceptable factor
plasma collected either from WB or by aphere-
levels, with the exception of Factor V and Factor
sis. PF24RT24 is prepared when WB or plasma VIII, and is typicallyused to avoid delays associ-
cannot be transported, processed, and frozen ated with thawing in emergency situations.
within 8 hours after phlebotomy due to geo-
graphic or logistical constraints. PF24RT24 is Cryoprecipitated AHF
equivalent to FFP,with the exception of Factor
V, Factor VIII, and protein S levels," Once Cryoprecipitated Antihemophilic Factor (AHF),
thawed, PF24RT24 has a shelf life of 24 hours or simply"cryoprecipitate" in Europe, is a con-
at 1 to 6 C (AABB Reference Standard centrated cryoglobulin fraction that is prepared
5.1.8A).4(Pp52-61)
Thawed plasma held for longer fromFFP. FFP can be thawed to prepare Cryo-
than 24 hours must be relabeled as Thawed precipitated AHF by placing the FFP in a refrig-
Plasma and can be stored for an additional 4 erator (at 1 to 6 C) overnight or in a circulating
waterbath at 1 to 6 C. Cold-insoluble protein
days at I to 6 C (AABBStandards 5.7.4.13 and
5.1.8A).4(PP29,52-61) that precipitates when FFPis thawed to 1 to 6 C
is collected by centrifugation, and the superna-
tant plasma is transferred into a satellite contain-
Thawed Plasma er. The precipitate is resuspended in a small
Thawed Plasma is FFP, PF24, or PF24RT24that amount of residual plasma, generally 15. mL,
has'been thawed and held at 1 to 6 C for >24 and the precipitate is refrozen. The Cryoprecipi·
hours (AABBStandard 5.7.4.13).4(P29)Thawed tated AHF is placed in a freezer within an hour
Plasma may be held at 1 to 6 C for up to 4 days of removal from the refrigerated centrifuge and
after the initial 24-hour postthaw period has can be stored at -18 C for 12 months from the
elapsed. Stable Factor II and fibrinogen and re- original collection date (AABBReference Stan-
duced amounts of other factors'(especiallyFac- dard 5.1.8A).4(PP52-61)In Europe, thawing is to be
tors V and VIII, for which transfusion of plasma performed at 2 to 6 C, and the component can
be stored for up to 3 months at -18 to -25 C or
is rarely indicated) have been observed in
for 36 months below -25 C.62
Thawed Plasma. Thawed Plasma prepared from
AABBStandard 5.7.4.15 requires that Cryo-
FFP and stored for up to 5 days after thawing precipitated AHFcontain at least 80 internation-
has reduced levels of Factor V, Factor VII, and al units (IU) of Factor VIII,and150 ~g of fibrin-
Factor VIII.89-91 Storage of thawed Plasma Cryo- ogen per unit,4(p29)although the average
precipitate Reduced for up to 5 daysdoes not af- fibrinogen content is generally 250 mg.'? Euro-
fect levels of fibrinogen, FactorVIII,or von Will- pean standards require at least 70 IU of Factor
ebrand factor (vWF)but can result in reductions VIII, 140 mg of fibrinogen, and 100 IU.of vWF
in ADAMTSI3, Factor V, and Factor VII.92 per untt." Current preparations are reported to
Thawed plasma has acceptable factor levels, have much higher amounts of fibrinogen (medi-
with the exception of Factor V and Factor VIII, an: 388 rng/unit)." Rapid freezing of FFP is
and is typically used to avoid delays associated found to increase the Factor VIII yield in Cryo-
with thawing in emergency situations. precipitated AHF.95Anti-Aand anti-Bare known
to be present in Cryoprecipitated AHF, but the Recovered and Source Plasma

combined amount of these antibodies from the Blood centers often convert plasma and Liquid
unit of plasma is only 1.15% of the total.96 Plasma to an unlicensed Recovered Plasma or
Thawed Cryoprecipitated AHF should be "plasma for manufacture" component, which is
used as soon as possible but may be held at usually shipped to a fractionator and processed
room temperature (20-24 C) for 6 hours as a sin- into derivatives, such as albumin, coagulation
gle unit or as a pool prepared in a closed system factors, and/or immune globulins. To ship Re-
using an approved sterile connection device, or covered Plasma, the collecting facility must
conversely for 4 hours if pooling was with an have a "short supply agreement" with the man-
open system (AABB Reference Standard ufacturer (21 CFR 601.22). Because Recovered
5.1.8A).4(Pp52.61)Pooling may be accomplished Plasma has no expiration date, records for this
with the aid of a diluent, such as 0.9% sodium component should be retained indefinitely. Stor-
chloride (USP),to facilitate removal of material age conditions and expiry dates for Recovered
from individual bags. Plasma are established by the plasma fraction-
At room temperature, the mean declines in ator. FFPused as human plasma forfractionation
Factor VIII levels at 2, 4, and 6 hours are ap- in Europe must comply with the applicable Eu-
proximately 10%, 20%, and 30%, respectively." ropean Pharmacopoeia guidelines.
Source Plasma is collected by apheresis and
Cryoprecipitated AHF from blood groups A and
stored at the appropriate temperature required
B has higher levels of Factor VIII compared to for further manufacturing of plasma protein
that derived from blood group 0 donors (about products. Source Plasma donors are typically
120 vs 80 IU per bag, respectively)." Thawed paid and have different donor qualificationcrite-
cryoprecipitate should not be refrozen.f ria compared to infrequent plasmapheresis do-
nors (AABBStandard 5.5.2).4(p19) Source Plasma
Plasma Cryoprecipitate Reduced donors can donate a maximum of two times in a
Plasma Cryoprecipitate Reduced (United States) 7-day period, and the interval between two col-
or Plasma, Fresh Frozen Cryoprecipitate- lections must be at least 2 days (AABBStandard
5.5.2.2.1 ).4(P20)In addition to required testing
Depleted (Europe) is the residual fluid after re-
for infectious agents on each donation, Source
moval of Cryoprecipitated AHF. If prepared us-
Plasma donors require physical examination and
ing a closed system, Plasma Cryoprecipitate Re-
testing to determine total plasma or serum pro-
duced must be refrozen with 24 hours of tein and immunoglobulin composition on the
thawing the FFP from which it is derived, and day of initial plasmapheresis and at least every 4
stored at less than -18 C (AABBReference Stan- months thereafter [21 CFR640.65(b)(I)].
dard 5.1.8A).4(pp52.61)The storage temperatures
and expirations stipulated in US and European
regulations that apply to FFP also apply to this S OLlECTi N
component. The component contains a normal R CESSmNG/B OD
level of FactorV (85%), Factor I, Factor VII, Fac- COMPONENT MODIFICATION
tor X, antiplasmin, antithrombin, protein C, and
protein S. Even after the removal of cryoprecipi- Prestorage Leukocyte Reduction by
tate, the component has a fibrinogen level of Filtration
about 200 mg/dl.," Levels of Factor VIII, the Blood collection systems can include in-line fil-
vWF antigen, vWF activity, fibrinogen, and Fac- ters for removal of leukocytes from WB, RBC
tor XIII are decreased.92,lOo Plasma Cryoprecipi- units, and/or platelets. Many WB leukocyte re-
tate Reduced is used almost exclusivelyfor plas- duction systems allow filtration at ambient tem-
ma exchange or transfusion in patients with perature for up to 24 hours after collection. WB
thrombotic thrombocytopenic purpura.' and RBC filtration may also be started and/or
completed at refrigerator temperatures. The re- through the filter, the residual leukocyte content
sidual number of leukocytes in 95% of leuko- may be higher than allowablelimits.102
cyte-reduced, WB-derived, single-donor platelet Becauselevels of residual leukocytes in leu-
units tested must be less than 8.3 x lOs per unit kocyte-reduced components are below the level
(AABBStandard 5.7.4.19).4(P30) The Council of of detection for most standard hematology ana-
Europe requires that the residual number of leu- lyzers, Nageotte hemocytometry and flow cy-
kocytes be less than 1 x 106per unit in 90% of tometry have historically been used to quantify
RBCsand pooled platelet and apheresis platelet white cell content in blood components. A Na-
components tested." The FDArequires that the geotte hemocytometer is a fixed-volume (50-ilL)
residual number of leukocytes be less than 5 x device that contains an etched grid to facilitate
106per unit in 95% of RBCsand pooled platelet manual counting of cells under a microscope.103
and apheresis platelet components tested, with Flow-cytometry methods involve labeling of
95% confidence.3s,36 fresh or fixed cells with a fluorescent DNA-
In the United States, leukocyte reduction by binding dye such that the leukocytes can be
filtration of RBCs must result in a component counted relative to an internal calibration bead.
that contains at least 85% of the original red cell In a multicenter study, flow cytometry gave
(AABB Standard 5.7.4.6) or platelet con- more precise results than Nageotte hemocytom-
tent.4(p27),101 After filtration, single-donorplatelet etry when freshly prepared samples (within 24
units must have :?:5.5x 1010platelets per unit in hours) were tested.'?' In general, Nageotte he-
75% of units tested, with 90% of units having a mocytometry tends to underestimate the num-
pH :?:6.2at the end of allowable storage (AABB ber of white cells compared to flow cytometry.
Standard 5.7.4. 19).4(p30) Also, 95% of leukocyte- Automated optical systems using image analysis
reduced apheresis platelets must contain :?:3.0x are now available to measure white cell counts,
1011 platelets per unit and have a pH :?:6.2,with reducing the technical burdens associated with
95% confidence, at the end of allowable stor- both Nageotte and flow cytometry.!"
age." The Council of Europe's standards require
that 90% of units tested have a minimum of 40 Pooling of Blood Components
g of hemoglobin in each RBCunit and >2 x l O" Pooling of platelet or plasma components from
platelets in each platelet unit after leukocyte re- multiple blood donors can be used to increase
ductlon.f the therapeutic dose of cell or plasma protein
Prestorage leukocyte reduction is generally components in a product. When pooled by the
performed soon after WB collection and is al- component manufacturer or hospital service
ways performed within 5 days of collection. In- and the connections are performed using a ster-
line WB filters that remove both WBCs and ile connection device, the component sterility
platelets permit preparation of leukocyte- and expiry time are not compromised (AABB
reduced RBCs and FFP. In-line WB filters that Standard 5.7.2.1.1 ).4(p24)
spare platelets permit preparation of leukocyte- For open systems, pooled platelets have an
reduced RBCs,FFP, and platelets. If WB is col- expiration time of 4 hours from when the sys-
lected without the in-line leukocyte reduction tem was opened for pooling (Reference Stan-
filter, a filter can be attached to the tubing using dard S.1.8A).4(pPS2-6l)
Closed systems for prestor-
an FDA-cleared sterile connection device.101 age pooling of platelets licensed by the FDA
Apheresis devices are designed to leukocyte- permit storage of pooled platelets for up to 5
reduce plasma and platelets during the collec- days from collection of the oldest units in the
tion such that the components do not require pool (Reference Standard S.1.8A).4(pPS2-6l) Four
postcollection leukocyte reduction. to 6 leukocyte-reduced or non-leukocyte-
Sickle cell trait in red cells is the most com- reduced platelet units (generally, ..from ABO-
mon cause of WBC filtration failure. Approxi- identical units) are pooled using a set consisting
mately 50% of the RBC units with sickle cell of a multi-lead tubing manifold for sterile con-
trait fail to filter. Although the other 50% pass nection. If non-leukocyte-reduced units are
pooled, they can be leukocyte-reduced by filtra- glycerol/rapid coohng'" and high-glycerol/slow

tion as part of the pooling process. The shortest cooling.108,109 A less commonly used method in-
expiration date of the pooled units determines the volves low concentrations (15-20%) of glycerol,
expiration date of the pool [21 CFR610.53(B)]. rapid cooling (>100 C/minute), storage in liq-
In the United States, each pool prepared uid nitrogen (-196 C) or nitrogen vapor (-165 C),
from leukocyte-reduced platelets must have and rapid thawing in a 42 to 45 C waterbath.
<5.0 x 106residua11eukocytes. Approved pool- The more common cryopreservation method
ing sets must also allow sampling of the pool for found in the United States and most internation-
bacteria detection. A record of the ABO/Rh al blood centers is the use of a high concentra-
type, DIN, and collecting facilityfor each unit in tion of glycerol (40%) in conjunction with slow
the pool must be maintained by the component cooling (7 1 C/min), storage at :::;-65C, and rap-
manufacturer (Standard 5.7.3.3).4(P25) id thawing in a 37 C waterbath. In each meth-
Many international blood manufacturers pre- od, controlled addition and removal of glycerol
pare prestorage pools of buffy-coatplatelets that are required to prevent osmotic lysis of the red
are preserved in PASor in the plasma from one cells and to minimize the transfusion recipient's
of the units from which platelets are preexposure to the chemical cryoprotectant. RBC
pared.l'" Instruments that automate the pooling components must be cryopreserved within 6
process are being used globally to improve effi- days of collection unless they have been bio-
ciency in the blood component laboratory. Out- chemically rejuvenated or are rare RBC units,
side of the United States, systems for prestorage which can be cryopreserved without rejuvena-
pooling of buffy-coat-derivedplatelets can be fol- tion up to the date of expiration (AABBStan-
lowed by pathogen inactivation treatment. dard 5.7.4.2.1 ).4(p26)
Cryoprecipitated AHF units pooled immedi- Cryopreserved RBCs must be stored at tem-
ately before transfusion in an open system have peratures equal to or less than -65 C and expire
an expiration time of 4 hours at 20 to 24 C stor- after 10 years. Rare frozen units may be used be-
age (AABBReference Standard 5.1.8A).4(pP52-6I) yond the expiration date, but a policy must be in
Prestorage pools can also be prepared in an open place for release of these units (AABBReference
system and stored for 12 months at -18 C Standard 5.1.SA).4(pp52-6I} European reguiations
(AABBReference Standard 5.1.8A).4(pP52-6I) Af- permit the cryogenic storage of cryopreserved
ter thawing, the component expires in 4 hours RBCs for up to 30 years." Cryopreserved RBC
(AABB Reference Standard 5.1.8A).4(Pp52-6I) units should be handled with care because the
Prestorage pools prepared using an FDA-cleared PVC or polyolefin freezing storage containers
sterile connection device can be stored for 12 may crack during shipment or if handled rough-
months at -18 C and have a postthaw expira- ly. Rare RBCcomponents that were thawed and
tion time of 6 hours (AABBReference Standard deglycerolized can be refrozen and rethawed
5.1.8A).4(PP52-6I)
The number of units pooled may when needed without adversely affectingthe re-
vary and can consist of 4, 5, 6, 8, or 10 units. covery of the components. I10
Prestorage pools must be placed in a freezer Glycerol must be removed after thawing by a
within 1 hour after removal from a refrigerated method that allows for the addition and removal
centrifuge (AABB Reference Standard of sodium chioride solutions. Addition and re-
5.1.8A).4(PP52-6I)
The potency of the pool is calcu- moval of glycerol (deglycerolization)can be per-
lated by assuming that each unit in the pool con- formed in an open system, with postthaw stor-
tains SO IU of coagulation Factor VIII and 150 age limited to 24 hours at 1 to 6 C (AABB
Reference Standard 5.1.SA).4(Pp52-6I) With open
mg of fibrinogen multiplied by the number of
units in the pool (AABBStandard 5.7.4.l5).4(p29) system processing, the final solution in which
cells are suspended is 0.9% sodium chloride and
0.2% dextrose. Dextrose provides nutrients and
Cryopreservation of Red Blood Cells
has been shown to support satisfactory post-
Currently, there are two methods used for transfusion viability for 4 days of storage after
the cryopreservation of RBC components: low deglycerollzatlon.!"
The commerclal availability of automated, Irradiation of Blood Components

closed-system cell processors for the addition
Cellular blood components can be irradiated to
and removal of glycerol allows extended post- reduce the risk of transfusion-associated gratt-vs-
thaw storage of cryopreserved red cells. Glycer- host disease (GVHD).123 Irradiation s0lll'c~sin-
ol is added within 6 days ofWB collection, and clude gal11ma rays-from elther cesium-l S?
the postthaw storage can be for up to 14 days at blood component irradiators or cobalt-of
1 to 6 C in AS-3in the United States or 7 days in sources-or Hays produced by radiation thera-
SAGM in Europe.111,112 Non-leukocyte-reduced py linear acceleratorsor stand-.aloneunits. Both
RBC components processed using a closed-sys- sources achieve satisfactory results in inactivat-
tem cell processor and cryopreserved have a he- ing T lymphocytes. Free-standing irradiators are
matocrit of 51 %to 53% and contain a mean 9.0 commercially availablefor blood bank use.
x I06leukocytes per untt.!" European standards In the United States, the radiation dose tar-
require that cryopreserved red cell products geted to the midplane of the container must be
have a minimum of 36 g of hemoglobin per at least 25 gray (Gy) [2500 centigray (cGy)]and
unit, a hematocrit of 35% to 70%, and a super- no more than 50 Gy (5000 cGy).62,124 Moreover,
natant hemoglobin level <0.2 g per unit." the minimum delivered dose to any portion of
the blood components must be at least 15 Gy in
Cryopreservation and Cold Storage of a fully loaded canister (MBB Standard
Platelets 5.7.3.2).4(p25)The European standard requires a
higher dose, with no part of the comp?nent re-
Cryopreserved platelets treated with 4% to 6%
ceiving <25 Gy and a maximum dose to any
dimethyl sulfoxide (DMSO) and stored at less
part of the component of 50 Gy.62Both US and
than -80 C for 2 to 4 years have been shown to
European standards are effective for prevention
maintain hemostatic tunctton.!'"!" Removal of
of GVHD.
DMSO before freezing allows platelets to be
Each instrument must be routinely moni-
thawed and reconstituted immediately with tored to ensure that an adequate dose is deliv-
plasma, making these units suitable for military ered to the blood container during irradiation
and civilian trauma use,u6,117Platelet cryopres- (MBB Standard 5.7.3.2.1 ).4(p25) Dose mapping
ervation and the subsequent thawing processes using irradiation-sensitive films or badges that
are; however, time-consuming and more expen- monitor the delivered dose are used for quality
sive than standard room-temperature storage. control of the irradiators.P' Irradiation-sensitive
Clinical trials are underway to support cryopres- labels are a requirement in Europe.f FDAregu-
ervation of platelets.118-120 lations require verification of the delivered dose
Cold storage of platelets at 1 to 6 C has a annually for the cesium-137 source and semian-
number of advantages over room-temperature nually for the cobalt-on source (MBB Standard
storage, including prolonged shelf life due to re- 5.7.3.2.1 ).4(P25)
For x-ray irradiators, the dosime-
duced metabolism, enhanced hemostatic activi- try should be performed in accordance with the
ty, improved bacteriologic safety, and ease of manufacturer's recommendations (MBB Stan-
storage arid transport.!" Storage of platelets at 1 dard 5.7.3.2.1).4(P25)Dose verification is also re-
to 6 C can result in increased cell activation and quired after major repairs or relocation of the ir-
procoagulant function, which has raised interest radiator (MBB Standard 5.7.3.2.1 ).4(P25)For
in u~ing this product to treat actively bleeding gamma irradiators, the turntable operation, tim-
patients." In the United States, apheresis plate- ing device, and lengthening of irradiation time
lets can be stored at I to 6 C without agitation caused by source decay should also be moni-
for up to 3 days (21 CFR 64024 al1d21.c:FR tored periodically.
640.25). Cold storage of platelets can result in In the United States, RBCsmay be irradiated
reduced in-vivo recovery and survival com- up to the end of their storage shelf life. The post-
pared with room-temperature-stored plate- irradiation expiration date is 28 days from the
lets.71,122 date of irradiation or the original expiration
date, whichever is earlier (AABB Reference pended in saline rather than an additive solution
Standard 5.1.8A).4(PP52.61)In Europe, RBCs may in the postwash period, a 24-hour expiry is ap-
be irradiated only up to day 28 followingcollec- plied to these components (AABB Reference
tion, and irradiated cells may not be stored lon- Standard 5.1.8A).4(Pp52.61) The use of closed-
ger than the earlier of 14 days after irradiation system, semiautomated cell processors and
or 28 days after collection.f Platelets may be ir- modern additive solutions can extend the post-
radiated until their expiration date, and their wash storage solution.134European standards re-
postirradiation expiration date is the same as the quire washed components to have a minimum
original expiration date (AABBReference Stan- of 40 g of hemoglobin per unit, a hematocrit of
dard 5.l.8A).4(PP5261) 0.50 to 0.70, less than 0.8% hemolysis, and less
Irradiation of RBCs followed by storage re- than 0.5 g of supernatant protein per unit.62
sults in a decrease in the in-vitro quality of the
blood components. The pre- and postirradiation Volume Reduction of Blood
storage period has been shown to affect the lev- Components
el of hemolysis and supernatant potassium level
in RBCs,which may pose a risk to certain pa- Volume-reduced platelets may be needed to re-
tient populations.Fv'" For example, in neo- duce the amount of plasma transfused to pre-
nates and in rapid transfusions to small children, vent cardiac overload, to minimize ABO anti-
high potassium levels may cause cardiac compli- body infusion, for intrauterine transfusion, or to
cations.!" prompting the use of a shorter expiry minimize the risk of repeated transfusion reac-
for irradiated RBCs or washing of components tions. The volume of a platelet unit (single-
before transfusion. Platelets are not damaged by donor, WB-derived) can be reduced to 10 to
an irradiation dose as high as 50 Gy.128 15 mLiunit by centrifugation just before trans-
fusion. In-vitro characteristics such as platelet
Washing of Red Cell and Platelet morphology, mean volume, hypotonic shock re-
Components sponse, synergistic aggregation, and platelet fac-
tor 3 activity appear to be maintalned when vol-
Washed RBCs are recommended for large- ume reduction is performed on storage day 5,
volume transfusions to potassium-sensitive pa- and platelet increments after transfusion have
tients, particularly neonates when fresh units been satisfactory.136The in-vitro recovery rate of
are not available.l" for recipients who have a platelets is roughly 85% after volume reduction.
history of plasma-related transfusion reac- Prepooled single-donor platelets from WB that
tions.!" and to mitigate TRALI.131The use of are volume reduced by centrifugation (580 x g
washed RBCshas been proposed as a means to for 20 minutes) from an approximate volume of
remove the accumulation of proinflammatory 60 mL to between 35 and 40 mL yield a high
cells and molecules in stored RBCs to reduce platelet count (>2.3 x 109/L).137 Lowering the
the bioactivity of the component'" and may be pH of platelets prepared from WB avoidsplatelet
an effective means to address red cell degrada- aggregates visible to the unalded eye (macroag-
tion during storage.l" gregatesl.!" The addition of 10% ACD-A to
RBCscan be washed in the blood bank using platelets to lower the pH before centrifugation
manual methods, semiautomated cell proces- can help with resuspension of high-concentra-
sors, or blood recovery devices just before trans- tion platelets and avoid aggregatlon.!" If an
fusion to reduce the concentration of plasma or open system is used, the maximum allowable
immunogiobulins.129,134,135 Methods must en-
storage time is 4 hours.
sure RBCsare washed with a volume of compat-
ible solution that will remove almost all of the
Pathogen Inactivation
plasma (AABBStandard 5.7.4.5).4(p27) Because of
the potential for bacterial contamination during Blood is the safest it has ever been. However,
washing of RBCs using an open processing sys- despite extensive blood donor history screening
tem and decreased red cell viabilitywhen resus- and testing, the risk of transfusion-transmitted
infection from both known and emerging patho- transfusion, with the same platform used to
gens persists.!" Blood components can be treat- treat platelets and plasma.
ed with pathogen inactivation technologies to The THERAFLEX UV-Platelet System (Maco-
inactivate pathogens such as viruses, bacteria, Pharma and German Red Cross) is CE marked
and parasites and hence reduce the risk of trans- for treatment of platelets. It is based on applica-
fusion-transmitted tnlections.!" The majority of tion of UVC light (Iv = 254 nm) to damage nu-
pathogen inactivation methods damage the nucleic acids of pathogens, combined with intense
cleic acids of viruses, bacteria, and parasites, agitation of the platelet unit to ensure a uniform
preventing their replication.139.These methods treatment of the entire volume of the unit. 145No
are also known to inactivate residual white cells photoactive compounds are added to the blood
and hence can be used to replace gamma or x- components; therefore, there is no chemical re-
ray irradiation to prevent GVHD.Because plate- moval step.!"
lets, plasma, and RBCsdo not contain genomic The THERAFLEXMB-Plasma System (Maco-
nucleic acids, they are relatively unaffected by Pharma) is CE marked for treatment of plasma.
pathogen inactivation treatment. 139 Methylene blue (MB)has been shown to inacti-
INTERCEPT (Cerus Corporation) employs vate pathogens by damaging nucleic acids.!"
amotosalen and ultraviolet A (UVA) light to MB is added to thawed FFP in pill form (85 ug
damage nucleic acids of pathogens. It has regu- anhydrous MB chloride) followed by activation
latory approval in Europe (CE mark), Canada using white light. Thawing of frozen plasma is a
(Health Canada), and the United States (FDA) precursor step for MB treatment in order to lyse
for treatment of platelets and plasma. Treatment white cells that may harbor viral partlcles.!" Af-
consists of the addition of 150 pM amotosalen ter activation, MB is removed with a filter (re-
followed by illumination with 3.0 Ioules/cnr' sidual concentration: 0.3 11M), and plasma can
UVAlight (Iv = 320-400 nm). After illumination, be refrozen. 146MB-treated plasma contains 10%
amotosalen is removed using the integral chemi- to 35% less Factor VIII and fibrinogen than un-
cal adsorption device (residual concentration of treated plasma, depending on varying analytical
amotosalen plus photoproducts: 0.02 mg/ml, procedures and laboratories.146
for platelets and 0.01 mg/ml.for plasma).140Av- Octaplas (Octapharma AG) is solvent/deter-
erage activity values for coagulation and anti- gent (SD)-treated, pooled human plasma ap-
thrombotic factors are reported to be within ref- proved in Europe (CE mark), Canada (Health
erence ranges for treated plasma.141In-vitro Canada), and the United States (FDA).SD plas-
studies have reported some loss in platelet prop- ma does not damage nucleic acids; rather, it dis-
erties and function. 142 rupts viral envelopes, cells, and most protozoa.
The Mirssol PRT System (Terumo BCT) em- It is not effective against nonenveloped virus-
ploys riboflavin (vitamin B2) and UV light. to es.147SD plasma is prepared from plasma pooled
damage nucleic acids of pathogens. It has a reg- from many donors that is tested for standard
ulatory CE mark for the treatment of platelets transfusion-transmitted agents and (nonenvel-
and plasma. Treatment consists of adding 35 mL oped) parvovirus B19 DNA, hepatitis A, and
of riboflavin followed by illumination for 6 to 10 hepatitis E virus RNA. It undergoes treatment
minutes with 6.24 Joules/mL UV light (Iv = with the solvent 1% tri-n-butyl phosphate and
280-400 nm). Riboflavinis a naturally occurring the detergent 1% Triton X-I00 to disrupt viral
vitamin that does not require removal. The lipid envelopes."? SDplasma is manufactured in
platelet or plasma product does not require any facilitiesthat can manage large-scaleproduction,
posttreatment processing and hence is ready for rather than in blood centers. Final transfusable
transfusion. Coagulation and anticoagulant pro- units consist of 200 mL of ABO-group-specific
teins are well preserved in plasma treated with plasma that is stored frozen at -18 C with an ex-
the Mirasol system.143,144 In-vitro studies have piration date of 12 months. 148Most coagulation
reported some loss in platelet properties and factors are reduced by approximately 10%in SD
function.'? The Mirasol PRT System is also ap- plasma, except for Factor VIII,which is reduced
proved under CE mark for treatment of WB for by 20%.149Also, levels of protein S and alpha;
antiplasmtn, which are labile to SD treatment, ponent for which any disqualifyinginformation
are controlled to ensure levels within the range has been generated.
of normal human plasma (>0.4 IU/mL).lso The Some blood components require emergency
component is labeled with the ABOblood group release because they have a very short storage
and, once thawed, should be used within 24 time. Emergency release requires physician ap-
hours. proval and a label or tie tag to indicate that test-
ing was incomplete at the time of release (AABB
Standard 5.27.3).4(P44)
IN Despite the widespread use of software to
IN control manufacturing processes, instances of
failure resulting in the distribution of unsuitable
All units of blood collected should be immedi- components continue to be reported to the
ately placed in quarantine in a designated area FDA, the majority of which are due to human
until donor information and donation records error or process control failures.lSI
have been reviewed, the current donor informa-
tion has been compared to the previous informa-
tion, the donor's previous deferrals have been
examined, and all laboratory testing has been
completed (21 CFR 606.100). Because of the
limited amount of time after collection that is The FDArequirements for labeling of blood and
avallable for component separation, WB units components are available in several publica-
may be separated into components before all of tions. The "Guideline for the Uniform Labeling
the aforementioned processes have been com- of Bloodand BloodComponents" was published
pleted. Separated components are quarantined in 1985.1S2 Detailed requirements for the label-
at the appropriate temperature until all of the ing of blood components are described in the
suitability processes have been completed and CFR(21 CFR606.120,606.121, and 606.122).
reviewed. Often, physical and electronic quar- AABB Standards requires that accredited facili-
antine are used simultaneously. ties label blood and blood component containers
Certain blood components from previous do- in accordance with the most recent version of
nations by donors whose more recent dona- the "United States Industry Consensus Standard
tions test positive for infectious agents also re- for the Uniform Labeling of Blood and Blood
quire quarantine and appropriate disposition, as Components Using ISBT 128" (AABBStandard
do units identified as unsuitable for transfusion 5.1.6.3.1 ).4(P12) This document outlines the in-
because of postdonation information. Other formation that must appear in the text of blood
components may need to be quarantined so that bag labels. Specifically,it defines the data identi-
QC samples can be taken and analyzed. For in- fiers use in transfusion and transplant settings,
stance, if a sample is obtained for bacteria detec- including the layout and precise placement of
tion, the component is held in quarantine for bar codes.
some predetermined time and then released if Base labels and any additional labels that are
the test result is negative. placed directly on the container must use ap-
A thorough understanding of the quarantine proved adhesives. In accordance with the 1985
process is needed to prevent erroneous release FDA guideline, only those substances that are
of unsuitable blood components. Components FDAapproved as "indirect food additives" may
may be removed from the quarantine area, la- be used in adhesives and coating components
beled, and released for distribution if all of the for labels placed over the base label.lS2 The FDA
donor information, previous donor records, and has additional standards for labels that are ap-
current test results are satisfactory.Optimally,la- plied directly on plastic blood containers. Tie
beling and release from quarantine are tightly tags may be used as an extension of the label if
controlled using the blood establishment com- there is insufficient label space, particularly for
puter system to prevent distribution of any com- informational items that do not have to be di-
rectly affixedto the container. National regulato- elude one or more of the following indications:
ry requirements should be verified when select- 1) hold for further manufacturing, 2) for emer-
ing labels and establishing labelingpolicies. gency use only, 3) for autologous use only, 4)
The FDA rulethatrequires all blood compo- not for transfusion,5) irradiated, 6) biohazard,
nents to be labeled with a bar-coded label be- 7) from a therapeutic phlebotomy, and 8)
came effective on April 26, 2006. The rule re- screened for special factors leg, HLAtype or cy-
quires that, at a minimum, the label contain the tomegalovirus (CMV) antibody status]. ISBT
following bar-coded information: 1) the unique 128 allows incorporation of special attributes of
facilityidentifier (ie, registration number), 2) lot the component, such as CMV antibody status,
number relating to the donor, 3) product code, from the previous donation.
and 4) ABO group and Rh type of the donor. As mentioned above, additional information
These pieces of information must be present in on the container can be conveyed using a tie
eye-readable and machine-readable format. The tag. Tie tags are especially useful for autologous
rule applies to blood establishments that collect and directed donations. Tie tags include the pa-
and prepare blood components, including hospi- tient's identifying information, name of the hos-
tal transfusion services that perform manufac- pital where the patient will be admitted for sur-
turing steps such aspreparation of pooled cryo- gery, date of surgery, and other information that
precipitate and/or divided units or aliquots of may be helpful to the hospital transfusion ser-
RBCs,platelets, and plasma for pediatric use. vice.
Another major part of labeling in the United Each component must also bear a unique
States is the Circular oj Injormation,2 which DIN that can be traced back to the blood donor.
must be made available to everyone involved in If components are pooled, a pool number must
the transfusion of blood components. The Circu- allow tracing to the individual units within a
lar is produced by AABB, the American Red pool.
Cross, America's Blood Centers, and the Armed An important source of information is ICCB-
Services Blood Program and is recognized as ac- BA (formerlyknown as the International Coun-
ceptable by the FDA. It provides important in- cil for Commonality in Blood Banking Automa-
formation about each blood component and tion). The ICCBBA website (www.iccbba.org)
should be consulted for information not includ- features updates and a revised list of product
ed in this chapter. codes. ISBT128 technical speclftcationssupport
Special message labels may also be affixed to the use of radiofrequency ID tags and other
blood component containers. The labels may in- means of electronic data transmission. 153,154
whole blood are typically defined based on the method used to separate the platelets from the
whole blood. .
6. Apheresis devices are continuous systems that remove blood from a donor, separate the blood
into the desired components, and return the remaining blood to the donor. Blood component
combinations are dependent on regulatory approval for each of the apheresis devices and indi-
vidual donor data.
7. Plasma preparations are defined and regulated through extensive combinations of differences
in collection methods, storage temperatures, freezing methods, secondary processing, timing,
and storage after thawing.
8. Postcollection modification to blood components may include prestorage leukocyte reduction
by filtration, pooling, cryopreservation, irradiation, washing, volume reduction, or pathogen in-
activation.
9. All units of blood collected should be immediately placed in quarantine in a designated area
until donor information and donation records have been reviewed, the current donor informa-
tion has been compared to the previous information, the donor's previous deferrals have been
examined, and alllaboratory testing has been completed.
10. The blood component identification process uniformly uses both a bar-coded and an eye-read-
able, unique DIN that is assigned to each sample tube and each component prepared from the
donation.
11. Bar-coded and eye-readable container labels follow the ISBT symbology (ISBT 128), which al-
lows identification of the manufacturer throughout the world, more product codes, better ac-
curacy as a result of reduced misreads during scanning, and enhanced conveyance of other la-
beling information.
REFERENCES
1. Chandler MH, RobertsM, SawyerM, Myers G. 5. Goldman M, Roy G, Frechette N, et al. Evalua-
The US military experience with fresh whole tion of donor skin disinfection methods. Trans-
blood during the conflicts in Iraq and Afghani- fusion 1997;37(3):309-12.
stan. Semin Cardiothorac Vase Anesth 2012; 6. Buchta C, Nedorost N, Regele H, et al. Skin
16(3):153-9. plugs in phlebotomy puncture for blood dona-
2. MBB, American Red Cross, America's Blood tion. Wiener klinische Wochenschrift 2005;
Centers, Armed ServicesBloodProgram. Circu- 117(4):141-4.
lar of information for the use of human blood 7. de Korte D, CurversJ, de KortWL, et al. Effects
and blood components. (October 2017) Bethes- of skin disinfection method, deviation bag, and
da, MD: AABB, 2017. [Available at https:// bacterial screening on clinical safety of platelet
www.aabb.org/tm/coi/Documents/coil 017.
transfusions in the Netherlands. Transfusion
pdf (accessedOctober 1, 2019).J
2006;46(3):476-85.
3. Code of federal regulations. Title 21, Food and
8. McDonald CP, RoyA, Mahajan P, et al. Relative
Drugs: Part 606, Current good manufacturing
values of the interventions of diversion and im-
practice for blood and blood components; Part
610, General biologicalproducts standards; Part proved donor-arm disinfection to reduce the
630, Requirements for blood and blood compo- bacterial risk from blood transfusion. Vox Sang
nents intended for transfusion or for further 2004;86(3):178-82.
manufacturing use; Part 640, Additional stan- 9. de Korte D, MarcelisJH, Verhoeven AJ, Soeter-
dards for human blood and blood products. boek AM. Diversion of first blood volume re-
Washington, DC: US Government Publishing sults in a reduction of bacterial contamination
Office,2019. (revisedannually). for whole-blood collections. Vox Sang 2002;
4. Gammon R, ed. Standards for blood banks and 83(1):13-16.
transfusion services. 32nd ed. Bethesda, MD: 10. Eder AF, Dy BA,KennedyJM, et al. The Ameri-
AABB,2020. can Red Cross donor hemovigilance program:
Complications of blood donation reported in chromatosis patients. Transfus Med 2015;

2006. Transfusion 2008;48(9): 1809-19. 25(4):239-42.
11. Newman BH. Blood donor complications after 25. Evers D, KerkhoffsJL,Van Egrnond L, et al. The
whole-blood donation. Curr Opin Hematol efficiency of therapeutic erythrocytapheresis
2004;11(5):339-45. compared to phlebotomy: A mathematical tool
12. Sorensen BS,Johnsen SP,JorgensenJ. Complica- for predicting response in hereditary hemochro-
tions related to blood donation: A population- matosis, polycythemia vera, and secondary
based study. Vox Sang 2008;94(2):132-7. erythrocytosis.J Clin Apher 2014;29(3):133-8.
13. Eder AF, Notari EP 4th, Dodd RY.Do reactions 26. Jolivet-GougeonA, Ingels A, Danic B, et al. No
after whole blood donation predict syncope on increased seroprevalence of anti-Yersima anti-
return donation? Transfusion 2012;52(12): bodies in patients with type 1 (C282Y/C282Y)
2570-6. hemochromatosis. Scand J Gastroenterol 2007;
14. Bravo M, Kamel H, Custer B, Tomasulo P. Fac- 42(11):1388-9.
tors associated with fainting: Before, during and 27. Sanchez AM, Schreiber GB, Bethel}, et al. Prev-
after whole blood donation. Vox Sang 2011; alence, donation practices, and risk assessment
101(4):303-12. of blood donors with hemochromatosis. JAMA
15. RiosJA, FangJ, Tu Y, et al. The potential impact 2001 ;286(12):1475-81.
of selective donor deferrals based on estimated 28. Leitman SF, BrowningJN, YauYY, et al. Hemo-
blood volume on vasovagal reactions and donor chromatosis subjects as allogeneic blood donors:
deferral rates. Transfusion 2010;50(6):1265-75. A prospective study. Transfusion 2003;43( 11):
16. Kamel H, Tomasulo P, Bravo M, et al. Delayed 1538-44.
adverse reactions to blood donation. Transfusion 29. McDonnell SM, Grindon AJ, Preston BL, et al.
2010;50(3):556-65. A survey of phlebotomy among persons with he-
17. France CR, France JL, KowalskyJM, et al. As-
mochromatosis. Transfusion 1999;39(6):651-6.
sessment of donor fear enhances prediction of 30. Phlebotomy guidelines for patients with heredi-
presyncopal symptoms among volunteer blood
tary hemochromatosis. Greenville, SCi Iron Dis-
donors. Transfusion 2012;52(2):375-80.
orders Institute, 2011. [Available at http://
18. Tomasulo P, Kamel H, BravoM, et al. Interven-
www.irondisorders.org/Websites/idi/files/
tions to reduce the vasovagal reaction rate in
young whole blood donors. Transfusion 2011; Contentl854256/Physician%20Chart%20phle
51(7):1511-21. botomy%20detail2011.pdf (accessed October 1,
19. Eder AF, KissJE. Adverse reactions and iron de- 2019).]
ficiencyafter blood donation. In: Simon TL, Mc- 31. Keohane C, McMuilin MF, Harrison C. The di-
Cullough J, Snyder EL, et al, eds, Rossi's princi- agnosis and management of erythrocytosis. BMJ
ples of transfusion medicine. 5th ed. Chichester, 2013;347:f6667.
UK:John Wiley and Sons, 2016:43-57. 32. Wagner SJ. Whole blood and apheresis collec-
20. Lee G, Arepally GM. Anticoagulation tech- tions for blood components intended for transfu-
niques in apheresis: From heparin to citrate and sion. In: Fung MK, Eder AF, Spitalnik SL,West-
beyond. J Clin Apher 2012;27(3):117-25. hoff CM, eds. Technical manual. 19th ed.
21. Food and Drug Administration. Fatalitiesreport- Bethesda, MD: AABB;2017:125-60.
ed to FDA following blood collection and trans- 33. Button LN, Orlina AR, KevySV,Josephson AM.
fusion - Annual summary for fiscal year 2016. The quality of over- and undercollected blood
SilverSpring, MD: CBEROfficeof Communica- for transfusion. Transfusion 1976;16(2):148-54.
tion, Outreach, and Development, 2016. [Avail- 34. Davey RJ,Lenes BL,Casper AJ, Demets DL.Ad-
able at https://www.fda.gov/media/111226/ equate survival of red cells from units "under-
download.] collected" in citrate-phosphate-dextrose-
22. Chin-YeeB, Lazo-LangnerA, Butler-FosterT, et adenine-one. Transfusion 1984;24(4):319-22.
al. Blooddonation and testosterone replacement 35. Prowse CV, de Korte D, Hess JR, van der Meer
therapy. Transfusion 2017;57(3):578-81. PF. Commercially available blood storage con-
23. Kim KH, Oh KY.Clinical applications of thera- tainers. Vox Sang 2014;106(1):1-13.
peutic phlebotomy. J Blood Med 2016;7:139- 36. Pietersz RN, de Korte D, Reesink HW, et al.
44. Storageofwhole blood for up to 24 hours at am-
24. Marrow B, Clarkson J, Chapman CE, Masson S. bient temperature prior to component prepara-
Facilitation of blood donation amongst haemo- tion. Vox Sang 1989;56(3):145-50.
37. Hardwick J. Blood processing. ISBTScience Se- 50. Food and Drug Administration. Guidance for in-
ries 2008;3:148-76. dustry: Recommendations for collecting red
38. Devine DV, Howe D. Processingofwhole blood blood cells by automated apheresis methods -
into cellular components and plasma. ISBTSci- technical correction February 2001. Silver
ence Series2010;5:78-82. Spring, MD: CBEROffice of Communication,
39. Levin E, Culibrk B, Gyongyossy-IssaM, et al. Outreach, and Development, 2001. [Available
Implementation of buffy coat platelet compo- at https://www.fda.gov/vaccines-blood-biolog
nent production: Comparison to platelet-rich ics/biologics-guidances/blood-guidances.]
plasma platelet production. Transfusion 2008; 51. Food and DrugAdministration. Guidance for in-
48(11):2331-7. dustry and FDAreview staff:Collection of plate-
40. Holme S, Heaton WA, Moroff G. Evaluation of lets by automated methods. (December 2007)
platelet concentrates stored for 5 days with re- SilverSpring,MD: CBEROfficeof Communica-
duced plasma volume. Transfusion 1994; tion, Outreach, and Development, 2007. [Avail-
34(1):39-43. able at https://www.fda.gov/vaccines-blood-
41. Ali AM, Warkentin TE, BardossyL, et al. Plate- biologics/biologics-guidances/blood-guidances.]
let concentrates stored for 5 days in a reduced 52. van der Meer PF, de Korte D. Platelet additive
volume of plasma maintain hemostatic function solutions: A review of the latest developments
and viability.Transfusion 1994;34(1):44-7. and their clinical implications. Transfus Med
42. van der Meer PF, CancelasJA, Cardigan R, et al. Hemother 2018;45(2):98-102.
Evaluation of overnight hold of whole blood at 53. Slichter SJ,CorsonJ, Jones MK, et al. Explorato-
room temperature before component process- ry studies of extended storage of apheresisplate-
ing: Effectof red blood cell (RBC)additive solu- lets in a platelet additive solution (PAS).Blood
tions on in vitro RBC measures. Transfusion 2014;123(2):271-80.
2011;51(Suppll):15s-24s. 54. RingwaldJ, Walz S, Zimmermann R, et al. Hy-
43. Perez-Pujol S, Lozano M, Perea D, et al. Effect perconcentrated platelets stored in additive
of holding buffy coats 4 or 18 hours before pre- solution: Aspects on productivity and in vitro
paring pooled filtered PLTconcentrates in plas- quality.VoxSang2005;89(1):11-18.
ma. Transfusion2004;44(2):202-9. 55. Cap AP, BeckettA, BenovA, et al. Whole blood
44. AckerJP, Hansen AL, KurachJD, et al. A quality transfusion. Mil Med 2018;183(SuppI2):44-51.
monitoring program for red blood cell compo- 56. YazerMH, Cap AP, SpinellaPC, et al. How do I
nents: In vitro qualityindicators before and after implement a whole blood program for massively
implementation of semiautomated processing. bleeding patients? Transfusion 2018;58(3):622-
Transfusion 2014;54(10):2534-43. 8.
45. Snyder EL,Whitley P, KingsburyT, et al. In vitro 57. Yazer MH, Spinella PC. The use of low-titer
and in vivo evaluation of a whole blood platelet- group 0 whole blood for the resuscitation of ci-
sparing leukoreduction filtration system. Trans- vilian trauma patients in 2018. Transfusion
fusion 2010;50(10):2145-51. 2018;58(11):2744-6.
46. PlazaEM, Cespedes P, Fernandez H, et al. Qual- 58. Nair PM, Pidcoke HF, Cap AP, Ramasubramani-
ity assessment of buffy-coat-derivedleucodeplet- an AK. Effect of cold storage on shear-induced
ed platelet concentrates in PAS-plasma, pre- platelet aggregationand clot strength. J Trauma
pared by the OrbiSac or TACSI automated Acute Care Surg 2014;77(3 SuppI2):S88-93.
system. Vox Sang2014;106(1):38-44. 59. van der Meer PF, Reesink HW, Panzer S, et al.
47. Lagerberg ]W, Salado-JimenaJA, Lof H, et al. Should DEHPbe eliminated in blood bags?Vox
Evaluation of the quality of blood components Sang 2014; 106(2):176-95.
obtained after automated separation of whole 60. LagerbergJW, Gouwerok E,Vlaar R, et al. In vi-
blood by a new multiunit processor. Transfusion tro evaluation of the quality of blood products
2013;53(8):1798-807. collected and stored in systems completely free
48. Johnson L, Winter KM, Kwok M, et al. Evalua- of di(2-ethylhexyl)phtha1ate-plastlcizedmateri-
tion of the quality of blood components pre- als. Transfusion2015;55(3):522-31.
pared using the Reveos automated blood pro- 61. Graminske S, Puca K, Schmidt A, et al. In vitro
cessingsystem.Vox Sang2013;105(3):225-35. evaluation of di(2-ethylhexyl)terephthalate plas-
49. Rajbhandary S, Whitaker BI, Perez GE. The ticized polyvinyl chloride blood bags for red
2014-2015 AABBBlood Collection and Utiliza- blood cell storage in AS-1 and PAGGSM addi-
tion Surveyreport. Bethesda, MD: AABB,2018. tive solutions. Transfusion 2018;58(5): 1100-7.
62. European Directorate for the Quality of Medi- 76. Food and Drug Administration. Guidance for in-
cines and HealthCare. Guide to the preparation, dustry: Bacterialrisk control strategies for blood
use and quality assurance of blood components. collection establishments and transfusion ser-
Strasbourg, France: Council of Europe, 2017. vices to enhance the safety and availability of
63. HessJR.Measures of stored red blood cell quali- platelets for transfusion. Silver Spring, MD:
ty. Vox Sang 2014;107(1):1-9. CBEROffice of Communication, Outreach, and
64. Kim DM, Brecher ME, Bland LA, et al. Visual Development, September 2019. [Available at:
identification of bacterially contaminated red https:llwww.fda.gov/regulatory-information/
cells. Transfusion 1992;32(3):221-5. search-fda-guidance-documents/]
65. Jordan A, Chen D, Yi Q-L, Acker JP. Assessing 77. McDonald C, Allen J, BrailsfordS, et al. Bacteri-
the influence of component processing and doal screening of platelet components by National
nor characteristics on red cell concentrates us- Health Service Blood and Transplant, an effec-
ing quality control data. Vox Sang 2016;111 (1): tive risk reduction measure. Transfusion 2017;
8-15. 57(5):1122-31.
66. Hansen AL, KurachJD, Turner TR, et al. The ef- 78. Berseus 0, Hogman CF, Johansson A. Simple
fect of processing method on the in vitro charac- method of improving the quality of platelet con-
teristics of red blood cell products. Vox Sang centrates and the importance of production con-
2015; 108(4):350-8. trol. Transfusion 1978;18(3):333-8.
67. Sweeney JD. Standardization of the red cell 79. McCullough J, Dodd R, Gilcher R, et al. White
product. Transfus Apher Sci 2006;34:213-18. particulate matter: Report of the ad hoc industry
68. Hogman CF, Meryman HT. Red blood cells in-
review group. Transfusion2004;44(7):1112-18.
tended for transfusion: Quality criteria revisited. 80. Welch M, Champion AB. The effect of tempera-
Transfusion 2006;46: 137-42.
ture and mode of agitation on the resuspension
69. Janatpour KA,PaglieroniTG, Crocker VL, et al.
of platelets during preparation of platelet con-
Visual assessment of hemolysis in red blood cell
centrates. Transfusion 1985;25(3):283-5.
units and segments can be deceptive. Transfu-
81. van der Meer PF, Dumont LJ, Lozano M, et al.
sion 2004;44(7):984-9.
Aggregates in platelet concentrates. Vox Sang
70. Kurach JD, Hansen AL, Turner TR, et al. Seg-
ments from red blood cell units should not be 2015;108(1 ):96-100.
used for quality testing. Transfusion 2014; 82. Moroff G, George VM. The maintenance of
54(2):451-5. platelet properties upon limited discontinuation
71. Murphy S, Gardner FH. Effect of storage tem- of agitation during storage. Transfusion 1990;
perature on maintenance of platelet viability- 30(5):427-30.
deleterious effect of refrigerated storage. N Engl 83. Moroff G, Holme S, George VM, Heaton WA.
J Med 1969;280(20): 1094-8. Effect on platelet properties of exposure to tem-
72. Dumont LJ, Gulliksson H, van der Meer PF, et peratures below 20 degrees C for short periods
al. Interruption of agitation of platelet concen- during storage at 20 to 24 degrees C. Transfu-
trates: A multicenter in vitro study by the BEST sion 1994;34(4):317-21.
Collaborativeon the effectsof shipping platelets. 84. Gottschall JL, Rzad L, Aster RH. Studies of the
Transfusion 2007;47(9):1666-73. minimum temperature at which human plate-
73. Dumont LJ,AuBuchonJP, GullikssonH, et al. In lets can be stored with full maintenance of via-
vitro pH effects on in vivo recovery and SUrvival bility. Transfusion 1986;26(5):460-2.
of platelets: An analysisby the BESTCollabora- 85. Smith JF, Ness PM, Moroff G, Luban NL. Reten-
tive. Transfusion 2006;46(8):1300-5. tion of coagulation factors in plasma frozen after
74. Wagner SJ,VassalloR, SkripchenkoA, et al. The extended holding at 1-6 degrees C. Vox Sang
influence of simulated shipping conditions (24- 2000;78(1 ):28-30.
or 30-hr interruption of agitation) on the in vitro 86. Dumont LJ, Cancelas JA, Maes LA, et al. The
properties of apheresis platelets during 7-day bioequivalence of frozen plasma prepared from
storage. Transfusion 2008;48(6): 1072-80. whole blood held overnight at room tempera-
75. VassalloRR,Wagner SJ,EinarsonM, et al. Main- ture compared to fresh-frozen plasma prepared
tenance of in vitro properties of leukoreduced within eight hours of collection. Transfusion
whole blood-derived pooled platelets after a 24- 2015;55(3):476-84.
hour interruption of agitation.Transfusion 2009; 87. Hmel PJ, Kennedy A, Quiles JG, et al. Physical
49(10):2131-5. and thermal properties of blood storage bags:
Implicationsfor shipping frozen components on 100. Yarranton H, LawrieAS,Mackie IJ, et al. Coagu-
dry ice. Transfusion2002;42(7}:836-46. lation factor levels in cryosupernatant prepared
88. Roth WK. Quarantine plasma: Quo vadis? from plasma treated with amotosalen hydro-
TransfusMed Hemother 2010;37(3}:118-22. chloride (S-59) and ultraviolet A light. Transfu-
89. Scott E, Puca K, Heraly I, et al. Evaluation and sion 2005;45(9}:1453-8.
comparison of coagulation factor activity in 101. Food and Drug Administration. Guidance for in-
fresh-frozenplasma and 24-hour plasma at thaw dustry: Pre-storage leukocyte reduction of
and after 120 hours of 1 to 6 degrees C storage. whole blood and blood components intended
Transfusion2009;49(8}:1584-91. for transfusion. (September 2012) SilverSpring,
90. Sheffield WP, Bhakta V, Mastronardi C, et al. MD: CBER Office of Communication, Out-
Changes in coagulation factor activity and con- reach, and Development, 2012. [Available at
tent of di(2-ethylhexyl}phthalatein frozen plas- https://www.fda.gov/downloads/Biologics-
ma units during refrigerated storage for up to BloodVaccines/ GuidanceComplianceRegulato
five days after thawing. Transfusion 2012; ryInformation/Guidances/Blood/UCM3206
52(3}:493-502. 41.pdf.]
91. SheffieldWP, BhaktaV,YiOL,Jenkins C. Stabili- 102. Schuetz AN, HillyerKL,RobackJD, HillyerCD.
ty of thawed apheresis fresh-frozen plasma Leukoreduction filtration of blood with sickle
stored for up to 120 hours at 1 degrees C to cell trait. TransfusMed Rev2004; 18(3}:168-76.
6 degrees C. J Blood Transfus 2016;2016: 103. Lutz P, DzikWHoLarge-volumehemocytometer
6260792. chamber for accurate counting of white cells
92. Bhakta V, Jenkins C, Ramirez-ArcosS, Sheffield (WBCs) in WBC-reduced platelets: Validation
WP. Stability of relevant plasma protein activi- and application for quality control of WBC-
ties in cryosupernatant plasma units during re- reduced platelets prepared by apheresis and fil-
frigerated storage for up to 5 days postthaw. tration. Transfusion 1993;33(5}:409-12.
Transfusion 2014;54(2}:418-25. 104. Dzik S, Moroff G, Dumont L. A multicenter
93. Ness PM, Perkins HA. Fibrinogenin cryoprecip- study evaluating three methods for counting re-
itate and its relationship to factor VIII (AHF)lev- sidual WBCs in WBC-reduced blood compo-
els. Transfusion 1980;20(1}:93-6. nents: Nageotte hemocytometry, flow cytome-
94. Callum JL, Karkouti K, Lin Y. Cryoprecipitate: try, and microfluorometry. Transfusion 2000;
The current state of knowledge. Transfus Med 40(5}:513-20.
Rev 2009;23(3}:177-88. 105. Strobel], Antos U, Zimmermann R, et al. Com-
95. Farrugia A, Prowse C. Studies on the procure- parison of a new microscopic system for the
ment of blood coagulation factor VIII: Effectsof measurement of residualleucocytes in apheresis
plasma freezing rate and storage conditions on platelets with flow cytometry and manual
cryoprecipitate quality. J Clin Pathol 1985; counting. Vox Sang2014;107(3}:233-8.
38(4}:433-7. 106. van der Meer PF, de Korte D. The buffy-coat
96. Smith JK, Bowell PJ, Bidwell E, Gunson HH. method. In: BJajchman M, Cid J, Loranzo M,
Anti-A haemagglutinins in factor VIII concen- eds. Bloodcomponent preparation: Frombench-
trates. J Clin PathoI1980;33(10}:954-7. top to bedside. Bethesda, MD: MBB Press,
97. Pesquera-LepatanLM, Hernandez FG, Lim RD, 2011 :55-81.
Chua MN. Thawed cryoprecipitate stored for 6 107. Rowe AW, Eyster E, Kellner A. Liquid nitrogen
h at room temperature: A potential alternative preservation of red blood cells for transfusion.
to factor VIII concentrate for continuous infu- Cryobiology1968;5(2}:119-28.
sion. Haemophilia2004; 1O( 6}:684-8. 108. Meryman HT, Hornblower M. A method for
98. Gunson HH. Variables involved in cryoprecipi- freezing and washing red blood cells using a
tate production and their effect on factor VIIIac- high glycerol concentration. Transfusion 1972;
tivity. Report of a working party of the Regional 12:145-56.
Transfusion Directors Committee. BrJ Haema- 109. Valeri CR, ZaroulisCG. Rejuvenationand freez-
tol1979;43(2}:287-95. ing of outdated stored human red cells. N EnglJ
99. Smak Gregoor PJ, Harvey MS, Briet E, Brand A. Med 1972;287(26}:1307-13.
Coagulation parameters of CPD fresh-frozen 110. ValeriCR.Viability,function and rejuvenation of
plasma and CPD cryoprecipitate-poorplasma af- previously frozen washed red blood cells. In:
ter storage at 4 degrees C for 28 days. Transfu- Chaplin HJ, JaffeER, Lenfant C, Valeri CR, edi-
sion 1993;33(9}:735-8. tors. Preservation of red blood cells. Washing-
ton, DC: National Academy of Science, 1972: 122. StollaM, FitzpatrickL, Gettinger I, et al. In vivo
265-97. viability of extended 4 degrees C-stored autolo-
111. Valeri CR, Ragno G, Pivacek LE, et al. A multi- gous apheresis platelets. Transfusion 2018;
center study of in vitro and in vivo values in hu- 58(10):2407-13.
man RBCs frozen with 40-precent (wt/vol) 123. Moroff G, Luban NLC. The irradiation of blood
glyceroland stored after deglycerolizationfor 15 and blood components to prevent graft-versus-
days at 4 degrees C in AS-3:Assessment of RBC host disease: Technical issues and guidelines.
processing in the ACP 215. Transfusion 2001; TransfusMed Rev 1997;11(1):15-26.
41(7):933-9. 124. Moroff G, Leitman SF, Luban NL. Principles of
112. Lelkens CC, de Korte D, Lagerberg JW. Pro- blood irradiation, dose validation, and quality
longed post-thaw shelf life of red cells frozen control. Transfusion 1997;37(10): 1084-92.
without prefreeze removal of excess glycerol. 125. Serrano K, Chen D, Hansen AL, et al. The effect
Vox Sang 2015;108(3):219-25. of timing of gamma-irradiationon hemolysis and
113. Valeri CR, Pivacek LE, Cassidy GP, Ragno G. potassium release in leukoreduced red cell con-
The survival, function and hemolysis of human centrates stored in SAGM. Vox Sang 2014;
RBCsstored at 4°C in additive solution (AS-I, 106(4):379-81.
AS-3or AS-5)for 42 days and then biochemical- 126. de Korte D, Thibault L, Handke W, et al. Timing
ly modified, frozen, thawed, washed and stored of gamma irradiation and blood donor sex influ-
at 4°C in sodium chloride and glucose solution ences in vitro characteristics of red blood cells.
for 24 hours. Transfusion2000;40(11): 1341-5. Transfusion 2018;58:917-26.
114. Valeri CR, FeingoldH, Marchionni LD.A simple 127. Strauss RG. RBC storage and avoiding hyperka-
method for freezing human platelets using 6 per lemia from transfusions to neonates and infants.
Transfusion 2010;50(9):1862-5.
cent dimethylsulfoxide and storage at -80 de-
128. Voak D, Chapman J, Finney RD, et al. Guide-
grees C. Blood 1974;43:131-6.
lines on gamma irradiation of blood components
115. Cid J, Escolar G, Galan A, et al, In vitro evalua-
for the prevention of transfusion-associated
tion of the hemostatic effectiveness of cryopre-
graft-versus-host disease. BCSH Blood Transfu-
served platelets. Transfusion 2016;56(3) :580-6.
sion TaskForce. TransfusMed 1996;6(3):261-
116. Valeri CR, Ragno G, Khuri SF. Freezing human
71.
platelets with 6 percent dimethyl sulfoxidewith
129. O'Leary MF, Szklarski P, Klein TM, Young PP.
removal of the supernatant solution before Hemolysis of red blood cells after cell washing
freezing and storage at -80 degrees C without with different automated technologies: Clinical
posthtaw processing. Transfusion 2005 ;45: implications in a neonatal cardiac surgery popu-
1890-8. lation. Transfusion2011 ;51(5):955-60.
117. Noorman F, van Dongen TT, Plat MJ, et al. 130. Tobian M, Savage WJ, Tisch DJ, et al. Preven-
Transfusion: -80 degrees C frozen blood prod- tion of allergic transfusion reactions to platelets
ucts are safe and effective in military casualty and red blood cells through plasma reduction.
care. PLoSOne 2016; 11(12):eOI68401. Transfusion 2011;51 (8):1676-83.
118. Slichter SJ,Dumont LJ,CancelasJA, et al. Safety 131. SillimanCC, Moore EE,Johnson JL, et al. Trans-
and efficacyof cryopreserved platelets in bleed- fusion of the injured patient: Proceed with cau-
ing patients with thrombocytopenia. Transfu- tion. Shock 2004;21 (4):291-9.
sion 2018;58(9):2129-38. 132. Cholette JM, Henrichs KF, Alfieris GM, et al.
119. Dumont LJ, Cancelas JA, Dumont DF, et al. A Washing red blood cells and platelets transfused
randomized controlled trial evaluating recovery in cardiac surgery reduces postoperative inflam-
and survival of 6% dimethyl sulfoxide-frozenau- mation and number of transfusions: Results of a
tologous platelets in healthy volunteers. Transfu- prospective, randomized, controlled clinical tri-
sion 2013;53(1 ):128-37. al. Pediatr Crit Care Med 2012; 13(3):290-9.
120. Reade MC, Marks DC, Johnson L, et al. Frozen 133. Cortes-Puch I, Wang D, Sun J, et al. Washing
platelets for rural Australia: The CLIP trial. An- older blood units before transfusion reduces
aesth Intensive Care 2013;41(6):804-5. plasma iron and improves outcomes in experi-
121. Waters L, Cameron M, Padula MP, et al. Refrig- mental canine pneumonia. Blood 2014; 123(9):
eration, cryopreservation and pathogen inactiva- 1403-11.
tion: An updated perspective on platelet storage 134. Hansen A, YiOL, Acker JP. Quality of red blood
conditions. Vox Sang 2018; 113(4):317-28. cells washed using the ACP 215 cell processor:
Assessment of optimal pre- and postwash stor- 145. Seltsam A, Muller TH. UVC irradiation for
age times and conditions. Transfusion 2013; pathogen reduction of platelet concentrates and
53(8):1772-9. plasma. Transfus Med Hemother 2011;38(1):
l35. Gruber M, Breu A, Frauendorf M, et al. Wash- 43-54.
ing of banked blood by three different blood sal- 146. Seghatchian J, Struff WG, Reichenberg S. Main
vage devices. Transfusion 2013;53(5):1001-9. properties of the THERAFLEXMB-PlasmaSys-
136. MoroffG, Friedman A, Robkin-KlineL,et al. Re- tem for pathogen reduction. Transfus Med
duction of the volume of stored platelet concen- Hemother 2011 ;38(1):55-64.
trates for use in neonatal patients. Transfusion 147. Hellstern P, Solheim BG. The use of solvent/
1984;24(2):144-6. detergent treatment in pathogen reduction of
137. Pisciotto PT, Snyder EL, Napychank PA, Hopfer plasma. Transfus Med Hemother 2011 ;38(1):
SM. In vitro characteristics of volume-reduced 65-70.
platelet concentrate stored in syringes. Transfu- 148. Hellstern P, Haubelt H. Manufacture and com-
sion 1991;31(5):404·8. position of fresh frozen plasma and virus-inacti-
138. Aster RH. Effectof acidificationin enhancing vi- vated therapeutic plasma preparations: Correla-
ability of platelet concentrates current status. tion between composition and therapeutic
Vox Sang 1969;17(1):23-7. efficacy.Thromb Res 2002; 107(Suppl 1):S3-8.
139. Seltsam A. Pathogen inactivation of cellular 149. Sharma AD, Sreeram G, Erb T, Grocott HP. Sol-
blood products-an additional safety layer in vent-detergent-treated fresh frozen plasma: A
superior alternative to standard fresh frozen
transfusion medicine. Front Med 2017;4:219.
plasma? J Cardiothorac Vasc Anesth 2000;
140. LinL, Conlan MG, Tessman I, et al.Amotosalen
14(6):712-17.
interactions with platelet and plasma compo-
150. Octaplas package insert and label information.
nents: Absence of neoantigen formation after
Hoboken, NJ:Octapharma USAInc, 2015:1-8.
photochemical treatment. Transfusion 2005; 151. Espinola R, O'Callaghan S, Weir L, et al. Quar-
45(10):1610-20. antine release errors (OREs)in blood establish-
141. Irsch J, Pinkoski L, Corash L, Lin L. INTER- ments: Summary of a public workshop. Transfu-
CEPTplasma: Comparabilitywith conventional sion 2012;52S:269A.
fresh-frozen plasma based on coagulation func- 152. Food and Drug Administration. Guideline for
tion-an in vitro analysis. Vox Sang 2010; the uniform labeling of blood and blood compo-
98(1):47-55. nents. (August 1985) Silver Spring, MD: CBER
142. Marks DC, Faddy HM, Johnson L. Pathogen re- Officeof Communication, Outreach, and Devel-
duction technologies. ISBTScience Series2014; opment, 1985. [AVailableat https://www.fda.
9:44-50. gov/vaccines-blood-biologics/biologics-guidanc
143. Larrea L, CalabuigM, RoldanV, et al. The influ- es/blood-guidances.]
ence of riboflavinphotochemistry on plasma co- 153. ISBT 128 standard technical specifications
agulation factors. Transfus Apher Sci 2009; v5.9.0. (March 2018) San Bernardino, CA: IC-
41(3):199-204. CBBA,2018:185.
144. Rock G. A comparison of methods of pathogen 154. KnelsR,Ashford P, Bidet F, et al. Guidelinesfor
inactivation of FFP.Vox Sang 2011;100(2): 169- the use of RFIDtechnology in transfusion medi-
78. cine. Vox Sang 2010;98(Suppl2):1-24.
7
lnfectleus Disease Screening
Lauren A. Crowder, MPH; Whitney R. Steele, PhD, MPH; and Susan L. Stromer, PhD
LOOD COMPONENTS,LIKEALLOTH- rus (HBV) accounted for only 25% of PTH cas-
er medications in the United States, are es.' Both HBV and non-A, non-B (NANB) hepa-
regulated by theFood and Drug Adminis- titis occurred more frequently in recipients of
tration (FDA).The FDArequires drug manufac- blood from commercial (paid) blood donors than
turers to verify the suitability of every raw in recipients of blood from volunteer donors. By
material in their products.' For biologicpharma- the mid-1970s, implementation of sensitive
ceuticals, the donor is the key ingredient whose tests for hepatitis B surface antigen (HBsAg) and
suitability must be verified. conversion to a volunteer donor supply resulted
A sample of blood from each donation must in a dramatic reduction in the incidence of both
be tested with screening tests approved by the HBV and NANB PTH. Still, NANB PTH contin-
FDA to identify donors and donated compo- ued to occur in approximately 6% to 10% of re-
nents that might harbor infectious agents. This cipientsof multiple transtustons.f
screening process is criticallyimportant because In the absence of a specific test for the
most blood components (eg, red cells, platelets, causative agent of .NANB PTH, investigators
plasma, and cryoprecipitate) are infused without searched for surrogate markers that could be
other treatments to inactivate infectious agents. used to identify donations associated with
Thus, infectious agents in a donor's blood at the NANB hepatitis. The presence of antibody to
time of donation that are not detected by the hepatitis B core antigen (anti-HBc) and/or the
screening process may be transmitted directly to presence of elevated alanine aminotransferase in
recipients. blood donors was shown to be associated with
an increased risk of NANB PTH.4-7 However,
concerns about the nonspecific nature of these
tests led to a delay in their implementation for
donor screening.
The concept of surrogate testing was revisit-
Table 7-1 shows the progression over time of do- ed in the early 1980s when concerns arose
nor testing for infectious diseases in the United about the transmission of AIDS by transfusions
States. Initially, donors were screened only for before the identification of its causative agent.
syphilis. In the 1960s, studies showed that In an effort to reduce the potential transfusion
greater than 30% of patients who received mu1- transmission of AIDS, some blood banks imple-
tip1e transfusions developed posttransfusion mented donor testing for anti-HBc,because this
hepatitis (PTH).2 Studies in the early 1970s antibody was highly prevalent in populations at
found that the newly discovered hepatitis B vi- increased risk of AIDS, and/or donor screening
Lauren A. Crowder, MPH, Epidemiologist, Scientific Aftatrs, American Red Cross, Rockville, Whitney
R. Steele, PhD, MPH, Director of Epidemiology, SCientificAffairs,American Red Cross, Rockville, Maryland; and
Susan L. Stramer, PhD, Vice President, Scientific Affairs,American Red Cross, Gaithersburg, Maryland
173
.s>.
"'0
o
..0
+=
<=
cU
"-
.E
1i)
.lB
"§
:.;:::
Q)
.c;
I-
vi
+=
~Cl.
Q)
.c;
:>
c:.:>
CC :J:
vi
2
« .s>.
o 2 "'0
LO
0)
..-
o
o
..0
- EC'd
(/)
Q) Q)
.s: (/)
cU
<= Q t)
*-
1i)
«
Cl
o "'0
LJ.. E o
>. '0
..0
"'0
Q)
(/) ~
Q)
:::! zs
~
"'0 rs C'd
"'0
<= Q)
cU
:5 ~
Ul
E
(/)
(/)
C'd
e
Cl.
Q)
Ul ~ "'0 E
Q)
~
Ul
is
Ul
:::l
o
-
1i)
Q)
~
:E
Cl.
x:
Ol Cl.
:.;:::
~Q)
:!2
(/)
iil
x:
o
<=
.~
Q)
~ ::::> 0 ~ >
-
~Q)
c:
Q)
s:
I-
:5~
-
«"'0
Q)
>
c:.:>
:J:
(/)
:E
I-
-"-
o
Ul
t)
~
OJ
c: >
c:.:>
c:
Q) :J:,
ec.::> ~
en ~
"-
o <=
c: o
a
o .~
en :g
::::> >-
"0 c:.:>
Q)
:J:
-
Ul
c:
Q)
o
c.::> >. Q
CC
=
"'0
:::J o
c: ..0
".j::l
<= +=
<=
Ul
Q) « «
OJ
c:
c;l
..c:
c..:> (/)
o I"-
LO
0) eo
,
r-
(/) (/)
0)
r-
o o LO
eo
eb
eo
eo
eo
o N
'8; I"-
0) 0) 0) 0)
0)
0)
0)
0)
CHAPTER 7 Infectious Disease Screening 175
>
<..:>
:c
.s
.>,
"C
o
..c
3
I-
:;::;
c: :c
C'<l o
+-'
1:5 c:
o
*
"C
o
N
o
+-'
E
"C
"C
C'<l
c:
G
:cI
~C'<l
(/)
>
g
zs
ro
"C
Q)
>
e
0..
E
(/)
C'<l
..c::
c:
o
'@
Q)
>
(/)
~
I
c..:>
C'<l
.c
1:5
*
-0
a
(/)
+-'
c:
Q)
c:
o «
2:
0..
Cl
E
o
c..:>
>
co
:c
10 §
Q3:p 1:5
ttl
*
C'<l
CiS
:::::: E "C
:;
.....I
'+-
o ro
OlE
.s
I- c:
= 0 ~
.,..... :cI 0.. _c..:> 2:
2: :;::;
c: ~ .~ >
co
:c « (/)$ :c
I'- 00
o o
o o
N <;J
<.0 <.0 0) c.b I'-
0) 0) 0) o o
0) 0) 0) o o
N N
-"'C
CL'l
::::l
c::
:;:::;
c::
o
8
(/)
CL'l
(/)
C'<:S
CL'l
(/)
o(/)
::::l
o
~
~c::
-t5
'-
o
~
Cl
c::
c::
CL'l
eo
en
'-
o
c::
o
Cl
en
~
"'C
CL'l
(/)
c::
CL'l
o
:.::J
.S
(/)
CL'l
Cl
c::
C'<:S
..c:
(_)
to
.,.... 0>
.,....
o o
N N
C HAP T E R 7 Infectious DiseaseScreening 177
for inverted CD4/CD8 T-cell ratio, which is an CA, clearly documented the efficacy of this ap-
immune abnormality found both in AIDS pa- proach. i2 In 1990, the FDA recommended ask-
tients and in people during the pre-AIDS incuba- ing each donor directly about each risk activity.
tion period of what was subsequently identified In 1992, the FDA issued comprehensive guid-
as human immunodeficiency virus (HN) infec- ance describing this questioning process.
tion." However, it was initially believed that the In the years since the discovery of HN, the
risk of transmitting AIDS by transfusion was too risk of transfusion-transmitted disease has been
low to warrant surrogate interventions." After progressively reduced through a variety of mea-
HIV was isolated and identified as the causative sures:
agent of AIDS, donor screening tests for anti-
body to HIVwere rapidly developed and imple- 1. Use of donor education and screening ques-
mented in 1985. tionnaires to exclude donors at increased risk
Once the HN antibody test became available
of blood-borne/window-period infections and
and cases of HN were recognized in both prior
donors and transfusion recipients, it became transfusion-transmitted diseases for which no
clear that the risk of transmitting HIV via blood tests are available.
transfusion had been greatly underestimated. io 2. Improving and/or adding tests to shorten the
The HIV experience highlighted the fact that an window period for specific agents by detect-
infectious agent associated with a lengthy ing earlier stages of infection and implement-
asymptomatic carrier state could be present in ing new donor screening tests, when
the blood supply for years without being recog- necessary.
nized. 3. Adoption of current good manufacturing
In the wake of this realization, the approach practice (cGMP) regulations to ensure that
to donor screening was extended beyond
unsuitable units are not collected and/or dis-
known agents. Current donor history evalua-
tributed.
tions include screening for, and the exclusion of,
donors with an increased risk of exposure to 4. Surveillance of the known and emerging
blood-borne or sexually transmitted infections. transfusion-transmissible diseases.
The intention is to reduce the likelihood that
the blood .supply will be subject to other as-yet- The approach used to screen potential do-
unidentifled agents that are potentially transmis- nors for an agent depends on whether specific
sibleby blood. risk factors are identifiable and whether donor
Transfusion transmission of HN persisted screening tests are available. Table 7-2 lists the
even after implementation of donor testing be- screening approaches used for different types of
cause of a delay of weeks or months between infectious agents.
the time a person is infected with HN and the
time the screening test for HIVantibody shows
positive results." Blood donated during this se- D N NIN T5
ronegative "window period" contains lnfec-
tious HN that is not detected by the donor The donor infectious disease tests required by
screening tests. the FDA are specifiedin Title 21, Part 610,40,
The most straightforward means of protect- of the ...Code of Federal Regulations (CFR).i .In
ing the blood supply from HN window-period addition to the CFR, the FDA communicates
donations is to exclude potential donors with an changes in its recommendations by issuing guid-
increased likelihood of exposure to HN. The ance publications. Although FDA guidance doc-
FDA initially recommended in 1983 that blood uments do not constitute legal requirements,
banks provide donors with informational materi- they define the standard of practiceintne Vnit-
als listing HN risk activities and requesting that ed States, and many blood collectors consider
individuals not donate if they had engaged in them legal imperatives. AABBalso issues recom-
these behaviors. Experience from San Francisco, mendations and requirements to the blood
TABLE7-2. Approaches to Donor Screening
Approach Context for Use ExampJe(s)
Questioningonly Infectious agentswith defined risk Malaria, prions

factors and no sensitiveand/or
specific test
Testing only Donor test is available,but no ques- West Nile virus

tion to distinguish individuals at risk
of infection
Questioningand testing Agents for which there are both Human immunodeficiency virus,
identified risk factors and effective hepatitis Band Cviruses, Babesia
tests
Useof blood components Agents with a high prevalencein Cytomegalovirus(universalleuko-

that test negativefor specific donors but for which an identifiable cyte reduction of cellular blood
recipients subset of recipients can benefit components has nearly replaced
from blood componentsthat test the use of cytomegalovirus-
negative seronegativeblood except in
selectedpatient populations)
Testing of blood components Infectious agent not detectablein Bacteria

donor samples; detection in blood
component (platelets) required
banking community. These are communicated collected at the time of donation and sent to the
either by AssociationBulletinsor by inclusion in testing laboratory. In addition, most platelet
AABBStandards for Blood Banks and Transfu- components are tested for bacterial contamina-
sion Services (Standards). AABBrecommenda- tion typicallyby the component manufacturing
tions and Standards do not have the force of facility.
law, except in California, where some sets of Laboratories that perform donation sample
AABBstandards have been incorporated into testing mandated by the FDAmust be registered
state law. with the FDA as biologics manufacturers be-
Since 1985, the FDAand AABBhave issued cause this "qualification of raw materials" is
a series of recommendations, regulations, and/ considered part of the blood component manu-
or standards for additional screening tests in ad- facturing process. The infectious disease tests
dition to the long-standing donor screens for and testing equipment used to screen donation
syphilis and HBsAg.Table 7-1 summarizes the samples must be approved (licensed or cleared)
chronology of changes in donor infectious dis- for this purpose by the FDA's Center for Biolog-
ease testing, and Table 7-3 lists the licensed do- ics Evaluation and Research. The FDAwebsite
nor screening tests that are performed by US maintains lists of assays approved for donor
blood banks. screening.13 The tests must be performed exact-
ly as specifiedin the manufacturers' package in-
Logistics of Testing serts. Tests and test platforms that are approved
All infectious disease testing for blood donor only for diagnostic use may not be used for
qualification purposes is performed on samples screening blood donors.
:c
Q.)
:0-
«
0
:::l
e::
eu
en
W-
...... ""
e::
0
c: 0 8-
g
eu
.s
en
« ......
0
0
W- zs
......
0
0
c:
.s
en ~
:::J
E
...... 0 .§
zs
0
~ Q)
of- 0
......
0 :§
"C ~ ~ c:
:::J
of-
~..o
« c:
-
"C
c:
c: 0 0 E o ....
c>S W- ~
~« ~ .§ W- eu
~ti)
c>S
>.
"C «0 « eu « ~
0
0
..0
2:~
0
2:
0::: :§ 2: ~
~ 0 ~
'';:::
> c:
>~
0::: ....
0 ~
c: cc 0
c..:l
« ;::: +-'
0
Li: «
c>S
..c: :c ~ :c 0 :c zs
e eu .!::::!
> "§ >
eu ~
W- eu
> ..-- N
UJ E :0-
Q)
en
c>S
:-2en
en «
*
eu 0::: :08 +-' E I I c:
en en :::J cc > ;::: eu Q)
eu c..:l eu
0::: c,
0
c, 2:
0
o, 0:::
0
c, s: :c $ g
0::: « « 0::: :;:!; 0::: -s 0::: «

c..:l UJ UJ c..:l LW c..:l LW c..:l UJ
c,
.....
....
0
....
0
c,
._
....0 n,
.... ....
0
n,
....0 0
(J)
«
0 « « 0
« 0
« «
~
(l)
:::J :::J « :::J « :::J « :::J
_. 2: s: ..c: 2: s: 2: ..c: 2: ..c:
c..:l c..:l c..:l c..:l
en I- I- I- I- c..:l
-
"0
(l)
·C
::J
c:
eu
en
c:
-(l)
..t::
.!:
"0
(l)
C>
:;::;
c:
c>S
e
0
c.;,
'Q3
+-'
ec.
......
c:
c>S
c:
E cc :0
'- en E
0 E 0
't:: 1; c.;,
(l)
e
-
c.... c.
eu
-
..c:
(J) "C
c: ~
0
(J) c>S ..--
(l)
>. en I
I-
c: "C eu
"C
;:::
0 :c
.=
Cl
t:: «
eu
C>
:;::;
c:
..0
:;::;
c:
:;::;
C.
eu
c.
0
+-'
>.
.,
"
(l) 2:
(l)
'- 0::: es
eu
~
~
«l
>
c..:l
"C
0
::J
l-
e..:>
en ~ c.;,
£' :c ..0
:;::; :c
2:
'- 0 ~:::J "C .s» c:
c>S
0
+-'
0 c: »
t:: ci. en c>S ~ -++
0 C. CC
"C
£' « "C
Cl en
.~ en
:;::;
2:
C> "<
2:
0
..0
:;::;
"<
2: "C
2:
0:::
0
..0
:;::;
"0 ::::::.. 0 c: 0::: c: c:
0 ~ ~c. c>S ro ..-- c>S
0 ..t::l :§ > ~C> > 2: >
I
~C>
co &5 eu
:c ~
CC
:c
c..:l
:c C>
s:
"0
(l)
(J)
c:
(l)
e..:>
:.:J
C?
r-- .~ N
--.. .,
"
I.U
-I '"
(I.)
..t::l > >
..--
2::
>
_J
= &5 CC
:c
c..:l
:c :c
I-
:c
~
=0-
Q.)
::::I
c::
:;:::;
c:: ::=; a:
o LJ.J (..)
S .....
o
a...
.....
en o
<:(
::J <:(
~ ..c:
(..)
:2
I-
"0
~
c::
:::J
Q.)
.s:
......
c::
"0
Q.)
E
o
't:
Q.)
0...
~
~
.=c::
Ol
Q.)
eu
U) '<c
z
._
o a:
c::
o s;:
o N
"0 .....
o o
o >
iii Z
"0 $:
Q.)
en
c::
Q.)
u
:.::i
c
II)
=
<
Serologic Testing Process blot assay) is no longer available, but revised

FDA guidance provides an alternate HCV sup-
Most serologic screening tests (assays for the de-
plemental testing pathway." Table 7-3 dis-
tection of antibody or antigen) are enzyme lm-
plays the available supplemental assays. If no li-
munosorbent assays (EIAs) or chemilumines-
censed supplemental assayis available, the FDA
cent immunoassays (ChLIAs). Typically, the
requires retesting of the donor sample using
screening process involves performing the re-
another FDA-licensed, -approved, or -cleared
quired test once on each donor sample. If the
test to provide additional information for donor
screening test is nonreactive, the test result is
counseling; unlicensed supplemental testing
considered negative. If a test is "initially reac-
may also provide useful information but cannot
tive," the package insert for the test typicallyre-
be used to requalify a donor or donation.
quires that the test be repeated in duplicate. If
A donation that is repeatedly reactive on a
both repeat results are nonreactive, the final in-
screening test may not be used for allogeneic
terpretation is nonreactive or negative, and the
transfusion, regardless of the results of further
unit may be used. If one or both repeat results
testing. Syphilis is the only disease for which
are reactive, the donor sample is characterized
negative results on supplemental tests can, in
as "repeatedly reactive," and the blood unit is
some circumstances, enable use of a screening-
not permitted to be used for allogeneic transfu-
test-reactive unit; this applies only to donations
sion. In the case of cellular therapy products,
screened using a nontreponemal assay. This is
there are some circumstances in which repeat-
discussed in the "Syphilis"section.
edly reactive donations may be used. (See "Con-
siderations in Testing Donors of Human Cells,
Nucleic Acid Testing
Tissues, and Cellular and Tissue-Based Prod-
ucts" later in this chapter.) Nucleic acid testing (NAT)was implemented to
The infectious disease tests approved for do- reduce the seronegative window periods de-
nor screening have performance characteristics scribed earlier. Screening for viral nucleic acid
chosen to make them highly sensitive. They are [ribonucleic acid (RNA) or deoxyribonucleic
designed to detect infected individuals and mini- acid (DNA)]is somewhat different from the se-
mize false-negativeresults. However, the assays rologic screening process. NAT requires the ex-
also react with samples from some individuals traction of nucleic acid from a donor plasma or
who are not infected (false-positiveresults). Be- serum sample followed by nucleic acid amplifi-
cause the blood donor population is preselected cation and detection of viral genetic sequences.
by questioning to be at low risk of infection, the The test systems that were implemented in
vast majority of repeatedly reactive results in do- 1999 to screen donors for HN and HCV RNA
nors do not represent true infections. To deter- were semiautomated and had insufficient
mine whether a repeatedly reactive screening throughput to allow individual testing of each
result represents a true infection rather than a donor sample. Testing of seroconversion panels
false-positive result, additional, more specific showed little loss of sensitivity if donor plasma
testing should be performed on the donor sam- samples were tested in small pools [minlpools
ple. (MPs)],because levels of HIVand HCV RNAare
The FDA requires that repeatedly reactive typicallyhigh in the blood of infected individu-
donor specimens be further evaluated by FDA- als, and NATassaysare exquisitely sensitive.
approved supplemental assayswhen such assays Thus, in the initially approved NAT donor
are available.' The FDA has approved supple- screening systems, MPs of 16 to 24 donor
mental assaysfor HBsAg;HN type 1 (HN-1) an- samples were prepared and tested. Current sys-
tibodies; hepatitis C virus (HCV)antibodies; an- tems use pools of 6 to 16. If a pool is negative,
tibodies to human T-cell lymphotropic virus, all donations in that pool are considered nega-
types I and II (HTLV-IlII);and antibodies to Try- tive for HN and HCV RNA. If a pool showed
panosoma cruzi The original HCV antibody NAT reactivity, further testing to the individual
confirmatory test (the recombinant immuno- samples was performed to determine which
donation was responsible for the reactive test ing is reactive and repeat testing is nonreactive;
result. Donations that were nonreactive on this in many countries, such donations are discarded
additional testing could be released for transfu- because these results would not rule out the
sion. Donations that were reactive at the presence of a low concentration of virus. Donor
individual-sample level were considered reac- management for this scenario varies and may in-
tive for viral nucleic acid and could not be re- clude a deferral and reentry algorithm.
leased for transfusion. In the case ofWest Nile virus (WNV),ID-NAT
In recent years, fully automated NATsystems screening rather than MP-NATscreening is rec-
have been developed. The automated test plat- ommended when WNV activity is occurring in a
forms approved by the FDAfor donor screening specific geographic area. Zika virus (ZIKV)test-
use multiplex assaysthat detect HN RNA,HCV ing by investigational NATwas introduced in re-
RNA, and HBVDNA in one reaction. Some di- sponse to the explosive ZIKV outbreak in the
rectly discriminate the reactive target during the Americas during 2015-2017. Initially ID-NAT
initial screen; others require separate discrimina- was required, but recently MP-NAThas been al-
tory testing to identify which virus is present. lowed, using 6 to 16 donations per pool and
These systems are approved for testing of indi-
with triggers for converting to ID-NATthat are
vidual donations and pools of 6 to 16 donor
similar to those used for WNV.26
samples, depending on the platform. The avail-
ability of fully automated NATplatforms creates
Implications of Reactive Test Results
the possibility of performing routine screening
on individual donor samples [individual dona- A repeatedly reactive result on a screening test
tion screening (ID-NAT)]rather than testing of (orindividuallyreactive NATresult) typicallyre-
pools (MP-NAT). However, the predicted fre- sults in discard of the reactive donation. Linkage
quency of reactive donations identified solely by of laboratory information systems to blood bank
ID-NATas compared with MP-NAT at present computer systems prevents labeling and/or re-
does not justify this practice in the United lease of components from donations with reac-
States. It has been estimated that ID-NAT tive test results. A reactive test result may also
screening would minimallyincrease detection of indicate that the donor should be prohibited
infected donors, whereas the associated testing from making future donations, because many in-
cost would be significantly higher than with fections are persistent. Furthermore, past dona-
MP-NAT.15An additional concern is that donors tions may also be considered suspect because
might be deferred for false-positiveresults more
the exact onset of a donor's infection cannot be
frequently with ID-NAT screening than with
determined.
pooled screening.
Both the FDAand MBB have issued recom-
In contrast to serologic testing policies, re-
peat testing is not permitted by the FDAfor an mendations regarding whether reactive test
individually reactive NAT sample to determine results affect a donor's eligibility for future
whether the initiallyreactive result represents a donations, whether components from prior do-
true-positive result. If an individual (unpooled) nations should be retrieved (and if so, how far
sample is reactive by NATfor HN, HCV,or HBV, back in time), and whether recipients who pre-
the FDA requires that the corresponding blood viously received components from that donor
component be discarded and the donor be de- should be notified. These recommendations are
ferred for specified periods. FDA-approveddo- often guided by the results of supplemental or
nor reentry algorithms are available (Table 7- confirmatory testing performed on the donor
4).27 In countries where donor screening for sample.
HN IHCV IHBV is routinely performed by ID- Because these recommendations are com-
NAT, it is common practice for initially reactive plex, creating checklists detailing each action to
specimens to be subjected to repeat testing. be performed after a specificreactive test result
Practices vary regarding the management of do- is obtained may be helpful. Staff use these
nors and components in cases where initial test- checklists to document completion of each
CHAPTER 7 Infectious DiseaseScreening 183
(].)
(/) +-'
ctl <=
Q)
(].) ::I
(/) C'
i:3 Q)
(/)
(/) .0
:::l :::l
o en
'.g eo
.l!2
c:
'$
.0
» ~0.
..a
E
c:i ctl
en
::::l
0.
:i;: ::I
..Q :l:
o S?
U-
,_ £
o COS'
c:
o o ~
Cl
OJ
,~ '§.
+-' Q)
c: en ....
'C Q)
......
0
.....
:::l <= <=

0" o ,~
(].)
0:: ~<= g
+-'
o .g~
:2: ~ <=
(/) e::,..g
'§
,_
~
Cl
2:'
~
o en
e. en
E -'"
~
"0
Q)
Q)
~ <0 :s:
"0 l.(') co
c:
ctl
't:l en
OJ <=
c: co ~
"0
~(].)
00
N Q)
l- <=
e. o
::::l <=
co
:i;: ..c::
......
..Q eo
o
U-
,_ E
o <=
c: o
o a:
Cl
OJ
c:
'C
1
!Xl 0
I'u;
c:
:::l I co
0" +=
c: (.):
U··
(].)
0:: « 0
~
..a
:§l
iIi
2:'
......
c:
(].)
(].)
0::
._
o
c:
o
Cl
E
"C
Q.)
E
;;::::
<=
o
'-'
<=
:::l
"5
.s::;,
a:
a:
kVJ
CD
I
"C
<=
ro
<= a:
o
.~ a:Z
'-'
ZI
:;:;
.s
J_~
:§
~;:!:
<=
Q.)
,-=7
E
Q.) ~~
C.
0- ® II>
:::l
VJ
ro
Q.)
~
J::
'0
VJ
<=
..c:::
1::
o
E
co
a: a:
a:
1
'-'
CD
1
CD
=7
'+'
I
I
<=
ro
~ro
"C "C
~
<= <= ~
ro ro "';"
a: a: >
1
Z
1
~
N
~
CD
I
,-CD
OI
'Vi =
"c:I
ca
.,...
~
~
II)
I:
>-
a: :>
I- i:'
Z
LU
LU
a:
=
CI)
'r:;
a:
o
=s
"c:I
u; Q
LU
_J
CD
==e
'" (.'J .§
~ :::::i = ,-
Z LU ca
> I- e ;:!:
CD
I
o
Z =
:c I
a:: a::
a::Z a::z
zi zl
J~ J_~
~2: ~2:
..-=f ..-=f
>-' .-
::c~
2::E
::C«
@ ®
Ul Ul
..c:: s:
1:: 1::
o o
E E
co co
<J.)
c::
o
"C
<J.)
"5
c::
> ....
o
~roa::
a:: a::
E:z za::
J_J
«0 JJ
«0
z::c Z.::c
6:§ 6:§
::C« ::C«
>-
a:: >-
a::
I- I-
z
UJ
z
UJ
UJ UJ
cc a::
a:: a::
a
LL.
a
LL.
UJ UJ
--'
CJ:l
--'
CJ:l
C!J C!J
::J
>
o ::J
UJ
I-
,
::c UJ
I-
a ""§ a
z « z
-e
II)
'>-
o ts '>-
o
'-'
e .>:2
.!:
.>:2
.!: :~
en
:::l
e
'S
c:; zs c:;
o CI" E
~eo II) ro
-E
c:
a:
...
II)
.c
."
x:
ci.
:::>
..
S
CI
II)
'E.
~ e
.E
o '"
U)
u;
......
o
=
CI
:;;
c:
o
o =
CI
-=~
Cl
c: CI.
·C
:::l
0-
=
CI
:!:!.
eo
c::: =en
II)
-
a:
eII)
n
....
"'in "'in "'in

.s::: x: .s:::
1:: 1:: 1::
o o o
E E E
C') C') C')
Cl
r::::
'c:."
en
e'-' C!l
w
2:
o
0..
rn
C<l
E
1
rn
1
rn
."
r::::
.l!:l .l!:l
o C<l C<l
c. E E
e en
+-'
."
r::::
."
r::::
.,,0 o o
ttjo.. c. c.
e
E4 ~
.l!:l v, r::::
o
+-'
r::::
o
<e.l!:l 2: 2:
C<l
E
."
r:::: C<l
o E
c. ."
e
+-'
r::::
o
c.
r:::: ."
o
2: ~
E
x:
~ ·E
1:: 0
Q;) Cl
ECii
.!!2>
c.. _J
c..1-
::s:c
(/)
"C
Q;)
(/)
c::
Q;)
g
(/) io
:E ..c::
o
c:: 1::
o
E E
to to
~ >-
r::c
I- I-
Z
LJ.J
z
LJ.J
LJ.J LJ.J
r::c r::c
r::c r::c
a
u;
a
u..
LJ.J LJ.J
_J Q) _J
co (..) co
C!:l ~ C!l
:::::i i::: ::J
LJ.J ('1:1 w
I- E I-
a
z =
::c
a
z
'"
a
~~
c:::
o
c:-
c:- =...
II)
=...
II)
II)
a:
II)
a:
-
=
"CI
.s: I.!!
"CI
.~ '=
CI"
II)
=
CI"
a:
II)
a: '5
z
'5 en
z
en :=z;
.::
;;;
II)
I-
II)
I-
=
.S!
== ca=
~
== =I.!!
"CI
"CI e,
e,
I.!!
~.
!;;:
2:
>
2:
3:
action as they perform it, and they should be even though the screening test results were neg-
part of standard operating procedures. ative.
Table 7-5 lists federal regulations, FDA guid- For HIV and HCV tests, the algorithms for
ance documents, AABB standards, and AABB managing prior donations and recipients of prior
Association Bulletins with recommendations re- donations are detailed in 21 CFR 610.46 and
garding management of blood donors with reac- 610.47. These requirements are replicated in
tive test results, retrieval of other components, the Centers for Medicare and Medicaid Services
and notification of prior recipients. These regu- regulations (Title 42, CFR Part 48?.27) to en-
lations and recommendations are described sure hospital transfusion service compliance
briefly below. with recipient notification requirements. For
other agents, recommendations for the manage-
Donor Eligibility
ment of previously donated components may be
found in the FDAguidance documents orMBB
FDA regulations in Title 21, CFR Part 610.41 AssociationBulletins (orboth) listed in Table 7-5.
address the deferral of donors with reactive In most cases, the FDA and AABB recom-
screening test results. FDAguidance documents mend retrieval and quarantine of any remaining
and AABBAssociation Bulletins contain more components from prior donations of that donor.
detailed recommendations regarding additional It is essential that the retrieval orm-date compo-
testing, donor eligibility, and donor counseling nents be initiated immediately after..the repeat-
for these and other tests. Donors must be noti- edly reactive result is obtained. This prevents
fied of any test results that affect their eligibility transfusion of these components while confir-
to donate or that could have important implica- matory testing is performed. The FDA requires
tions for their health. Systems that prevent fu- initiation of retrieval within 3 calendar days of a
ture collections from ineligible donors and the reactive HIV, HCV,WNV, ZIKV,or T. cruzitest
and within 1 week of a reactive HBsAg, anti-
release of any components inadvertently collect-
HBc, or anti-HTLVscreening test. Babesia tests
ed from such individuals are required.
also require component retrieval; however, no
The FDA has issued guidance documents
specific time-frame has been mandated by the
(referenced in Table 7-4) that define reentry FDA. If confirmatory test results on the current
pathways for donors deferred for reactivity on donation are negative, the FDA, in some cir-
anti-HIV, anti-HCV, HBsAg,and anti-HBc tests; cumstances, permits re-release of the prior dona-
serologic tests for syphilis and T. cruzi; and tions. In many cases, some or all components
HIV/HCV/HBV, WNV, Babesia, andZIKVNAT. from prior donations will have been transfused.
Most of the pathways require that the donor For some infectious agents, the FDA and AABB
have negative results on specified tests after a recommend that the recipients of prior dona-
defined waiting period. Reentry of donors must tions from confirmed-positive donors be noti-
follow the FDA-definedalgorithms explicitly." fied within 12 weeks via their physicians of
their possible exposure to the Infectious agent.
Retrieval of Prior Donations and Notification Recommendations for notification of recipi-
of Prior Recipients ("Look-Back") ents of prior donations ("look-back") are usually
issued by AABB,the FDA,or both, at the time a
The FDA and AABB offer guidance regarding new test is implemented, but these recommen-
the appropriate management of previously col- dations may evolve as supplemental (confirma-
lected blood components from donors whose tory) tests become available or medical treat-
current donation is repeatedly reactive (or, in ments are developed for the infection in
the case of NAT, individually reactive) on an in- question. Look-backis required by law only for
fectious disease screening test. These recom- HIV and HCV tests (21 CFR 610.46 and
mendations address the concern that at the time 610.47). For an HIV look-back investigation in-
of the previous donation(s), the donor could volving a deceased prior recipient, the next of
have been in the window period of an infection kin must be notified. The CFRspells out specific
lii,,;
I:)
i:!-
= '5 x x x x
I:) I»
CI I»
a:
..........
=
= ,S!
cu ......
'- .. 1'1:1
-
CI. Co:) X X X
"C:; !E
I»
a: I:)
z
1:; 'ii
":!::
..= a;
(II
(/) Co:) > I»
:::l
's, 'CI
I:) ';:: X X X
I:)
en I- a.. a:
-
Q.)
0:
(/)
Q.)
t-
'5
Q.)
>
+=i ... e I»
=
I:)
u I»
ctl I:) co X X X X X
Q.)
CI 1'1:1
0: =
1'1:1
Cl
c::
'§
==
..Q co
(5
L.l..
= .....a:
:;;
... a..=
(/) I»
c:: l- X X X X X X X X
'CI
0 I:)
+=i
U =
I:)
1'1:1
< CI
't:l
c::
ctl
Cl
c::
+=i
(/)
Q.)
t-
....
0 ~
c:: .e
0;
~
0
Cl c.o 70 '"
a
'j:!.. 00 00 R.
't:l
0 ..n '" ..n 5
0 ~ T-
C)
~ ~ N LO N
C) N ....
-
T-
CO a T- <0 KN C)
N
.... OC!
C)
C) '0 C)
N ex:! 0;>
0 .... 0;>
N ~ ....
'<I:
"'(0" LO LO ..Q
"'(0" ..Q '<I:

E
-
"0
ci C) ci N 0;>
E ....
'0 t5 C) C) 0;>
....
'0
(0 co ..Q
0;> ro :::l (0 (0
..Q
ro ~c..
Q.)
-e- 0 0
- -
(.) '0 Ol '0
~ t::: t::: t::: t::: (.)
0
0;> c:::
ro
:::l t::: t::: t> c:::
ro
0;>
CD ro ro ro ro 0 c:( ro ro 0 Cf)
n, c, e, o, c, c,
0:
cr: cr: cr: cr:
as as Cf)
as cr: cr:
as Cf) 0;>"
en LL LL LL LL
(.)
c:::
(.)
c:::
Cf) (.)
c::: LL LL
(.)
c:::
(.)
c:::
....
't:l
ctl
0 0 0 0
~""
ro
'0
I::::
co ""
'0
0 0
~""
~
co ""
'0
N '::; co '::; co '::;
-
't:l T- :::l T- ::::l
c:: N N N N
ctl 0;>
:;:;
0;>
:;:;
0;>
:;:; "'"
0;>
:;:;
Ol
c:(
Ol
-c
co
co
Ol
c:(
N
0;> 0;>
Ol
c:(
co
co
Ol
c:(
C/)
i= i= i= i=
0
LL
0
LL ~ 0
LL [§ [§ 0
LL ~ 0
LL
't:l
c::
ctl
en
c::
0
~:::l
Cl
0;>
0: 0
c..
:::l
on ....
0
r..:.. ~ Ol
L.LI T-
.....
CCI ~
T-
I
2:: >
co
:c :c :c
15
£'
Q.)
::J
c::
x x x x x x Eo
2..
x x x x x
x x x x x x
x x x x
x x x x
<0
F- ~
e;
~ ~
00
en (0 ?o ~
._
00 00 "-
~0 ~r-, CC!
'"E
..0
u1
LO
'enen"
(0
ee
u1
LO
T"""
T"""
0
N
a
~0 5
N
5
N
~
5
LO
LO
'" cc! ,.-

>- 0 cc! ._
'"
0 ; F- R 0
N
._ ._
N
,__
cc!
'-"
'"
LO
::; 5
LO
._
""0
..0 o::t: '<t
c::)
N
~
._
'"E
..0 '"E
..0 Q)
LO
._
Cl
E
1:
rn
J
E-
N
rn
E
'"
>
0
to
0
to to 00
'<t '"
.c
0
'" '"
.c
0
""0
rn
1';
- -
""0 ""0 '-" '-" ""0
c: c: 0 1:: 1:: 1:: 1:: ts ts c:
::I
""0
rn
U5
::I
""0
2: rn 2: ro
c,
rn rn es 0 '"
Cl '"
Cl 0 rn
c: c: as U)
as c, c, c,
as as ",- as U)
~ U) e '-"
c:
U) '-"
c:
CC
u...
CC
u...
CC
u...
CC
u... '-"
c: '-"
c: '-"
c:
U
c: U)
0
E t:::: 0
E
rn
""0
t:::: rn c.J c.J c.J c.J rn rn rn rn
""0
t::::
CO
.:; CO ""0
.:; ""0
.:; :E ""0
.:; .:; CO
'" CO
'" CO ,.- N ::I CO
E CO E r::n CO r::n N N N -;r r::n r::n r::n 0 CO
«
Cl
CO «
Cl
«
Cl
CO «
Cl
Q)
:;=l '" '"
:;=l '"
E
«
Cl
«
Cl
«
Cl
c:(
Cl
CO
u... ~ u... u... ~ u... i= ~ i= I- u... u... u... u... ~
+-
'-"
=.
CO
.;:::;
c: Q;'
ro c:
""0 .<3
c: '-"
ro '-"
CO rn
r::n r::n ::cI 2:-
«
en
«
en :;:::: > >
CO CO c: CO (.)
::c ::c « ::c ::c
...
= -=
i:::;'•
CI
CI II' x x x X
Q II'
a:
~·.·.··.c
=.S
.-CD ....ca
'0 .So .!::! X X
a:>
::::I
u:=
II' _
I:: a:.z
CI
:0:::;
I::
0
S
~C/) till 'S 'ii
.±::
::::I
C/)
u
'Ci.
CI
=.. :s
"CI
CI
>
II'
X X
-
a:> II'
0:::: I- a.. a:
C/)
a:>
I-
a:>
>
:0:::;
c:.;> .. e C
II'
'"
a:> =
CI
CI
II'
='
ca
X X X X X
0::::
C'l
I::
Q
=
ca
o§ ==
.£
"0
LJ....
==
:;::;
C/) till ~
... a..
II'
I::
0
I- X X X X X X X X
==
"CI
:0:::; CI
c:.;> ca
-c CI
"0 Q
I::
'"
C'l
I::
:0:::;
C/)
a:>
I-
'-
0
I::
0 ~
g K,.- ~
Cl '(0
0:;
70 0 70
co ex:! '" N cO
"0
0
0
~
0
t.r:i
'<:t
,.- tn 't!-.
,.-
....
<J.) t.r:i
0 .c
N
-
co 0
CCi 0
N
....
tn
N E
tn"
b N ex:! b ex:! .... ex:!

0
"'= ~ i::' io '<:t ::;; <J.)
.c co <J.)
.c
<J.)
<..>
<J.)
co
"0
1: 0 0 E 1: ....
"C
a:>
1ti C
II'
0
(0
0
(0 '"
::::I
l:i '"c:::co
"C
,.-
co
,.-
co
<J.)
15. '"
"C
E
<J.)
<..>
0
as
<..> '"
"C
t t t c::: c:::
(i5
e t <J.) <J.) <J.)
c:::
=
u; 0
0::::
U '"
c,
'"
c,
as
+-'
(f) '"
c,
'"
c,
(f)
as '"
+-'
(f)
as '"
"C
os '"
+-'
(f)
C/)
._
"0
CI a:
u;
a:
u...
<..>
c:::
(f) a:
u...
a: <..>
c:::
(f) <..>
c:::
Q u; c:n
.1:.. I:::: I:::: ~
'"
"0
(.) (.)
'"
:so! CD
(.) (.)
'"
:so! CD '"
:so!
~
CD
-'"
::::I CD ::::I CD ::::I CD
I::
!:
II'
N N c:n CD N N c:n CD c:n "C CD
<J.) <J.) <:( <J.) <J.)
C/) a; E 0
CD
E :;::;
<:(
0
CD <:(
0
<:(
0
CD
«
"0 a: I- ~ u, ~ f- i= u; ~ u, u; <:(
I::
'"
C/)
I::
0
~
::::I
C'l
o§
a:>
0:::: t5
Lt) in
f2
0
,.:.. II'
::::::: ~
S ~ c::
~
loLl
....I
CCI
<
==
II'
~
....J
f-
x:
C.
>. ~
I- < :r: (f) f-::
>< >< ><
>< >< >< >< ><

~
g
c::
8~
'" c:
E
~
E
:=J
a: g
>< >< >< >< >< >< "0"0
et=
~ ~
o 0-
U <J.)
Q) (J)
0::.0
.5 ~
~E
0"0
~ cu
>< >< >< >< >< >< u"O
~ Q)
"0 ~
Q) <J.)
Cl.Cl.
(J) :=J
(J) (J)
'" c:
.c: Q)
u Q)
:::>.0
(J) Q)
[,:;
.c:
>< >< >< >< >< >< >< >< >< >< ><
<0
en
C)
C)
C'\J R. ~
TI-
C)
.._
Q)
C)
I
(.0
C)
I
LO
C) .c
C'\J E
Q)
~ ~
Q)
c:: > c:: c::
o
:::l
-, 2: ~ ~:::l
::;
ai
(.)
ai
(.) CO co
c:: c:: c:: c::
ctS ctS o o
"t:J "t:J
":::; ":::; ~ ~
0> 0> 'zs
o
'zs
o
<:( <:( (J) (J)
Cl Cl (J) (J)
LL LL <:( <:(
~
:c
o
"t:J
c::
ctS
>
2:
LJ.J
Cl
"t:J
c:
.e
>
2: >
::.:::
5: N
timelines for component retrieval and recipient contained in cellular blood components. Frozen/
notification. It also specifies how far back in thawed plasma components do not appear to
time (ie, to which donations) the retrieval and transmit CMV infection. Immunocompromised
notification should extend. For other agents, patients who are at increased risk of transfusion-
such as WNV, ZIKV,Babesia spp, and T. cruzi, transmitted disease include fetuses, low-
recommendations regarding retrieval and recipi- birthweight premature infants who are born
ent notification are included in FDA guidance to CMV-seronegative mothers, and CMV-
documents and AABBAssociation Bulletins. In- seronegative recipients of solid organ or alloge-
vestigational protocols for unlicensed screening neic hematopoietic stem cell (HSC) transplants
assays may also include these requirements. Ta- from seronegative donors."
ble 7-5 indicates which of these documents ad- Most blood donors have had prior exposure
dress product retrieval or recipient notification. to CMV,indicated by the presence of CMVanti-
In the absence of published guidance, it is bodies. Therefore, it would not be possible to
not always obvious whether or when it is appro- produce an adequate supply of blood if all CMV
priate to notify prior recipients of their possible antibody-positivedonations were discarded.
exposure to infection. If there is no supplemen- It is possible, however, to minimize CMV
tal assay available or further testing performed, transmission to patients at risk of severe CMV
it is not possible to determine whether a repeat- disease, such as those described above. These
edly reactive screening test result for a donor patients should be supported with cellular blood
represents a true infection. Furthermore, if there components that have a reduced risk of trans-
is no effectivetreatment for that infection, there mitting CMV. These reduced-risk options in-
may be no medical benefit to the recipient of be- clude using blood components from donors who
ing told that he or she might have been ex- are CMV antibody negative or components that
posed. There could, however, be a public health have been effectivelyleukocyte reduced. The lit-
benefit from such a notification; specifically, a erature suggests that these two methods have
recipient who is alerted of a potential exposure similar but not identical efficacy, with an
can be tested and, if the results are positive, take estimated transmission risk by seronegative
precautions to avoid further spread of the infec- components of 1% to 2% vs a risk of 2% to 3%
tion. with leukocyte-reduced components.v? Recent
studies, however, found no CMV transmissions
Cytomegalovirus Testing of Components among a total of 176 carefullymonitored alloge-
for Immunocompromised Recipients neic stem cell transplant recipients who re-
ceived CMV-untested, leukocyte-reduced com-
Some common infections cause relatively innoc- ponents.f-" Because many at-risk patients
uous illnesses in immunocompetent individuals receive leukocyte-reduced components and are
but can cause severe disease in immunocompro- monitored closely for CMV infection, treated
mised patients; such is the case with cytomega- earlywith anti-CMVdrugs, or both, it is difficult
lovirus (CMV). to measure a benefit from also providing CMV-
CMV is a lipid-enveloped DNA virus in the seronegative components to these patients.
Herpesvitidee family. Like other herpesviruses, There has been a transition to reduced use of
CMV causes lifelong infection, typically in a CMV-seronegativeblood in favor of universally
latent state, with the potential for reactivation. leukocyte-reduced blood, particularly because
Primary CMVinfection in immunologicallycom- there is no evidence that CMV continues to be
petent individuals is mild, with symptoms rang- transmitted by leukocyte-reduced components.
ing from none to an infectious mononucleosis- However, there remains a small but theoretical
type syndrome. In immunocompromised pa- risk of transmission from blood components col-
tients, however, both primary infection and re- lected before seroconversion or following early
activated disease can be overwhelming and antibody production because such units contain
even fatal. CMV can be transmitted by blood the highest levels of CMV DNN°; CMV DNA
transfusion, primarily through intact white cells screening or pathogen inactivation (when uni-
versally available) would likely eliminate this ing and transplantation may be associated with
risk. The major risk of transmission occurs in transmission of infectious agents.59Autologous
low-birthweight infants as a result of breastfeed- tissues and reproductive tissues from recipients'
ing from a CMV-infected mother.51,52 sexually intimate partners may be exempt from
some testing requirements.
Autologous Donations Laboratories that test samples from HCTIP
donors must be registered with the FDA and
The FDA requires infectious disease testing of must use tests approved for testing of these do-
autologous donations that are shipped from one nors, when such tests are available. HCT/P test-
facility to another. If the receiving facility does ing requirements and approved tests may be
not permit autologous donations to be crossed found on the FDA website." Testing laborato-
over to the general inventory, the FDArequires ries must take care to check package inserts for
testing of only the first donation in each 30-day HCT/P testing methods; a package insert may
period [21 CFR610.40(d)(3)].The labelingof the require a different testing method for HCTIP
unit must be consistent with its testing status. donors than for blood donors. For example, NAT
Units from donors with repeatedly reactive tests for most types of HCT/P donors must be per-
must be labeled with biohazard labels. Some formed on individual donor samples;MP-NATis
hospitals have policies that prohibit acceptance not permitted for most HCT/P donor categories.
of autologous units with positive results on some In some cases, FDA regulations permit the
tests because there is a potential for an infec- use of HCT/P donations that are reactive on in-
tious unit to be transfused to the wrong patient. fectious disease screening tests. These excep-
tions are listed in 21 CFR 1271.65. The FDA
Considerations in Testing Donors of has issued specific labeling, storage, and notifi-
Human Cells, Tissues, and Cellular and cation requirements for these tissues. Testing of
Tissue-Based Products HCT/P donors for antibody to To cruziis not re-
Both the questions and tests required by the quired by the FDA (at the time of writing this
FDA to screen donors of human cells, tissues, chapter).
and cellular and tissue-based products (HCT/Ps)
differ from those for blood donors, and the re- International Variations in Donor Testing
quirements vary by type of tissue. The general Although this chapter focuses on infectious dis-
requirements are spelled out in 21 CFR 1271 ease screening in the United States, the general
and an August 2007 FDA guidance document approach to donor screening is similar in other
and are summarized in Table 7_6.54,55 The FDA countries. However, the specific donor ques-
has issued additional guidance regarding testing tions and tests vary from country to country
of HCT/Ps for syphilis" and WNV,57as well as based on the regional epidemiologyof infections
draft guidance documents on other infectious and tests available. For example, most countries
agents, including ZIKV.An up-to-datelist of FDA where WNV is not endemic do not test for this
HCT/P guidance documents may be foundon agent, although they may question donors and
the "Tissue Guidances" page of the FDA web- defer them for travel to WNV-affectedcountries.
site." Countries where HBVis hyperendemic cannot
The time frames for testing. HCT/P donors exclude donations from individuals who test
are specified in 21 CFR 1271 and the August reactive for anti-HBc, without adversely affect-
2007 FDAguidance document.=" In most cas- ing the adequacy of their blood supply. The
es, the samples for infectious disease testing AABB Subcommittee for the Evaluation of
must be obtained within 7 days before or after Variances considers variance applications from
the tissue donation. Samples from donors of pe- facilities desiring accreditation in countries
ripheral blood hematopoietic progenitor cells or where national practices and available testing
marrow may be tested up to 30 days before do- methods are different from those used in the
nation; however, longer intervals between test- United States. Some blood donations outside the
TABLE 7-6. FDA Testing Requirements for HCT/Ps (as of October 2018)
HIV Antibody to HIV-1 and HIV-2*

HIV-1 RNA*
HBV Hepatitis B surface antigen*

All tissues Antibody to hepatitis B core
antigen*
HCV Antibody to hepatitis C*

HCVRNA*
Treponemapal/idum FDA-licensed,-approved, or
-cleared donor screening test
All living donorst WNV WNVRNA*
For donors of viable leukocyte-rich HCT/Ps HTLV-IIII Antibody to HTLV-I/II*

(eg, hematopoietic progenitor cells or semen),
test for the following in addition to the above: CMV FDA-clearedscreening test for
anti-CMV (totallgG and IgM)
For donors of reproductive tissues, test for the Chlamydia FDA-licensed,-approved, or

following in addition to the above: trachomatis -cleared diagnostic test
Neisseriagonorrhea FDA-licensed,-approved, or
-cleared diagnostic test
*These tests must be FDAlicensedfor donor screening.

tUving donors must meet clinical and behavioraleligibility requirements for ZIKV exposure, but no testing is required to be
considered eligible.53
FDA= Foodand Drug Administration; HCT/Ps= human cells, tissues, and cellular and tissue-based products; HIV = human
immunodeficiency virus; HBV = hepatitis B virus; HCV = hepatitis C virus; WNV = West Nile virus; HTLV = human T-cell
Iymphotropic virus; CMV = cytomegalovirus; Ig = immunoglobulin.
United States are tested for agents not included of the infection in the donor population and the
in routine US donor testing. Hepatitis E virus nature of the donor screening processes in
(HEV),discussed later in this chapter, is an ex- place.
ample.
Agents for Which Blood Is Tested
Transfusion transmissions of HN, HCV, and

u HBVare now so rare that the rates of transmis-
RI K sion cannot be measured by prospective clinical
studies. The risk can only be estimated by mod-
Despite donor screening, blood components eling.
may still transmit infections. The residual risk of One theoretical source of risk is a virus strain
transmission varies according to the incidence that the current test kits do not detect. The Cen-
ters for Disease Control and Prevention (CDC), the average duration of the window period is es-
blood providers, and the test manufacturers con- timated to be 9.0 to 9.1 days for HN and 7.4
duct surveillance for such emerging strains." days for HCV.15,63 The window period for HBVis
Over time, the.FDAhas required that test manu- longer, currently estimated to be 18.5 to 26.5
facturers expand their detection capabilities to days, the reasons for which will be discussed lat-
include new strains. Asecond potential cause of er in this chapter.
transmission is a quarantine failure (ie, a blood The likelihood that a blood donation has
bank's failure to quarantine a unit that tests re- been obtained from a donor in the window peri-
active). Quarantine errors are thought to be rare od can be estimated mathematically using the
in facilitiesthat use electronic systems to control incidence x window-period model'":
blood component labeling and release because
these systems are designed to prevent the label- Residualrisk of window period donation =
ing and release of any unit with incomplete test- length of window period x incidence of
ing or a reactive test result. Erroneous releases infections in repeat donors
appear to occur more frequently in facilitiesthat
rely on manuai records and quarantine process- The incidence of infections in repeat donors
es; facilitiesusing manual processes are rare." can be calculated from the observed number of
The primary cause of residual transmissions
donors with a negative test result on one dona-
is thought to be donations from individuals in
tion but a positive result on a subsequent dona-
the window period of early infection, before test
results are positive. Figure 7-1 displays the se- tion (ie, seroconverting donors), divided by the
quence in which different types of donor screen- amount of time (usually in person-years) of sur-
ing tests demonstrate reactivity. Over time, the veillance or follow-up of the sampled popula-
window periods have been shortened by the im- tion. This method is limited by considering inci-
plementation of donor screening tests that de- dence rates only in repeat donors and does not
tect earlier infections. However, because no test permit assessment of the l~kelihood that first-
gives a positive result immediately after an indi- time donors might be in the window period.
vidual acquires an infection, the infectious win- This method also does not consider NAT-
dow period remains a concern. With MP-NAT, converting donors.
Nucleic Acid
or Antigen
r-.
'"0
o
o
ID /\
f '\
IgG...b'1!!body
,/ .......
/ .......
<,
/ <, . .
.................................................. :-:,....::.::._
\. ;/ (chronic infection)
\1.
/ .....
.......
r, ·· 1
; -,
/ ······..· ..(r!?~olvedinfection)
. .
Time
FIGURE7-1. Time sequenceof the appearanceof various markers of infection.

Other methods permit measurement of new might be transmitted by transfusion. It is impos-

infection rates in both first-time and repeat do- sible to estimate the risk of transmission for each
nors using tests that differentiate new from es- of the infectious agents for which donors are not
tablished infections. Such tests include NAT (ie, tested. The infections that are most likely to be
with NAT, donor blood that contains HN or recognized as transmitted by transfusion are
HCV RNA or HBV DNA, but not antibodies to those that have a distinctive clinical presenta-
these infections, can be interpreted as represent- tion and are otherwise rare in the United States.
ing a very early infection) and "sensitive/less The likelihood that an infection will be recog-
sensitive," "detuned" antibody, or limiting anti- nized as transfusion transmitted is enhanced if
gen (LAg)avidity testing. 15,63,65,66 When these al- the infection is usually associated with a clinical
ternative methods have been used, new HN or behavioral risk that the transfusion recipient
and HCV infections were two to four times lacks (eg, when malaria develops in a transfu-
more common among first-time donors than sion recipient who has not traveled to a known
among repeat donors.63,65·67 However, both of malaria-endemic region).
these donor populations have significantlylower If a life-threatening agent is recognized as a
infection rates than the general population. The potential threat to the blood supply, both AABB
continued importance of using donor education and the FDAtypically consider whether a donor
and questioning to select donors with a low inci- screening question could be used to exclude po-
dence of infection is explored in more detail in tentially exposed donors in the absence of a do-
the HN section below. nor screening test. Donor questioning regarding
The current estimated risks of HN, HCV, and travel to, and residence in, areas where an agent
HBVtransmission in donors, based on window- is endemic is currently the only means of pro-
period and incidence calculations, are shown in tecting the US blood supply from malaria and
Table 7_7.63,68
variant Creutzfeldt-Iakob disease. Most infec-
tious agents, however, do not have such clear
Agents for Which No Donor Screening
geographic risk areas. In general, it is difficult to
Tests Are Available
design donor questions that are both sensitive
Essentially any infectious agent that can circu- (ie, detect most infected individuals) and specif-
late in the blood of an apparently healthy person ic (ie, exclude only infected individuals).
TABLE 7-7. Estimated Risks of Transfusion-Transmitted Infection in the United States Based on the
Incidence/window-Period Model
Incidence. per 105 Infectious Window Residual Risk per

Study Period Agent Person-Years Period (days) Donated Unit
2007 -200863* HIV 3.1 9.1 1:1,467,000
2007 -200863* HCV 5.1 7.4 1:1,149,000
2009-201168t HBV 1.6 26.5-18.5 1:843,000 to

1:1,208,000
~risk~~sed ~AT in pools of ~ ....--~ ..- ....~. ~ ..~~~--.~~. ~-. ----
tHBV risk estimates are basedon MP-NATin pools of 16 using the Grifols Ultrio Plus assay.The range indicated for the
HBV window period reflects uncertainty regarding the minimum infectious dose of HBV (1 copy in 20 mL plasma vs 10
copies in 20 mL plasma).
HIV = human immunodeficiencyvirus; HCV = hepatitis C virus; HBV = hepatitis B virus; MP-NAT = minipool nucleic acid
testing.
An alternative method of protecting the ed virus identified initially in West Africa, only
blood supply from infectious agents is pathogen rarely seen in the United States."
inactivation.Heat-inactivation,solvent/detergent HN is transmitted by infected body fluids ei-
(SD)treatment, nanofiltration, chromatography, ther through sexual or parenteral contact (eg,
cold ethanol fractionation, and other approaches injection drug use), or vertically (fromInfected
have been used with remarkable success to inac- mothers to their infants). Although heterosexual
tivate or remove residual pathogens in plasma and vertical HN· transmission predominates. in
derivatives. Pathogen Inactivation systems for some parts of the world, new HIV cases in the
platelets and transfusable plasma have been United States continue to be concentrated in
available outside the United States for several men who have sex with men (MSM),injection
years, and one manufacturer's system was FDA drug users, and individuals with high-risk het-
approved for use in the United States in Decem- erosexual contact (defined as contact with an in-
ber 2015 (INTERCEPT Blood System). SD- dividual who is HN positive or in an identified
treated plasma (SD plasma) is also available for risk group for HIV, such as MSM or injection
transfusion in the United States; Octaplas (Octa- drug users).73
pharma) is manufactured from pools of human Current donation screening for HIVincludes
source plasma (630-1520 individual donors]." NATfor HIV-1RNAand serologictesting for an-
Pathogen inactivation systems are discussed lat- tibodies to HIV.The antibody tests approved for
er in this chapter. donor screening detect both immunoglobulin M
The AABBTransfusion-Transmitted Diseases (IgM)and IgGantibodyto both HIV-1and HIV-2.
Committee published an extensive review of in- Current-generation assays also detect antibody
fectious .agents that are possible threats to the to HIV-I group 0, a strain of HN-1 found pri-
blood supply.70 Potential mitigation strategies marily in Central and West Africa.With this de-
were discussed for each agent, including the tection claim, donor centers no longer need to
documented or theoretical efficacyof pathogen exclude individuals who have resided in, re-
inactivation processes. AABB periodically up- ceived medical treatment in, or had sex partners
dates. this information via its website, adding from HIV-1group 0 endemic areas."
materials for new potential threats as they are The average window period after HIV-1in-
ldentified." The agents deemed to pose the fection to test detection is currently estlmatedto
highest threat from either a scientific or public be 9.0 to 9.1 days for MP_NAT.15,63 Based on
perspective are briefly discussed in this chapter. window-period and incidence-rate calculations,
(See the 2009 supplement to TRANSFUSION the current risk in the United States of acquiring
and updates on the AABBwebsite for a more HN from transfusion from repeat blood donors
thorough review of these potential infectious is estimated to be approximately 1 in 1.5 million
risks_7°,71
) units (Table 7-7). Unpublished data indicate that
the residual risk has further declined to 1 in
2 million to 1 in 3 million units. However,
NIN p residual risks would be two to four times higher
if first-timedonor risk was included in these esti-
NT
rnates/" Also, it should.be noted that these risks
Human ImmunodefiCiency Virus are based on the transtustor, of only 1 unit, and
risk increases if multiple units are transfused.
HN-I, a lipid-enveloped, single-stranded RNA In the United. States, blood.donor screening
sphericalretroviruscontainingtwo linear,positive- questions exclude very broadly defined popula-
sense strands of RNA,was identified in 1984 as tions at increased risk of HIV. Given the short
the causative agent of AIDS. Blood donation delay of only days between infection.an.d.detec-
screening for antibodies to this virus was imple- tion of infection by NAT, experts have ques-
mented in the United States in 1985. In 1992, tioned whether donor interviews and exclusion
donor screening tests were modified to include of donors with increased risk remain medically
detection of antibodies to HN-2, a closelyrelat- necessary. The continued importance of a low-
risk donor population becomes evident if differ- nome with overlapping reading frames. Like
ent HN incidence figures are used for the blood HIV, HBVis transmitted by contact with body
safety calculation. For example, HN incidence fluidsfrom HBV-infectedindividuals.Jaundice is
rates as high as 1%to 8% have been observed in noted in only 25% to 40% of adult cases and in a
some high-risk populations, such as young ur- smaller proportion of childhood cases. A large
ban MSM.74,75 If an individual from a population percentage of perinatally acquired cases result in
with a 1% incidence of HIV donates blood, the chronic infection, but most HBVinfections ac-
likelihood that this individual is in the window quired in adulthood are cleared. HBVis highly
period and that the component will transmit prevalent in certain parts of the world, such as
HN can be calculated as follows: eastern Asia and Africa, where perinatal trans-
mission and resultant chronic infection have
Riskthat the donation is in window period = amplified infection rates in the population. In
length of window period x incidence of the United States, the incidence of acute HBV
infection in donor population = infection has decreased by at least 80% with the
(9.0 days/365 days/year) x implementation of routine vaccination pro-
(11100 person-years) = 114100. grams. Maternal screening for HBV and new-
born prophylaxis have also been effective in re-
This is the likelihood that a unit from this ducing perinatal transmission.
high-riskdonor would harbor HN but be missed During HBVinfection, DNA and viral enve-
by the current donor screening. This risk is lope antigen (HBsAg)are typically detectable in
clearly much higher than the estimated HN circulating blood. Antibody to the core antigen
transmission risk of 1 in 1.5 million for a unit of is produced soon after the appearance of HBsAg,
blood obtained from the current donor popula- initiallyin the form of IgM antibody, followed by
tion. Thus, despite the short window period IgG.As infected individuals produce antibody to
with current testing, inclusion of donors with a the surface antigen (anti-HBsAg), HBsAg is
high risk of HN would have a profoundly ad- cleared.
verse impact on blood safety.Accordingly, ques- The FDArequires donor screening for HBsAg,
tioning of donors for risk and temporarily ex- HBVDNA, and total anti-HBc (IgM and IgG an-
cluding those at increased risk to minimize tibody). Measurements of HBVincidence in do-
window-period donations continue to be criti- nors have been complicated by the transience of
cal for preserving blood safety. HBsAgand false-positiveresults on the HBsAg
Although there has been great interest in de- test." Published estimates of the infectious win-
veloping a more specific donor-screening algo- dow have varied because of differences in the
rithm based on specific behaviors instead of sensitivityof various HBVassays and lack of cer-
sexual preference that would exclude only indi- tainty regarding the level of virus in a blood
viduals who are truly at increased risk of HN, component that is required for infectivity.76,77
the FDAguidance issued in December 2015 in- Recent publications provide window-period esti-
dicated that the efficacyof a more specific algo- mates for different potential infectious doses of
rithm had not yet been established. An April virus (eg, 10 copies/20 mL of plasma vs 1 copy/
FDAguidance lists the current definitions of po- . 20 mL of plasma). The infectious window be-
tential risks for HIV exposure in the United fore a positive result on the Abbott PRISM (Ab-
States that require donor deferral. Most of these bott Laboratories) HBsAgtest has been estimat-
risk categories, including MSM behavior, now ed to be 30 to 38 days." With the addition of
require a 3-month deferral." HBVDNAtesting in MPs of 16, the window pe-
riod is estimated to have been reduced to 18.5
Hepatitis B Virus to 26.5 days (range dependent on whether the
infectious dose is assumed to be 1 copy or 10
HBVis a lipid-enveloped, spherical virus in the copies)." Using these MP-testing estimates, US
Hepedneviridee family.It is unique in that it has HBVtransfusion-transmissionrisk has been esti-
a partially double-stranded circular DNA ge- mated to be between 1 in 843,000 donations
and 1 in 1.2 million donations (Table 7-7).68 Re- screening in 1992, but the risk factors for the re-
sidual risks have been declining for HBV and are mainder of the.infectlons are not clear." Sexual
likely now in the 1 in 2 million to 1 in 3 million and vertical transmissions are uncommon, al-
range based on unpublished data. though coinfection with HN increases transmis-
Donor screening for HBV DNA is of value in sion rates by these routes.
detecting HBV-infected donors. HBV DNA is de- Current donor screening for HCV includes
tected during the infectious window period be- NATfor HCVRNAand serologic testing for anti-
fore HBsAg detection; however, DNA levels bodies to HCV. FDA guidance released in Octo-
may be below the limits of detection of MP-NAT ber 2019 recommends further testing of repeat-
assays_76,77Later in infection, following the clear- reactive donations by HCVNAT and, ifnonreac-
ance of HBsAg, HBV NAT may detect persistent tive, by a second licensed screening test." The
(ie, "occult") infection_76,77Such infections are average window period between infection and
interdicted in the United States by the donor detection of infection by MP-NAT is estimated
screening test for anti-HBc, with about 1% of to be 7.4 days." The serologic test detects only
anti-HBc repeat-reactive donations considered IgG antibody, a relatively late marker of infec-
to be from donors with occult HBV infection tion. Therefore, there may be a significant lag
due to the presence of HBV DNA in the abs.ence (1.5 to 2 months) between detection of RNA
of detectable HBsAg.68 High-sensitivity NAT is and. detection of antlbody." Donor questioning
required to detect occult HBV infections. be- has limited potential to exclude individuals who
cause viral loads are typically low. HBV NAT can may be harboring HCVinfection because a large
also detect acute HBV infections in individuals proportion of infected individuals are asymp-
who have previously been vacctnated.Y" Such tomatic and admit to no risk factors or possible
individuals may never develop detectable exposure. Despite this limitation, the current es-
HBsAg, but they may have detectable DNA. The timated US risk of HCV transmission by transfu-
infectivity of such donations is not known be- sion is extremely low-approximately 1 in 1.1
cause these units contain vaccine-induced anti- million (Table 7_7).63As already mentioned for
bodies to HBsAg. Routine HBV DNA screening HN and HBV,the residual risk of HCV has been
of US blood donations detects at least some of declining as well, possibly as low as the 1 in 2
these infections. There may come a time when million to 1 in 3 million range.
HBsAg testing is deemed unnecessary due to the
use of HBV NAT; the additional risk, if HBsAg Human T-CellLymphotropic Virus,
were discontinued with testing only by NAT and Types I and II
anti-HBc, has been projected to be 1 per 4.4 mil-
lion donations. 80
HTLV-I is a lipid-enveloped RNA retrovirus.
It was the first human retrovirus identified,
isolated in 1978 from a patient with cutaneous
Hepatitis C Virus
T-celllymphoma. A closely related virus, HTLV-
HCV is a small, lipid-enveloped, single-stranded II, was later isolated from a patient with hairy
RNA virus in the family Hsvtvtridee. HCV was cell leukemia. Both viruses are highly cell associ-
shown to be the cause of up to 90% of cases pre- ated, infect .lymphocytes, and cause lifelong
viously called NANB transfusion-related hepatl- infections, but most of these infections remain
tis." Most HCV infections are asymptomatic. asymptomatic. Approximately 2% to 5% of
However, HCV infection is associated with a HTLV-I-infectedindividuals develop adult T-cell
high risk of chronicity, which can result in liver leukemia/lymphoma after a lag of 20 to 30
cirrhosis, hepatocellular carcinoma, and a vari- years. A smaller percentage develop a neurolog-
ety of extrahepatic syndromes. ic disease called HTLV-associatedmyelopathy or
HCV is transmitted primarily through blood tropical spastic paraparesis. HTLV-IIdisease asso-
exposure. In the United States, about 55% of ciations remain unclear. Both infections are
HCV infections are associated with injection thought to be spread through blood, sexual con-
drug use or receipt of transfusion before donor tact, and breastfeeding.
HTLV-Iinfection is endemic in certain parts nation remain qualified as HTLV-negativefor all

of the world, including regions of Japan, South future donations.88,89
America, the Caribbean, and Africa.In the Unit-
ed States, infections are found in immigrants Hepatitis EVirus
from areas where it is endemic, injection drug
One agent that has received recent attention
users, and the sexuai partners of these individu-
globallyis HEV,a small, nonenveloped, icosahe-
als. Approximately one-half of the HTLVinfec-
dral, single-stranded RNAvirus in its own fami-
tions in US blood donors are with HTLV-II.84,85
ly, Hepeviridae. HEVwas first recognized in the
The only FDA-approveddonor tests for HTLV
1980s in Afghanistan among soldierswith unex-
infection are screening assays for IgG antibody
plained hepatitis. There is a single serotype but
to HTLV-Iand HTLV-II.Units that are reactive at least four genotypes with differinggeographic
on the screening assay may not be released for distributions and epidemiologic patterns. Geno-
transfusion. Until recently, only screening assays types 1 and 2 are generally associated with
were licensed by the FDA for HTLV-I1IIanti- large, waterborne (fecal-oral transmitted) out-
body detection. The screening tests do not dif- breaks in less-developedtropical countries. Gen-
ferentiate between HTLV-Iand HTLV-IIantibod- otypes 3 and 4 appear to be animal viruses that
ies. A Western blot licensed in December 2014 result in zoonotic infection of humans, most of-
(MP Biomedicals,version 2.4) uses recombinant ten through consumption of inadequately
and peptide antigens in addition to HTLV-Iviral cooked pork products. Genotype 3 is widely dis-
lysate to detect and differentiate between HTLV- tributed and is present in developed countries,
I and HTLV-IIantibodies. In February 2020, the and genotype 4 seems to be more common in
FDAreleased guidance allowing a reentry algo- certain Asian countries.
rithm for certain categories of donors who were The incubation period is 3 to 8 weeks, with
previously deferred due to reactive screening re- resulting illness that is generally self-limitedbut
sults." can result in fulminant hepatitis in those who
Risk estimates for transfusion-transmitted are immunosuppressed or patients with chronic
HTLV are somewhat uncertain, given the ab- liver disease. Genotypes 1 and 2 can be lethal in
sence of well-defined window periods for the pregnant women and their fetuses. Transfusion
current HTLVantibody tests. However, there is transmission, mostly of genotype 3, has been
no evidence of residual HTLV transfusion- well documented in Japan, France, England, the
transmission risk, which is estimated at less than Netherlands, and Spain, with greater than 30
1 per several million. Because HTLVis cell asso- transfusion transmissions documented to date."
ciated, leukocyte reduction likely reduces infec- Recent studies suggest a wide range of seroprev-
tivity; in addition, infectivity is reduced with in- alence rates of 20% to 40% in areas where HEV
creased refrigerated red cell storage.86,87 Like is endemic, but some of the variability may be
CMV,HTLVis thought to be transmitted only by attributable to the differences in performance
white-cell-containingblood components and not characteristics of the tests used, and some to di-
by frozen/thawed plasma components.rv" The etary habits. Most studies indicate a cohort ef-
low frequency of HTLV transfusion transmis- fect, with prevalence rates increasing with age.
sion (even in the absence of testing) is related to Transfusion infectivity is associated with the
the low incidence of infection in the United presence of viral RNA in plasma; reported fre-
States, for example. Because of this, the nearly quencies of RNA prevalence in donations have
universal use of leukocyte reduction, and the ranged from approximately 1 in 1000 to 1 in
poor persistence of HTLVin stored blood com- 10,000. In a large study in England, 225,000
ponents, some countries (eg, the United King- donations were tested, and 79 (1 in 2848) were
dom and the Netherlands) have transitioned or positive when tested for HEV RNA by an in-
are considering transition from testing each do- house NAT assay." Tracing was possible for 43
nation to testing donors only one time, so that recipients, of whom 18 (42%)showed evidence
donors with negative results on their initial do- of transfusion-transmitted infection, with 10
having prolonged infection; three were immu- tial screening is performed using a nontrepone-
nosuppressed individuals requiring treatment to mal test (eg, RPR)or a treponemal-specific test.
clear the virus, and another had clinical hepati- If screening is performed with a nontreponemal
tis. In contrast, in a smaller, blinded study in the test, additionaltesting with a treponemal-speciflc
United States, 7.7% of donors were anti-HEY test can be used to guide donor and component
positive and only two of approximately. 19,000 management. The FDA permits release of units
were RNA positive (1:9500).92 The main con- from donors who have reactive nontreponemal
cern is the finding that highly immunosup- screening test results and negative treponemal-
pressed patients (such as solid-organ transplant specificresults if the units are labeled with both
recipients) develop chronic HEV infections with test results.':" If screening is performed using a
long-term clinical sequelae, although these have treponemal-specific test, the FDA recommends
not been specifically linked to infection via additional testing by a second FDA-cleared
transfusion in the United States. treponemal test. If the second test is nonreac-
As a nonenveloped agent, HEVis not suscep- tive, the donor can be reentered, although the
tible to SD treatment or to current-generation component cannot be released. If the result of
pathogen inactivation technologies. Break- the second treponemal-speciflc test is reactive,
through infection following treatment has been the donor must be deferred for at least 12
documented, and commercial pooled lots of SD months, and before reentry must have docu-
plasma (Octaplas) are screened for HEV mented evidence from his or her physician or a
RNA.69,93 Asking donors about risk would not be public health laboratory of effective treat-
an effectivescreening measure because most do- ment." The same is true if a nontreponemal
nors would be considered at risk due to dietary screening test was used and if subsequent test-
factors (eg, consumption of pork or wild game). ing using a treponemal test is reactive.
NAT screening is the only effective protective The current value of donor screening for
measure. Routine NAT has been implemented syphilis is controverslal.r':" Although numer-
in several European countries (eg, England, Ire- ous cases of transfusion-transmitted syphilis
land, and the Netherlands) and areas of Japan were reported before World War II, no cases
with documented higher HEV incidence; other have been reported in the United States in over
countries are evaluating such testing, particular- 40 years. The low transmission risk is probably
ly for patients at increased risk, including solid- related to a declining incidence of syphilisin do-
organ transplant and HSC recipients. nors as well as the limited survival of the T. pall-
idum spirochete during blood storage." Syphilis
Syphilis rates have been increasing since 2000, particu-
larly in the MSM population, dampening hopes
Syphilisis caused by the spirochete Treponema
for elimination of donation testing."
pallidum. Donor screening for syphilis has been
One issue that has been considered is wheth-
performed for almost 80 years. Donors were ini-
er the syphilis screen improves blood safety by
tially screened by nontreponemal serologic tests
serving as a surrogate marker of high-risk sexual
that detect antibody to cardiolipin leg, rapid
activity. However, studies have demonstrated
plasma reagin (RPR)].In recent years, tests that
that donor screening for syphilis does not pro-
detect specific antibodies to T. pallidum have re- vide incremental value in detecting other blood-
placed RPR because these tests can be per-
borne and sexually transmitted infections, such
formed with automated testing instruments. as HIV,HBV,HCV, or HTLV.96
Most reactive donor test results do not repre-
sent infectious syphilis. Most reflect either bio-
Other Baderia
logicfalse-positive results or persistent antibody
in previously treated individuals (the former has Bacterial contamination of blood components
negative and the latter has positive treponemal- (malnlyplatelets) continues to cause transfusion-
specific antibody screening tests). FDA recom- related morbidity and fatalities;4o,99As defined
mendations vary depending on whether the ini- by FDAguidance, platelets are associated with a
higher risk of sepsis and related fatality than Standards requires blood centers to use a bacte-
any other transfusable blood component and ria detection method that is approved by the
are a leading cause of infection from blood FDA or validated to provide sensitivity equiva-
transfusion, with a rate of approximately 1 per lent to FDA-approved methods.31(pll)None of
100,000 transfused apheresis platelet unitS.40,100 these methods is sensitive enough to detect bac-
Bacteria are detected in approximately 1 in teria immediately after collection. All methods
6000 apheresis platelet donations by routine require a waiting time for bacteria contaminants
quality-control culture, but some escape detec- to multiply before the component is sampled.
tion to cause septic transfusion reactions.40,100,101 The process most commonly used in the
The source of the bacteria is most commonly United States to screen apheresis platelets is a
the donor's skin but can also be asymptomatic single-step, culture-based system inoculated at
bacteremia in the donor. least 24 hours after phlebotomy. After that time,
The level of bacteria in components just after a sample is withdrawn and inoculated into one
collection is generally too low to detect or to or more culture bottles. The bottles are then in-
cause symptoms in the recipient. However, bac- cubated in the culture system. Some blood cen-
teria can multiply during component storage, ters continue to hold the platelet components
particularly in platelet components, due to stor- during the first 12 to 24 hours of culture and re-
age at room temperature. Bacteria proliferate to lease them for use only if the culture is negative
a lesser extent in refrigerated red cells, and sep- at the end of that time. In all cases, the culture
tic reactions occur much less commonly with is continued for the shelf life of the unit. If the
these components. In rare cases, Red Blood Cell culture becomes positive after the component is
(RBC)units contaminated with bacteria capable released, the blood center attempts to retrieve
of growth at cold temperatures during the stor- it. If the component has not been transfused,
age period have caused life-threatening sep- resampling of the component for culture is infor-
SiS.99,101In 2004, to reduce the risk of septic mative because approximately two-thirds of the
transfusion reactions associated with platelets, initiallypositive signals are caused by either con-
AABBimplemented a standard for facilities to tamination of the bottle (and not the compo-
limit and detect bacterial contamination in all nent) or false signals from the culture sys-
platelet components. AABBrecognized in 2009 tem.100,101 All positive cultures should be tested
the availabilityin some countries of pathogen in- to determine the identity of the organism. If a
activation methods, which can replace bacteria true-positiveresult is related to an organism that
detection methods forplatelet components,31(Pll) is not a common skin contaminant but could in-
To limit blood component contamination by dicate asymptomatic bacteremia, the donor
bacteria from donor skin, two elements of the should be notified and advised to seek medical
blood collection process are critical. Beforeveni- consultation."
puncture, the donor skin must be carefully dis- Alltestingmethods are approvedfor leukocyte-
infected using a method with demonstrated effi- reduced apheresis platelets, and some are ap-
cacy. Most of these methods involve iodophors, proved for testing pools of leukocyte-reduced,
chlorhexidine, or alcohol.l'" Second, diversion whole-blood-derivedplatelets.103Single-stepcul-
of the first 40 mL or more of donor blood con- ture testing applies to whole-blood-derived
taining the contaminated skin plug into a sam- platelets and poststorage pools of whole-blood-
ple pouch and away from the platelet compo- derived platelets. Low-technology methods for
nent further reduces the likelihood that skin screening platelets just before issue, such as vi-
contaminants will enter the component.40,100.102 sually inspecting the platelets for swirling or
Since 2008, AABBhas required that collection testing them for low glucose or pH, lack both
sets with diversion pouches be used for all plate- sensitivity and specificity and do not fulfill the
let collections, including whole blood collec- AABBstandard for bacteria detection.31,102 How-
tions from which platelets are made.31(p22) ever, visual inspection of all platelets for signs of
A variety of technologies are availablefor de- bacterial contamination should occur before
tection of bacteria in platelet components. AABB transfusion regardless of the method of platelet
C HAP T E R 7 InfectiousDiseaseScreening 205
testing. FDA·cleared point-of-issue assays can be al culture for platelets, with no documented
used for detection of bacterial contamination of confirmed bacterial breakthrough infections in
platelet concentrates that are pooled immediate- over 2 million units of inactivated platelets
ly before issue. 104,105 transfused through 2017 .•Replacement of bacte-
Since the implementation of routine bacteria rial culture by pathogen inactivation has not
screening of apheresis platelets, the frequency been widely adopted in the United States due to
of FDA-reported fatalities from contaminated operational issues with the approved process.
apheresis platelets has declined." However, ap- The 2019 FDA guidance for lndustry" recom-
proximately 50% .of .contaminated apheresis mends interventions beyond those included. in
platelets escape detection by this early testing, AABBStandards. These additional measures in'
presumably because bacterial concentrations reo elude: secondary testing with a point-of-issue
main below the limits of detection at the time of test of previously cultured apheresis platelets or
sampling; thus, septic, and even fatal, reactions prestorage pooled platelets within 24 hoursof
occur. AABBhas recommended consideration of transfusion, reculture of platelets on day 4 with
policies to further reduce the risk of bacterially a hold for at least the first 12 hours after recul-
contaminated platelets, and the FDA finalized ture, or pathogen inactivation per device in'
guidance in September 2019, which classified structions for use. Enhanced primary culture
methods as either .slngle-step strategies or parts methods as used in England and in Canada (ie,
of two-step strategies.40,100 102 The point-of-Issue delayed-sampling, large-volume culture) are
assays mentioned above are cleared by the FDA now also recommended by the FDA These
as adjunct tests for apheresis platelets that have methods use a culture no sooner than 48 hours
been screened by another method. In a large from the time of collection with a sample vol-
clinical trial, one of these assays detected 9 ume of 1'6 mL inoculated evenly into aerobic
bacterially contaminated components among and anaerobic culture bottles. This practice al-
27,620 apheresis platelet units (lip 3069 com- lows for 7·day platelet dating and has demon-
ponents) that were negative by an early-storage strated significantreductions in the frequency of
culture-based assay.l'" There were also 142 septic transfusion reactions to rates of 'approxi-
false-positive results. At the time of writing, reo mately I in 1.2 milliontransfused platelets,
sults. for pomt-of-issue retesting of apheresis although one Staphylococcus sureus break·
platelets in routine use have been presentedIn a through infection and,threeotherS. aurelis near
single report, but without demonstrating addi- misses caught by visual platelet inspection' be-
tional yield. As of yet, point-of-issueretesting us- fore transfusion were documented. Another en-
ing FDA-approvedtests has not been widely lm- hanced culture method has been proposed (min-
plemented in the United States.!" imal proportional sampling volume) involving
The culture approaches used to date in the proportional culture of a minimum of3.8%of
United States provide incomplete assurance of collectedapheresis platelets vs a fixed volume of
bacteria detection. Enhancements include,.sec- 1.1%to 2.7% (le, 8 to 10 mL into only a single
ondary culture at days 3 or 4, use of point-of- aeroblc?ottle), but there are no dataon out-
issue secondary testing, or primary culture using comes cOI~parapletothose from England.104,109
delayed-sampling, large-volume methods. lOB. In
addition, some of these testing options may al-
Vector-Bor~~Infections
low extension of platelet dating to 7 days, pro·
vided that platelets are collected in deviceswith Before the late 1990s, malaria was the most
the appropriate 7·day labeling. Licensed patho- common vector-borne infection' recognized as
gen inactivation methods for platelets, which having the potential for transfusion transmis-
impair proliferation of bacteria, have been docu- sion. Since 2002, additional vector-borne diseas-
mented by active hemovlgilance in Switzer- es have been recognized as having the potential
land, France, and Belgium to reduce the risk of for secondary transmission by blood transfusion.
septic transfusion reacttons.!" In such coun- Each one has been evaluated as it emerged and
tries, pathogen inactivation has replaced bactert- has been addressed for the threat it posed to the
blood supply. For some, interventions such as retrospective cohort studies of blood donors
new blood donor screening assays were devel- with positive WNV RNA tests, 29% to 61% de-
oped and implemented, while for others donor scribed symptoms before or after donation, com-
screening questions were added. Not all re- pared to 3% to 20% of control uninfected donors,
quired immediate intervention, and surveillance demonstrating that screening with donor history
continues. The extent to which these vector- questionnaires is neither sensitive nor specific
transmitted infections have been perceived as enough to prevent transfusion transmission.112
threats to the US blood supply, as well as the To maximize efflciency,donation samples are
ease of developing a reasonable intervention tested by MP-NATin pools of 6 to 16 donations,
strategy,has ofteninfluencedwhat actionis taken. asis done for HCV,HBV,and HIV.However, be-
cause circulating levels of WNV RNA are low
West Nile Virus (the highest viral load documented has been
720,000 copies/mL), a donor sample may con-
WNV is a lipid-enveloped RNAvirus in the Fla-
tain a low concentration of RNAthat will not be
vivirtdse family.The first WNV cases were iden- detected by MP-NAT.l13It has been observed
tified in the West Nile district of Uganda in
that MP-NAT screening for WNV RNA fails to
1937. Afterward, outbreaks occurred in the
detect 50% or more of donations from infected
Middle East, South Africa, and Europe. First de-
donors due to dilution of low levels of RNA
tected in the United States in 1999, WNV sub- during pooling.38,113,114 Therefore, both the FDA
sequently spread throughout North America, ap-
and AABBrecommend testing by ID-NAT,not
pearing in virtually every US state in cyclical,
MP-NAT, during periods of WNV activity in a
annual epidemics every summer and autumn. particular geographic collection region.26,38 Re-
Transmission primarily occurs in a bird-to-bird
gionalWNV activity is monitored through active
transmission cycle by culicine mosquitoes, and
communications between neighboring blood
human infections occur incidentally. WNV viral
collection agencies and includes monitoring
loads in humans are too low to re-infect mosqui-
RNA-reactive donations, public health surveil-
toes. Approximately 80% of human cases are
lance of clinical WNV cases, and animal and
asymptomatic, 20% are associated with a self-
mosquito surveillance in the area.
limited febrile illness, and less than 1% are
Following the 23 cases of transfusion-
associated with severe neuroinvasive disease, transmitted WNV documented before the initia-
such as meningoencephalitis or acute flaccid tion of NATscreening in 2003, an additional 14
paralysis.
cases were reported, most related to donations
During the summer of 2002, a model sug-
having only low levels of RNA.92 All implicated
gesting a significant risk for WNV infection in
donations, except one in 2002, have been IgM
blood donors was published. 110 Concurrently,
negative.l13,llS Blood centers must remain vigi-
four recipients of organ allografts from a single
lant to make the rapid conversion from MP- to
organ donor infected by transfusion developed
ID-NATneeded for this seasonal strategy to be
neuroinvasive WNV infection. 111 Transfusion
effective.
transmission was documented in 23 recipients
in 2002 compared with a background of 2946
Zika Virus
cases of WNV-meningoencephalitisin the gener-
al population. 111 A multidisciplinary effort led to ZIKV is a tropical arbovirus of the flavivirus
the implementation of WNV investigational group, closely related to dengue viruses and
screening assays by the following summer. Be- transmitted to humans through the same Aedes
cause RNA-positiveunits, not antibody-positive mosquito vector. ZIKVwas first recognized in
units, were considered to pose the greatest risk Africa in 1947, then moved through Asia and
from this acute infection, NAT,rather than sero- subsequently the Pacific Islands. In 2007, the
logic testing, was required to protect the blood first large-scalehuman outbreak was recognized
supply. Donor tests for WNV RNAare required on the island of Yap in Micronesia, with subse-
for use by both the FDA and AABB.26,30,38 In quent spread to French Polynesia and other Pa-
cific Island groups in 2013, to Brazil in May The duration of ZIKVviremia is believed to
2015, and to the Caribbean in December be 1 to 2 weeks, consistent with other mosquito-
2015.115,116At the time of writing, 86 countries borne viruses. Viral clearance was estimated, us-
or areas had reported active transmission, in- ing diagnostic assays, to be 19 days for 95% of
cluding 48 countries or territories in the Ameri- affected patients (95% confidence interval of 13
cas, all involved in the pandemic from 2013 to to 80 days in a pooled analysisof published stud-
2017.117 Most cases (-80%) of ZIKV infection ies).134ZIKVRNA has been reported to persist
are asymptomatic.!" However, ZIKV infection longer in whole blood, semen, vaginal fluid, and
causes fetal loss, congenital ZIKV-related syn- urine, vs serum and plasma. One report found
dromes including microcephaly, and Guillain- persistence of ZIKVRNA (as opposed to infec-
Barre syndrome and other neurologic complica- tious virus) in whole blood for 5 to 58 days after
tions in adults.1l6,1l9-124 symptom onset despite RNA-negativefindingsin
The primary route of infection in US resi- corresponding serum samples; RNA positivity
dents is travel to areas with active vector-borne for 5 to 26 days occurred in urine from these
ZIKV transmtsston.Pv'" At the time of writing, same mdivtduals.f" The longest persistence of
there were 5430 travel-associated ZIKV cases in ZIKVhas been in semen. In one case, RNAwas
the continental United States reported to the detected for 62 days, and in others for 92 to 93
CDC, 52 sexually transmitted cases, and two days with the longest at 188 days, all in travel-
laboratory-acquired cases.l" In addition, a fatal, ers returning from ZIKV-activeor previously ac-
rapidly progressive infection acquired outside of tive areas.136-139ZIKV RNA duration was esti-
the United States resulted in a secondary local mated from following 150 ZIKV patients in
transmission to a caregiver without other ZIKV Puerto Rico positive by polymerase chain reac-
risk tactors.F' During 2016 and 2017, local tion (PCR).140The medians and 95th percentiles
mosquito-borne transmission was reported in (plus 95% confidence limits) for ZIKVRNAper-
two states: Florida (226 cases) and Texas (11 sistence were 15 (14-17) and 41 (37-44) days in
cases); none in 2018. In contrast, 37,115 locally serum, 11 (9-12) and 34 (30-38) days in urine,
transmitted cases of ZIKVclinical disease were and 42 (35-50) and 120 (100-139) days in se-
reported in the US territories, mostly in Puerto men."?
Rico. An additional 13 cases of Guillain-Barre Because of concern about severe disease as-
syndrome associated with ZIKVinfection have sociations, rapid virus spread in the Americas,
been reported to the CDC in the continental recovery of RNA from blood of asymptomatic
United States, and 51 in US territories.125 donors, and reports of probable transfusion
ZIKVRNA can be recovered from blood do- transmission, blood centers initially asked do-
nors, as demonstrated in the 2013-2014 ZIKV nors about travel to or residence in ZIKV-active
outbreak in French Polynesia, in which 2.8% of areas using a specific question with an associat-
donors tested RNApositive by NAT, and subse- ed 28-day deferral.141Because of increased
quently in Martinique and Puerto Rico with concern about the probability of autochthonous
rates of 1.8% in blood donors.118,128-130 At the transmission and travel-related and sexual trans-
time of writing, there have been four probable mission, the FDArevised its guidance in August
transfusion transmissions, all reported in Bra- 2016 to require universal ZIKVID-NAT or the
zil.131,132
These were identified when three do- use of approved pathogen inactivation technolo-
nors reported postdonation information of gies.39,142
dengue/ZIKV-like symptoms. None of the recip- Two NAT assays were used under investiga-
ients developed ZIKV-relatedsymptoms follow- tional new drug (IND) applications; both were
ing transfusion. In addition to mosquito-borne subsequently licensed in late 2017 and mid-
transmission and transfusion transmission, sexu- 2018. These assays have been used in the conti-
al transmission has been documented, mostly nental United States and in Puerto Rico. The do-
from an infected male to his partner (male or fe- nor screening NAT assays are more sensitive
male)116;female-to-malesexual transmission has than FDA-cleareddiagnostic assays.l" RNAhas
also been reported in one case.!" been reported in donor plasma samples as long
as 71 and 97 days after donor return from ZIKV- nosis; donors testing NATreactive and/or with
endemic areas.144,145 A recent report using one a clinical diagnosis are deferred for 120 days,
assay presented data from 15 months of investi- with subsequent reinstatement allowed after
gational testing of 4.3 million donations with that period." Predonation screening of donors
nine confirmed-positive donors identified (of for possible exposure to ZIKVthrough travel ac-
160 initially reactive, or a positive predictive tivityis no longer recommended.
value of 5.6%), yielding 99.997% specifidty.!"
Upon further investigation of the nine Other Arboviruses
confirmed-positive donors, two infections were
locally acquired, six were acquired while travel- There are other vector-transmitted infections
ing to ZIKV-endemicareas, and one donor test- that could be secondarily transmitted by transfu-
ed positive followingreceipt of an experimental sion. These agents and potential intervention
ZIKV vaccine. RNA levels were detectible in strategies are reviewed in the AABBemerging
RBCs for up to 154 days after donation and in infectious disease resources.Y" Two of these
plasma for up to 80 days after donation. The agents, dengue and chikungunya viruses (DENV
costs for identifyingthe eight donors infected by and CHIKV), have received attention because
mosquito (assumingassociated donations would donations containing viral nucleic acid were
be infectious) was $5.3 million/infected donor. documented during epidemics outside the conti-
This is true primarily because there was little nental United States; however, to date no specif-
vector-borne transmission in the United States ic interventions for either agent have been intro-
during the 2015-2017 outbreak, and subse- duced. Like ZIKV,the magnitude and clinical
quently, cases of ZIKVinfection have appeared Significanceof transfusion transmission for these
to be declining worldwide. Whether ZIKVwill agents remains unclear.
return in epidemic proportions remains to be Forty percent of the world's population lives
seen. There have been no local transmissions re- in areas with risk for DENV,including many ar-
ported in the United States during 2018, and all eas visited by US travelers. It has spread rapidly
68 cases reported in Florida (at time of writing) in Latin America and the Caribbean since the
have been from travelers to areas that remain 1980s. DENVis endemic in PuertoRico,the US
ZIKVactive.125 Virgin Islands, and American Samoa, and there
The FDAhas approved pathogen inactivation have been outbreaks in Hawaii, Texas, and Flor-
for platelets and plasma, which has been shown ida during the last 10 years.!" DENVis caused
to be effectivefor reducing relevant arbovirus ti- by four related flavivirusesspread person to per-
ters. Published data using the FDA-approved son by Aedes aegypti and A. albopictus (these
method demonstrated more than a 6 log., re- are the same vectors that transmit ZIKV).Most
duction in ZIKVinfectivity in plasma and plate- infections are asymptomatic, but illness ranges
lets, with similar reductions observed for RBCs. from undifferentiated fever to classic breakbone
ZIKVis also readily inactivated in SD plasma fever and severe dengue (dengue hemorrhagic
and other plasma-derivedproducts. 147-150 fever and dengue shock syndrome). An approxi-
An amended FDA guidance, published in mately 7-day viremia is a feature of both asymp-
July 2018, allows for the use of MP-NATuniver- tomatic and symptomatic infection, and asymp-
sallyin place of ID-NAT.27Conversion from MP- tomatic blood donors from Hong Kong,
NAT to ID-NATis required when triggered by Singapore, Brazil, and Puerto Rico have trans-
local vector-borne ZIKVactivity in a collection mitted DENV to blood recipients in seven
region or if a donation tests NAT reactive for clusters. Although the number of reports of
ZIKV,and local vector-borne transmission is pos- transfusion-transmitted infections is limited
sible (ie, other modes of transmission have been compared to the high rates of vector-borne infec-
excluded). Pathogen inactivation for plasma and tion, the lack of systematic surveillance for
platelet collections remains an alternative to transfusion-transmitted DENV makes its recog-
testing. In addition, the guidance requests do- nition in the face of widespread outbreaks prob-
nors to self-deferbased on a ZIKVdisease diag- Jemattc.!" RNA-positive,asymptomatic donors
have been identified in Brazil, Central America, cells (providing for the islands' needs by supply-
and Puerto Rico using NAT and antigen detec- ing blood from the French mainland) and by im-
tion tests. Rates of donor RNA positivity in Puer- plementation of limited NAT and the use of
to Rico are comparable to those found in US do- pathogen inactivation for platelets.l'" Other pre-
nors during the most active WNV seasons.153,154 cautions that have been used include strength-
A study in Brazil documented transfusion trans- ening requirements of postdonation informa-
mission from RNA-positive donors; however, tion from donors, a process enhanced by the
when the medical charts of infected recipients high (50%-80%)frequency of symptoms among
of blood components from those donors were infected subjects, along with deferrals tor resi-
compared to control recipients who did not re- dence in affected areas. Chikungunya symptoms
ceive an RNA-positive unit, there was no mea- are like those of dengue, but without the impact
surable apparent clinical illness.!" The patho- on the circulatory system. Arthralgia is a promi-
genesis and phenotypic expression of disease
nent symptom and may be prolonged. Although
may differ after transfusion vs infection from
routine tests are currently available, as combina-
mosquitoes, due to the absence of. promoters
tion NAT assaysWith DENV,they are not widely
that are present in mosquito saliva and other al-
used. Rates of CHIKV RNA detection in un-
terations of the virus that occur after mosquito
infection. 156,157 linked bloop donation samples reached 2.1 % in
In the absence of sustained outbreaks of
Puerto. Rico during the 2014 outbreak.P'' Viral
locally transmitted DENY in the continental titers in positiyedonatlons ranged from 104 to
United States, transfusion risk relates mainly to
109 RNAcopies per mL.158-160
return of infected, asymptomatic or presymp-
tomatic travelers. A 3- to 14-day incubation peri- Trypanosoma cruzi
od precedes symptom onset. Deferral for travel T. cruzt is the protozoan parasitic agent of Cha-
to malaria-endemic areas (12 weeks for US resi- gas disease. Chagas disease is endemic in parts
dents) offers some protection, but a large proof Mexico, Central America, and South Ameri-
portion of dengue-affected areas frequently visit- ca. It is transmitted to humans most commonly
ed from the United States are malaria free, and by an infected triatomine or reduviid bug, but
travelers to those areas could potentially intro- human-to-human transmission is possible via
duce the virus into the community and the
blood transfusion, organ/tissue transplantation,
blood supply. The conditions for sustained
congenitally, and from ingestion of contaminat-
spread of DENV exist in large areas of the Unit-
ed foodor beverages. The insect vector is associ-
ed States: a source of infection from travelers, a
ated with a wide variety of mammalian reservoir
susceptible population, and competent mosqui-
to vectors.
hosts in rural areas of countries where T. cruzi
CHIKV is another tropical arbovirus also
is endemic; human infections most commonly
transmitted by Aedes spp. mosquitoes. It is a occur when the vector cannot find an alterna-
togavirus of the alphavirus group first recog- tive mammalian host for a blood meal. Acute in-
nized in Africa. It has been responsible for ex- fection is usually self-limited,inVolvingl<;>ccll~z~d
plosive outbreaks in the islands of the Indian swelling atthe bite site and fever, but may ])~se-
Ocean, followed by spread to the Caribbean, vere 111lrnmunocompromised patients. Most.ln-
where more than 1.7 million clinical cases were fections become.chronic but remain asymptom-
reported from the end of 2013 to the middle of atic. Decades after the initial infection, 10% to
2015, including RNA positivity in blood do- 40% of infected individuals develop late-stage
nors.158,159There have been no reported cases of manifestations, including intestinal dysfunction
transmission by transfusion, but the similarity of or cardiac disease, which can be fatal. Transfu-
early infection to that of dengue has resulted in sion transmission of T. ctuzi from the blood of
significant concern. Notably, French authorities chronically infected, asymptomatic donors has
responded to the outbreak in the islands of the been recognized for decades in areas where it is
Indian Ocean by halting local collection of red endemic, although it has become uncommon
with decreasing use of fresh whole blood and had been identified in the United States and
implementation of donor serologic screening. Canada; cases with available data were linked to
The first blood donation screening EIA for platelet transfusion, and none were from a do-
antibodies to T. cruzi, using parasite lysate, was nor with recent infection. Twenty cases of trans-
approved by the FDA for use in the United fusion transmission in the United States, Cana-
States in December 2006. Although not initially da, and Spain have now been documented-
required by the FDA,the test was widely imple- again, all related to platelets from remotely in-
mented by US blood centers during 2007. Sub- fected donors who had formerly resided in areas
sequently, a second licensed screening test using where T. cruzi was highly endemtc.I" Since
ChUA was approved using recombinant anti- the implementation of US donor screening,
gens for antibody capture instead of parasite ly- confirmed-positivedonors have been identified,
sate as used for the EIA. Initially there was no and recipients of their prior donations have been
FDA-approvedsupplemental assay, but addition- notified and tested. Only two prior recipients of
al testing of reactive donations using an unli- platelets from one remotely infected donor born
censed radioimmunoprecipitation assay (RIPA) in a Chagas-endemlc country appear to have
was helpful for guiding donor counseling. Based been infected by transfusion since 2007, and
on the results of the latter assay, about 25% of they were identified only through look-backpro-
reactive US donors appear to be truly infect- cedures.l" Despite historically reported trans-
ed.161,162 An enzyme strip assay (ESA)using the mission rates of 10% to 20% from whole blood
same T. cruzi recombinant antigens as used in from infected donors in T. cruzi-endemic areas,
ChUA is now approved by the FDAas a supple- no T. cruzitransmissions by red cellsin the Unit-
mental assay but yields high rates of false posi- ed States have been documented (only one case
tivity. has ever been reportedj.!" The lower infectivity
Most US donations with reactive T. cruzi of red cell components compared to platelets or
screening test results and positive supplemental fresh whole blood is likely attributable to the
results are from donors born in T. cruzi-endemic limited survival of the parasite in refrigerated
areas of Latin America. The minority have con- components.
genital acquired infections transmitted from a
mother from a T. cruzi-endemic area. Only a Babesia
small number of donor infections appear to have
been acquired from vector exposure within the Babesia are intraerythrocytic parasites that are
United States (autochthonous cases). In the first the causative agent of babesiosis. Well over 100
2 years of universal donor screening in the Unit- species have been described worldwide. Human
ed States, no donor seroconversions (ie, incident Babesia infections are zoonotic, usually ac-
infections) were identified. 162 In December quired through the bite of an infected tick. In
2010, the FDAissued guidance recommending the Northeastern and Midwestern United
one-time testing of each US donor for T. cruzi States, the most common Babesia species is
antibodies.163 A subsequent 2017 guidance B. mtcrott. The vector is Ixodes scepulsris, the
maintalns one-time screening of donors in the same tick that transmits Lyme disease. In the
United States. For blood donors not found to be Western United States, Babesia infections are
positive or indeterminate on supplemental test- less common, and a different species, B. dun-
ing, the guidance provides an algorithm for do- cent, predominates. The vector for B. duncani
nor reentry." Finally, the FDArecommends re- in the United States is likely 1. pac(ficus. Report-
moval of the question on the Donor History ed human infections with Babesia are becoming
Questionnaire (DHQ) about a history of Chagas more frequent, and in 2011, the CDC made ba-
disease, due to the lack of specificityof the ques- besiosis nationally notifiable, although not all
tion and the efficacy of one-time donor screen- states require reporting. Babesia infection is usu-
ing.23,164 ally asymptomatic, even though parasites can
Before implementation of donor screening, circulate for months to years. In some individu-
seven cases of transfusion-transmitted T. cruzi als, however, Babesia infection presents as a se-
vere malaria-like illness that can be fatal. Gener- least 2 years and can donate only if the donor
ally, fatalities range from 6% to 9% but may be has not had a positive Babesia test in the last 2
as high as 21 % in immunocompromised pa- years and the donation tests NATnegative using
tients.25,168 Immunocompromised, elderly, and a licensed test. The licensed NAT assay and an
asplenic patients are at increased risk of severe additional NAT assay in use investigationally
disease, but following a review of published both have increased analytic sensitivity com-
transfusion-transmitted babesiosis (TTB)cases, it pared to the initial licensed NAT (PCR)assay.l72
has become clear that any recipient is suscepti- Licensed pathogen inactivation technology is ca-
ble to infection and serious illness.169 Treatment pable of significantlyreduclng B. microti infectivi-
with antibiotics is very effective, most common- ty and may be used as an alternate to a licensed
ly including oral atovaquone with azithromycin NAT.25,m
for 7 to 10 days; more severe cases often receive Babesia infection is most commonly diag-
red cell exchange transfusion. The key is prompt nosed when the intraerythrocytic parasites are
recognition and diagnosis of infection. seen on a blood smear; tetrads of parasites in in-
TTB is being identified with increasing fre- fected red cells, referred to as a Maltese cross,
quency. From 1979 to 2009, 162 TTB cases are characteristic of B. micrott (vs malaria) infec-
were described, but there were greater than tion. If a patient is suspected of having acquired
200 through October 2018.168,170,171 This is an the infection by transfusion, donors of the pa-
underestimate because most infections are as- tient's components can be recalled and may be
ymptomatic. In March 2018, the FDA ap- tested by research assays for Babesia antibodies,
proved two independent donor screening tests by NAT,or both; typically,thepresence of B. mi-
for B. mtcroti (one antibody and one PCR assay), croti nucleic acids or high-titer antibody in the
but neither remain commercially available." donor is suggestive of recent infection. Most of
Subsequently in January 2019, the FDAlicensed the donors implicated in TTB cases have been
a NAT assay for multiple Babesia spp. (B. micro- residents of Babesia-endemic areas, although
ti, B. duncani, B. divergens, and B. venato rum) rarely nonresidents of these areas are infected
that because of enhanced sensitivity can be used during travel to such areas and implicated in
without antibody testing. In May 2019, the TTB cases.25,168,!70,!74 The frequency of TTB
FDAreleased final guidance for Babesia via test- from travelers is estimated at 1 per 10 million
ing for nucleic acids using licensed NAT. donations."!
Through 2016, 95% of Babesia cases occurred The results of investigational blood donation
in the seven states considered to be Babesia- screening describing the testing yield, the dura-
endemic areas (all in the Northeast and upper tion of donor positivity, the relationship of test-
Midwest)." However, the FDAguidance consid- negative units to transfusion transmission, the
ers 14 US states and the District of Columbia to infectivity of test-positive units, and residual
be at risk (defined as being either a Babesia- risks have been reported.'?' Approximately
endemic state/district or. adjacent to states 90,000 donations from consenting donors were
where Babesia are highly endemic) and thus tested by two investigational assays: automated
recommended for testing. The past FDArequire- immunofluorescence for antibody detection and
ment that donors be asked if they have had a ID-NAT by PCR to detect the parasite's DNA
history of babesiosis and permanently deferred (these were the first two FDA-licensed assays
for affirmative responses can be replaced in ar- that are no longer commercially available]. Do-
eas of the country with testing, now that the nors reactive by either test were further tested
guidance is final. Blood establishments must ei- to confirm infection. The frequency of infected
ther test by licensed NAT in the 14 states plus donors in an area where Babesia are highly en-
DC (or other areas if added by the FDA in the demic and where screening occurred was 1 per
future) or retain a history of the babesiosis risk 300 donations, with 1 per 10,000 donations in
question regarding prior test reactivity or a med- the PCR-positive,antibody-negativewindow pe-
ical diagnosis. In either case (test reactivity or a riod. The current risk of TTB from an un-
medical diagnosis), the donor is deferred for at screened RBCunit in Babesia-endemic states is
1 per 100,000 but as high as 1 per 18,000 in ar- approved test is available to screen USblood do-
eas where Babesia are highly endemic. Receipt nations for malaria infection. Screening is ac-
of unscreened RBCsin areas where Babesia are complished solelyby donor questioning. Donors
highly endemic was nine times more likely to are excluded temporarily from donating blood
result in TTB than screened blood (screened after traveling to malaria-endemic areas, after re-
blood was not associated with any case of TTB). siding in malaria-endemic countries, or after re-
DNA clearance in 95% of infected donors oc- covery from clinical malaria. Donor questioning
curred after 1 year, but antibody clearance oc- has been effective at preventing transfusion-
curred in less than 10%during that interval. transmitted malaria (TTM) in the United States
The current versions of NAT assays (includ- with only 11 cases reported between 2000 and
ing the licensed test in use and an additional in- 2017, eight of which were due to P.jalciparum.
vestigational test) are ultrasensitive and run on Risk of TTM continues in the United States at
fully automated platforms. These assays detect rates of fewer than 1 per million collected blood
and amplify the parasite's ribosomal RNA units. Approximately 70% of TTM cases occur
(rRNA),of which thousands of copies per infect- due to failure to appropriately defer a donor
ing parasite are present, vs amplifying and de- during the screening interview, most notably be-
tecting the single chromosomal DNA template; cause the donor may not complete the DHQ
DNA will also be amplified and detected. ThUS, correctly. In a large majority of cases, this is
if even one infected red cell is sampled, it likely linked to donors with a history of residence (as
will be detected. In addition, these assays are opposed to short-term travel) in Africa.175-177
able to detect multiple Babesia species, in addi- This level of transfusion safety has been
tion to B. microti, that are human pathogens. achieved at a substantial cost in terms of donor
The use of these u1trasensitiverRNANATassays loss; malaria-related questions have excluded
is also consistent with the recommendations of hundreds of thousands of otherwise acceptable
the AABB Ad Hoc Babesia Policy Working US donors annually. Since 2013, FDAguidance
Group. 172 has redefined malaria-endemic areas as only
those for which chemoprophylaxis is recom-
Malaria mended.!" With this new definition, travel to
many popular tourist locations is no longer con-
Malaria is caused by an intraerythrocytic para- sidered a malarial risk; for example, in the past,
site of the genus Plasmodium. Infection is trans- Mexico has accounted for the largest proportion
mitted to humans through a mosquito bite. Five of deferred donors while having risk that is ex-
species account for most human infections: ceedingly low. 179 However, the 2013 guidance
P.jalciparum, P. vivsx, P. malariae, P. ovate, and adds a complex algorithm for evaluating travel
P. knowlesi. Disease symptoms include periodic by donors who have lived for more than 5 years
fever, rigors and chills, and hemolytic anemia. in malaria-endemic countries, because of a con-
Plasmodium parasites are present in the cir- cern about partial immunity causing prolonged
culation during asymptomatic infection and are asymptomaticparasitemia in such donors.
readily transmitted by blood transfusion. Recog- Outside of the United States, some countries
nition of infected recipients, as is true of Babesia that exclude donors after travel to malaria-
infection, may be complex and requires a high endemic areas permit reentry of these donors if
index of suspicion and identification of the para- they test negative for malaria 4 to 6 months af-
site in red cell blood smears, detection of para- ter completion of travel using an antibody test
site DNAby PCR,or detection by antibody tests. relying on recombinant antigens to two Plasmo-
Transfusion transmission is common in tropical dium spp. (P.fslciperum and P. vivax; other spe-
malaria-endemic areas and occurs elsewhere cies are detected at lower rates due to cross-
when infected travelers return from these areas reactivity). This is done routinely in France, En-
or especiallywhen residents of these areas with gland, and Australia, with only one suspected
partial or incomplete immunity donate in areas TTM case associated with a reentered donor (in
where malaria is not endemic. No FDA- France in 2012). A review of practices in five
countries where malaria is not endemic report- transmission of sporadic CJD through blood
ed 11 TTM cases from 2002 to 2013 (three in transfusion.182,183Nevertheless, blood donations
France, one in England, and seven in the United are not accepted from donors who are perceived
States). In the absence of all approved assay, at increased risk for this disease or from family
such a "test-in" reentry strategy has not been ac- members. of those who have died of sporadic
cepted by the FDA.180 CJD.18~
Pathogen inactivation methods have been Genetic prion disease makes up approximate-
shown to be effective at reducing transfusion ly 15% of all forms of human prion disease and
transmission in malaria-endemic areas. 181FDA is inherited from one or both parents through an
guidance in April 2020 allowed for an alterna- autosomal dominant mutation of the prion pro-
tive procedure whereby platelets and/or plas- tein gene.
ma components may be collected from residents In contrast, variant CJD (vCJD),is transmissi-
of countries where malaria is not endemic after ble by blood transfusion. It is caused by the pri-
travel to or through malaria-endemic areas. on that causes bovine spongiform encephalopa-
These residents may become donors without a thy (BSE),also known as "mad cow disease,"
deferral period, provided the components are with human infection occurring after ingesting
pathogen reduced with an appropriate device, neural tissues from infected animals. vCJD dif-
and the donor meets all other eligibility require- fers from C]D in that infected individuals are
ments.!" A randomized clinical trial involved young~r, present with psychiatric symptoms,
214 patients in Africa(Kumasi, Ghana) who and have' .a longer course from diagnosis to
completed a blinded evaluation comparing the death t:h~ those with sporadic CJD. Postmor-
efficacyofpathogen-reduced vs untreated whole tem diagnosisinvolves unusual florid plaques in
blood for prevention of TTM .(107 treated and the brain. Clinicalcases have occurred primarily
107 untreated patientsl.!" overall, 65 nonpara- in the United Kingdom,but cases have occurred
sitemic recipients (28 treated and 37 untreated) in other areas worldwide due to the export of
were exposed to parasitemic blood. The inci- contaminated animal tissues. The number of
dence of TTMwas significanqylower for the pa- cases reported has declined since the peak of the
tients treated with pathogen-reduced blood epidemic at the turn of the century. Approxi-
[1 (4%) of 28 patients] than the untreated group mately 230 cases have been seen in 12 coun-
[8 (22%)of 37 patients]. tries, with over 95% of these occurring in the
United Kingdom and Europe. Of those, four
Prions were cases of vC]D transmission by transfusion
in the United Kingdom.!" Three of these four
Prions are proteinaceous infectious particles that resulted in the development of clinical vCJD;
induce disease by triggering conformational the fourth was identified at autopsy of an indi-
changes in their naturally occurring protein vidual who died of underlying disease but
counterparts. These agents causefatal infections whose spleen and one lymph node contained
of the nervous system called transmissible spon- vC]D prions. In addition, one latent vC]D infec-
giform encephalopathies (TSEs). tion was identified in a patient with hemophilia
Sporadic Creutzfeldt-]akob disease (CJD)is a in the United Kingdomwho died of othercaus-
TSE that occurs sporadically throughout the es. This patienthadreceived Uk-plasrna-derived
world at an incidence rate of approximately 1 Factor VIII, including material from a donor
case per million population. Iatrogenic CJD has who later developed vCJD, suggesting that
been transmitted by injection of human-derived vC]D might have been transmitted by clotting
growth hormone or implantation of products factor concentrates.F" vCJD infection is ex-
from infected central nervous system tissues, in- tremely rare in the United States. The few re-
cluding dura mater grafts, corneal tissue, and ported cases have been in individuals who most
pituitary-derivedhormones. Surveillance studies likely acquired their infections elsewhere, and
in the United Kingdom and United States since no US transfusion-transmitted cases have been
the mid-1900s have shown no evidence of reported.
Although all but a single vC]D case reported Screening of Plasma Derivatives
to date have occurred on the genetic back-
Commercial plasma derivatives are prepared
ground of methionine homozygosity at the con-
from pools of plasma derived from thousands of
don 129 sequence of the prion protein gene, the
donors. Before the incorporation of specific
occurrence of a single case in a methionine-
pathogen inactivation processes, contamination
valine heterozygote raises questions about a po-
of these large pools with viral agents was com-
tential prolonged incubation period and suggests
mon. Today, plasma-derivative manufacturing
that vC]D deferrals will be maintained for the processes incorporate methods such as pro-
foreseeable future.l" longed heat or SD treatment to remove or inac-
There are no FDA-approveddonor screening tivate most known pathogens. SD treatment in-
tests for prion infections. Recent reports indicate activates lipid-enveloped agents, such as HN,
that some research technologies may be adapt- HCV,and HBV.Pathogen infectivitymay alsobe
able for prion diagnostics"; however, none of reduced by nanofiltration, chromatography, or
these technologies as of yet has adequate perfor- cold ethanol fractionation, which are used in
mance characteristics, including testing turn- the production of certain products. Not all infec-
around time suitable for blood donation screen- tious agents, however, are removed or inactivat-
ing. If tests became available, the issue of donor ed by these processes.
acceptance of screening for a fatal, incurable dis- One agent that can persist in plasma-
ease would be an ethical and social challenge.!" derivative products is human parvovirus B19.
However, the need to screen for prion-related This small, nonenveloped DNA virus is ex-
diseases may not be necessary considering that tremely resistant to physical inactivation. Acute
sporadic C]D has not been demonstrated to be infection is typically mild and self-limited;clini-
transmitted by transfusion and the incidence of cal manifestations include "fifth disease" (ery-
vC]D is miniscule and continues to decline. It is thema infectiosum) and polyarthropathy. Acute
thought that plasma-derivative manufacturing infection is associated with transient red cell
processes remove substantial amounts of TSEin- aplasia that may be clinically significant in im-
Iectlvity.!" Blood donors in the United States munodeficient individuals and those with un-
are screened solely by questioning and are ex- derlying hemolytic processes. The aplasia in im-
cluded if they have an increased risk of either munodeficient individuals can be prolonged.
C]D or vCJD. In April 2020 guidance, the FDA Intrauterine infection is associated with severe
revised long-standing deferral criteria, broaden- fetal anemia and hydrops fetalis. Parvovirus BI 9
ing the potential donor base. The new recom- infection is very common; most adults have anti-
mendations include removing deferrals for 1) bodies to this agent, indicating previous expo-
cumulative residence in Europe outside of the sure. Levelsof viral DNA during acute infection
United Kingdom, France, and Ireland, and 2) may exceed 1012 IU/mL, decreasing over weeks
military personnel who lived on any US military to months in association with antibody produc-
base anywhere in Europe. In addition, donors tion. Viral DNA, mostly at low concentrations,
are no longer asked about receipt of human has been detected in approximately 1% of blood
growth hormone derived from pituitary glands donations and in essentially all lots of pooled
or for receipt of UKbovine insulin, among other plasma derivatives. Transmission of parvovirus
changes. C]D exclusions are based on familyhis- B19 by transfusion has been linked only to
tory of the disease or on receipt of a cadaveric blood components or plasma products that con-
dura mater tissue graft. vC]D exclusions are tain high concentrations of viral DNA; only one
now for residence in the United Kingdom transmission has been documented with a prod-
(1980-1996) or Ireland or France (1980-2001) uct containing less than 104 IU/mL.71
during times when BSEwas endemic, as speci- Currently, there is no FDA-approvedtest to
fied, or for receipt of a transfusion in these coun- screen fresh blood donations for parvovirus B19
tries. 184 Ireland represents a new addition to the infection, although automated assays are avail-
residence criteria. able outside of the United States and cleared for
C HAP T E R ·7 Infectious DiseaseScreening 215
use by European regulatory agencies. However, test that would be effectivein detecting infec-
plasma-derivative manufacturers require screen- tious donations (eg, serologyvs NAT), devel-
ing of incoming plasma units for the presence of opment of a test suitable for donor screening,
high-titer parvovirus B19. This is accomplished performance of clinical trials.of the test,' and
by performing NAT on pools of samples from regulatory approval.' During this development
plasma units, with sensitivity adjusted to detect process, infections can be transmitted.
only units with a high concentration of virus. By
5. Testing cannot interdict unknown patho-
excluding high-titer units from the plasma pools,
gens or pathogens not yet recognized.or sus-
the final titer in the plasma pool is kept below
l04IU/mL. pected to be transfusable.
Other Agents Pathogen inactivation provides an attractive

and proactive alternative to relying on donor
MBB.maintains a publicly accessible electronic questioning and testing to prevent infectious do-
resource containing expert analyses of emerging nations from reaching recipients. Pathogen inac-
infectious disease agents that have received at- tivation processes reduce the infectivity of resid-
tention as potential threats to the United States ual pathogens in blood components. This
or global blood supply." This digital resource approach could reduce the transmission of infec-
contains up-to-date fact sheets on a variety of tious agents. for which there are no donor
agents. Each fact sheet includes information screening tests and further reduce the residual
about clinical manifestations and epidemiology transmission risks of known agents. Once ap-
of infection, evidence of transfusion transmissi- proved and implemented, pathogen inactiva-
bility, and analyses of the potential effectiveness tion could theoretically enable discontinuation
of various mitigation strategies (eg, donor ques- of some testing that is currently performed (eg,
tioning, serologictesting or NAT,or pathogen in- CMV testing and bacteria testing of platelets,
activation). Readers are encouraged to use this ID-NAT, others), and pathogen inactivation
rich resource. methods have been shown to obviate the need
for irradiation, potentially offsetting some of its
cost.
PATHOG rIVAl_ON As discussed above, pathogen inactivation
TECHNO methods are an essential component of the
plasma-derivative manufacturing process. An
Donor screening reduces, but cannot eliminate, SD-treated pooled plasma product has been ap-
the infectious risks of blood transfusion. The effi- proved for transfusion in the United States
cacy of blood donor testing is limited by several (Octaplas)." Because SD treatment and methy-
factors, including the following": lene blue/visible light treatment used on plasma
damage cell membranes, these methods are not
1. It is not logisticallyfeasible to test donors for used for platelets or red cells. The pooling of
every infection that is conceivably transmissi- plasma components for use with these technolo-
ble by transfusion. gies results in an increased risk of transmitting
2. For every test, there is a lag time (ie, window an agent that lacks a lipid envelope and is partic-
period) between when a person becomes ularly resistant to inactivation. Hence, incom-
ing lots of plasma intended for manufacture of
infected and when the test detects infection.
plasma derivatives are prescreened to eliminate
3. Every test has limited sensitivity (concentra- the risk of contamination by these agents, such
tion of the target marker that can be detected as parvovirus B19, HEV, and hepatitis A virus
by the test). (another nonenveloped hepatitis agent rarely
4. Developinga donor test can be a long, multi- transfusion transmitted). Such prescreening also
phase process that includes identification of occurs for SD plasma lots used for transfusion in
the infectious agent, selection of the type of the United States.
One manufacturer's pathogen inactivation ternationally on pathogen-reduced red cells us-

process is now approved in the United States for ing amustaline plus glutathione (INTERCEPT)
treatment of individual units of plasma and and whole blood using riboflavin and UV light
platelets (INTERCEPT) using amotosalen (a (Mirasol). Processes that are available or in de-
psoralen) and ultraviolet A (UVA) light. Addi- velopment have been recently reviewed in de-
tional technologies are in use outside the United tail and are summarized in Table 7_8.191-193
States. Riboflavin (vitamin B2) and UVB and Pathogen inactivation technologies that tar-
UVA light are being used for plasma and plate- get nucleic acids usually do so by generation of
lets (Mirasol). The nucleic-acid-damaging patho- nucleic acid crosslinking, preventing pathogen
gen inactivation technologies provide signifi- replication. Platelets treated with these technol-
cant activity against all agents for which tests ogies have somewhat lower l-hour posttransfu-
are performed currently: HN, HBV,HCV, HTLV, sion corrected count increments.'?' In clinical
WNV, CMV, ZIKV,Babesia spp, and parasites, as trials, mild and moderate bleeding frequency is
well as syphilis and agents causing bacterial con- increased, but not severe bleeding complica-
tamination of platelets. However, pathogen inac- tions; the time between transfusions and the to-
tivation methods appear to differ greatly in their tal number of platelet transfusions have not gen-
capacity. Amotosalen/UV treatment also inacti- erally been different. Pulmonary toxicity, like
vates white cells to prevent transfusion-associated transfusion-related acute lung injury (TRALI),
graft-vs-host disease, decreases formation and was reported in clinical trials and in animal
release of cytokines during storage, and reduces model experiments with amotosalen/UV, but
febrile nonhemolytic transfusion reactions. does not appear to be a material issue in general
Amotosalen/UV treatment appears to have no in the European Union. Previous clinical trials of
effect on the rate of white-ceIl-induced alloanti- one red cell inactivation method were halted
body (eg, HLA antibody) formation.!" Clinical because of asymptomatic immunoreactivity
trials are under way in the United States and in- against the red cell neoantigens believed to be
TABLE 7·8. Pathogen Inactivation Technologies for Transfusable Blood Components
Component Technology Manufacturer
Plasma:commerciallypreparedpools Solvent/detergenttreatment Octapharma(FDAcleared)
Plasma:individualunits Amotosalen(psoralen)+ UV light Cerus(FDAcleared)
Riboflavin(vitaminB2)+ UV light TerumoBCT
Methyleneblue + light Macopharma
Platelets Amotosalen(psoralen)+ UV light Cerus(FDAcleared)
UV light Macopharma
RedBloodCelisIWholeBlood Arnustallne and glutathione Cerus

FDA= Foodand Drug Administration; UV = ultraviolet.
the result of treatment, and trials are being re- such as DENV,CHIKV,and ZIKV;and 4) proac-
sumed with a reformulated process (amustaline tively decrease threats from unknown, emerging
plus glutathione). Preliminary reports suggest pathogens. Again, it should be noted that inacti-
riboflavin/UV causes functional impairment in vation capabilitiesdiffer greatly between the var-
red cells stored nearest the 42-day expiration. ious technologies, and each must be evaluated
Although there have been discussions about the for its intended use.
potential for adverse reactions to treated compo-
nents, extensive reviews of European data on
pathogen-reduced platelets and plasma do not U A
support the additional concern. Nevertheless,
the FDA has required Phase N postrnarketmg The current level of safety of blood components
studies in the United States as part of the imple- is based on two critical elements of donor
mentation of the approved system, and this is screening: donor education and questioning,
likely for other methods approved in the .future. which is the sole method of screening for cer-
The benefit to be gained from pathogen inac- tain agents, such as malaria and prions, and do-
tivation in the United States is primarily the mit- nor testing. Testing must be performed carefully
igation of emerging pathogens and platelet- and in accordance with manufacturers' instruc-
associatedbacterial sepsis. Currently, the quanti- tions, FDA regulations, and MBB Standards,
fiable infectious risks of transfusion in the Unit- and facilitiesmust have robust systems for quar-
ed States are low. Therefore, it is critically antining components of donations that test reac-
important to demonstrate that inactivating treat- tive and for retrieving prior donations from do-
ments do not introduce new hazards to patients. nors whose. sarnpleshave tested positive.
Rigorous preclinical and clinical studies are re- Current quantifiable risks of infectious dis-
quired for US regulatory approval. Extensive ease transmission through the US blood supply
toxicology studies have been critical because are very low; the published estimated risk of
most of the pathogen inactivation agents inter- HN transmission by transfusion in the United
act with nucleic acid, raising the theoretical po- States is approximately 1 in 1.5 million units,
tential of carcinogenicity and .mutagenicity. the risk of HCV transmission is approximately 1
Treated components should be assessed for neo- in 1.1 million units, and the risk of HBVtrans-
antigen formation and the impact of the inacti- mission is approximately 1 in 800,000 to 1 in
vating process on the final product's clinical effi- 1.2 million units,63,68and as noted, these risks
cacy. The evaluation processes required for are further decreasing. However, it is critical to
approval of pathogen inactivation methods in remain vigilant for changes in rates of known
North America have been reviewed. 191·193 agents as donor eligibility policies change, and
Interest in pathogen inactivation remains for evidence of new infectious agents so that
high because it has the potential to 1) reduce mitigation measures are implemented as quickly
sepsis-related platelet transfusion complications as feasible or required.141,142,61,194
Pathogen inac-
and eliminate the need for complex testing pro- tivation technologies show promise in replacing
cedures to reduce risk related to bacterially con- or augmenting current screening strategies such
taminated platelets; 2) inactivate parasites such as for bacteria in platelets and may be effective
as B. microti and P. fslciperutn; 3) mitigate risks in providing protection against infectious agents
associated with recognized emerging pathogens for which no screening is currently in place.
2. There is a delay between the time when an individual is exposed to an infection and the time
:e~~:O~~~~i~~~~~:~~~~~~ti~~:~tion yields a positive result. Blood donated during
3. ~~~e:~~~:: ::~;d~~:i~~r~: ~P-NAT of donor samples is less than 10 days for HIVand
4. The residual risk of transfusion-transmitted infection is a function of the length of the window
period and the incidence of infection in the donor population. Maintaining a donor population
that has a low incidence of infection continues to playa key role in preserving blood safety.
5. The residual risk of HIV, HCV, and HBV is very low and declining. Based on window-period
and incidence calculations, the current risk of HIV transmission by transfusion in the United
States is approximately I in 1.5 million units, the risk of HCV transmission is approximately
1 in 1.1 million units, and the risk of HBVtransmission is approximately 1 in 800,000 to 1 in
1.2 million units.
6. There are no donor screening tests approved by the FDA for malaria or vCJD. Donor question-
~~;e~~~t potential exposure is the sole means of protecting the US blood supply from these
7. MBB requires blood banks to have processes that limit and detect or inactivate bacteria in
platelet components. Either pathogen inactivation or bacteria testing can satisfy this require-
::~rh~~~~:r~:;:C~~~~~~~~Oe~~~~iSting bacterial culture method are being considered to
8. Infections transmitted to humans by vectors are increasingly recognized as a potential source
~:n~~l~fu~~f~~=i!~~;~!~~~~~. ir~~include WNV, T. cruzi, Babesia species, DENV,po-
9. Pathogen inactivation may reduce the transmission of infectious agents for which there are no
donor screening tests and may further reduce the residual transmission risks of known agents.
Pathogen-reduced products include commercially manufactured plasma derivatives and pooled
~~:~~tldetergent-treated plasma, as well as pathogen-reduced platelet and plasma compo-
10. Blood banks must have processes in place to ensure that donations with positive test results are
not released for transfusion. In some circumstances, 1) prior donations from those donors must
also be retrieved and quarantined, and 2) recipients of prior donations must be notified of their
possible exposure to infection.
REFERENCES
1. Code of federal regulations. Title 21, CFRParts 5. Aach RD,SzmunessW, Mosley]W, et at. Serum
211 and 610. Washington DC: US Government alanine aminotransferaseof donors in relation to
PublishingOffice,2019 (revised annually). the risk of non-A, non-B hepatitis in recipients:
2. Alter HI, Klein HG. The hazards of blood trans- The transfusion-transmitted viruses study.
fusion in historical perspective. Blood 2008; N Eng!I Med 1981;304:989-94.
112:2617-26. 6. Alter HI, Holland PV.Indirect tests to detect the
3. SeeffLB,Wright EC, Zimmerman HI, McCollum
non-A, non-B hepatitis carrier state. Ann Intern
RW. VA cooperative study of post-transfusion
hepatitis, 1969-1974: Incidence and character- Med 1984;101:859-61.
7. Stevens CE, Aach RD, HollingerFB,et al. Hepa-
istics of hepatitis and responsible risk factors.
Am I Med Sci 1975;270:355-62. titis B virus antibody in blood donors and the
4. Alter HI, Purcell RH, Holland PV, et al. Donor occurrence of non-A, non-B hepatitis in transfu-
transaminase and recipient hepatitis. Impact on sion recipients. An analysis of the Transfusion-
blood transfusion services. lAMA 1981 ;246: Transmitted Viruses Study. Ann Intern Med
630-4. 1984;101:733-8.
8. Galel SA, Lifson JD, Engleman EG. Prevention (Anti-Hlsc).Silver Spring, MD: CBEROffice of
of AIDS transmission through screening of the Communication, Outreach, and Development,
blood supply. Annu Rev ImmunoI1995;13:201- 2010.
27. 19. Food and Drug Administration. Guidance for in-
9. Joint statement on acquired immune deficiency dustry: Use of nucleic acid tests on pooled and
syndrome (AIDS) related to transfusion. Transfu- individual samples from donors of whole blood
sion 1983;23:87-8. and blood components, including source plas-
10. Busch MP, Young MJ, Samson SM, et al. Risk of ma, to reduce the risk of transmission of hepati-
human immunodeficiency virus (HN) transmis- tis B virus. Silver Spring, MD: CBER Office of
sion by blood transfusions before the implemen- Communication, Outreach, and Development,
tation of HN-l antibody screening. The Transfu- 2012.
sion Safety Study Group. Transfusion 1991 ;31: 20. Food and Drug Administration. Guidance for in-
4-11. dustry: Nucleic acid testing (NAT) for human
11. Ward JW, Holmberg SD, Allen JR, et al. Trans- immunodificiency virus type 1 (HN-l) and hep-
mission of human immunodeficiency virus atitis C virus (HCV): Testing, product disposi-
(HN) by blood transfusions screened as negative tion, and donor deferral and reentry. Silver
for HIV antibody. N EnglJ Med 1988;318:473-8. Spring, MD: CBEROffice of Communication,
12. Perkins HA, Samson SM, Busch MP. How well Outreach, and Development, 2017.
has self-exclusion worked? Transfusion 1988; 21. Food and Drug Administration. Guidance for in-
28:601-2. dustry: Recommendations for screening, testing
13. Food and Drug Administration. Infectious dis- and management of blood donors and blood and
ease tests. Silver Spring, MD: CBER Office of blood components based on screening tests for
Communication, Outreach, and Development, syphilis. Silver Spring, MD: CBER Office of
2019. [Available at https://www.fda.gov/vac Communication, Outreach, and Development,
cines-blood-biologics/blood-donor-screening/ 2014.
infectious-dis ease-tests. J 22. Food and Drug Administration. GUidancefor in-
14. Food and Drug Administration. Information for dustry: Revised recommendations for reducing
blood establishments: Unavailabilityof CHIRON the risk of human immunodeficiency virus
RIBAHCV 3.0 SIA (RIBA).Silver Spring, MD: transmission by blood and blood products.
CBEROffice of Communication, Outreach, and (April2020) Silver Spring, MD: CBEROffice of
Development, 2012. Communication, Outreach, and Development,
15. Busch MP, Glynn SA, Stramer SL, et al. A new 2020.
strategy for estimating risks of transfusion- 23. Food and Drug Administration. Guidance for in-
transmitted viral infections based on rates of de- dustry: Use of serologicaltests to reduce the risk
tection of recently infected donors. Transfusion of transmission of Trypanosoma cruz! infection
2005 ;45:254-64. in blood and blood components. Silver Spring,
16. Food and Drug Administration. Memorandum MD: CBER Office of Communication, Out-
to all registered blood establishments: Recom- reach, and Development, 2017.
mendations for the management of donors and 24. Food and Drug Administration. Guidance for in-
units that are initially reactive for hepatitis B dustry: Use of serologicaltests to reduce the risk
surface antigen (HBsAg). Silver Spring, MD: of transfusion-transmitted human Tlymphotro-
CBEROffice of Communication, Outreach, and pic virus types I and II (HTLV-IlII).SilverSpring,
Development, 1987. MD: CBER Office of Communication, Out-
17. Food and Drug Administration. Guidance for in- reach, and Development, 2020.
dustry: Requalification method for reentry of 25. Food and Drug Administration. Guidance for in-
donors who test hepatitis B surface antigen dustry: Recommendations for reducing the risk
(HbsAg)positive following a recent vaccination of transfusion-transmitted babesiosis. Silver
against hepatitis B virus infection. Silver Spring, Spring, MD: CBEROffice of Communication,
MD: CBER Office of Communication, Out- Outreach, and Development, 2019.
reach, and Development, 2011. 26. Food and Drug Administration. Guidance for in-
18. Food and Drug Administration. Guidance for industry: Use of nucleic acid tests to reduce the
dustry: Requalification method for reentry of risk of transmission of West Nile virus from do-
blood donors deferred because of reactive test nors of whole blood and blood components in-
results for antibody to hepatitis B core antigen tended for transfusion. (November 2009) Silver
Spring, MD: CBER Office of Communication, ther testing, product disposition, and notifica-
Outreach, and Development, 2009. tion of transfusion recipients based on donor
27. Food and Drug Administration. Guidance for in- test results indicating infection with HCV. Silver
dustry: Revised recommendations for reducing Spring, MD: CBEROffice of Communication,
the risk of Zlka virus transmission by blood and Outreach, and Development, 2010.
blood components. (July 2018) Silver Spring, 36. Food and Drug Administration. Guidance for in-
MD: CBER Office of Communication, Out- dustry: Further testing of donations that are re-
reach, and Development, 2018. [Available at active on a licensed donor screening test for an-
https://www.fda.gov/vaccines-blood-biolog tibodies to hepatitis c virus. Silver Spring, MD:
ics/biologics-guidances/blood-guidances.] CBEROffice of Communication, Outreach, and
28. Eder AF. Donor reentry. In: Eder AF, Goldman Development, 2019.
M, eds. Screening blood donors with the donor 37. Food and Drug Administration. Guidance for in-
history questionnaire. Bethesda, MD: AABB dustry: Assessing donor suitability and blood
Press, 2019:209-47. and blood product safety in cases of known or
29. Code of federal regulations. Title 42, CFR Part suspected West Nile virus infection. Silver
482.27. Washington, DC: US Government Pub- Spring, MD: CBEROffice of Communication,
lishing Office,2019 (revised annually). Outreach, and Development, 2005.
30. Food and Drug Administration. Guidance for in- 38. West Nilevirus nucleic acid testing - revised rec-
dustry: Use of nucleic acid tests on pooled and ommendations. Association bulletin #13-02.
individual samples from donors of whole blood Bethesda, MD: AABB,2013.
components (including Source Plasma and 39. Updated recommendations for Zika, dengue,
Source Leukocytes) to adequately and appropri- and chikungunya viruses. Association bulletin
ately reduce the risk of transmission of HN-1 #16-07. Bethesda, MD: AABB,2016.
and HCV. Silver Spring, MD: CBER Office of 40. Food and Drug Administration. Draft guidance
Communication, Outreach, and Development, for industry: Bacterial risk control strategies for
2004. blood collection establishments and transfusion
31. Gammon R, ed. Standards for blood banks and services to enhance the safety and avallabilityof
transfusion services. 32nd ed. Bethesda, MD: platelets for transfusion. Silver Spring, MD:
AABB,2020. CBEROffice of Communication, Outreach, and
32. Food and Drug Administration. Guidance for in- Development, 2019.
dustry: Recommendations for management of 41. Guidance on management of blood and platelet
donors at increased risk for human immunodefi- donors with positive or abnormal results on bac-
ciency virus type 1 (HN-1) group 0 infection. terial contamination tests. Association bulletin
Silver Spring,MD: CBEROffice of Communica- #05-02. Bethesda, MD: AABB,2005.
tion, Outreach, and Development, 2009. 42. Recommendations to address residual risk of
33. Food and Drug Administration. Guidance for in- bacterial contamination of platelets. Association
dustry: Requalification of donors previously de- bulletin #12-04. Bethesda, MD: AABB,2012.
ferred for a history of viral hepatitis after the 43. Actions following an initial positive test for pos-
11th birthday. Silver Spring, MD: CBEROffice sible bacterial contamination of a platelet unit.
of Communication, Outreach, and Develop- Association bulletin #04-07. Bethesda, MD:
ment, 2017. AABB,2004.
34. Food and Drug Administration. Memorandum 44. Food and Drug Administration. Guidance for in-
to all registered blood and plasma establish- dustry: Nucleic acid testing (NAT)to reduce the
ments: Recommendations for the quarantine possible risk of parvovirus B19 transmission by
and disposition of units from prior collections plasma-derived products. Silver Spring, MD:
from donors with repeatedly reactive screening CBEROffice of Communication, Outreach, and
tests for hepatitis B virus (HBV),hepatitis C vi- Development, 2009.
rus (HCV), and human T-lymphotropic virus 45. BlajchmanMA, Goldman M, Freedman]J, Sher
type I (HTLV-I).Silver Spring,MD: CBEROffice GD. Proceedings of a consensus conference:
of Communication, Outreach, and Develop- Prevention of post-transfusion CMV in the era
ment, 1996. of untversal leukoreductlon. Transfus Med Rev
35. Food and Drug Administration. GUidancefor in- 2001;15:1-20.
dustry: Lookback for hepatitis C virus (HCV): 46. Vamvakas E. Is white blood cell reduction
Product quarantine, consignee notification, fur- equivalent to antibody screening in preventing
C HAP T E R 7 InfectiousDiseaseScreening 221
transmission of cytomegalovirus by transfusion? nema pallidum (syphilis). Silver Spring, MD:

A review of the literature and meta-analysis. CBEROffice of Communication, Outreach, and
Transfus Med Rev 2005; 19: 181-99. Development, 2015.
47. Bowden RA, Stichter SJ, Sayers M, et al. A com- 57. Food and Drug Administration. GUidancefor in-
parison of filtered leukocyte-reduced and cyto- dustry: Use of nucleic acid tests to reduce the
megalovirus (CMV) seronegative blood products risk of transmission of West Nile virus from liv-
for the prevention of transfusion-associated ing donors of human cells, tissues, and cellular
CMV infection after marrow transplant. Blood and tissue-based products (HCTIPs). Silver
1995;86:3598-603. Spring, MD: CBEROffice of Communication,
48. Nash T, Hoffmann S,· Butch S, et al. Safety of Outreach, and Development, 2016.
leukoreduced, cytomegalovirus (CMV)-untested 58. Food and Drug Administration. Tissue guidanc-
components in CMV-negative allogeneic human es. Silver Spring, MD: CBEROffice of Commu-
progenitor cell transplant recipients. Transfusion nication, Outreach, and Development, 2016.
2012;52:2270-2. [Available at https://www.fda.gov/vaccines-
49. Hall S, Danby R, Osman H, et al. Transfusion in blood-biologics/biologics-guidances/tissue-guid
CMV seronegative T-depleted allogeneic stem ances.]
cell transplant recipients with CMV-unselected 59. Mezochow AK, Henry R, Blumberg EA, Kotton
blood components results in zero CMV trans- CN. Transfusion transmitted infections in solid
missions in the era of universal leukocyte reduc- organ transplantation. Am J Transplantation
tion: A U.K. dual centre experience. Transfus 2015; 15:547-54.
Med 2015;25:418-23. 60. Food and Drug Administration. Testing donors
50. Ziemann M, Thiele T. Transfusion-transmitted of human cells, tissues, and cellular and tissue-
CMV infection - current knowledge and future based products: Specific requirements. Silver
perspectives. Transfus Med 2017;27:238-48.
Spring, MD: CBEROffice of Communication,
51. Furui Y, Yamagishi N, Morioka I, et al. Sequence
Outreach, and Development, 2015. [Available
analyses of variable cytomegalovirus genes for
at https://www.fda.gov/vaccines-blood-biolog
distinction between breast milk- and transfu-
ics/ safety-availability-biologics/ testing-donors-
sion-transmitted infections in very-low-birth-
human -cells-tissues-and -cellular-and -tissue-
weightinfants. Transfusion 2018;9999: 1-9.
based-products-hctp-specific-requirements.]
52. Josephson CD, Caliendo AM, Easley KA, et al.
61. Custer B,Stramer SL,Glynn S, et al. Transfusion-
Blood transfusion and breast milk transmission
of cytomegalovirus in very low-birth-weight in-
transmissible infection monitoring system: A
fants: A prospective cohort study. JAMA Pediatr
tool to monitor changes in blood safety. Transfu-
2014; 168: 1054-62. sion 2016;56: 1499-502.
53. Food and Drug Administration. Guidance for in- 62. Anderson SA, Yang H, Gallagher LM, et al.
dustry: Donor screening recommendations to Quantitative estimate of the risks and benefits of
reduce the risk of transmission of Zika virus by possible alternative blood donor deferral strate-
human cells, tissues, and cellular and tissue- gies for men who have had sex with men.
based products. Silver Spring,MD: CBEROffice Transfusion 2009;49: 1102-14.
of Communication, Outreach, and Develop- 63. Zou S, Dorsey KA,Notari EP, et al. Prevalence,
ment, 2018. incidence, and residual risk of human immuno-
54. Code of federal regulations. Title 21, CFR Part deficiency virus and hepatitis C virus infections
1271 21. Washington, DC: US Government among United States blood donors since the in-
Publishing Office, 2019 (revised annually). troduction .of nucleic acid testing. Transfusion
55. Food and Drug Administration. Guidance for in- 2010;50: 1495-504.
dustry: Eligibility determination for donors of 64. Schreiber GB,Busch MP, Kleinman SH, Korelitz
human cells, tissues, and cellular and tissue- JJ. The risk of transfusion-transmitted viral infec-
based products (HCT/Ps). Silver Spring, MD: tions. The Retrovirus Epidemiology Donor
CBEROffice of Communication, Outreach, and Study. N Eng!J Med 1996;334: 1685-90.
Development, 2007. 65. Stramer SL, Glynn SA, Kleinman SH, et al.
56. Food and Drug Administration. Guidance for in- Detection of HN-l and HCV infections among
dustry: Use of donor screening tests to test do- antibody-negativeblood donors by nucleic acid-
nors of human cells, tissues and cellular and amplification testing. N Engl J Med 2004;351:
tissue-based products for infection with Trepo- 760-8.
66. Dodd RY, Notari EP 4th, Stramer SL. Current 77. Stramer SL. Pooled hepatitis B virus DNA test-
prevalence and incidence of infectious disease ing by nucleic acid amplification: Implementa-
markers and estimated window-period risk in tion or not. Transfusion 2005;45:1242-6.
the American Red Cross blood donor popula- 78. Linauts S, Saldanha], Strong DM. PRISMhepa-
tion. Transfusion 2002;42:975-9. titis B surface antigen detection of hepatits B vi-
67. Duong YT, Kassanjee R, Welte A, et al. Recall- rus minipool nucleic acid testing yield samples.
bration of the limiting antigen avidity ElA to de- Transfusion 2008 ;48:1376-82.
termine mean duration of recent infection in di- 79. Stramer SL,Wend U, Candotti D, et al. Nucleic
vergent HIV-1 subtypes. PLoS One 2015;10: acid testing to detect HBVinfection in blood do-
eOl14947. nors. N Engi] Med 2011;364:236-47.
68. Stramer SL, Notari EP, Krysztof DE, Dodd RY. 80. Dodd RY,Nguyen ML, KrysztofDE, et al. Blood
Hepatitis B virus testing by minipool nucleic donor testing for hepatitis Bvirus in the United
acid testing: Does it improve blood safety? States: Is there a case for continuation of hepati-
Transfusion 2013;53:2449-58. tis B surface antigen detection? Transfusion
69. Octaplas, pooled plasma (human), solvent/de- 2018;58:2166-70.
tergent treated solution for intravenous infusion 81. Alter H], Descartes before the horse: I clone,
insert (prescribing information). Hoboken, N]: therefore I am: The hepatitis C virus in current
Octapharma USA. perspective. Ann Intern Med 1991;115:644-9.
70. Stramer SL, Hollinger FB, Katz LM, et al. 82. Smith BD, Morgan RL, Beckett GA, et al. Rec-
Emerging infectious disease agents and their po- ommendations for the identification of chronic
tential threat to transfusion safety. Transfusion hepatitis C virus infection among persons born
2009;49(Suppl2): 1S-29S. during 1945-1965. MMWR Recomm Rep
71. Emerging infectious disease agents and their po- 2012;61 :1-32.
tential threat to transfusion safety. Bethesda, 83. Page-ShaferK, Pappalardo BL, Tobler LH, et al.
MD: AABB, 2017. [Available at http://www. Testing strategy to identify cases of acute hepati-
aabb.org/tm/ eid/Pages/ default.aspx (accessed tis C virus (HCV) infection and to project HCV
October 1, 2019).J incidence rates. ] Clin Microbiol 2008;46:499-
72. Stramer SL, Yu G, Herron R, et al. Two human 506.
immunodeficiency virus Type 2 cases in US 84. Guidelines for counseling persons infected with
blood donors including serologic, molecular, human 'l-lymphorropic virus type I (HTLV-I)and
and genomic characterization of an epidemio- type II (HTLV-II).Centers for Disease Control
logically unusual case. Transfusion 2016;56: and Prevention and the U.S.P.H.S. Working
1560-8. Group. Ann Intern Med 1993; 118:448-54.
73. Centers for Disease Control and Prevention. 85. Vrlelink H, ZaaijerHL, Reesink HW. The clinical
HN surveillance report, 2016. Vol. 28. Atlanta, relevance of HTLV type I and II in transfusion
GA:CDC, 2017. [Availableat https://www.cdc. medicine. Transfus Med Rev 1997;11 :173-9.
gov/hiv /library/reports/hiv-surveillance.html. J 86. Aubron C, Nichol A, Cooper D], Bellomo R.Age
74. Centers for Disease Control and Prevention. of red blood cells and transfusion in critically ill
HIV prevalence, unrecognized infection, and patients. Ann Intensive Care 2013;3:2.
HIV testing among men who have sex with 87. Eikelboom]W, Cook R], Liu Y, Heddle NM. Du-
men-five U.S. cities, June 2004-April 2005. ration of red cell storage before transfusion and
MMWR Morb Mortal Wkly Rep 2005;54:597- in-hospitalmortality. Am Heart] 2010;159:737-
601. 43.
75. Truong HM, Kellogg T, Klausner ]D, et al. In- 88. UK BTS Joint Professional Advisory Commit-
creases in sexually transmitted infections and tee's (JPAC)HTLVWorking Group. Options for
sexual risk behaviour without a concurrent in- human T-1ymphotropicvirus (HTLV)screening
crease in HN incidence among men who have within the UK Blood Services (updated October
sex with men in San Francisco: A suggestion of 2015). [Available at https://www.transfusion
HIV serosorting? Sex Transm Infect 2006;82: guidelines.org/ document-library / options- for-
461-6. human-t-Iymphotropic-virus-htlv-screening-
76. Kleinman SH, Busch MP. Assessing the impact with-the-uk-blood-services-updated-october-
of HBV NAT on window period reduction and 2015-r.J
residual risk. ] Clin Virol 2006;36(Suppl 1):S23- 89. Hewitt PE, Davison K, Howell DR, Taylor GP.
9. Human 'l-lymphotroptc virus lookback in NHS
Blood and Transplant (England) reveals the effi- 103. Benjamin RJ, Kline L, Dy BA, et al. Bacterial
cacy of leukoreduction. Transfusion 2013;53: contamination of whole-blood-derived platelets:
2168-75. The introduction of sample diversion and
90. Petrik J, Lozano M, Seed CR, et al. Hepatitis E. presto rage pooling with culture testing in the
VoxSang2016;110:93-130. American Red Cross. Transfusion 2008;48:
91. Hewitt PE, Ijaz S, Brallsford SR, et al, Hepatitis 2348-55.
E virus in blood components: A prevalence and 104. McDonald C, Allen JM, BrailsfordS, et al. Bac-
transmission study in southeast England. Lancet terial screening of platelet components by Na-
2014;384: 1766-73. tional Health Service Blood and Transplant, an
92. Stramer SL,Moritz ED, Foster GA, et al. Hepati- effective risk reduction measure. Transfusion
tis E virus: Seroprevalence and frequency of vi- 2017;57:1122-31.
ral RNA detection among US blood donors. 105. Benjamin RJ, Braschler T, Weingand T, Corash
Transfusion 2016;56:481-8. LM. Hemovigilancemonitoring of platelet septic
93. Hauser L, Roque-AfonsoAM, Beyloune A, et al. reactions with effective bacterial protection sys-
Hepatitis E transmission by transfusion of Inter- tems. Transfusion2017;57:2946-57.
cept blood system-treated plasma. Blood 2014; 106. Jacobs MR, Smith 0, Heaton WA, et al. Detec-
123:796-7. tion of bacterial contamination in prestorage
94. Orton S. Syphilis and blood donors: What we culture-negative apheresis platelets on day of is-
know, what we do not know, and what we need sue with the Pan Genera Detection test. Trans-
to know, Transfus Med Rev2001;15:282-91. fusion 2011 ;51:2573-82.
95. Katz LM. A test that won't die: The serologic 107-.Ruby KN,Thomasson RR,Szczepiorkowski lM,
test for syphilis.Transfusion 2009;49:617-19. Dunbar NM. Bacterial screening of apheresis
96. lou S, Notari EP, Fang CI, et al. Current value platelets with a rapid test: A 113-month single
of serologictest for syphilisas a surrogate mark- center experience. Transfusion 2018;58: 1665-
er for blood-borne viral infections among blood 9.
donors in the United States. Transfusion 2009; 108. Erony SM, Marshall CE, Gehrie EA, et al. The
49:655-61. epidemiology of bacterial culture-positive and
97. Orton SL, Liu H, Dodd RY,et al. Prevalence of septic transfusion reactions at a large tertiary ac-
circulating Treponema pallidum DNA and RNA ademic center: 2009 to 2016. Transfusion
in blood donors with confirmed-positivesyphilis 2018;58: 1933-9.
tests. Transfusion 2002;42:94-9. 109. Kamel H, Townsend M, Bravo M, Vassallo RR.
98. Centers for Disease Control and Prevention. Improved yield of minimal proportional sample
Sexually transmitted diseases (STDs): Syphilis volume platelet bacterial culture. Transfusion
statistics.Atlanta, GA: CDC, 2016. [Availableat 2017;57:2413-19.
http://www.cdc.gov/std/syphilis/stats.htm.j 110. Biggerstaff BJ, Petersen LR. Estimated risk of
99. Food and Drug Administration. Fatalitiesreport- transmission of the West Nile virus through
ed to FDAfollowing blood collection and trans- blood transfusion in the US, 2002. Transfusion
fusion: Annual summary for fiscal year 2015. 2003;43:1007-17.
Silver Spring,MD: CBEROfficeof Communica- Ill. Pealer L,Marfin A, Petersen LR.Transmission of
tion, Outreach, and Development, 2015. West Nile virus through blood transfusion in the
100. Eder AF, Kennedy JM, Dy BA, et al. Limiting United States in 2002. N EnglJ Med 2003;349:
and detecting bacterial contamination of aphere- 1236-45.
sis platelets: Inlet-line diversion and increased 112. lou S, Foster GA,Dodd RY,et al. West Nile fe-
culture volume improve component safety. ver characteristics among viremic persons iden-
Transfusion 2009;49: 1554-63. tified through blood donor screening. J Infect
101. Ramirez-Arcos SM, Goldman M, Blajchman Dis 2010; 202: 1354-61.
MA. Bacterial infection: Bacterial contamina- 113. Dodd RY,Foster GA, Stramer SL.Keepingblood
tion, testing and post-transfusion complications. transfusion safe from West Nile virus: American
2nd ed. Philadelphia: Church Ltvingstone, Red Cross experience, 2003 to 2012. Transfus
2007. Med Rev 2015;29:153-61.
102. Suggested options for transfusion services and 114. O'Brien SF,ScaliaV, luber E, et al. West Nilevi-
blood collectors to facilitate implementation of rus in 2006 and 2007: The Canadian Blood Ser-
BElTS Interim Standard 5.1.5.1.1.· Association vices' experience. Transfusion 20 I 0;50: 1118-
bulletin #10-05. Bethesda, MD: AABB,2010. 25.
115. Groves JA, Shafi H, Nomura JH, et al. A proba- Polynesia, November 2013 to February 2014.
ble case of West Nile virus transfusion transmis- Euro Surveill2014;19:20761.
sion. Transfusion 2017;57:850-6. 129. Kuehnert MJ, BasavarajuSV,Moseley RR,et al.
116. Petersen LR,Jamieson DJ, Powers AM, Honein Screening of blood donations for Zika virus in-
MA. Zika virus. N Engl I Med 2016;374:1552- fection - Puerto RiCO,April 3-June 11, 2016.
63. MMWR Morb Mortal Wkly Rep 2016;65:627-
117. Pan American Health Organization, World 8.
Health Organization. Zika - epidemiological 130. Benjamin RJ. Zika virus in the blood supply.
update. (August 25, 2017) Washington, DC: Blood2017; 129:144-5.
PAHO/WHO, 2017. [Available at http:// 13l. Barjas-CastroML, Angerami RN, Cunha MS, et
www.paho.org/hq/dmdocu mentsl2017/ al. Probable transfusion-transmitted Zika virus
2017 -aug-25-phe-epi-update-zika-virus.pdf in Brazil.Transfusion2016;56: 1684-8.
(accessedOctober 1,2019).] 132. Motta II, Spencer BR, Cordeiro da Silva SG, et
118. GallainP, Cabie A, RichardP, et al. Zikavirus in al. Evidence for transmission of Zika virus by
asymptomatic blood donors in Martinique. platelet transfusion. N Engl J Med 2016;375:
Blood2017; 129:263-6. 1101-3.
119. Rasmussen SA, Jamieson DJ, Honein MA, Pe- 133. Davidson A, SlavinskiS, Komoto K, et al. Sus-
tersen LR.Zikavirus and birth defects-review- pected female-to-male sexual transmission of
ing the evidence for causality. N Engl J Med Zikavirus -New YorkCity, 2016. MMWRMorb
2016;374: 1981-7. Mortal Wkly Rep 2016;65:716-17.
120. BrasilP, PereiraJP Ir, Moreira ME, et al. Zikavi- 134. LesslerJ, Ott CT, Carcelen AC, et al. Times to
rus infection in pregnant women in Riode Janei- key events in Zika virus infection and implica-
tions for blood donation: A systematic review.
ro. N EnglJ Med 2016;375:2321-34.
BullWorld Health Organ 2016;94:841-9.
121. Franca GV,Schuler-PacciniL,OliveiraWK, et al.
135. LustigY,Mendelson E, Paran N, et al. Detection
Congenital Zikavirus syndrome in Brazil:A case
of Zika virus RNAin whole blood of imported
series of the first 1501 livebirthswith complete
Zika virus disease cases up to 2 months after
investigation. Lancet 2016;388:891-7.
symptom onset, Israel, December 2015 to April
122. Johansson MA, Mier-y-Teran-RomeroL, Reefhu-
2016. Euro Surveill2016;21(26).
is J, et al. Zika and the risk of microcephaly.
136. Allinson B,Hearn P, Afrough B,et al. Detection
N EnglJ Med 2016;375:1-4. of Zikavirus in semen. Emerg Infect Dis 2016;
123. Garcez PP, LoiolaEC, Madeiro da Costa R, et al.
22:940.
Zika virus impairs growth in human neuro- 137. Mansuy JM, Pasquier C, Daudin M, et al. Zika
spheres and brain organoids. Science 2016;352: virus in semen of a patient returning from a non-
816-18. epidemic area. Lancet Infect Dis 2016;16:894-
124. Cao-Lormeau VM, Blake A, Mons S, et al. 5.
Guillain-Barre Syndrome outbreak associated 138. Gaskell KM, Houlihan C, Nastouli E, Checkley
with Zikavirus infection in French Polynesia:A AM. Persistent Zikavirus detection in semen in
case-controlstudy. Lancet 2016;387: 1531-9. a traveler returning to the United Kingdomfrom
125. Centers for Disease Control and Prevention. Brazil,2016. Emerg Infect Dis 2017;23: 137-9.
Zika virus: Statistics and maps. Atlanta, GA: 139. Medina FA,Torres G, AcevedoJ, et al. Duration
CDC, 2019. [AVailableat http://www.cdc.gov / of the presence of infectious Zikavirus in semen
zika/ reportinglindex.html.] and serum. J Infect Dis 2019;219:31-40.
126. Porse CC, Messenger S, Vugia DJ, et al. Travel- 140. Paz-BaileyG, Rosenberg ES, Doyle K, et al. Per-
associated Zika cases and threat of local trans- sistence ofZikavirus in body fluids -final report.
mission during globaloutbreak, California,USA. N EnglJ Med 2018;13:1234-43.
Emerg Infect Dis 2018;24:1626-32. 14l. Food and Drug Administration. Guidance for in-
127. Swaminathan S, Schlaberg R, LewisJ, et al. Fa- dustry: Recommendations for donor screening,
tal Zikavirus infection with secondary nonsexu- deferral, and product management to reduce
al transmission. N EnglJ Med 2016;375:1907- the risk oftransfusion transmission of Zikavirus.
9. SilverSpring,MD: CBEROfficeof Communica-
128. Musso D, Nhan T, RobinE. Potential for Zikavi- tion, Outreach, and Development, 2016.
rus transmission through blood transfusion 142. Food and Drug Administration. Guidance for in-
demonstrated during an outbreak in French dustry: Revised recommendations for reducing
the risk of Zika virus transmission by blood and 155. Sabino EC, Loureiro P, Lopes ME, et al.
blood components. Silver Spring, MD: CBER Transfusion-transmitted dengue and associated
Office of Communication, Outreach, and Devel- clinical symptoms during the 2012 epidemic in
opment, 2016. Brazil.J Infect Dis 2016;213:694-702.
143. Stone M, Lanteri MC, Bakkour S, et al. Relative 156. Matos D, Tomashek KM, Perez-PadillaI, et al.
analytical sensitivity of donor nucleic acid ampli- Probable and possible transfusion-transmitted
fication technology screening and diagnostic dengue associated with NS1 antigen-negative
real-time polymerase chain reaction assays for but RNA confirmed-positive red blood cells.
detection of Zika virus RNA. Transfusion 2017; Transfusion 2016; 56:215-22.
57:734-47. 157. Vogt MB, Lahon A, Arya RP,et al. Mosquito sali-
144. Ga1el SA, Williamson PC, Busch MP, et al. First va alone has profound effects on the human im-
Zika-positive donations in the continental Unit- mune system. PLoS Negl Trop Dis 2018;12:
ed States. Transfusion 2017;57:762-9. e0006439.
145. Williamson PC, Linnen JM, Kessler DA, et al. 158. Chiu CY,BresV, Yu G, et al. Genomic assaysfor
First cases of Zika virus-infected US blood do- identification of Chikungunya virus in blood
nors outside states with areas of active transmis- donors, Puerto Rico, 2014. Emerg Infect Dis
sion. Transfusion 2017;57:770-8. 2015;21: 1409-13.
146. Saa P, Proctor M, Foster G, et al. Investigational 159. Brouard C, Bernillon P, Quatresous I, et al. Esti-
testing for Zika virus among U.S. blood donors. mated risk of Chikungunya viremic blood dona-
N EnglJ Med 2018;378:1778-88. tion during an epidemic on Reunion Island in
147. Aubry M, Richard V, Green J, et al. Inactivation the Indian Ocean, 2005 to 2007. Transfusion
of Zika virus in plasma with amotosalen and ul- 2008;48: 1333-41.
160. Simmons G, Bres V, Lu K, et al. High incidence
traviolet A illumination. Transfusion 2016;56:
of chikungunya virus and frequency of viremic
33-40.
blood donations during epidemic, Puerto Rico,
148. Laughhunn A, Santa Maria F, Broult J, et al.
USA,2014. Emerg Infect Dis 2016;22: 1221-8.
Amustaline (S-303) treatment inactivates high
161. Otani MM, Vinelli E, KirchhoffLV,et al. WHO
levels of Zika virus in red blood cell compo-
comparative evaluation of serologic assays for
nents. Transfusion 2017;57:779-89.
Chagas disease. Transfusion 2009;49: 1076-82.
149. Blumel J, Musso D, Teitz S, et al. Inactivation
162. Food and Drug Administration. Blood Products
and removal of Zika virus during manufacture of
Advisory Committee, 94th meeting. Silver
plasma-derived medicinal products. Transfusion
Spring, MD: CBER Office of Communication,
2017; 57:790-6. Outreach, and Development, 2009. [Available
150. Kuhnel D, Muller S, Pichotta A, et al. Inactiva-
at https://wayback. archive-it.org/7993/
tion of Zika virus by solvent/detergent treat- 20170111012330/http://www.fda.gov/ Advi
ment of human plasma and other plasma- soryCommittees/Committees MeetingMateri
derived products and pasteurization of human als/BloodVaccinesandOtherBiolo gics/Blood-
serum albumin. Transfusion 2017;57:802-10. ProductsAdvisoryCommittee/ucm 121612.
151. Anez G, Rios M. Dengue in the United States of htm.]
America: A worsening scenario? Biomed Res Int 163. Food and Drug Administration. Guidance for in-
2013;2013:678645. dustry: Use of serologicaltests to reduce the risk
152. Tomashek KM, Margolis HS. Dengue: A poten- of transmission of Trypanosoma cruzi infection
tial transfusion-transmitted disease. Transfusion in whole blood and blood components intended
2011;51:1654-60. for transfusion. Silver Spring, MD: CBEROffice
153. Stramer SL, Linnen JM, Carrick JM, et al. Den- of Communication, Outreach, and Develop-
gue viremia in blood donors identified by RNA ment, 2010.
and detection of dengue transfusion transmis- 164. Steele WR, Hewitt EH, Kaldun AM, et al. Do-
sion during the 2007 dengue outbreak in Puerto nors deferred for self-reported Chagas disease
Rico. Transfusion 2012;52:1657-66. history: Does it reduce risk? Transfusion 2014;
154. Matos D, Tomashek KM, Perez-Padilla J, et al. 54:2092-7.
Probable and possible transfusion-transmitted 165. Benjamin RJ, Stramer SL, Leiby DA, et al. Try-
dengue associated with NSI antigen-negative panosoma cruziinfection in North America and
but RNA confirmed-positive red blood cells. Spain: Evidence in support of transfusion trans-
Transfusion 2016; 56:215-22. mission. Transfusion 2012;52: 1913-21; quiz 2.
166. Kessler DA, Shi PA, Aveciila ST, Shaz BH. Re- to malaria-endemic areas. Transfusion2009;49:
sults of lookbackfor Chagas disease since the in- 2335-45.
ception of donor screening at New York Blood 180. O'Brien SF, DelageG, Seed CR,et al. The epide-
Center. Transfusion2013;53: 1083-7. miology of imported malaria and transfusion
167. Blumental S, Lambermont M, Heijmans C, et al. policyin 5 nonendemic countries. TransfusMed
First documented transmission of Trypanosoma Rev 2015; 29:162-71.
cruzi infection through blood transfusion in a 181. AllainJP, Owusu-OforiAK,Assennato SM, et ai.
child with sickle-cell disease in Belgium. PLoS Effect of Plasmodium inactivation in whole
NeglTrop Dis 2015;9:e0003986. blood on the incidence of blood transfusion-
168. Herwaldt BL, Linden N, Bosserman E, et al. transmitted malaria in endemic regions: The Af-
Transfusion-associated babesiosis in the United rican investigation of the Mirasol System
States: A description of cases. Ann Intern Med (AIMS) randomised controlled trial. Lancet
2011; 155:509-19. 2016;387: 1753-61.
169. FangDC, McCulloughJ. Transfusion-transmitted 182. Crowder LA, Schonberger LB, Dodd RY,Steele
Babesia microti. Transfus Med Rev 2016;30: WR. Creutzfeldt-Jakob disease lookback study:
132-8. 21 years of surveillancefor transfusion transmis-
170. Linden N, Prusinski MA, Crowder LA, et al. sion risk. Transfusion2017;57:1875-8.
Transfusion-transmitted and community- 183. Urwin PJ, Mackenzie JM, Llewelyn CA, et al.
acqulred babesiosisin New York,2004 to 2015. Creutzfeldt-Jakobdisease and blood transfusion:
Transfusion 2018;58:660-8. Updated results of the UKTransfusionMedicine
171. Moritz ED, Winton CS, Tonnetti L, et al. Epidemiology Review Study. Vox Sang 2016;
Screening for Babesiamicrotiin the U.S. blood 110:310-16.
supply. N EnglJ Med 2016;375:2236-45. 184. Food and Drug Administration. Guidance for in-
172. Ward SJ, Stramer SL, Szczepiokowski ZM. As- dustry: Recommendations to reduce the possi-
sessing the risk of Babesiato the United States ble risk of transmission of Creutzfeldt-JakobDis-
blood supply using a risk-baseddecision-making ease (CJD)and variant Creutzfeldt-Jakobdisease
approach: Report ofAABB'sAd Hoc BabesiaPol- (vCJD) by blood and blood products. Silver
icyWorklng Group (originalreport). Transfusion Spring, MD: CBEROffice of Communication,
2018;58:1916-23. Outreach, and Development,2020.
173. Tonnetti L, Laughhunn A, Thorp AM, et al. In- 185. Seed CR, Hewitt PE, Dodd RY, et al.
activation of Babesia microti in red blood cells Creutzfeltd-Jakobdisease and blood transfusion
and platelet concentrates. Transfusion 2017;57: safety.Vox Sang2018;113:220-31.
2404-12. 186. MoleT, Iaunmuktane Z, Joiner S, et al. Variant
174. Tonnetti L, Eder AF, Dy B, et al. Transfusion- Creutzfeldt-Jakob disease in a patient with het-
transmitted Babesia microti identified through erozygosity at PRNP codon 129. N EnglJ Med
hemovigilance.Transfusion2009;49:2557-63. 2017;379:292-4.
175. Mali S, Steele S, Slutsker L,Arguin PM. Malaria 187. BougardD, BrandelJP, BelondradeM, et al. De-
surveillance - United States, 2007. MMWR Sur- tection of prions in the plasma ofpresymptomat-
veill Summ 2009;58: 1-16. ic and symptomatic patients with variant
176. Mali S, Tan KR, Arguin PM. Malaria surveil- Creutzfeldt-Jakobdisease. SciTransl Med 2016;
lance-United States, 2009. MMWR Surveill 8:370ra182.
Summ 2011;60:1-15. 188. CooperJK,Andrews N, Ladhani K,et al. Evalua-
177. Anand A, Mace KE,Townsend RL,et al. Investi- tion of a test for its suitabilityin the diagnosisof
gation of a case of suspected transfusion- variant Creutzfeldt-Iakob disease. Vox Sang
transmitted malaria. Transfusion2018;58:2115- 2013;105:196-204.
21. 189. Snyder E, Stramer S, Benjamin RJ.The safety of
178. Food and Drug Administration. Guidance for in- the US blood supply - time to raise the bar.
dustry: Revisedrecommendations to reduce the N EnglJ Med 2015;372:1882-5.
risk of transfusion-transmitted malaria. Silver 190. Norris PJ, KaidarovaZ, Maiorana E, et al. Ultra-
Spring, MD: CBEROffice of Communication, violet light-basedpathogen inactivation and allo-
Outreach, and Development,2020. immunization after platelet transfusion: Results
179. Spencer B,SteeleW, Custer B, et al. Riskfor ma- from a randomized trial. Transfusion 2018;58:
laria in United States donors deferred for travel 1210-17.
191. Webert KE, Cserti CM, Hannon J, et al. Pro- red cells and platelet concentrates. Transfus
ceedings of a Consensus Conference: Pathogen Apher Sci 2011 ;45:75-84.
inactivation-making decisions about new tech- 194. Food and Drug Administration. Guidance for in-
nologies. Transfus Med Rev 2008;22: 1-34. dustry: Recommendations for assessment of
192. Klein HG, Anderson D, Bernardi MJ, et al. blood donor eligibility,donor deferral and blood
Pathogen inactivation: Malting decisions about product management in response to Ebolavirus.
new technologies. Report of a consensus confer- Silver Spring,MD: CBEROffice of Communica-
ence. Transfusion 2007;47:2338-47. tion, Outreach, and Development, 2017.
193. Seghatchian J, Hervig T, Putter JS. Effect of
pathogen inactivation on the storage lesion in
olecular Biology and Immunology in
Transfusion Medicine
Sean R. Stowell, MD, PhD, and James D. Gorham, MD, PhD
HIS CHAPTER REVIEWSFUNDAMEN- because many assays exist, specific assay proto-
tal principles and approaches for analyzing cols are not provided here.
nucleic acids and proteins, particularly an-
tibodies, and introduces basic concepts in hu-
moral (antibody-mediated)immunity. ANA NA
In the practice of transfusion medicine, anal-
yses of nucleic acids and antibodies are exten- The practical application of nucleic acid analysis
sively employed to 1) detect infectious patho- in transfusion medicine lies in two principal ar-
gens in donated components; 2) predict the eas: 1) the detection of infectious pathogens and
phenotypic expression of antigens on the sur- 2) the genotyping of blood donors and recipi-
face of cells (red cells, platelets, and neutro- ents. The human genome is made up of DNA,
phils); 3) detect and identify red cell and platelet organized into chromosomes, which are present
antibodies; 4) determine HLA type; and 5) in the cell nucleus. Whereas most cells in the
perform relationship testing. Although the last is human body are nucleated, red cells and plate-
somewhat outside transfusion medicine prac- lets lack nuclei and therefore DNA. The geno-
tice, it is a component of AABB'sstandards and type of an individual is his or her genetic make-
accreditation programs. up, encoded in the DNA;in practical transfusion
Entire textbooks explain molecular biology medicine, the term "genotype" is commonly
and immunology. This chapter focuses on topics used to. refer to the specific allele present at a
immediately relevant to the practice of transfu- single gene locus. Bacteria, fungl, protozoa, and
sion testing. Moreover, within immunology, the many viruses also use DNA to encode their ge-
primary focus is humoral immunity, which is nomes. Some viral pathogens use RNA instead.
more relevant to most of the practice of transfu- Either DNA or RNA encodes all known genetic
sion medicine than is cellular immunity. Finally, material relevant to transfusion medicine, with
Sean R. Stowell, MD, PhD, Director of Apheresis, Emory University Hospital, and Medical Director of the Blood
Bank and Transfusion Services, Emory University Orthopedic and Spine Hospital, Emory University School of
Medicine, Atlanta, Georgia; and James D. Gorham, MD, PhD, Professor of Pathology, Chief, Division of Labora-
tory Medicine, and Medical Director of the Blood Bank and Transfusion Medicine Services, University ofVirginia
Health System, Charlottesville, Virginia
229
the notable exception of prions,which appear to fer to the end with a free phosphate attached to
lack nucleic acids. C5 and the opposite end with a free hydroxyl
group attached to carbon 3 (C3), respectively
Basic Chemistry and Structure of [Fig8-1 (B)].
Nucleic Acids The human genome consists of double-
A brief overview of this topic is presented here, stranded DNA (dsDNA).The bases within one
while more details are available elsewhere.1 strand of DNA form noncovalent hydrogen
DNAis a nucleic acid polymer containingchains bonds with complementary bases on the other
of nucleotides. Nuc1eotides are composed of strand. Specifically,T alwayspairs with A, using
three moieties: 1) a pentose sugar (five carbon two hydrogen bonds, and G always pairs with
atoms); 2) a phosphate group linked to carbon 5 C, using three hydrogen bonds. When two
(C5); and 3) a base group attached to carbon 1 strands have complementary sequences, they
(CI) [Fig8-1 (A)]. can "hybridize" via hydrogen bonding between
There are four different nucleotides in complementary base pairs to form a dsDNA
DNA-adenine (A), guanine (G), cytosine (C),
molecule [Fig8-1 [Cl].The two complementary
and thymine (T)-which differfrom one anoth-
er in the chemical structure of the CI base strands align such that the 5' and 3' ends have
group. Polymersof DNAconsist of repeats of co- opposite orientations and form a double helix in
valently bound sugars and phosphates that form which the phosphodiester backbone is on the
the outside backbone of the double helical outside of the helix and the hydrogen-bond-
strand of DNA. DNA strands are described as paired bases are on the inside. DNA molecules
having a "5' end" and a "3' end," terms that re- vary from each other based on the sequence of
d>
5' End
A. B. -0
- I
ft < -base
Phosphate-
o-P-O-CH,
-6 5
4
0
1
O_~JO0 (11,\,
3 2 - sugar
4 1
OH
c. 3' TACTCCAGO.AACGA.TT 5'

o
I
2
DNA
5' ATGAGGTCTTTGCTAA 3'
O-j~~O ",yi 0
3' TACTCCA~CGA.TT 5'

Unwinding
ofDNA S' AUGAGGUCUUUGCUAA j' RNA 4 1
and transcription
3 2
{
5' ATGAGGTCTTTGCTAA 3' o
• I
3' TACTCCAGO.AACGA.TT 5'

DNA
O_j~~OoiiiiC!
5' ATGAGGTCTTTGCTAA 3'
4 1
5' ".~UCUU~ 3' RNA 3 2
•
Translation into Protein
OH
3' End
FIGURE 8-1. Chemical structure of nucleic acids and DNA.

C HAP T E R 8 Molecular Biology and Immunology in TransfusionMedicine 231
nucleotides, determined by the bases incorpo- The Polymerase Chain Reaction

rated into the polymer.
Nucleic acid detection and analysiswere revolu-
When genes are expressed, the DNA encod-
tionized by the invention of the polymerase
ing a given gene is used as a template to pro-
chain reaction (PCR)in the 1980s. PCRwas the
duce RNA,which is further processed to form a
first amplification-based technique for generat-
messenger RNA (mRNA)that is then exported
ing many copies of nucleic acid fragments for di-
from the nucleus to the cytoplasm and used as a
rect analysis." A PCRrequires: 1) a DNAsample
guide to produce proteins, the workhorses of
to be analyzed (the "target" or "template"); 2)
most cellular activity. The structure of RNA is
gene-specific primers of around 20 nucleotides
similar to that of DNA, with some differences:
in length; 3) a thermostable DNA polymerase
1) ribonucleotides have an additional hydroxyl
enzyme that recognizes primer bound to target
group on carbon 2; 2) uracil (U)is used in place
DNA and sequentially adds the complementary
of thymine (T); and 3) RNA is typically single
nucleotide building blocks to extend the length
stranded.
of the primed DNA strand; 4) the four nucleo-
Several classes of RNA exist in human cells;
tides (ATCG); and 5) the proper buffer. PCR
the class that is used as a blueprint for protein
involves repeat cycles of heating! cooling
synthesis is mRNA.When a gene is expressed, it
("thermocycling"), allowing exponential DNA
is used as a template to generate a copy of itself
in mRNA form, a process termed "transcrip- amplification of the fragment of interest, carried
out in a thermocycler that rapidly changes tem-
tion." The transcription machinery unwinds the
DNA double helix, synthesizes a new mRNA perature with accuracy and precision. A single
strand of complementary sequence, and recycle involves 1) heating to denature the tem-
plate dsDNAto allow strand separation, 2) cool-
anneals the DNA in its wake [Fig8-1 (C)]. Thus,
the mRNA represents a copy of the gene se- ing to allow annealing of primer to complemen-
tary regions on the template DNA, and 3)
quence present in the DNA. mRNA is always
extension and synthesis of DNA on the primer
synthesized in the "5' to 3' direction," and thus
strand. Typically,20 to 40 cycles are performed
for a given gene, only one of the two DNA
in .total, depending on the abundance of the
strands is transcribed into mRNA.After synthe-
sis in the nucleus, mRNA is processed and ex- template and the required sensitivity of the as-
ported to the cytoplasm,where ribosomes use it say.
Figure 8-2 shows an example, beginning
as a template to synthesize a new protein mole-
with a single copy of a dsDNA template. The
cule, a process called "translation."
dsDNA is denatured by heating to near boiling
(~95 C), disrupting the hydrogen bonds be-
Isolation of Nucleic Acid
tween complementary bases, thereby separat-
The first step in most DNA and RNAanalysesis ing the two strands. The temperature is then
the isolation of nucleic acids. Because essentially lowered to allow gene-specificprimers in the re-
all nucleated cells of an individual contain iden- action mix to anneal to their complementary
tical.genomic DNA (gDNA),gDNA can be iso- targets. One primer is designed to anneal "up-
lated ..from any readily obtainable cellular stream" (aka "just 5"') to the region of interest,
source, .such as peripheral blood leukocytes.or whereas the other is designed to anneal "down-
buccal swabs (epithelialcells).However, mRNAs stream" (aka "just 3"') to this region. The tem-
differbetween cell types, because their differen- perature is then raised to 72 C, the temperature
tial expression patterns are important in defin- at which the thermostable polymerase functions
ing the phenotypes of these cells. The implica- optimally, and the primers are extended along
tion of this is that for mRNAanalysis,the choice the length of the DNA through incorporation of
of source cells is critical. Kits are readily avail- complementary nucleotides. Thus, at the end of
able from multiple manufacturers for the simple extension, there are two copies of the DNA.
and rapid isolation of high-quality cellular DNA The process of denaturing, annealing, and ex-
or mRNA,or of nucleic acids from plasma. tension then repeats itself. Each subsequent
Copies
Melting/annealing
- =
Extension 2
-
__:__=-=
=
__,=
Melting/annealing
- =
4
Extension
FIGURE 8-2. Overview of the polymerase chain reaction.
cycle yields a doubling in DNA copy number. thermostable RNAseenzymes found in many bi-
peR results in an exponential expansion of a se- ologicspecimens.
lected DNA "amplicon," defined as the se-
quence bound by the two chosen primers. Inhibitors
Because peR amplification depends on the en-
PCR Considerations zymatic activity of DNA polymerase, peR can
Although peR is a robust method of detecting be inhibited by substances that negatively affect
nucleic acids, technical considerations can affect the activity. Heparin can inhibit peR, and hemo-
peR and other ampllflcation-basedtechniques. globin or lactoferrin released from erythrocytes
or leukocytes also inhibits this process.' Most
Specimen Processing and Template analytic systems minimize risk of interference
by an inhibitor, but deviations from established
Degradation
protocols may introduce unintended inhibitory
DNA is stable and can usually withstand varia- substances. To detect the presence of inhibitors,
tions in storage temperature and handling be- controls include amplifyingubiquitous target se-
fore being processed for genomic analysis. Ex- quences (conserved regions of gDNA) and/or
ceptions include samples in which the target spiking the specimen with a positive control.
DNA is present in low quantity, such as fetal
typing from maternal plasma and viral testing. Primer Design
Bycontrast, RNAis far less stable, as it is suscep- Although inferior performance of primers is typi-
tible both to spontaneous autodegradation and cally not a concern for commercially available
to catalytic degradation mediated by abundant tests, understanding primer design is important
in assay troubleshooting and in developing PCR- Another effective method to avoid contami-
based assays for new targets. Although the ideal nation involves the addition of deoxyuridine tri-
primer hybridizes to a target found in only one phosphate (dUTP) before PCR amplification.
location in the entire genome, given the com- Polymerases incorporate dUTP in place of de-
plexity of gDNA, primer annealing to unintend- oxythymidine triphosphate (dTTP).The enzyme
ed targets can occur, resulting in both the poten- uracil-DNA glycosylase (UNG) is also added to
tial to amplify unintended targets and the specificallycleave DNA containing uracil"; thus,
ongoing consumption of primers. In addition, UNG in PCR reactions destroys contaminating
primers can anneal to each other to form a short amplicons from previous amplifications but not
amplicon, in a so-called prlmer-dlmer forma- native DNAin the specimen. The initial PCR de-
tion." naturation step inactivates the heat-labile UNG.
Finally,it is standard practice to include a wa-
Contamination ter control, for which water, rather than a DNA
sample, is used as the input sample. The water
One of the greatest strengths of PCRis its ability control (no-DNAcontrol) reaction should yield
to amplifyvery small amounts of genetic materi- no detectable signal above background. Signal
al. In theory, single-copy sensitivity can be in the water control indicates probable contami-
achieved. In practice, about 10 copies of target nation of a reagent or instrument and invali-
DNA are the lower limit of detection, depending dates the test run.
on the sensitivity of the assay readout. The sen-
sitivity makes PCR susceptible to false-positive Reverse Transcriptase PCR
results due to contamination either from other
specimens being analyzed or, more commonly, Messenger RNA is unsuitable as a template for
from amplicons generated in previous PCRruns. PCR. When analysis of mRNAis desired, an ad-
Beginning with. just 10.molecules of DNA, 30 ditional step is employed to generate a single-
rounds of PCR amplificationyields> 1010 ampli- stranded complementary DNA (cDNA), using
cons. Thus, as little as 0.0000001 % of a previ- mRNA as the template. The enzyme reverse
ous reaction inadvertently introduced into a pi- transcriptase (RT) synthesizes cDNA from an
pette, picked up from a laboratory surface, or RNAtemplate (in the 5' to 3' direction), requir-
from the thermocycler can lead to a false- ing the annealing of a primer to initiate tran-
positiveresult in a subsequent reaction. scription. The cDNAis then a suitable substrate
To minimize contamination, PCR laborato- for PCR.
ries routinely process samples in one geograph-
ic direction. DNA extraction is located away Transcription-Mediated Amplification
from testing, and PCR reactions are assembled and Sequence-Based Amplification
in one room (orin a positive pressure hood) and There are additional non-PCR amplification
amplified in a second room; any downstream techniques; among these, transcription-mediated
analysis is performed in a third room. There amplificationlTMA) and nucleic acid sequence-
should be no retrograde flow, and no materials based amplification (NASBA) are further de-
or instruments used in the amplificationor anal- scribed (Fig8-3).7,8
ysis (post-PCR) rooms should ever make their TMAplays a large role in nucleic acid testing
way into the DNA extraction and PCR setup (NAT) for human immunodeficiency virus
(pre-PCR)room. Filtered pipette tips are routine- (HN), hepatitis C virus (HCV),and West Nile vi-
ly used to minimize carryover contamination rus, for which viral RNAis the target. The reac-
and sample aerosols. Recent developments in tion contains two primers, RT,DNA polymerase,
which PCR extraction is avoided altogether and RNAse H, and a sequence-specific RNA poly-
the PCR process uses an entirely closed system' merase called T7 polymerase. A sequence-
may soon obviate the need for geographic direc- specificdownstream primer hybridizes to the 3'
tional flow in clinicallaboratories. end of the target RNA, and RT synthesizes a
Hybridization 01 Primer 1
5' - - - - DegradedRNA
RNA 5'--------------------------------3' Primer 1 containing
Reverse transcriptass ! Step 1 --

T7promoter
Primer2
5'
eDNA 3'----------------------------~r
RNA 5'---------------------------------3'
!
eDNA 3' RNAse H
y::~'
Step 2
RNA 5' - - - - - - - - - - - - - - - - - - - - -3'

5'} RNAseH is only used in NASBA
InTMA,RNAisdegradedbythe
reverse transcnptase.
eDNA 3'
Hybridization 01 Primer 2 ! Step 3
~~~~9~
5'
Primer2 5'-
L DNA polymerase
! Step4
3' 5'
NewDNA 5,;;;:======================~,~-~J3'
Template
T7polymerase ! Step 5
1. Direct detection
Multiple
copies _
of RNA
antisenseto 2. Additional
original
mRNA --------------- amplification
FIGURE 8-3. Overview of transcription-mediated amplification and nucleic acid sequence-based amplification.
RNA = ribonucleic acid; cDNA = complementary deoxyribonucleic acid; mRNA = messenger RNA; RNAse =
ribonuclease.
cDNAcopy (Fig8-3, step 1). Primer 1 contains a vantage of NASBAover PCR is that repeat nu-
sequence at its 3' end that hybridizes to the tar- cleic acid denaturation is unnecessary; amplifi-
get RNA and a specific sequence at its 5' end cation of RNA is isothermal and thus does not
that serves as a promoter for the T7 polymerase. require a thermocyc1er.
The RNA template is degraded either by RT it-
self (TMAassay) or by RNAseH (NASBAassay) Detection of Amplification Products
(step 2). A second primer (primer 2) then binds
to the newly synthesized cDNA (step 3) and uti- Traditionally, amplicons were detected through
lizes DNA polymerase to synthesize a dsDNA separation by gel electrophoresis, followed by vi-
molecule (step 4). This molecule now has a T7 sualization of fragment length using a fluores-
promoter at one end (from primer 1), and thus cent DNA intercalating agent such as ethidium
T7 polymerase drives transcription of new RNA bromide. Because gel electrophoresis is time-
(step 5). The numerous RNAtranscripts synthe- consuming and not readily automated, more ad-
sized from a single DNA template can reenter vanced techniques for detecting amplicons have
the amplification cycle, with primer 2 initiating become routine in the clinicallaboratory.
reverse transcription, followed by RNAdegrada-
tion and subsequent synthesis of DNA using Real-Time ten
primer 1 and DNA polymerase. This leads to ad-
ditional amplification, with ongoing cycles of Real-time PCR employs one or more "probes"
transcription and template synthesis. One ad- that fluoresce only when the target amplicon is
present. A probe is a short oligomer of DNA that cence does not occur unless the two tags are
hybridizes to a specific cDNA sequence located near to one other. If the amplicon is present, the
within the amplicon. For real-time PCR ap- probes anneal so that the two tags are in close
proaches, the thermo cycler used also incorpo- proximity, and fluorescence ensues [Fig8-4 (C)].
rates a built-in fluorescence spectrophotometer, A fourth method (not shown) uses a dye
allowing for the detection of amplification prod- called SYBRgreen (Thermo Fisher Scientific),
ucts with each cycle. Real-time PCR is highly which fluoresces only when bound to dsDNA.
sensitive and quantitative and (through the use Unlike the above approaches, SYBRgreen is not
of multiple reporter dyes with distinct fluores- sequence-specific but detects all dsDNA in the
cence spectra) permits the detection of multiple reaction tube. It is thus less specific than ap-
targets in a single reaction tube (multiplex am- proaches that use sequence-specific probes and
plification). In addition, because detection al- is more prone to false-positiveresults. Fortunate-
lows analysis without opening the reaction tube, ly, authentic amplicons typically can be distin-
the risk of laboratory contamination with ampli- guished from aberrant products by making use
cons (and subsequent false-positive results) is of a "melting curve" analysis, which assesses the
greatly minimized.
temperature profile at which the amplicon dena-
Several approaches can be used to detect flu-
tures ("melts"). Because the melting curve is a
orescence resulting from amplicon production.
The TaqMan system (Applied Biosystems) em- function of amplicon size and of GC content,
ploys a probe built with a fluorescent chemical and the size and sequence of the correct ampli-
tag (a "fluorochrome") at one end and a fluores- con is known, the melting curve is useful to
cence quencher tag at the other [Fig 8-4 (A)]. confirm the identity of the amplicon.
Tethered to the probe, the two tags are close These fluorescent-probe techniques are appli-
enough that the quencher effectively blocks the cable not only to real-time PCR, but also to oth-
fluorescent signal. During amplicon synthesis, er amplification technologies, such as TMA
DNA polymerase, moving along the template (above).
strand, encounters the hybridized probe. DNA
polymerase also has a nuclease activity, which DNA Arrays
causes probe degradation. The fluorochrome is Evaluation at the DNA level of differences in the
now liberated from the probe and, no longer in genes encoding protein blood groups is becom-
proximity to the quencher, begins to fluoresce. ing much more commonly employed in the
Because fluorochrome liberation occurs only practice of transfusion medicine." The vast ma-
when 1) the probe hybridizes to its target and 2)
jority of blood group antigens are the result of
DNA polymerase degrades the bound probe, flu-
small differences in membrane proteins, often a
orescence generation is a highly specific func-
single amino acid residue, that are encoded by
tion of amplicon generation.
A second approach uses a molecular "bea- single nucleotide polymorphisms (SNPs) at the
con" that, like the TaqMan probe, has a fluoro- level of gDNA.
chrome tag at one end and a quencher tag at the Methods for detecting blood group SNPs pri-
other. The beacon probe is built such that the marily use PCR amplification followed by analy-
target sequence is flanked by complementary se- sis by DNA-array technologies. After the region
quences. Unbound probe forms a hairpin loop, of interest in a blood group gene is amplified by
thus closely juxtaposing the fluorochrome with PCR, specific products are detected by primers
the quencher. As the amplicon is generated, the (or probes) designed such that hybridization and
beacon hybridizes to its target, and the hairpin signal output depends on the presence of one al-
loop unfolds; the quencher is now sufficiently lele but not the other. Multiplex PCR and DNA-
distant from the fluorochrome that fluorescence array systems can determine the genotypes of
is generated [Fig8-4 (B)]. blood group antigens in individual specimens
A third approach uses two probes, each of and offer high throughput and automated read-
which has a distinct tag tethered to it. Fluores- out. 10-12
A.
Amplicon 11111 I 11111111 III II I 1111 II I
Amplicon IIIIIIIIII~IIIIIIIIIIIIIII
Amplicon
Amp,,:":""j"li 11111111I1 ;IIIIII~IIII ~
B.
Amplicon 111111111111111111111 1111111111111111111
Fluorescence
c.
Amplicon 1111111111111111111111111111111 111111111
Fluorescence
Amplicon
FIGURE 8-4. Methods of detection by sequence-specific probes during real-time polymerase chain reaction.
Pol = polymerase enzyme.
Genotyping to predict red cell antigens can RhCE allele sequences is particularly important,
be more efficient than traditional serologic typ- and genotyping platforms are often incapable of
ing.13.16In the case of the patient who has had resolving the underlying allelic composition."
multiple transfusions,where it may not be possi- Because current FDA platforms require inclu-
ble to distinguish the patient's own red cells sion of specific targets a priori, a new (hitherto
from transfused red cells, genotyping patient unknown) polymorphism in a coding region
DNA can reliably predict the patient's red cell that alters protein structure, or in a noncoding
phenotype. 17When serologicreagents inadequate- region (such as the gene promoter) that regu-
ly characterize certain antigens (eg, partial D), lates expression, would not be detected. As the
genotyping can provide detailed information" relationship between genetic •polymorphisms
and is helpful in identifying which patients with and antigen expression continues to be refined,
a weak D identified using serologic testing the prediction of antigen expression will contin-
would benefit from Rh Immune Globulin ue to improve.
(RhIG) immunoprophylaxis." Genotyping can
also be helpful when assessing alloantigens that
Next-Generation Sequencing
may not be detected serologically, such as those
of the DO (Dombrock] system, some forms of The entire human genome was sequenced by
weak D (eg, DEL), and FyX2o;identifying these traditional sequencing methods in 2001 at a
antigens is particularly helpful when investigat- cost of nearly 13 billion US dollars. Defining the
ing patients manifesting hemolytic transfusion sequence of the human genome was expected
reactions in the absence of detectable alloanti- to dramatically change approaches to genetic
bodies." Genotyping to predict multiple blood analysis of clinical disease and to biomedical re-
group antigens is especially useful in sickle cell search, but the high cost and excessively long
disease (SCD) patients and other disease states turnaround time of whole genome sequencing
in which frequent transfusion is expected and by traditional methods prohibited its use in rou-
antigen-matching to prevent alloimmunization tine clinical practice. However, over the last de-
is an important component of long-term care.22·25 cade or so, more robust sequencing technolo-
Genotyping platforms currently approved by gies have emerged, collectively referred to as
the Food and Drug Administration (FDA)focus next-generation sequencing (NGS), allowing the
on known single nucleotide and other common realization of the practical potential of DNA se-
polymorphisms, including those that regulate
quencing."
antigen expression reg, GATAmutation impact- Whereas early sequencing approaches ad-
ing FY (Duffy)expression]," and allow the iden-
dressed one stretch of DNA at a time, NGS in-
tification in parallel of multiple polymorphisms,
volves sequencing massive amounts of distinct
greatly facilitating the selection of appropriate
stretches of DNA in parallel. Although different
Red Blood Cell (RBC) units for transfusion for
patients with multiple antibodies.25,26However, NGS approaches use a variety of platforms and
some haplotypes., such as .RHCE*ce alleles that technologies, all involve sequencing individual
express Rhce variants. may be inferred based on sections of target DNA multiple times as
results generated on current platforms, although "reads." A bioinformatic algorithm incorporates
confirmatory studies are typically required. overlapping reads with a prior sequencing data-
Some blood group antigens arise from complex base to determine the location of a relevant read
genetic interactions or from alleles not fully rep- within a reference genome. NGS aligns reads in
resented on FDA-approved platforms. Genotyp- silico to produce the genetic signature of a given
ing platforms do not clarify complex antigen de- individual..Although the fidelity of a smgl~ N.GS
terminants such as ABO, GLOB, RHD, RHCE, read is less robust than older sequencing ap-
GYPA, and GYPB.17Despite prophylactic match- proaches, an increased read number greatly en-
ing, alloimmunization toward Rh variants con- hances statistical confidence. The ability of NGS
tinues to be a significant challenge in transfusion to simultaneously sequence multiple stretches of
practice. The reliable determination of RhD and DNA has dramatically reduced the time and
cost of sequencing an entire genome, with costs bind more than one target molecule, allowing
approaching 1000 US dollarsper genome." antibodies to crosslink antigens present in multi-
Despite significant advantages over tradition- ple copies on red cells. A standard serologic
al sequencing technologies, NGS has limita- method for detecting antibody-antigen interac-
tions. First, although the cost of acquiring the tions, agglutination, is used extensively in trans-
raw sequence has dropped dramatically," the fusion medicine.
bioinformatics infrastructure required to store The antigen copy number and density vary
and process the very large dataset is substantial depending on the blood group. Agglutination is
and can be cost prohibitive. These costs may be used for serologic crossmatching (donor red
partially offset by mining extant sequence data cells incubated with recipient plasma or se-
previously acquired for other clinical indica- rum), screening for unexpected antibodies (re-
tions. Indeed, some groups have begun to agent red cells of known blood group antigen
develop promising bioinformatic methods composition incubated with recipient plasma or
interrogating NGS data for blood group charac- serum), and blood group antigen phenotyping of
terization, with high accuracy.29.32Second, due the donor or recipient (test red cells incubated
to relatively short reads, NGS is often limited in with monoclonal antibodies or reagent-quality
its abilityto distinguish highly similar genetic se- antisera of known specificity).
quences, such as the paralogs RHD and RHCE, Agglutination can be detected by several
which are 93% identical.P However, recent methods. With manual tube testing, agglutina-
paralog-specificNGS approaches can successful- tion is visually detected by the adhesion of red
ly detect variations in RHD and RHCE.34 Al- cells to one another in the postcentrifuge pellet.
though such strategies are still at an investiga- Agglutlnation in microtiter plates is visualized
tive stage, they illustrate the potential impact of by the spread pattern of red cells in individual
genomic approaches on the practice of transfu- wells. Gel-based testing is used widely; after ago
sion medicine. For example, coupling robust glutination is allowed to take place, the reaction
paralog-specificNGS approaches in donors and mixture is centrifuged through a gel matrix, typ-
recipients with targeted efforts to increase mi- ically composed of dextran-acrylamide. Un-
nority blood donor participation holds Signifi- agglutinated red cells pass through the gel,
cant promise in reducing the deleterious conse- while the larger, agglutinated complexes are re-
quences of RhD and RhCE alloimmunization in tained at the top of, or within, the matrix. Ad-
patients with SCD.35 vantages of gel-based testing over tube testing
include standardization of reaction strength,
sensitivity,and streamlined throughput, because
AN IN of both the use of automated platforms and the
opportunity to eliminate some wash steps."
Much laboratory testing in transfusion medicine Although agglutination reaction tests are sen-
involves the detection and identification of anti- sitive and easy to perform, the formation of ag-
bodies in patients' plasma. In the following sec- glutinates depends on the proper stoichiometric
tions, the principles of the more common labo- ratio of antibody to antigen. The "zone of equiv-
ratory assays used to detect antibodies are alence" refers to the ratio of antigen to antibody
described. Less routinely used technologies that that permits ready agglutination. Each arm of
are nevertheless important to understand are the antibody binds to a different particle, and a
also presented. network (lattice)of linked particles results in ag-
glutination [Fig8-5 (B)].
Fluid-Phase Assays (Agglutination- A false-negative result is generated if the
Based Methods)
stoichiometric ratio is outside the zone of equiv-
alence at either extreme. A prozone effect can
Depending on antibody isotype, immunoglobu- occur with an unusually high antibody concen-
lins contain two (IgG) to 10 (IgM) antigen- tration, diminishing the likelihood of an anti-
binding sites per molecule. Each antibody can body binding to two separate particles or red
A.
No agglutination
Prozone effect (antibody excess)
B.
Agglutination
Zone of equivalence
c.
No agglutination
Postzone effect (antigen excess)
FIGURE 8·5. Effects of relative concentrations of antigen and antibody on the outcome of agglutination
reactions.
cells [Fig8-5 (A)].Although unusual in classical protein from solution and irreversibly binds
red cell serology, prozone has been observed protein to the plastic. The well is washed, the
when titers of red cell antibodies are very high analyte is added and incubated with the protein-
and can cause discrepant reverse ABO typing." coated solid phase, and its adherence is mea-
Diluting the serum being tested, or using dilu- sured. Several combinations of adherence and
ents containing EDTA,decreases the likelihood detection approaches have been described.
of prozone." False-negative agglutination reac-
tions can also be the result of a postzone effect, Solid-Phase Assays for Phenotyping Red Cells
which occurs when there is excess antigen [Fig Antibodies specific for a known blood group an-
8-5 (C)] and each. antibody binds to multiple tigen are coated onto round-bottom microtiter
epitopes on the same particle, thereby prevent- plates [Fig8-6 (Al]. Red cells to be analyzed are
ing crosslinldngand agglutination. added to the wells and allowed to adhere; then
the plate is centrifuged. If no binding occurs
Solid-Phase Assays (negativereaction), the red cells cluster together
In solid-phase assays, a specific antigen or anti- as a "button" at the well bottom. In contrast,
body is immobilized on a solid matrix, typically specific binding results in dispersion of the red
made of plastic. A solution containing the pro- cells over the surface of the entire well (positive
tein of interest is placed into a well; the polysty- reaction), which indicates the presence of the
rene (or other plastic employed) directly absorbs antigen on the red cells.
A.
Negative
Red cells
Antibody- Positive
coated well
B.
Negative
Serum
e/A
Red cells coated with anti·lgG
Positive
eO Antigen·containing particle
FIGURE 8·6. Schematic of (A) phenotyping red cells and (8) detecting antibodies by solid-phase assay.
Solid-Phase Assays for Detecting Antibodies such as HPA-la, as well as to detect antibodies
to Red Cell Antigens againstplatelet antigens."
Antigen-coated particles, typically red cells (or
Enzyme-Linked Immunosorbent Assay
red cell fragments), are coated onto microtiter
plate wells [Fig8·6 (B)]. Patient serum is then Enzyme-linked immunosorbent assay (ELISA,
added, followed by incubation and washing. If aka "enzyme immunoassay") can detect either
the patient serum contains antigen-speciflcanti- antibodies or antigens. To generate a signal, a
bodies, they will bind to the red cells. Indicator secondary antibody is used that is tethered to an
red cells (coated with antihuman IgG) are then enzyme, that modifiesan added enzyme-specific
added. A positive reaction is demonstrated by substrate to generate either color (chromogenic
reaction) or photons (chemiluminescent reac-
diffuse adherence of the indicator red cells to
tion). The use of secondary antibodies and enzy-
the well, whereas a negative reaction is demon-
matic signal generation renders ELISAscapable
strated by clustering of indicator cells in a but-
of robust signal amplification; ELISAsare there-
ton.
fore much more sensitive than fluid-phaseagglu-
tination or SPRCAassays.
Solid-Phase Assays for Platelet Testing
ELISAstypicallyuse purified or recombinant
Using the approaches described above, solid- antigens or antibodies, depending on the analyte
phase red cell adherence (SPRCA)technology to be detected. However, intact red cells can be
has been adapted to detect antigens on platelets, used to screen for red cell antibodies, and this is
referred to as the enzyme-linked antiglobulin through use of a standard curve. Sometimes

test." samples need to be diluted to ensure that they
Detection of Antibodies by the Indirect yield absorbance values in the linear range of
ELISA. To detect antibodies against a specific the assay. Detection of antigens carried on
antigen, the antigen is coated onto microtiter multipass transmembrane proteins can be diffi·
plate wells [Fig8-7 (All.The test sample is then cult in ELISAs,because epitopes may not retain
added, incubated, and washed. Antigen-specific antigenic conformation when deposited onto a
antibodies bound to the antigen-coatedwell are solidsurface such as a microtiter well.
then detected by incubating with an antibody Detection of Antigens by Sandwich
(eg,anti-IgG)that serves as a reporter, as it is en- ELISA. The sandwich ELISAis used to detect
gineered to be tethered to an enzyme such as al- and quantify a specific soluble antigen. For this
kaline phosphatase or horseradish peroxidase. assay, two different antibodies [typicallymono-
After further washing, enzyme substrate is add- clonal antibodies (MoAbs)] are used, each of
ed and enzymatically converted to a detectable which binds separate epitopes on the same tar-
color. A spectrophotometer is used to measure get antigen without cross-interference. Microti-
light absorbance at the wavelength specific for ter plate wells are first coated with one MoAb
that enzyme/substrate. Color intensity is posi- (the "capture" antibody) [Fig8-7 (B)].The sam-
tively related to the amount of antibody bound ple is then added, and antigen in solution binds
to the antigen. Quantification is possible to the capture antibody.· Next, the plate is
Serum Secondary antibody
A.
)..J..A).J.. /'
Incubate
$i~~
;;\ i''S- , -+-
Incubate
and wash
Substrate
~~~~
o Purified antigen
Substrate Substrate
Analyte (no color) (withcolor)
t7 Incubate Incubate () D r--+
-~
B. and wash and wash :f -----
~~
Y Capture antibody
c. 't). .. . Jy I NoanligenInanalyle
Analyle_
o Purified antigen
~~
;"o;..AA
A :J
~ - - <:1"0
--
Antigenin analyle
Secondaryantibody
linked to enzyme
and detection
~
FIGURE8-7. (A) indirect enzyme-linked immunosorbent assay (ELISA), (8) sandwich ELISA, and (C)
competitive ELISA.
washed and incubated with a second MoAb added simultaneously. The hook effect can be
linked to a reporter enzyme. Becauseit is specif- readily overcome by diluting the antigen. Final-
ic for the target antigen, the reporter antibody ly, patients exposed to mice or to mouse-based
binds to the well only if antigen is alreadybound biologic drugs may develop human anti-mouse
via the capture antibody. After additionalwash- antibodies (HAMAs)that crosslink the capture
ing, enzyme substrate is added, and it is convert- antibody and/or detection antibody in sandwich
ed to a detectable color if antibody with enzyme ELISAs,resulting in very high signals.
has been bound.
Detection of Antigens by Competitive Protein Assays Encountered Less
ELISA. Competitive ELISAbegins similarly to Commonly in Transfusion Medicine
indirect ELISA,using wells in which target anti-
gen is alreadybound. The test sample is preincu- A number of other technologicalapproaches can
bated with solution-phase antibody, and this be used in the analysisof proteins. For a variety
mixture is added to the well [Fig8-7 (Cl], If no of reasons, these approaches have not been
antigen is in the specimen, then reagent widely adopted in blood banking and transfu-
antibodies bind unimpeded to the solid-phase sion medicine, although some reference labora-
antigen. If antigen is present in the specimen, tories use such technologies as laboratory-
the antigen binds to reagent antibodies, prevent- developed tests (LDTs).
ing them from binding to the solid-phase anti-
gen. As the amount of soluble antigen in the Protein Microarrays
specimen increases, the amount of reagent anti- Microarray technology dramatically increases
body free to bind to the solid-phaseantigen de- the number of substrates that can be simultane-
creases. Thus, the signal is inversely related to ously assayed by solid-phase methods. When
the amount of soluble antigen in the specimen. numerous different proteins are placed (spotted)
Competitive ELISAis also used for antibody on a small chip, a single specimen can be as-
detection. In this case, the test sample is added sayed for binding activityto multiple analytes si-
to an antigen-coated well along with a labeled multaneously. For example, a microarray chip
antigen-specific reagent antibody. Patient anti- spotted with different blood group antigens can
body and labeled reagent antibody compete be used to assess a singie patient specimen for
with each other for antigen-binding sites in the several alloantibodies simultaneously. Like
well. Again, a higher signal is generated if the ELISAs,protein microarrays require that the
antibody level in the sample is low or absent. Al- structural conformation required for antibody
though more difficultto optimize than sandwich recognition be maintained. The practical appli-
ELISAs,competitive ELISAsdo not require two cation of protein microarrays to blood bank se-
separate antibodies against different epitopes on rologyhas yet to be realized.
the target antigen.
Technical Problems with ELISAs. ELISAs
Western Blotting
are usually straightforward and robust. False-
negative signals can result from enzymatic in- Although highly sensitive, ELISAsare prone to
hibitors in the sample, and false-positivesignals false-positiveresults if the antigen used to coat
from nonspeclfic enzymatic activity; however, the well is not pure (eg, cell lysates of viruses
the use of proper controls and thorough wash- grown in tissue culture), leading to cross-
ing typically prevents these problems. Falsely reactivity with other components in the antigen
low signals can occur if the amount of antigen preparation. In Western blot (WB) assays, the
exceeds the amount of antibody present, a phe- antigen mixture is first separated by high-
nomenon termed the "hook effect." Similar to resolution protein electrophoresis, using poly-
the prozone effect (see "Fluid-Phase Assays" acrylamidegels. The separated proteins are then
section above), excess antigen can cause the sig- transferred to a membrane to serve as the solid
nal to decrease in some sandwich ELISAsin phase for probing with an antibody-containing
which the antigen and detection antibody are patient sample. Using molecular-weight size
markers in an adjacent lane, one can determine arrays of microspheres capable of detecting mul-
the molecular weight of the antigens recognized tiple analytes at the same time.
by the antibodies. Alternatively, antigens can be For sample analysis, a suspension of beads
~?s~s~~~:snC~;g~jSiSof other physical.proper- covalently tethered with receptors is first incu-
bated with the solution of interest (eg, plasma)
Because of the low likelihood of .aiCross- to bind the target analyte (antigen or antibody).
reactive antigen sharing the same physical char- Next, the bead-receptor-analytesuspension is in-
acteristics as the intended analyte, WB provides cubated with a secondary MoAb. The secondary
more specificitythan ELISAs.WB can be used to (reporter) MoAb is labeled with its own fluoro-
confirm positive serologic screening assays of chrome, rather than with an enzyme (as done in
transfusion-transmitted infectious agents, such ELISA).Detection is rendered by sending bead
as HN or HeV, although.NAT testing has be- suspensions through a flow cytometer, which
come the approach of choice in confirmation of detects individual beads (asopposed to individu-
infection by these viruses. al cells). Software in the flow cytometer can rec-
ognize the specific capture antibody on each in-
Flow Cytometry dividual bead based on the unique fluorescent
Flow cytometry revolutionized the analysis of signature of the bead and can determine the
cell populations. The basic principle is that quantity of analyte bound based on the fluores-
fluorescent-tag-labeled antibodies against cell- cent intensity of the secondary MoAb. Multiple
surface molecules are incubated with a target specificanalytes can be measured simultaneous-
population of cells; these "stained" cells are ly. The system allows for high throughput using
then passed through a flow cytometer. As they very small quantities of sample.
travel, cells are exposed to lasers that excite the An example of SATapplicable to transfusion
fluorescent tags, causing emission of photons of medicine is the use of the Luminex system (Lu-
very specific wavelengths that can be detected minex) in the detection and identification of
by sensors in the flow cytometer. Fluorescence HLA-specificalloantibodiesfor screening platelet
is assessed on a cell-by-cellbasis, allowing for donors and for workup of platelet refractori-
the visualization and quantification of minor ness." This system is also used for blood group
populations of cellswithin a complex mixture." genotyping.
Although use in transfusion practice is limited,
some laboratories use flow cytometry to quanti-
fy RhD-positivefetal red cells in a sample of ma- s
ternal blood in order to determine appropriate
dosing of RhIG to prevent sensitization to the The process by which the immune system gen-
RhD antigen in RhD-negativepregnant females." erates antibodies against foreign antigens (yet
maintains tolerance to self-antigens) is compli-
Suspension Array Technology cated and elegant, with multiple cellular players
and intricate regulation. Of necessity, this sec-
Suspension array technology (SAT) combines tion is primarily limited to antibody structure,
the specificityof solid-stateantibody/antigen in- function, and role in transfusion complications.
teraction (ELISA)with the sensitivity and high
throughput of optical detection by flow cytome- Antibody Structure
try.43 Through selection of fluorescent dyes
during the manufacturing process, populations At its simplest, an antibody is a tetramer com-
of microspheres (beads) are generated with disposed of two identical heavy chains and two
tinct fluorescent properties and used as solid identical light chains (Fig8-8). Each heavy chain
supports for the initial binding of specificrecep- and light chain contains a variable region, which
tors (capture antigens or antibodies). Bypairing is the part of the molecule that varies between
beads of specific fluorescent optical properties antibodies and binds antigen, and a constant re-
with specific receptors, it is possible to prepare gion. Two light-chain families (kappa and lamb-
Antigen- Antigen-
••• Variable domains binding binding
site site
i CO""'"tdom""I'~ab {
domain,
Antigen binding
Fe
domain,
Effector biology
Antigen Antigen
FIGURE 8·8. General structure of a monomeric immunoglobulin.
da) are found in humans. A given antibody has vidual binding sites of IgM are of relatively low
either two kappa light chains or two lambda affinity, IgM demonstrates high antigen avidity
light chains. because it has 10 antigen-bindingsites. In IgM,
Immunogiobulins treated with the enzyme the five immunoglobulin molecules are held to-
papain can be digested into two functional frag- gether by an additional protein (the J chain) and
ments. The "Fab" fragment consists of the by extensive disulfide binding. Treatment with
heavy-and light-chainvariableregions, the light- dithiothreitol (DTT)can destroy IgMbinding be-
chain constant regions, and one heavy-chain cause it cleaves (reduces) disulfide bonds; DTT
constant region domain. The Fab fragment treatment is used in the clinical laboratory to
binds antigen but does not activate effector distinguish IgM antibodies from IgG antibodies.
mechanisms. In contrast, the Fc fragment, con- IgM potently activates complement by chang-
sisting only of heavy-chain constant regions, ac- ing its three-dimensional structure after antigen-
tivates effector mechanisms, allowing destruc- binding. In general, IgM (and some IgG) can
tion of the antibody target. Fc constant regions cause hemolysis during transfusion reactions
differbetween antibody molecules based on an- and in autoimmune hemolytic anemia. Impor-
tibody isotype and subclass. tantly, unlike IgG, IgM does not cross the pla-
There are five different antibody isotypes centa and is therefore not involved in hemolytic
(IgM, IgG, IgE, IgA, IgD), determined by the disease of the fetus and newborn.
constant region of the heavy chain. Antibodies Whereas IgM is generated early in the
of different isotypes differboth in the number of antigen-specificimmune response, IgG antibod-
antigen-binding sites per molecule and in the ies are important in mature humoral immune ef·
potency of their effector functions [Fig8-9 (A)]. fector functions, and are divided into four sub-
The "affinity" of an antibody for its antigen re- classes: IgG1, IgG2, IgG3, and IgG4. Each
flects the binding ability of a single binding site, subclass has a different constant region and a
whereas the "avidity" refers to the total binding different capacity to activate complement and/
strength conferred by the combined effects of or interact with Fc receptors on phagocytes [Fig
multiple binding sites. Thus, although the indl- 8-9 (B)]. IgG1 and IgG3 are the most potent,
J chain
IgM IgD IgG IgE IgA
B.
~( ~( ~( ~(
IgGl IgG2 IgG3 IgG4
C' activation Strong Weak Strong No
Binds FcRs
on phagocytes
Yes No Yes Weakly
FIGURE 8-9. Immune globulin (Ig) isotypes (A), IgG subclasses,and their relative activation of complement and
binding to Fc-gammareceptors (FcyRs) (8).
whereas IgG2 only weakly activates comple- should be considered when analyzing a patient
ment, and IgG4 largely lacks effector activity. with both hemolysis and a negative direct anti-
Consistent with these observations, patients globulin test result.
with. only IgG4 subclass red cell antibodies typi- IgE antibodies bind to Fc receptors on. mast
cally do not exhibit hemolysis. In contrast, cells, inducing histamine release upon antigen
IgG1, IgG2, and IgG3 red cell antibodies can all encounter, and are the predominant cause of al-
induce hemolysis. lergic and anaphylactic responses (TypeI hyper-
IgA is the primary antibody isotype secreted sensitivity). IgD primarily remains membrane
at mucosal surfaces; therefore, it Is largely rebound. on theBcell surface, with only minimal
sponsible for neutralizing pathogens encoun- levels in serum, but itsfunctions.remain unclear.
tered in the gastrointestinal, genitourinary, and Antibody screens use reagents that recog-
respiratory tracts..Although 19A exists in either nize IgG, the primary isotype produced follow-
monomeric or. dimeric forms (Fig 8-9 shows a ing red-cell-induced alloimmunization. In con-
dimer), in serum it is often monomeric. IgA is trast, naturally occurring .antibodies against
further divided into IgAl and IgA2 subclasses carbohydrate blood group antigens, such as A
(not shown). Dimeric IgAis composed of mono- and B, are typically, but not exclusively, IgM.
mers connected by the J chain (the same J chain The pentameric structure of IgM allows it to di-
found in IgM). Only rarely is immunoglobulin- rectly agglutinate alloreactive A and B red cells
mediated hemolysis due to IgA. Antiglobulin and represents the underlying principle of the
(Coombs) reagents do not detect IgA, and the forward and reverse typing reaction. However, it
potential presence of IgA red cell antibodies should be noted that IgM directed against other
antigens may be present but not cause agglutina- ous FcyRs [Fig8-9 (B)].A mixture of IgGs may
tion, possibly due to lower target antigen levels, bind a particle or cell bearing a foreign antigen.
and therefore can be missed using current assay The net effect on phagocytosis depends on the
systems. Similarly, 19A antibodies, which have relative binding of different IgG subclasses and
been shown to cause autoimmune hemolytic their interactions with different FcyRs.Thus, di-
anemia," are not directly detected using stan- rect binding of Fc domains to FcyRs promotes
dard clinical assays. Some anti-IgGreagents also red cell clearance in many but not all cases.
failto detect all IgG subtypes, leading to the pos-
sibility that some IgG alloantibodies in certain Complement in Target Cell
patients may be missed." What drives distinct Opsonization and Destruction
alloantibody isotypes to blood group antigens re-
mains largely unknown. Red-cell-induced allo- Fc regions of IgG antibodies can also activate
antibody formation and class switching is complement. The complement system consists
thought to require CD4 T-cell help," although of a cascade of proteases that, once activated,
recent studies suggest that CD4 T cells may not amplifiesthe initial signal,leading to the produc-
be required for all red-cell-induced IgG alloanti- tion of a large number of effector molecules. Al-
bodies." Naturally occurring antibodies can oc- though there are several pathways to comple-
cur in the absence of any exogenous stimulus," ment activation, this discussion focuses on the
suggesting that environmental exposure and im- "classicalpathway" initiated by Fc regions.
munologic mechanisms distinct from RBC-trans- IgM is highly efficient in activating comple-
fusion-induced alloimmunization likely drive A ment. However, to avoid indiscriminate activa-
and B antibody formation and may be responsi- tion, IgM must undergo a conformational shift,
ble for the persistence of IgM antibodies against which occurs only after binding antigen, there-
these antigens. by exposing complement-binding sites in the
heavy-chain constant region. This interaction is
so potent that, in theory, a single antigen-bound
Antibody Receptors (FcyRs)in Target
IgM is sufficientto lyse a target.
Clearance
In contrast, complement activation by IgG
The Fc regions of antigen-bound IgGs are recog- does not involve a conformational change. Rath-
nized by the gamma family of Fc receptors er, complement activation requires clustered
(FcyRs)found on the surface of cells. At least binding to the same target by multiple IgGmole-
four FcyRshave been described, each with dif- cules. This ensures against indiscriminate com-
ferent properties that can have opposite func- plement activation by unbound circulating IgG.
tions. FcyR2a and FcyR3promote phagocytosis Once activated, the complement system initi-
of targets. Because of the relatively low affinity ates at least two distinct mechanisms of target
of these receptors, monomeric IgG does not en- destruction. The first involves decoration of the
gage FcyR2aor FcyR3;these receptors are pref- target by complement components labeling it for
erentially engaged when an antigenic target is destruction; this process is known as "opsoniza-
bound by multiple IgGs simultaneously. In con- tion." Early in activation, a portion of one com-
trast, FcyR2bis an inhibitory receptor that pre- plement component (termed "C3") covalently
vents phagocytosis.FcyRI has an unusually high attaches (by cleavage of a thioester bond, gener-
affinity for IgG and binds monomeric IgG such ating a highly labile carbonyl group that reacts
that FcyRI binds IgG whether or not it is com- with free amines or hydroxyl groups on the cell
plexed with a target; the function of this activity surface) to the surface of the antigen. Multiple
is currently unclear. copies of C3b can be recognized by specific re-
FcyR biology is complex, as 1) a given IgG- ceptors on phagocytic cells. Upon encountering
bound cell or particle may simultaneously acti- a C3b-coated molecule, a phagocyte ingests and
vate multiple (potentially antagonistic) recep- destroys it. However, C3b can also rapidly de-
tors, and 2) each IgG subclass (IgG1, IgG2, grade to C3dg, which is not recognized by
IgG3, IgG4) has a different affinity for the vari- phagocytes, thus bypassing phagocytosis. C3dg
is recognized by B-cell complement receptors complement activation is complete, MAC inser-

and therefore may affect alloantibody develop- tion causes red cell lysis.
ment. The relative contribution of each pathway
In the second mechanism, activation of C3 varies based on the relative amounts of antibody
promotes the assembly of the membrane attack isotype .and subclass, and on the properties of
complex (MAC). The MAC consists of comple- the antigen (eg, antigen density and!or linkage
ment proteins (C5b-C9) arranged into a struc- to cytoskeleton), which likely converge to dic-
ture resembling a hollow tube that is inserted tate the overall outcome. of an incompatible
into the membrane of the target cell. This cre- RBC transfusion.50-52 The sections below de-
ates a channel between the inside of a target cell scribe what is known about these processes
and its external environment, typicallyresulting withregard to red cell destruction and the clini-
cal manifestations of hemolysis.
in osmotic lysisof the target.
Extravascuh.'ll'Hemolysis
Outcomes of Complement Activation
Extravascular hemolysis refers to the consump-
The effector mechanisms induced by antibody
tion of antibody- and! or C3b-bound red cells by
binding have different effects on bacteria, virus- phagocytes in the reticuloendothelial system
es, particles, and various human tissues. In gen- (RES),found predominantly in the spleen and
eral, once an IgG antibody binds to a red cell, liver. The term "extravascular" is used because
the target cell may undergo FcyR-mediated the red cells are destroyed outside of their nor-
phagocytosis (Fig8-10). If the antibody initiates mal compartment, the intravascular space. In
the complement cascade, C3b.deposition on the contrast, "intravascular" hemolysis (see below)
red cell surface contributes to opsonization, occurs rapidly during or following transfusion
leading to phagocytosismediated by the C3b re- and is often associated with an acute hemolytic
ceptors CRl, CR3, CR4, and CRIg. Finally, if transfusion reaction (AHTR).Unlike AHTRs,de-
A. B. D.
lysis
FIGURE 8,.10. Mechanismsof red cell destruction by antibody binding. Uponbinding (A), an immune globulin G
(lgG) represents a ligand for Fc-gamma receptors (FcyRs) on phagocytes (B). If the red cell avoids FcyR-
mediatedphagocytosis, opsonization may be increasedby activation ot complement with deposition of C3b (C).
It the combined opsonization of FcyR binding and C3b is not sufficient to mediateclearance,completion of the
complement cascade may lead to insertion of the membrane attack complex (MAC) into the red cell surface,
resulting· in lysis (D). These processes likely occur simultaneously, with the outcome representing the
aggregateeffect of competing pathways.
CR = complement receptor.
layed hemolytic transfusion reactions (DHTRs) Intravascular Hemolysis

demonstrate delayed kinetics because the anti-
In some cases of incompatible transfusion, the
bodies implicated are often absent or of low titer
MAC rapidly assembles and lyses the red cells
initially, requiring some time to develop and in-
before C3b and/or IgG opsonization can induce
crease in titer before the onset of significant red
phagocytosis. Because these red cells lyse while
cell destruction. This results in manifestations of
still circulating, this process is termed "intravas-
delayed hemolysis, which is often, but not ex-
cular hemolysis." In addition, because antibody-
clusively, extravascular in nature.
Extravascular hemolysis is markedly different mediated intravascular hemolysis occurs at a
from intravascular hemolysis, in which red cell brisker pace than extravascular hemolysis, the
contents are directly released into the circulating clinical signs and symptoms may be more quick-
blood. The term "hemolysis" in this context can ly noticed.
cause confusion for health-care providers not ac- As discussed, AHTRs are typically caused by
customed to transfusion medicine terminology; IgM antibodies, which efficiently activate com-
hemolysis is often thought of as just described- plement, leading to rapid formation of the MAC.
the rupture of red cells within the circulation. In Although an IgM-specificFc receptor has been
contrast, in extravascular hemolysis, the red described [Pcuk), its role in promoting the clear-
cells are destroyed by phagocytes, typicallywith- ance of IgM-coated red cells in the setting of in-
in the lysosomes. This is a very important compatible RBCtransfusion is not known. Com-
distinction because phagocytes in the RES con- plement activation by IgM red cell antibodies
sume a substantial number of senescent, autolo- can result in opsonization by C3b, leading to
gous red cells each day in the normal process of some receptor-mediated phagocytosis. Overall,
red cell turnover. Thus, consumption of red cells however, IgM antibodies are thought to pre-
by this pathway follows a process that evolved dominantly induce intravascular hemolysis.
specifically to break down and recycle red cell Intravascular hemolysis, unlike the removal
contents (hemoglobin and iron) in a manner of senescent red cells through extravascular
that avoids tissue damage. mechanisms, does not occur at any appreciable
This does not mean, however, that extravas- level under normal conditions. The release of
cular removal of antibody-coated red cells is bio- red cell contents directly into the circulation can
logically equivalent to clearance of normal, se- be highly toxic, with free hemoglobin inducing
nescent red cells. On the contrary, DHTRs can perhaps the greatest insult. Although much free
cause substantial morbidity (and occasional mor- hemoglobin is scavenged by the circulating mol-
tality). ecule haptoglobin, this system can be easily
It is not clear why some red cell antibodies overwhelmed. AHTRsoften result in tea-colored
preferentially promote opsonization and phago- urine (hemoglobinuria) and can induce renal
cytosis instead of osmotic lysisby the MAC. The dysfunction, which likely reflects heme-induced
antibody type and the topology and copy num- cell injury that results from a confluence of com-
ber of the target antigen on the red cells are im- promised mitochondrial function, altered oxida-
portant. In addition, although complement may tive stress, perturbed metabolism, and increased
be activated, the aggregate opsonization of red inflammation. 53 Moreover, the signs and symp-
cells by C3b and antibody may result in phago- toms of AHTRscan be dramatic and include dis-
cytosis before MAC-induced lysis occurs. Con- seminated intravascular coagulation, shock, and
sistent with these explanations, extravascular death. This type of reaction most often occurs as
hemolysis is typicallyinduced by IgGred cell an- a result of a clericalerror with ABO-incompatible
tibodies, whereas intravascular hemolysisis typi- transfusion. To prevent such an occurrence,
cally induced by IgM red cell antibodies. IgM transfusion service practices have evolved multi-
can be much more efficient at activating comple redundant checkpoints to prevent AHTRs
plement and promoting MAC formation. from ABOincompatibility.
Hyperhemolysis and Antibody- are bound to their own red cells), yet there is no
Negative DHTRs evidence of hemolysis. For antigens known to be
frequently responsible for antibody-mediated he-
Occa~ionally, patients who ~xperience AHTRs molysis (in the RH, KEL,JK, FY, and MNS sys-
or DHTRs may also hemolyze th~irown (non-
tems), hemolysis is variable. Indeed, in patients
transfused) red cells, a process often referred to
who mistakenly received ABO-incompatible
as hyperhemolysis. Although hyperhemolysis
RBC units, clinically Significant hemolysis oc-
has beenrec0!Wiz~d, ..in.pa~ticular, in patients curs in only 50% of cases even for this robustly
with SCD, it is s0l11e~111esobserved lnother hemolytic antigenl antibody combination.
clinical settings. As hyperhem()lysis restllts in
Several explanations may account for the
the loss of the patient's own ted cells, some-
lack of hemolysis during incompatible transfu-
times with no detectable autoantibody, it may
sions. For antigens rarely implicated in hemoly-
be antibody-independent. Several studies indi-
sis, antigen cell surface density or topography
cate a role for complement; case series suggest
may not be conducive to initiating the hemolyt-
that patients experiencing hyperhemolysis may
ic pathway. For antigens variably involved in he-
benefit from eculizumab, an antibody that inhib-
molysis, the antibody response (thermal range,
its complement activation beyond C5.54,55
titer, affinity, isotype, or IgG subclass) may play
In addition to hyperhemolysis, some patients
roles. Distinct antibodies of the same antigenic
experience accelerated clearance of transfused
specificity may have different capacities for acti-
red cells, but wit1:10utany detectable alloanti-
vating complement. This is the rationale for in-
body, a process referred to asantibody-negatlve
cluding the anti-C3 component in the antiglobu-
DHTR (AN-DHTR).AN-DHTRscan b~ fatal and
lin (Coombs) reagent: it provides information
can be accompanied by hyperhemolysis, but the
on whether an antibody can activate (fix) com-
underlying mechanism(s) remain unknown.
plement. Recent studies also suggest that en-
Awareness of this phenomenon is critical, as tra-
gagement of some antigens by antibodies may
ditional approaches designed to identify red cell
result in the selective loss of the antigen from
incompatibility focus entirely 011the presence of
the cell surface, thereby rendering the cells inert
an alloantibody, and therefore fail to directly de-
to additional alloantibody-mediated hemoly-
tect AN-DHTR. Monitoring of hemoglobin A sis.59,6o
values immediately following transfusion and at
IgG blood group antibodies are typically poly-
can
thetimeof suspected AN-DHTR be especially
clonal responses, and a number of early studies
helpful when evaluating such caSeS.56Given the
demonstrated that clinically significant red cell
mortality rate associated with these reactions, a
IgG responses are primarily IgG1 and IgG3 iso-
greater understanding of the underlying biology
is needed in order to discoverand implement ef- types, with IgG2 and IgG4 isotypes much less
fective treatment options for AN-DHTRs,with or frequently observed61.63;indeed, some clinically
without accompanying hyperhemolysts.V'" insignificant blood group antibodies (eg, anti-
bodies to CH/RG antigens) show a converse re-
lationship, exhibiting a higher frequency of
Nonhemolytic Red Cell Jo',l1tibodies
IgG4.61In this regard, it is notable that one anti-
Given the redundant pathways leading to the human globulin MoAb in common use in pre-
destql<,:tipIl gf antipoclUo,?-teci red cells, .it transfusion testing fails to detect IgG4 isotypes,
makes sense that transfusions of. crossmatch- as well as some IgG3 allelic variants present in
incompatible red cells can produce hemolysis. specific populations." The studies establishing
Curiously, many red cell antibodies are actually that clinically significant antibodies are often
not hemolytic. For some blood group antigens, IgG1 and IgG3, and clinically insignificant anti-
hemolysis is only very rarely observed following bodies are IgG2 and IgG4, are now decades old.
incompatible transfusion (eg, JMH or CH/RG Using new exquisitely specific reagents could
antigens). Moreover, approximately 1% of healthy potentially establish, for each blood group anti-
blood donors have positive direct antiglobulin gen, the relationship between IgG isotype and
test results (indicating that IgG autoantibodies clinical significance.
Different genetic polymorphisms and/or defi- considered. Close and frequent communication
ciencies may regulate hemolysis vs red cell between a given patient's clinical care team and
clearance on a patient-by-patient basis, including the transfusion medicine physician should take
perturbations in complement, complement- place, so as to determine optimal approaches to
regulatory proteins, and allelic polymorphisms managing patients in these unfortunate circum-
in FcyRs.Furthermore, the status of the reticulo- stances.
endothelial system of a given patient may dic-
tate the extent to which antibody-coated cells New Approaches to Understanding the
are efficiently removed. The monocyte mono- Mechanistic Basis for Alloantibody
layer assay (MMA) provides the only available Development
method aimed at determining the potential clin-
ical significance of a given red cell alloantibody. Although research characterizing the structural
Although useful, the MMA does not address im- basis of blood group antigens and the regulation
portant additional recipient factors, such as al- of their expression has been extensive, the fac-
terations in complement within a potential re- tors responsible for alloantibody formation fol-
cipient. Recent data suggest that some cases of lowing red cell exposure have remained ob-
immune-mediated platelet refractoriness may scure. In solid organ transplantation, a plethora
occur independent of alloantibodies, facilitated of animal models described, defined, and re-
instead by CD8 T cells," and that regulation of fined the roles of major histocompatibility anti-
clearance may have alloantibody-independent gens and immune mechanisms in transplant re-
mechanisms. jection; by contrast, analogous animal models
It is fair to state that there is only a rudimenta- for transfusion-induced alloimmunization were,
ry understanding of the underlying basis for the for many years, not available.
development and manifestation of immune red Recent advances in the development of rele-
cell destruction." Because the testing approach vant animal models are providing important in-
for the presence of alloantibodiesis not an activi- sights into factors responsible for the develop-
ty assay, the historical significanceof a particular ment of alloantibodies in the setting of blood
alloantibodyis used as an indicator of its clinical transfusion. Studies using these systems, cou-
relevance. From a practical standpoint, cross- pled with compelling corollary studies in pa-
match-incompatible RBCunits may be issued for tients, are beginning to clarify the mechanisms
transfusion if the offendingentity is in the catego- responsible for this fundamental process in
ry of a "clinicallyinsignificantantibody," especial- transfusion medicine.47,66·71 Historically, a given
ly if the antigen is of very high frequency and an- blood group antigen becomes defined as clinical-
tigen-negative blood is difficult or impossible to ly relevant only after a patient generates an im-
obtain. The transfusion service must be prepared mune response to it and suffers a hemolytic
to address appropriate concerns from health-care event. Understanding the process of red cell al-
providers managing these patients. Although an loimmunization using these newly available re-
antibody deemed "clinicallysignificant"may not search tools opens up possibilities for the pre-
actually produce hernolysis in all patients, there vention of alloantibody development.
is no practical method of predicting hemolysis in
a given recipient in the acute setting. RBCtrans- Summary of Immunologic Responses
fusion strategies can include units that are anti- to Red Cells
gen-matched to the patient for common clinically
significant blood group antigens. RBCunits that In aggregate, when an antibody binds a red cell,
are crossmatch incompatible should not be issued multiple pathways are activated that can lead to
for clinically significant antibodies, except as a cell destruction. Complement activation pro-
lifesaving intervention. If compatible RBC units motes phagocytosis through the opsonizing
are unavailable, the potential for hemolysis may properties of C3b and through direct red cell ly-
be less deleterious than the patient's severe ane- sis via assembly of the MAC. The presence of
mia; in such cases, immunosuppression can be IgG Fc domains promotes red cell phagocytosis
by binding FcyRson the phagocyte surface. The beyond simple loss of efficacy of the transfused
relative contributions of these different path- red cells. Indeed, substantial toxicity can occur,
ways vary depending on the nature of the target leading to morbidity and, in some cases, mortali-
antigen and the properties of the cognate anti- ty. The reader is referred to a more in-depth ref-
bodies. The activation of immune pathways and erence if additional mechanistic details on im-
the toxicity associated with red cell hemolysis munobiology and the immune response are
can lead to negative clinical consequences far desired."
REFERENCES
1. Alberts B, Johnson A, Lewis J, et al, Molecular 3. Al-SoudWA, Radstrom P. Purificationand char-

biology of the cell. 6th ed. New York: Garland acterization of PCR-inhibitory components in
Science, 2014. blood cells.J Clin Microbio12001;39(2):485-93.
2. Mullis KB, Faloona FA. Specific synthesis of 4. RychlikW. Selection of primers for polymerase
DNA in vitro via a polymerase-catalyzed chain chain reaction. Mol Biotechnol 1995;3(2): 129-
reaction. Methods Enzymo11987;155:335-50. 34.
5. Wagner FF, Flegel WA, Bittner R, Doscher A. partial and weak D types with different serologlc
Molecular typing for blood group antigens with- techniques. Transfusion 2008;48(3):473-8.
in 40 min by direct polymerase chain reaction 19. Sandler SG, Flegel WA, Westhoff CM, et al. It's
from plasma or serum. Br J Haematol 2017; time to phase in RHD genotyping for patients
176(5):814-21. with a serologic weak D phenotype. College of
6. Pang J, Modlin J, Yolken R. Use of modified nu- American PathologlstsTransfusion Medicine Re-
cleotides and uracil-DNA glycosylase (UNG) for source Committee Work Group. Transfusion
the control of contamination in the PCR-based 2015;55(3):680-9.
amplification of RNA. Mol Cell Probes 1992; 20. Baumgarten R, van Gelder W, van Winter-
6(3):251-6. shoven J, et al, Recurrent acute hemolytic trans-
7. Compton J. Nucleic acid sequence-based ampli- fusion reactions by antibodies against Doa
fication. Nature 1991;350(6313):91-2. antigens, not detected by cross-matching. Trans-
8. Kwoh DY, Davis GR, Whitfield KM, et al, Tran- fusion 2006;46(2):244-9.
scription-based amplification system and detec- 21. Fasano RM, Sullivan HC, Bray RA, et al. Geno-
tion of amplified human immunodeficiency typing applications for transplantation and trans-
virus type 1 with a bead-based sandwich hybrid- fusion management: The Emory experience.
ization format. Proc Natl Acad Sci USA 1989; Arch Pathol LabMed 2017;141(3):329-40.
86(4):1173-7. 22. Svensson AM, Delaney M. Considerations of
9. Elkins MB, Davenport RD, O'Malley BA, Bluth red blood cell molecular testing in transfusion
MH. Molecular pathology in transfusion medi- medicine. Expert Rev Mol Diagn 2015;15(11):
cine. Clin Lab Med 2013;33(4):805-16. 1455-64.
23. Wheeler MM, Johnsen JM. The role of genom-
10. Denomme GA, Van Oene M. High-throughput
ics in transfusion medicine. Curr Opin Hematol
multiplex single-nucleotide polymorphism anal-
2018;25(6):509-15.
ysis for red cell and platelet antigen genotypes.
24. Chou ST, Westhoff CM. The role of molecular
Transfusion 2005;45(5):660-6.
immunohematology in sickle cell disease. Trans-
11. Bugert P, McBride S, Smith G, et al. Microarray-
fus Apher Sci 2011 ;44(1):73.
based genotyping for blood groups: Comparison
25. Wilkinson K, Harris S, Gaur P, et al. Molecular
of gene array and Sf-nuclease assay techniques blood typing augments serologlc testing and al-
with human platelet antigen as a model. Trans- lows for enhanced matching of red blood cells
fusion 2005;45(5):654. for transfusion in patients with sickle cell dis-
12. Hashmi G, ShariffT, Seul M, et al. A flexible ar- ease. Transfusion 2012;52(2):381-8.
ray format for large-scale, rapid blood group 26. KlapperE, Zhang Y,Figueroa P, et al. Toward ex-
DNA typing. Transfusion 2005;45(5):680. tended phenotype matching: A new operationaI
13. van der Schoot CE, de Haas M, Engelfriet CP, et paradigm for the transfusion service. Transfu-
al. Genotyping for red blood cell polymer- sion 2010;50(3):536-46.
phisms. Vox Sang 2009;96(2):167-79. 27. Chou ST, Jackson T, Vege S, et al. High preva-
14. Flegel WA, Castilho L, Delaney M, et aI. Molec- lence of red blood cell alloimmunization in sick-
ular immunohaemato1ogy round table discus- le cell disease despite transfusion from Rh-
sions at the AABB Annual Meeting, Denver matched minority donors. Blood 2013;122(6):
2013. Blood Transfus 2015;13(3):514-20. 1062-71.
15. FlegelWA, Johnson ST, Keller MA, et al, Molec- 28. Metzker ML. Sequencing technologles - the
ular immunohaematology round table discus- next generation. Nat Rev Genet 2010; 11(1):31-
sions at the AABB Annual Meeting, Boston 46.
2012. Blood Transfus 2014;12(2):280-6. 29. Lane WJ, Westhoff CM, Uy JM, et al. Compre-
16. FlegelWA, Chen Q, Castllho L, et al. Molecular hensive red blood cell and platelet antigen pre-
immunohaematology round table discussions at diction from whole genome sequencing: Proof
the AABB Annual Meeting, Orlando 2016. of principle. Transfusion 2016;56(3):743-54.
BloodTransfus 2018;16(5):447-56. 30. Lane WJ, Westhoff CM, GleadaIl NS, et al, Au-
17. Denomme GA. Molecular basis of blood group tomated typing of red blood cell and platelet an-
expression. Transfus Apher Sci 2011 ;44( 1):53- tigens: A whole-genome sequencing study. Lan-
63. cet HaematoI2018;5(6):e241-e51.
18. Denomme GA, Dake LR, Vilensky D, et al, Rh 31. Fichou Y,Audrezet MP, Gueguen P, et al. Next-
discrepancies caused by variable reactivity of generation sequencing is a credible strategy for
blood group genotyping. Br J Haematol 2014; for al1oimmune-refractory patients. Transfusion

167(4):554-62. 2015;55(2):235-44.
32. McBean RS,Hyland CA, Flower RL.Approaches 45. Chadebech P, Michel M, Janvier D, et al. IgA-
to determination of a full profile of blood group mediated human autoimmune hemolytic ane-
genotypes: Single nucleotide variant mapping mia as a result of hemagglutination in the
and massively parallel sequencing. Comput spleen, but independent of complement activa-
Struct Biotechno1]2014;11 (19):147-51. tion and FcalphaRI. Blood 2010; 116(20):4141-
33. Okuda H, Suganuma H, Kamesaki T, et al. The 7.
analysis of nucleotide substitutions, gaps, and 46. Howie HL, Delaney M, Wang X, et al. Serologi-
recombination events between RHD and RHCE cal blind spots for variants of human IgG3 and
genes through complete sequencing. Biochem IgG4 by a commonly used anti-immunoglobulin
Biophys Res Commun 2000;274(3):670·83. reagent. Transfusion 20 16;56( 12):2953-62.
34. Wheeler MM, Lannert KW, Huston H, et al. Ge- 47. Calabro S, Gallman A, Gowthaman U, et al.
nomic characterization of the RH locus detects Bridgingchannel dendritic cells induce immuni-
complex and novel structural variation in multi- ty to transfused red blood cells. J Exp Med
ethnic cohorts. Genet Med 2019;21 :477·86. 2016;213(6):887-96.
35. Chou ST, Evans P, Vege S, et al. RH genotype 48. Mener A, Patel SR, Arthur CM, et al. Comple-
matching for transfusion support in sickle cell ment serves as a switch between CD4+ T cell-
disease. Blood 2018;132(11): 1198-207. independent and -dependent RBC antibody re-
36. Harmening DM, Walker PS. Alternative tech- sponses. JCI Insight 2018;3(22).
nologies and automation in routine blood bank 49. Haury M, Sundblad A, Grandien A, et al. The
testing. In: Harmening DM, ed. Modern blood repertoire of serum IgM in normal mice is large-
banking and transfusion practices. 5th ed. Phila- ly independent of external antigenic contact.
delphia: FADavis, 2005:293-302. Eur IImmunoI1997;27(6):1557-63.
37. Judd WJ, Steiner EA, O'Donnell DB, Oberman 50. Stowell SR,Winkler AM, Maler CL, et al. Initia-
HA. Discrepancies in reverse ABO typing due to tion and regulation of complement during he-
prozone. How safe isthe tmmediate-sptn cross- molytic transfusion reactions. Clin Dev Immu-
match? Transfusion 1988;28(4):334-8. nol20 12;20 12:307093.
38. Salama A, Mueller-Eckhardt C. Elimination of 51. LiepkalnsJS, Hod EA, Stowell SR, et al. Biphasic
the prozone effect in the antiglobulin reaction clearance of incompatible red blood cells
by a Simple modification. Vox Sang 1982; through a novel mechanism requiring neither
42(3):157-9. complement nor Fcgamma receptors in a mu-
39. Procter JL,Vigue F, Alegre E, et al. Rapid screen- rine model. Transfusion 2012;52( 12):2631-45.
ing of platelet donors for PIAl (HPA-1a) alloanti- 52. Girard-Pierce KR, Stowell SR, Smith NH, et al.
gen using a solid-phase microplate immunoas- A novel role for C3 in antibody-induced red
say. Immunohematology 1998;14(4):141-5. blood cell clearance and antigen modulation.
40. LeikolaJ, Perkins HA. Enzyme-linked antiglobu- Blood 2013; 122(10):1793-801.
lin test: An accurate and simple method to 53. Tracz MJ, Alam J, Nath KA. Physiology and
quantify red cell antibodies. Transfusion 1980; pathophysiology of heme: Implications for kid-
20(2): 138-44. ney disease. J Am Soc NephroI2007;18(2):414-
41. Arndt PA, Garratty G. A critical review of pub- 20.
lished methods for analysis of red cell antigen- 54. Dumas G, Habibi A, Onimus T, et al. Eculizum-
antibody reactions by flow cytometry, and ap- ab salvage therapy for delayed hemolysis trans-
proaches for resolving problems with red cell ag- fusion reaction in sickle cell disease patients.
glutination. Transfus Med Rev 2010;24(3):172- Blood 2016;127(8):1062-4.
94. 55. Chonat S, Quarmyne MO, Bennett CM, et al.
42. Dziegiel MH, Nielsen LK, Berkowicz A. Detect- Contribution of alternative complement path-
ing fetomaternal hemorrhage by flow cytometry. way to delayed hemolytic transfusion reaction
Curr Opin Hematol2006; 13(6):490-5. in sickle cell disease. Haematologica 2018;
43. Nolan JP, Sklar LA. Suspension array technolo- 103:e483-5.
gy: Evolution of the flat-array paradigm. Trends 56. Mekontso Dessap A, Pirenne F, Razazi K, et al.
BiotechnoI2002;20(1 ):9-12. A diagnostic nomogram for delayed hemolytic
44. Kopko PM, Warner P, Kresie L, Pancoska C. transfusion reaction in sickle cell disease. Am J
Methods for the selection of platelet products HematoI2016;91(12):1181-4.
57. HabibiA, Mekontso-DessapA, Guillaud C, et al. 65. Flegel WA. Pathogenesis and mechanisms of
Delayed hemolytic transfusion reaction in aduit antibody-mediated hemolysis. Transfusion
sickle-celldisease: Presentations, outcomes, and 2015;55(SuppI2):S47-58.
treatments of 99 referral center episodes. Am J 66. Yazdanbakhsh K, Ware RE, Noizat-Pirenne F.
Hemato12016;91(10):989-94. Red blood cell alloimmunization in sickle cell
58. VidlerJB, Gardner K, Amenyah K, et al. Delayed disease: Pathophysiology,risk factors, and trans-
haemolytic transfusion reaction in adults with fusion management. Blood 2012;120(3):528-
sickle cell disease: A 5-year experience. Br J 37.
Haemato12015;169(5):746-53. 67. Stowell SR,Henry KL,Smith NH, et al. Alloanti-
59. Stowell SR, LiepkalnsJS, Hendrickson JE, et al. bodies to a paternally derived RBCKELantigen
Antigen modulation confers protection to red lead to hemolytic disease of the fetus/newborn
blood cells from antibody through Fcgammare- in a murine model. Blood 2013;122(8):1494-
ceptor ligation. J Immuno12013;191(10):5013- 504.
25. 68. Patel SR, Cadwell CM, Medford A, ZimringJC.
60. Sullivan HC, Gerner-Smidt C, Nooka AK, et al.
Transfusion of minor histocompatibilityantigen-
Daratumumab (anti-CD38) induces loss of
mismatched platelets induces rejection of bone
CD38 on red blood cells. Blood 2017;129(22):
3033-7. marrow transplants in mice. J Clin Invest 2009;
61. Devey ME, Voak D. A critical study of the IgG 119(9):2787-94.
subclasses of Rh anti-D antibodies formed in 69. BaoW, YuJ, Heck S,YazdanbakhshK.Regulato-
pregnancy and in immunized volunteers. Immu- ry T-cell status in red cell alloimmunized re-
nology 1974;27(6):1073-9. sponder and nonresponder mice. Blood 2009;
62. Szymanskl 10, Huff SR, Delsignore R. An au- 113(22):5624-7.
toanalyzer test to determine immunoglobulin 70. Evers D, van der Born JG, Tijmensen J, et al.
class and IgG subclass of blood group antibod- Absence of the spleen and the occurrence ofpri-
ies. Transfusion 1982;22(2):90-5. mary red cell alloimmunizationin humans. Hae-
63. Michaelsen TE, Kornstad L. IgG subclass distri- matologica 2017; 102(8):e289-e92.
bution of anti-Rh, anti-Kell and anti-Duffyanti- 71. Elayeb R, Tamagne M, Pinheiro M, et al. Anti-
bodies measured by sensitivehaemagglutination CD20 antibody prevents red blood cell alloim-
assays.Clin Exp Immunol1987;67(3):637-45. munization in a mouse model. J Immuno12017;
64. Arthur CM, Patel SR,Sullivan HC, et al. CD8+ 199(11):3771-80.
T cells mediate antibody-independent platelet 72. Murphy K, Weaver C. Janeway's irnmunobiolo-
clearance in mice. Blood2016; 127(14):1823-7. gy. 9th ed. New York:Garland Science, 2016.
Blood Group Genetics
Margaret A. Keller, PhD, and Sandy Wortman, MT(ASCpfMSBB
HE SCIENCE OF GENETICS IS THE vide safe blood transfusion. Therefore, an under"

study of heredlty=-that is, the mecha- standing of the principles of human genetics (in"
nisms by which particular characteristics eluding the patterns of inheritance and the
are passed from parents to offspring.1 A ge- language or terminology in use) is an important
nome refers to all of the genetic information of aspect of immunohematology and transfusion
a cell, and genomics is the study of the entire medicine. This chapter outlines the fundamen-
genome of an organism.' This chapter describes tal principles of genetics as they apply to blood
the genetics of blood groups. The term "blood group antigens and relates them to examples rel-
group" can be applied to any detectable, vari- evant to transfusion medicine. Thisrequires the
able characteristic of a component of the blood. use of numerous genetic terms; each term,
But in this chapter, the term applies primarily to when first used or when fully described, will be
antigens on the surface of the red cell mem- in bold type and usually is closely followed by a
brane that are defined serologically by an anti" definition.
body. Platelet and white cell antigens are dis" Molecular genetics is the study of genes at
cussed in Chapter 15. the nucleic acid level for a variety of clinical and
That blood groups are inherited characteris- research purposes. 1,6 Molecular genetic research
tics was first shown by von Dungern and has provided an understanding of the genes and
Hirszfeld in 1910, 10 years after Landsteiner's their.associated regulatory elements that control
discovery of the ABO blood group.3 Blood the expression of blood groups. With the links
groups became an ideal tool for geneticists be" between genetic variation and antigen status
cause they could be identified by specific anti" came the ability to predict blood group types us"
bodies in. simple hemagglutination tests and, ing DNA-basedmolecular methods," The use of
once identified, their inheritance could easily be molecular immunohematologic testing is be"
followed in family studies. Red cell antigens coming commonplace in clinical scenarios that
were (and still are) valuable as markers (detect" challenge transfusion mediciIl€ practitioners.
able characteristics to recognize a gene's pres, Whereas transfusion medicine has been "per"
ence and allelic forms) in genetic and anthropo- sonalized" since the advent of compatibility test"
logic studies." ing, the integration of genomics into the field is
Differences in antigen expression on the red elevating transfusion medicine to the status of
cells from different people can be used to pro" genome-informed precision medicine.
Margaret A. Keller, PhD, Senior Director, National Molecular and Genomics Laboratories, American Red Cross,
and Adjunct Associate Professor, Thomas Jefferson University, Philadelphia, Pennsylvania; and Sandy Wortman,
MT(ASCP)CMSBB, Reference and Transfusion Laboratory Services Director, Carter BloodCare, Bedford, Texas
255
GEN ANllATiON carry equivalent genes) and are referred to as

AND G UL ior~ the autosomes (any chromosome that is not a
sex chromosome). The remaining pair is nonho-
Many excellent texts offer greater insight into mologous and consists of the sex chromo-
classical genetics" and genomics.' The funda- somes; they determine an individual's gender.
mental principles of genetics and genomics out- Males carry X and Y chromosomes, whereas fe-
lined in this chapter are intended to serve as a males carry two X chromosomes. The karyo-
review of the inheritance and expression of type represents the chromosome complement
blood group antigens and for background on the of a person; this is written as "46,XY" and
mechanism of molecular methods that are com- "46,XX" for a normal male and female, respec-
monly used to predict antigen phenotypes. tively. The hereditary information carried by the
chromosomes is passed from a parent cell to a
Genes
daughter cell during somatic cell division (mito-
sis; see "Mitosis" section below), and from par-
A gene is a functional unit of heredity. It is a ents to offspring (children) by the gametes or
segment of deoxyribonucleic acid (DNA)within germ cells produced by meiosis during repro-
a chromosome that codes for a molecule that duction (see "Meiosis" section below).
has a function. Many but not all genes encode Cytogenetics is the study of chromosomes.
polypeptide chains. A gene is the basic unit of Chromosome structures can be studied during
inheritance of any trait (defined as a genetically mitosis when they can be visualized by various
determined characteristic or condition), includ- staining and imaging techniques. Each chromo-
ing blood group antigens, that is passed from some has two sections, or arms, that are joined
parents to offspring. A locus is a fixed position at a central constricted region called the cen-
on a chromosome, such as a gene or genetic tromere (Fig9-1). All chromosomes have some
marker. A locus may be occupied by one of sev- common morphologic features but differ in oth-
eral alternative forms of the gene, called alleles. er characteristics, including size, location of the
For example, the gene that encodes the protein centromere, DNA content, and staining proper-
carrying the Jk' antigen is an alternative form ties.
(allele) of the one that encodes the Jkb antigen. Staining techniques provide a means to dis-
The terms "gene" and "allele" can be used inter- tinguish individual chromosomes. Selected dyes
changeably. JK*O 1 and JK*02 are the allele do not stain chromosomes uniformly, and the
names of the SLCl4AI gene encoding the poly- different banding patterns can be used to distin-
morphic common antigens Jka and Jkb, respec- guish each human chromosome and detect
tively. (See the section on terminology for more gross rearrangements. Giemsa stains well the
information.) A group of alleles that tend to be chromosomal regions that are rich in adenine
inherited together are referred to as a haplo- (A) and thymine (T), also known as heteroch-
type. RHD and RHCE alleles in the RHlocus are romatic regions. Euchromatic regions that
an example of a haplotype. are rich in guanine (G) and cytosine (C) stain
poorly. This results in a chromosomal banding
Chromosomes
pattern that can be used to identify and charac-
terize chromosomes. The so-called G-bands are
Chromosomes contain the genetic material numbered from the centromere outward. Using
(DNA) necessary to maintain the life of the cell chromosome 1 as an example, the region closest
and the organism. A human somatic cell (any to the centromere on the short or long arm is
nonreproductive cell) contains 46 chromosomes numbered lp l or lql, respectively. With great-
made up of 23 pairs; each pair has one paternal- er resolution, there is further distinction into
ly- and one maternally-derived chromosome. In subbands (eg, lpl l and Ip12) that, in turn, can
humans, 22 of the pairs are homologous chro- be subdivided (eg, lp l Ll and Ipl1.2). Genes
mosomes (a pair of chromosomes in which pa- can be individually mapped to a specific band
ternally- and maternally-derived chromosomes location (Fig 9-2). Other stains can be used to
C HAP T E R 9 Blood Group Genetics 257
Base Pairs
Chromosome
Chromatid Chromatid
FIGURE 9-1. Chromosomes are made up of one linear double-stranded DNA.rnplecule, packaged in an
organized fashion. Eachchromosome has a centromere, or constricted region, with the regions on either side
referred to as "arms": the short arm, or p arm, on the top, and the long arm, or q arm, on the bottom. The
chromosomes differ in their overall size,the location of the centromere, and their DNAcontent. [Imagecourtesy
of National Human GenomeResearchInstitute (genome.gov).J
study chromosomes; fluorescent in-situ hybrid- Mitosis

ization (FISH)uses fluorescent dyes to identify Somatic cells divide for growth and repair by
specific regions and detect gross structural ab- mitosis (Fig9-3). Through this process, a single
normalities and trisomies. The chromosomallo- cell gives rise to two daughter cells with identi-
cations of the genes encoding the 39 red cell cal sets of chromosomes. The daughter cells,
systemsr" are listed in Table 9-1, and Fig 9-2 il- like the parent cell, are diploid (2N); that is,
lustrates some of these. they contain 46 chromosomes in 23 pairs and
have all the genetic information of the parent
Cell Division cell.
As a cell divides, the chromosomes replicate,
Meiosis
and each daughter cell receives a full comple-
ment of genetic material. In somatic cells, this Meiosis occurs only in germ cells that are in-
tended to become gametes (sperm and egg
occurs through mitosis; in reproductive cells, a
cells). Somatic cells are diploid (2N), whereas
similar process called meiosis takes place. A
gametes are haploid [having half the chromo-
feature common to both types of cell division is somal complement of somatic cells (1N)].Meio-
that, before the start of the process, each chro- sis is a process of cell division and replication
mosome replicates to form two identical daugh- that leads to the formation of haploid gametes.
ter chromatids attached to each other through During meiosis, diploid cells undergo DNA rep-
the centromere (Fig9·1). lication, followed by two cycles of cell division
tel" change of genetic material between homologous

p36.3
chromosome pairs. Such shufflingof genetic ma-
p3S.2 terial ensures diversity and produces genetically
p36.1
unique gametes that fuse to produce a unique
pf4~~
p34.2 zygote.
i>3p~1
p32.3
Short ~~H
p3l.3 X Chromosome Inactivation
arm p3l.2
p3l.1
(Lyonization)
(p) p22.3
p22.2 Based on the inheritance of the sex chromo-
p22.l
p21
somes, females have two copies and males have
p13.3
only one copy of genes carried by the X chromo-
p13.2
p13.1 some. Because most genes carried by the
~U Centromere X chromosome do not have a homolog on the
q12 Y chromosome, there is a potential imbalance in
q21.1
q21.2
the dosage of the gene products between males
q2l.3
q22 and females. Dosage compensation involves
q23
q24
X chromosome inactivation (also called ly-
Long q25 onization), a process through which most of the
arm q3l
genes on one of the two Xchromosomes in each
(q) female somatic cell are inactivated at a very ear-
q32.1
KN ly stage of embryonic development." It is a mat-
q32.2
q33.3
:JCROM ter of chance whether the maternal or paternal
q41
q42.1
q42.2
X chromosome is inactivated in anyone cell,
q42.3
q43 but once inactivation has occurred, all descen-
q44
dants of that cell will have the same inactive
tel" X chromosome. Some genes carried by the
FIGURE 9-2. The morphology and banding pattern of X chromosome escape inactivation; the first
a Giemsa-stained human chromosome 1. The loca- gene found to escape inactivation was XC, the
tions of the genes controlling the expression of gene encoding the antigens of the XG blood
antigens for the RH, SC, FY, KN, and CROM blood group system. Like XC, most of the genes that
group systems are shown. (Table 9-1 lists the ISBT escape inactivation are located on the extreme
symbols for the blood groups.) tip of the short arm of the X chromosome, but
several are clustered in regions on the short and
long arms of the chromosome.24(pp359·370),25
The XK gene, which encodes the XK blood
to produce four haploid gametes (Fig 9-4). group system, is the only other gene carried by
Sperm and egg cells fuse at fertilization such the X chromosome known to encode a red cell
that the gametes, each carrying a haploid (IN) antigen. Changes or deletions in XK result in
set of chromosomes come together to form a McLeod phenotype red cells that lack Kx anti-
zygote with 46 chromosomes. Eggsfertilized by gen and have reduced expression of KEL anti-
Xbearing sperm become females (XX) while gens.26,27 The XKgene, unlike XC, is subject to X
those fertilizedby Ybearing sperm become males chromosome inactivation, with the result that a
(XY). female who is a carrier (a person who carries
Meiosis ensures genetic diversity through one gene for a recessive trait and one normal
two mechanisms: independent assortment gene) of a gene that is responsible for the Me-
and crossing over. Through independent as- Leod phenotype can have a dual population of
sortment, each daughter cell randomly receives Kx- (McLeod phenotype) and Kx- (non-Me-
either maternally or paternally derived homolo- Leod phenotype) red cells. Flow cytometry, us-
gous chromosomes. Crossing over involves ex- ing selected KELantibodies, shows the weaken-
TABLE 9-1. Blood Group Systems
ABO ABO 9q34.2 Glycosyltransferase, A; B; A,B; Ai

(001 ) (ABO) carbohydrate [Group 0)
MNS MNS 4q31.21 Glycophorin A M, N, S, s, U, He, Mia,

(002) (GYPA (GPA) [CD235a) Vw, Me, and 40 more
[En(a-); U-; MkMk)
GYPB) Glycophorin B
(GPB) [CD235b)
P1PK Pl 22q13.2 Galactosyltransferase, Pi, pk, NOR

(003) (A4GALT) carbohydrate
RH I Rh RH 1p36.1 i RhO [CD240D) D,G

(004) (RHO RhCE [CD240CE) C, E, c, e, V, VS,
RHCE) and 47 more
[Rhnull)
LU I Lutheran LV 19q13.32 Lutheran glycoprotein, Lua, Lub, Lu3, Lu4,

(005) (BCAM) B-cell adhesion mole- Aua, Aub, and 19 more
cule [CD239) [Recessive Lu(a-b-))
KELI Kell KEL 7q34 Kell glycoprotein K, k, Kpa, Kpb, Ku,

(006) (KEL) [CD238) K11, Jsa, Js", and 28
more
[Ko or KnuU)
LE I Lewis LE i9pi3.3 Fucosyltransferase, Lea, Le", Leab,Lebh,

(007) (FUT3) carbohydrate Aleb, Bleb
(adsorbed from plasma) [Le(a-b-))
FY I Duffy FY 1q23.2 FY glycoprotein [CD234) Fy", Fyb, Fy3, Fy5, Fy6

(008) (ACKR1) [Fy(a-b-))
JK I Kidd JK(SLC14Al) 18q12.3 Human urea transporter Jka, Jkb, Jk3

(009) (HUT), Kidd glycoprotein [Jk(a-b-))
01 I Diego DI 17q21.31 Band 3, anion or, or, Wr", Wrb,

(010) (SLC4Al) exchanger 1 Wda, Rba, and 16
[CD233] more
(Continued)
TABLE 9·1. Blood Group Systems (Continued)

--~--.~-
ISBT .SYst8c111 Gene ~roduct Associated Blood
SylTlboll Name Gene Name: Chromosome (C9l11Ponent Name) 9roullAntig8cn~
(Number) ISBT (HGNC)* Location [CD number! [Null phenotype]
YT IYt YT 7q22.1 Acetylcholinesterase Yta, Ytb, YTEG, YTLI,

(011 ) (ACHE) YTOT
XG/Xg XG Xp22.33 Xga glycoprotein Xga, CD99

(012) (XG) Yp11.2 CD99 (MIC2 product)
SC I Scianna SC 1p34.2 Erythroid membrane- Sc1, Sc2, Sc3, Rd,

(013) (ERMAP) associated protein and 3 more
(ERMAP) [Sc:-1,-2,-3)
DO I Dombrock DO 12p12.3 Do glycoprotein, ART 4 Doa, Dob, Gya,Hy, Joa,

(014) (ART4) [CD297) and 5 more
[Gy(a-))
CO I Colton CO 7p14.3 Aquaporin 1 (AQP1) COa,COb,Co3, C04

(015) (AQPt) [Co(a-b-))
LW I Landsteiner- LW 19p13.2 LW glycoprotein, intra- Lwa, LWab,LWb

Wiener (ICAM4) cellular adhesion mole- [LW(a-b-))
(016) cule 4 (ICAM4) [CD242)
CH/RG I CHIRG 6p21.32 Complement compo- Ch1, Ch2, Rg1, and 6

Chido/Rodgers (C4A, C4B) nent: more
(017) C4A,C4B [Ch-Rg-)
H H 19q13.33 Fucosyltransferase, H
(018) (FUTt) carbohydrate [CD173) [Bombay (Oh))
XKI Kx XK Xp21.1 XK glycoprotein Kx

(019) (XK) [McLeod phenotype)
GE I Gerbich GE 2q14.3 Glycophorin C (GPC) Ge2, Ge3, Ge4, and 8

(020) (GYPC) [CD236) more
Glycophorin 0 (GPO) [Leach phenotype)
CROM I Cromer CROM 1q32.2 OAF cr, Tca, Tcb, Tce, Ora,
(021) (CD55) [CD55) Esa, IFC, and 13 more
[I nab phenotype)
KN I Knops KN 1q32.2 CR1 Kna, Knb, McCa, sr,

(022) (CRt) [CD35) Yka, and 4 more
TABLE 9·1. Blood Group Systems (Continued)
IN I Indian IN 11p13 Hermes antigen [CD44] Ina, Inb, and 4 more

(023) (CD44)
aKiak OK 19p13.3 Neurothelin, basigin aka, OKGV, OKGM

(024) (BSG) [CD147]
RAPH I Raph RAPH 11p15.5 CD151 MER2

(025) (CD 151) [Raph-]
JMH I John JMH 15q24.1 Semaphorin 7A JMH and 5 more

Milton Hagen (SEMA1A) [CD108] [JMH-]
(026)
I GCNT2 6p24.2 Glucosaminyltransfer-

(027) (IGNT) ase, carbohydrate [1- or i adult]
GLOB I Globoside GLOB 3q26.1 Transferase, P, PX2

(028) (B3GALNT1) carbohydrate [P-]
(Gb4, globoside)
GILl Gill GIL 9p13.3 Aquaporin 3 (AQP3) GIL

(029) (A QP3) [GIL-]
RHAG I Rh- RHAG 6p21.3 Rh-associated Duclos, ala, DSLK

associated glycoprotein
glycoprotein [CD241]
(030)
FORS14 FORS 9q34.2 Globoside 3-a-N- FORS1

(031 ) (GBGT1) acetylgalactosaminyl-
transferase 1
JR15,16 JR 4q22.1 Jr glycoprotein, Jra

(032) (A BCG2) ATP-binding cassette, [Jr(a- )]
subfamily G, member 2
(ABCG2) [CD338]
LAN17 LAN 2q36 tan glycoprotein, Lan

(033) (A BCB6) ATP-binding cassette, [Lan-]
subfamily B, member 6
(ABCB6)
(Continued)
TABLE 9-1. Blood Group Systems (Continued)
ISBTSystem Gene Product Associated Blood

Symbol/Name Gene Name: Chromosome (Component Name) Group Antigens
(Number) ISBT (HGNC)* Location (CD number] [Null phenotype]
VEL / Vej18-20 VEL 1p36 Small integral membrane Vel

(034) (SMIM1) protein 1 (SMIM1) [Vel-]
CD5921 CD59 11p13.33 CD59 CD59.1

(035) [CD59:-1]
AUG / Auqustlne" ENTt 6p21.1 Equilibrative nucleoside AUG1, Ata (AUG2),

(036) (SLC29A 1) transporter 1 (ENT1) and 2 more
[AUG:-1,-2]
KANNO PRNP 20p13 Prion protein (PRNP) KANN01

(037)
SID / Sid 84GALNT2 17q21.32 Beta-1,4-N-acetylgalac-

(038) tosaminyltransferase 2
CTL2t SLC44A2 19p13.2 Solute carrier family 44, CTL2.1 (global),

(039) member 2 CTL2.2 (Rif)
*If the genetic information is obtained by blood group typing, the gene name is the italicized form of the blood
group system ISBT name. For example, SLC14A 1 (HGNC terminology) would be written as JK*A and JK*8 or
JK*01 and JK*02 (ISBT terminology).
tAs the manual went to press, the ISBT Working Party meeting to approve CTL as blood group 039 was
delayed. See ISBT website for updates.
ISBT = International Society of Blood Transfusion; HGNC = Human Gene Nomenclature Committee; ATP = ade-
nosine triphosphate.
ing of KELantigens on red cells of the McLeod given locus than other systems, such as FYand
phenotype and demonstrates two red cell popu- CO.9 An allele that is polymorphic in one popu-
lations in carrier females. This mixed-cell population is not necessarilypolymorphic in all popu-
lation reflects the randomness in whether the lations; for example, the FY allele associated
maternal or paternal X chromosome is inactivat- with silencing of Fybin red cells (FY*02N.Ol) is
ed in any single somatic cell lineage. polymorphic in populations of African ancestry,
with a prevalence of >70%, but this allele is not
typically found in other populations. A gene
N TI HI N polymorphism may represent an evolutionary
advantage for a population, and a polymorphic
A polymorphism refers to the occurrence in population is likely to adapt to evolutionary
the population of genomes with allelicvariation change more rapidly than if the population had
(two or more alleles at one locus), each with ap- genetic uniformity. It is not yet understood
preciable (>1%) frequency. Some blood group what, if any, evolutionary advantages were de-
systems (RH and MNS, for example) are highly rived from the extensive polymorphism dis-
polymorphic and have many more alleles at a played by red cell antigens, but many publica-
Interphase: Individual chromosomes

are not easily distinguishable because
the chromatin is not contracted.
Prophase: Chromatin has condensed;

individual chromosomes can be seen; each
chromosome has duplicated and consists of
two sister chromatids joined at the centromere
Metaphase: Nuclear envelope has broken

down; homologous chromosomes, their
centromeres still joined, line up at the
equatorial plate as cell division begins.
Anaphase: Centromeres divide; sister

chromatids separate and move to
opposite poles.
Telophase: The nuclear envelope

reappears; the cytoplasm divides; the cell
membrane pinches inward; two identical
daughter cells have been formed.
FIGURE9-3. Diagram showing mitosis.
tions associate resistance to, or susceptibility for, phenotype has been estimated to be < 10'5 (<1
particular diseases with particular blood types." in 100,000) in humans.
Mutations can be inherited (germline) or ac- There are three types of genetic variation:
quired. Germline mutations are present in every single base-pair substitutions, also known as sin-
cell and will be passed on to offspring.Acquired gle nucleotide variants (SNVs);insertions or
or de-novo mutations can occur spontaneously deletions (in/dels) ofa single stretch of DNA,
or be brought about by agents such as radiation ranging from two to several hundred base pairs;
(eg, ultraviolet rays or x-rays) or chemicals. A and structural variants, which. involve lqrger
mutation may occur within a gene or in the in- stretches of DNA and include deletions, inser-
tergenic regions. It may be silent-that is, have tions, inversions, duplications, and copy number
no effect on the encoded protein-or it may al- variants. All three types of genetic variation.are
ter the gene product and potentially cause an seen in the genes that encode red cell antigen
observable effect in the phenotype. The muta- phenotypes. SNVshave given rise to most of the
tion rate of expressed genes resulting in a new diversity in the human genome.29'31 Accordingly,
2N r?If\f~~rt Interphase: Individual chromosomes

are not easily distinguishable because
the chromatin is not contracted.
Diploid
~Nucleus
! DNA Replication
Prophase I:
4N @!}f::~~:~~~ Chromosomes have duplicated and

homologous chromosomes have paired.
+
Crossing over occurs, resulting in an
exchange of genetic material between
4N~ homologous chromosomes.
4N @ Metaphase I: Homologous
chromosomes line up at the
equatorial plate.
t
([£)
Anaphase: First meiotic division occurs
as homologous pairs of chromosomes
4N separate. Process continues to
telophase and cell division.
2N wei)
I
i \
\
Formation of two daughter cells.
Di~~id®
I \
®I \
Second meiotic division: Chromosomal
duplication does not occur; previously
duplicated chromosomes separate.
Ha~oidCDCD CDCD Four daughter cells formed with

chromosomes reduced to half.
FIGURE 9-4. Diagram showing meiosis.
the majority of polymorphic blood group anti- nation of a polypeptide, often resulting in a non-
gens are the result of SNVs_7,32,33 SNVslocated functional protein product. Genotyping or
in gene-coding regions can be synonymous, DNA-basedassays designed to determine the se-
meaning that they encode the same amino acid quence at the location of an SNVcan be used to
as the reference sequence; or nonsynony- predict a red cell phenotype. This is discussed in
mous, meaning they result in substitution of a
more detail in the "Blood Group Genomics" sec-
different amino acid; or nonsense, meaning
they encode a stop codon. Because of the triplet tion. In the context of an allele that encodes a
nature of codons encoding amino acids, in/ dels protein that carries red cell antigens, any geneti-
can result in frameshifts if the number of nude- callyinduced change in the antigen must be rec-
otides inserted or deleted is not divisible by ognized by a specific antibody before an allele
three; frameshifts often cause premature termi- can be said to encode an antigen.
Alleles antigens. It is incorrect to refer to red cells that

are, for example, K-k+ or Kp(a-b+) as being ho-
A gene at a given locus on a chromosome may
mozygous for the k or Kpbantigen; rather, it
exist in more thaIl one form; that is, it may be
should be said that the cells have a double dose
allelic. Each person has two alleles for a trait,
of the antigen and that they are from a person
one that is maternally derived and another that
who is homozygous for the allele. Genes are al-
is paternally derived. At the simplest level, and
lelic,whereas some antigens are antithetical.
for the purpose of explaining the concept, the
The quantity of antigen expressed (antigen
°
ABO gene locus can be considered to have three
alleles: A, B, and (although genotypinghas re-
vealedmany variant alleles). With three alleles,
density) is influenced by whether a person is
heterozygous or homozygous for an allele; the
antigen density is generally greater when a per-
there are six possible genotypes (AlA, AID, AI son is homozygous. In some blood group sys-
B, BIB, BID, and OlD). Depending on the pa-
tems, this... difference in .antigen density is mani-
.
rental contribution, a person could inherit any fested by antibodies giving stronger reactions
combination of two of the alleles and express with cells that have a double dose of the anti-
the corresponding antigens on their red cells.
gen. Red cells with the Jk(a+b-) phenotype, en-
For example, inheritance of AlA and AIO coded by a JK*AIA genotype, have a double
would result in group A red cells, AIBwould re-
dose of the Ik" antigen and often are more
sult in group ABred cells, BIB and BIO would strongly reactive with anti-Ik' than those that
result in group B red cells, and 010would re-
are Jk(a+b+) and have a single dose of the anti-
sult in groupO red cells. gen. Similarly,M+N- red cells tend to be more
When identical alleles for a given locus are strongly reactive with anti-M than are M+N+
present on both. chrofilosomes, the person is red cells. Antibodies that are weakly reactive
said to be homozygous for the particular allele. may not be detected if they are tested with red
A person who is hemizygous for an allele has
cells expressing a single dose of the antigen.
only a single copy of an allele instead of the cus-
This observable difference in strength of reac-
tomary two copies; an example is the deletion of
tion, based on homozygosity or heterozygosity
one RHD in a D+ phenotype. When different for an allele, is termed the "dosage effect."
(ie, not identical) alleles are present at a particu-
(SeeTable 9-2.)
lar locus, the person is said to be heterozy-
gous. For example, a person who is homozy-
Genotype and Phenotype
gous at the KEL locus for the allele encoding
the k antigen (KEL *02) will have K-k+ red The genotype of a person is the complement of
cells. A person who is heterozygous for KEL*01 genes inherited from his or her parents; the
and KEL*02 (KEL*01/02 genotype), would term is frequently also used to refer to the set of
have red cells that are K+k+. alleles at a single gene locus. The phenotype is
Antigens that are encoded by alleles at the the observable expression of the genes inherited
same locus are said to be antithetical (meaning by a person and reflects the biologic activity of
opposite);thus, K and k are a pair of antithetical the gene(s). Thus, the presence or absence of
Homozygous JK*01/JK*01
Heterozygous JK*01/JK*02 Jk(a+b+) Single dose Single dose
JK*O 1IJK*O 1N.D1 (null) Jk(a+b ) Single dose N/A
expression.
antigens on the red cells, as determined by sero- and symbols used for the construction of pedi-
logic testing, represents the phenotype; the pres- grees are provided in Figs9-5 and 9-6.
ence or absence of antigens on the red cellspre-
dicted by DNA-based testing represents the Autosomal Dominant Inheritance
genotype. Sometimes the genotype can be pre-
An antigen (or any trait) that is inherited in an
dicted from the phenotype; for example, when a
autosomal dominant manner is always ex-
person's red cells are reactive with anti-lk' and
pressed when the relevant allele is present, re-
anti-Ik", which is a Jk(a+b+) phenotype, a
gardless of whether a person is homozygous or
JK*AIJK*Bgenotype can be inferred. Frequent-
ly, the phenotype provides only a partial indica- heterozygous for the allele. The antigen appears
tion of the genotype; for example, red cells that in every generation and occurs with equal fre-
are group B reflect the presence of a B gene, but quency in both males and females. A person
the genotype may be ABO*BIB or ABO*Bin who carries an autosomal dominant trait trans-
For decades, family studies were often used to mits it, on average, to half of his or her children.
infer a person's genotype, but now that most an- The pedigree in Fig 9-7 demonstrates autosomal
tigens and phenotypes can be defined at the dominant inheritance and shows that the B al-
DNA level, family studies to determine geno- lele is dominant over O.
type can be mostly replaced by DNA analysis.
(See "BloodGroup Genomics" section below.] Autosomal Codominant Inheritance
Blood group antigens that appear to be autoso-
mal dominant may be encoded by alleles that
INH N N
are inherited in a codominant manner-that
R ITS is, when two different alleles are present (the
heterozygous condition), the products of both
A genetic trait is the observed expression of one
or more genes. The inheritance of a trait (and
red cell antigens) is determined by whether the
gene responsible is located on an autosome or
on the X chromosome (sex-linked)and whether
the trait is dominant or recessive.
Pedigrees
A family study follows the inheritance of a ge- II
netic characteristic-for example, an allele en-
coding the expression of a red cell antigen-as it
is transmitted through a kinship. A diagram that
depicts the relationship of family members and
III
shows which family members express (are af-
fected), or do not express, the trait under study FIGURE 9-5. An example of a pedigree. Males are
is termed a pedigree. A review of a pedigree denoted by squaresand females by circles, and each
should reveal the pattern or type of inheritance different generationin a pedigreeis identified by Ro-
for the trait, or antigen, of interest. The person man numerals. Personsin eachgenerationare iden-
who first caused the familyto be investigated is tified by Arabic numbers; the numbering is sequen-
considered the index case and is often referred tial from left to right, with the eldest child for each
to as the proband or propositus/proposita family unit being placed on the left of any series of
(male singular form or gender unknownlfemale siblings. Closedsymbols representfamily members
singular form); propositi is the plural form re- affected by the trait, whereas open symbols are un-
gardless of gender. Details of the conventions affected members.
o Male alleles are expressed. Thus, when red cells have

the S+s+ phenotype, the presence of one allele
o Female encoding S and another allele encoding s [or an
<> Gender.not specified 51s (GYPB*51s) genotype]can be inferred.
D-O Mating (the male is
always on the left) Autosomal Recessive Inheritance
D=O Consanguineous mating A trait with autosomal recessive inheritance
is expressed only in a person who is homozy-
[] (5 (5 [] Siblings (in order of.birth)
gous for the allele and has inherited the reces-
Trait expressed sive allele from both parents. When a person in-
herits a single copy of a recessive allele in
[J o Heterozygote combination with a silent or deleted (null) al-
o Carrier (heterozygote) lele-that is, a nonfunctioning allele or one that
encodes a product that cannot be detected-the
for X-borne recessive trait
recessive trait is expressed and the person ap-
6 L 4 iJ Abortions or rniscarriages pears to be homozygous. It is difficult or impos-
~ Monozygotic twins
sible to distinguish such a combination from ho-
mozygosity for the recessive allele through
db c:5O Ob Dizygotic twins serologic testing, but DNA-based testing can
usually make this distinction.
o JZf Deceased A mating between two heterozygous carri-
ers results in one chance in four that the chil-
Indicates proband dren will be homozygous for the trait. The par-
ents of a. child who is homozygous for a
FIGURE 9-6. Symbols, and their significance, used in recessive trait are obligate carriers of the trait. If
the construction of pedigrees. the frequency of the recessive allele is low, the
condition is rare and usually found only among
B o
II
DorO = group0
III .orl e groupB

o B o o B
FIGURE 9-7. Autosomal dominant inheritance of the ABO alleles. Basedon the ABO groups of his children, 1-1
would be expected to have a BID rather than a BIB genotype (showing that the B allele is dominant over 0)
becausetwo of his children (11-6and 11-7)are group 0 and must have inherited an 0 allele from their father (1-1)
in addition to the 0 allele inherited from their mother (1-2). Similarly, 11-2and 11-3are BID, based on the ABO
type of their children, showing the dominance of B over O.
siblings (brothers and sisters) of the person and tance of antithetical antigens Lu"and Lub as well
not in other relatives. The condition is not as autosomal recessive inheritance of a null LU
found in preceding or successive generations gene, which in the homozygous state results in
unless consanguineous mating (Ie, between the Lu(a-b-) phenotype. The proband, U-3,
blood relatives) occurs. When a recessive allele who received multiple transfusions, developed
is very rare, the parents of an affectedperson are anti-Lu3 (an antibody to a high-prevalence
most likely consanguineous because a rare allele Lutheran antigen) in his plasma. Because his
is more likelyto occur in blood relatives than in Lu(a-b-) phenotype is the result of recessive in-
unrelated persons in a random population. heritance, his siblingsare potential donors, with
When a recessive trait is one that is common, a probability of one in four that the offspringof
consanguinity is not a prerequisite for homozy- the mating between I-I and 1-2would have the
gosity;for example, the 0 allele of the ABOsys- Lu(a-b-) phenotype. However, in this case, only
tem, although recessive, is not rare, and persons the proband had red cells with the Lu(a-b-)
who are homozygous for 0 are easily found in phenotype.
the random population.
In blood group genetics, a recessive trait al- Sex-Linked Inheritance
most always involves homozygosity for a si-
lenced allele that encodes no product such that A sex-linked trait is one that is encoded by a
the red cells express a null phenotype leg, the gene located on the X or Y chromosome. The Y
Lu(a-b-) or Rhnull or 0 phenotypes]. The family chromosome carries few functional genes, and
in Fig 9-8 demonstrates the codominant inheri- discussion of sex-linked inheritance generally is
1
I Lub/Lu Lub/Lu
Lu(a-b+) Lu(a-b+)
II Lub/Lub Lub/Lu Lu/Lu Lua/Lub Lub/Lub Lub/Lu Lub/Lu Not

Lu(a-b+) Lu(a-b+) Lu(a-b-) Lu(a+b+) Lu(a-b+) Lu(a-b+) Lu(a-b+) tested
4 6
III Lub/Lub Lub/Lu Lub/Lub Lu"/Lu Lub/Lu Not

Lu(a-b+) Lu(a-b+) Lu(a-b+) Lu(a+b-) Lu(a-b+) tested
• = homozygosity for Lu DorO = lacks Irail (Lu) IJor() = heterozygosity for Lu
FIGURE 9-8. Autosomal recessive inheritance. The offspring of 11-3, the Lu(a-b-) proband, and 11-4, his
Lu(a+b+) wife, demonstrate that Lu (depicting a Lutheran null allele, or LU*02N) is recessive to lu' (LU*A) and
Lub (LU*8) and that the presence of the silenced Lutheran allele is masked by the product of iu' (LU*A) or Lub
(LU*8) at the phenotype level.
synonymous with inheritance of genes carried Sex-linked Dominant Inheritance

by the X chromosome. In females (who have
Sex-linkeddominant inheritance describes a
two X chromosomes), the inheritance of genes
trait encod~d bya gene on a sex chr0J:Il9sQIlle
carried by the X chromosome, like the inheri-
showing dominant inheritance. For X-lip1<ed
tance of genes carried on the autosomes, can be
genes, the 1]',ait is expressed by males and by fe-
dominant or recessive. Males, in contrast, have males heterozygous or homozygous. for the
one X chromosome (always maternally derived) gene. A male passes his single X chromosome to
and one Y chromosome (always paternally de- all of his, daughters and all daughters will ex-
rived) and are hemizygous for genes on the X or press tl1e' cqndition or trait. When a female is
Y chromosome' because only one' chromosome heterozygous for an allele that encodes a domi-
(and thus one copy of a gene) ispresent. Most nant trait, each of her children, whether male or
genes carried by the X chromosome do not have female, has a 50% chance of inheriting the trait.
a homolog (a similar sequence of DNA) on the When a female is homozygous for an XIlnked
Y chromosome. As a consequence, inheritance trait with dominant inheritance, the encoded
of an X-linked dominant trait is the same in trait is expressed by all her children. (See Fig 9-
males and females. However, an X-linkedreces- 9.)
sive trait is expressed by all males who carry the The Xg' antigen (XGblood group system) is
gene for the trait. There is no male-to-male encoded by an allele on the X chromosome and
transmission of either dominant or recessive is inherited in a sex-linked dominant manner.
Xlinked traits; that is, Xfinked traits are not The first indication that the Xg' antigen was
transmitted from father to son. Xlinked came from the observation that the
2
I Xg{a+} Xg(a-)
Xg· Xg/Xg
II Xg(a-} Xg(a+}
Xg Xg"/Xg
III Xg{a+} Xg{a-} Xg(l:l+)

Xg/Xga Xg/Xg Xg"
OorO::; personshaveXg(a-) redcells .ore::; personshave Xg(a+)redcells

FIGURE 9·9. Sex-linked dominant inheritance. The xga antigen is encoded by an allele on the tip of the short
arm of the X chromosome. This family demonstrates the sex-linked dominant inheritance of the xga antigen.
prevalence of the Xg(a-) and Xg(a+)phenotypes anthocytosis and, frequently, compensated he-
differed noticeably between males and females, molytic anemia. More than 30 different XK gene
with the Xg' antigen having a prevalence of 89% mutations associated with a McLeod phenotype
in females and only 66%in males."Recently, dif· have been found. Different XKmutations appear
ferential expression of Xg' was found to be to have different clinical effects and may ac-
linked to disruption of GATA-bindingmotif in count for the variability in the prognosis." Se-
the XGgene.34 quencing of XK to determine the specifictype of
Figure 9-9 shows the inheritance of the Xg' mutation in individuals with McLeod pheno-
antigen in a three-generation family. In genera- types has clinical prognostic value. McLeod syn-
tion I, the father (I-I) is Xg(a-] and has transmit- drome is an X-linkedrecessive condition and, as
ted X( to all his daughters but to none of his demonstrated by the familyin Fig 9-10, is found
sons. His eldest daughter (II-2), for example, only in males.
must be heterozygous for X( /Xg, she received
the allele encoding the Xg" antigen from her The Principles of Independent
Xg(a+) father and a silent allele, Xg, from her Segregation and Independent
Xg(a-) mother. II-2 has transmitted X( to half Assortment
her children, regardless of whether they are
sons or daughters. The passing of a trait from one generation to the
next follows certain patterns or principles. The
Sex-Linked RecessiveInheritance principle of independent segregation refers to
the separation of homologous chromosomes and
Sex-linked recessive inheritance describes a their random distribution to the gametes during
trait encoded by a sex chromosome that is not meiosis. Only one member of an allelic pair is
expressed in carriers of a single copy, such as in passed on to the next generation, and each gam-
heterozygous females. A male inherits the trait ete has an equal probability of receiving either
from his mother, who is usually a carrier (or ho- member of a parental homologous allelic pair;
mozygous for the trait). An affected male trans-
these chromosomes are randomly united at fer-
mits the trait to all of his daughters, who in turn
tilization and thus segregate independently from
transmit the trait to approximately half of their
one generation to the next. The familyin Fig 9-
sons. Therefore, the prevalence of the expres-
11 demonstrates the independent segregation of
sion of an Xlinked recessive trait is much higher
the ABOalleles on chromosome 9.
in males than in females. A carrier female who
The principle of independent assortment
mates with a male lacking the trait transmits the
states that alleles determining various traits are
trait to one-half of her daughters (who will also
inherited independently from each other. In oth-
be carriers) and to one-halfof her sons (who will
er words, the inheritance of one allele (eg, a B
be affected). If mating is between an affected
male and a female who lacks the trait, all of the allele, on chromosome 9) does not influence the
sons will lack the trait and all of the daughters inheritance of another allele (eg, an M allele, on
will be carriers. If an X-linked recessive trait is chromosome 4). This is demonstrated by the
rare in the population, the trait is expressed al- familyin Fig9-11.
most exclusivelyin males.
The XK gene encodes the Kx protein and
demonstrates Xlinked recessive inheritance.
TRUCTURAL RI N
Mutations in or deletions that include XK result Linkage and Crossing Over
in red cells with the McLeod phenotype, in
which red cells lack Kx and have reduced ex- Linkage is the physical association between
pression of KELantigens. McLeod syndrome two genes that are located on the same chromo-
is associatedwith late-onset clinicalor subclinical some and are inherited together. Examples in-
myopathy, neurodegeneration, and central ner- clude RHD and RHCE encoding the antigens of
vous system manifestations, as well as with ac- the RH system, which are both on chromosome
II
8
III
o}normal
Unaffected; I McLeod phenotype
o
Kell
antigen expression
o Carrier of McLeod phenotype; mixed. red cell population
of McLeod phenotype and normal Kell antigen expression
FIGURE 9·10. Sex-linkedrecessiveinheritance.This family demonstratesthat a sex-linkedtrait that is recessive

in femaleswill beexpressedby any malewho inherits the trait. Homozygosityfor sucha trait is requiredfor it to
beexpressedin females.Thetrait skips one generationand is carriedthrough females.
1 and represent linked loci that do not assort in- Two gene loci carried by the same chromo-
dependently. some that are not closely linked are referred to
Crossing over is. the exchange of genetic as being syntenic. For example, the loci for RH
material between homologous chromosome and FY, both located on chromosome 1, are syn-
pairs (Fig 9·4). In this process, a segment from tenic because the distance between them (RH
one chromatid (and any associated genes) on the short arm and FY on the long arm) is
changes places with the corresponding part of great enough for them to undergo crossing over
the other chromatid (and its associated genes); and to assort independentiy.
the segments are rejoined, and some genes will The frequency of crossing over involving two
genes on the same chromosome is a measure of
have switched chromosomes. Thus, crossing
the distance [measured in centimorgans (cM)]
over is a means to shuffle genetic material. Be-
between them; the greater the distance be-
cause crossing over can result in new gene com- tween two loci, the greater the probability that
binations on the chrornosom~sinvolved, it is crossing over (and recombination) will occur. In
also.referred to .as recombination, and the reo contrast, genes located very close together
arranged chromosomes can be referred to as re- (linked) tend to be transmitted together with no
combinants. Crossing over and recombina- recombination. The degree of crossing over be-
tion, using chromosome 1 as an example, are tween two genes can be calculated by analyzing
explained in Fig 9·12. pedigrees of familiesinformative for the genes of
1
I Phenotype B A
Genotype BIO Ala
Phenotype M+N+ M+N-
Genotype MIN MIM
2
II Phenotype B 0 A AB
Genotype BIO ala AID AlB
Phenotype M+N+ M+N+ M+N- M+N-
Genotype MIN MIN MIM MIM
FIGURE 9·11. Independent segregation and independentassortment are illustrated by the inheritance of blood
group alleles in one family. Parental ABO alleles were randomly transmitted (independent segregation), and
eachchild has inherited a different combination. The family also illustrates that the alleles encoding antigens of
the ABOand MNS blood group systems are inherited independentlyfrom eachother.
interest and observing the extent of recombina- nized example of autosomal linkage in humans
tion. The traditional method of linkage analysis and is explained in Fig 9-13.
requires the use of LOD (logarithm of the odds) Although crossing over occurs readily be-
scores." Linkage analysis was the basis through tween distant genes, rare examples of recombi-
nation have been documented for genes that are
which chromosomes were mapped and the rela- very closely linked or adjacent on a chromo-
tive position and distance between genes estab- some. Such genes include those encoding the
lished. Linkage between Lutheran (LU) and MN (GYPA) and Ss (GYPB) antigens on chromo-
ABHsecretion (SE or FUT2) was the first recog- some 4 and are reviewed by Danie1s.24(pp96·142)
FYBJ(:;:E
KN [SI(a+)] KN [SI(a-)]
Fy;}{:~:E
c,~:::"''1-+ U ~=::~:,~~~
KN [SI(a-)]
•V J!
KN [SI(a+)] 2 recombinant
Homologous pair chromosomes
of chromosome 1
FIGURE 9·12. Crossing over and recombination. In the diagram, chromosome 1 is usedas an example.The very
closely linked RH genes, RHD and RHCE, are located nearthe tip of the short arm of chromosome 1. The loci for
FYand KN are on the long arm of the chromosome and are not linked. During meiosis, crossing over occurs
betweenthis homologous chromosome pair, and portions of chromosome break and become rejoined to the
partner chromosome. Crossing over of the long arm of chromosome 1 results in recombination betweenthe loci
for FYand KN such that the gene encoding Fybantigen is now traveling with a gene that encodes the SI(a-)
phenotypeof the KN system.
1
I Luo/Lub
Se/se
1 2
II Lua/Lub Lub/Lub Lub/Lub LUB/Lub Lua/Lub
se/se Se/se Se/se se/se se/se
FIGURE 9·13. Linkage between LU and Sf (FUT2). 1-2is homozygous for Lub (LU*8) and se and must transmit
these alleles to all her offspring. 1-1 is doubly heterozygous [Llfl Lub (LU*A/8) and Se/5eJ.He has transmitted
Lub (LU*8) with Se,and Lif (LU*A) with 5e,showing linkage between LU and Se(FUT2). Several such informa-
tive families would needto beanalyzedto statistically confirm linkage.
Linkage Disequilibrium HI R
As mentioned earlier, genes at closelylinked loci The observationthat a sample gives mixed-field
tend to be inherited together and constitute a agglutination is not an unusual one in transfu-
haplotype (a combination of alleles at two or
sion medicine. Often this is the result of artifi-
more closely linked loci on the same chromo- cially induced chimerism through the transfu-
some). The alleles encoding the MNS antigens sion of donor red cells or the result of a
are inherited as four haplotypes:MS, Ms, NS, or hematopoietic stem cell.transplant. On rare oc-
Ns [in the International Society of Blood Trans- casions, the observation of mixed-field aggluti-
fusion (ISBT) allele terminology, these haplo- nation identifies a true chimera, that is, a per-
types would be written as GYPA*M-GYPB*S, son with a dual population of cells derived from
GYPA*M-GYPB*s, GYPA*NGYPB*S, or GY- more than one zygote. Indeed, the first example
PA *NGYPB*s, respectively). Because Iinked of a human chimera was a female blood donor
genes do not assort independently, the antigens discovered through mixed-field agglutination
encoded by each of these haplotypes have a dif- during antigen typing. Most human chimeras
ferent prevalence in the population than would can be classified as either twin chimeras or
be.expected by random assortment. If M and S tetragametic (dispermic) chimeras. Chimerism
were not linked, the expected prevalence for is not a hereditary condition."
M+ and S+ in the population would be 17% Twin chimerism occurs through the forma-
(fromfrequency calculations),whereas the actu- tion of placental blood vessel anastomoses,
al or observed prevalence (obtained from testing which results in the mixing of blood between
and analyzing families) of the MS haplotype is two fetuses. Thisvascular bridge allows hemato-
24%.24(pp96.142) This constitutes linkage disequi- poietic stem cells to migrate to the marrow of
librium, which is the tendency of specificcom- the opposite twin. Each twin may have two dis-
binations of alleles at two or more linked loci to tinct populations of cells (red cells and leuko-
be inherited together more frequently than cytes), that of his or her true genetic type and
would be expected by chance. that of the twin. The percentage of the two cell
lines in each twin tends to vary; the major cell ple, are always inherited together, but C and E
line is not necessarily the autologous cell line, are not. The preceding explanation, which im-
and the proportions of the two cell lines may plies that one gene encodes C and c antigens
change throughout life. Chimeric twins have while another linked gene encodes E and e anti-
immune tolerance; they do not make antibody gens, was based on the Fisher-Race theory of
against the A or B antigens that are absent from three genes at the RHlocus. In contrast, genom-
their own red cells but are present on the cells ic analysis indicates that only one gene (RHCE),
of the engrafted twin. This tolerance extends be- with four alleles (RHCE*Ce, RHCE*cE,
yond red cells to negative mixed-lymphocyte RHCE*ce, and RHCE*CE), encodes one protein
cultures and the mutual acceptance of skin that carries the CcEe antigens. ThUS,for Rh,
grafts. DCe is an example of a haplotype, and the RHD
In twin chimeras, the dual cell population is allele is in cis to the RHCE*Ce allele.
strictly confined to blood cells. Tetragametic or The expression of red cell antigens may be
dispermic chimeras present chimerism in all tis- modified or affected by gene or protein interac-
sues and are more frequently identified because tions that manifest primarily as reduced antigen
of infertility than because of mixed populations expression. One example in which the haplo-
of red cells. The mechanism(s) leading to the de- type on one chromosome affects the expression
velopment of tetragametic chimeras are un- of the haplotype on the paired chromosome is
known, but they arise through the fertilization commonly referred to as the position effect
of two maternal nuclei by two sperm, followed and can be observed with Rh antigen expres-
by fusion of the two zygotes and development sion. When a Ce haplotype (note the absence of
into one person containing two cell lineages. RHD) is in trans to a D-antigen-encodinghaplo-
More commonly, chimeras occur through type, the expression of D is dramatically re-
medical intervention and arise from the transfer duced and a weak D phenotype can result.
of actively dividing cells, such as through hema- When the same D-encodinghaplotype is inherit-
topoietic cell transplantation." However, chime- ed with either ce or cE, D antigen is normally
rism may be more prevalent than once thought, expressed. The cause of this reduced antigen ex-
based on the discovery of people with dual red pression is not known, but it may involve differ-
cell populations when performing DNA analysis ences in gene expression levels or altered assem-
to predict red cell phenotypes, and chimerism bly of proteins in the membrane. In the
has been the cause of disputed maternity.38,39 presence of the KELsystem antigen Kp", expres-
sion of other KELsystem antigens encoded by
the same allele is suppressed to varying degrees
N I I N (cis-modifiereffect). This is best observed in per-
sons who have a silenced KEL gene (Ko)in trans.
Alleles that are carried on the same chromo- The amino acid change that results in the ex-
some are referred to as being in cis position, pression of Kp' adversely affects trafficking of
whereas those on opposite chromosomes of a the Kell glycoprotein to the red cell surface, so
homologous pair are in trans position. Alleles that the quantity of Kp' carrying Kell glycopro-
that are in cis and linked are always inherited tein that reaches the red cell surface is greatly
together on the same chromosome, whereas reduced.
genes in trans segregate independently.
Historically, the RH blood group system was
used to explain the meaning of cis and trans.
For example, the DCe/DcE genotype was de-
scribed as having C and e alleles in cis in the
DCe haplotype, with c and E alleles in cis in the
partner DcE haplotype, whereas C and E and A genetic modifier is a gene or locus that af-
also c and e are in opposing haplotypes and are fects the expression of another gene or genes.
in trans. In this alignment, C and e, for exam- Genetic modifiers can be unlinked or inherited
independently from the genes they modify. For ids. The genes responsible for carbohydrate-
example, KLFl, located on chromosome based antigens do not encode membrane pro"
I9pI3.3·pI3.I2, encodes erythroid Kruppel-like teins but instead encode a glycosyltransferase
factor (EKLF),which is a transcription factor es- that catalyzes the sequential transfer of the ap-
sential for expression of genes critical to termi- propriate immunodominant monosaccharide.
nal differentiation of red cells. Singleton et al40 The carbohydrate antigens are carried on oligo"
first discovered that heterozygosity for nucleo- saccharide chains that are assembled by the
tide variants in KLFI is responsible for the domi- stepwise addition of monosaccharides. Each
nant Lu(a-b-) phenotype," which is also known monosaccharide structure is transferred by a
as the In(Lu) phenotype. This heterozygosity separate glycosyltransferase, such that two
is characterized by reduced expression of anti" genes are required for a disaccharide, three for a
gens in the Lutheran system, PI, Inb, and AnWj trisaccharide, and so on. If the enzyme encoded
antigens. Xlinked forms of the In(Lu) pheno- by one locus is inactive due to a genetic variant,
type have been associated with mutations in the this can prevent or modify the expression of the
transcription factor GATAI, encoded by the X- other gene products. The product encoded by
borne GATA-I gene, known to be essential for the H gene is the biosynthetic precursor for A
erythroid and megakaryocyte differentiation." and B antigen production. Thus, if the H gene is
Typing of red cells for one or more of these oth-
silenced, A or B antigens cannot be produced.
er antigens can help differentiate the In(Lu) phe-
Genetic variation in ABO*A and ABO*B alleles
notype from Lu(a-b-). Sequencing of the rele-
can result in expression of a glycosyltransferase
vant genes also can be used to make the
distinction. that is less functional or inactive. Details on the
Another example of an unlinked modifier of biosynthesis of the ABO, H, LE, and I antigens
a blood group system is the regulator type (as may be found in Chapter IO.
opposed to the amorph type) of Rhnull• Family
studies have been employed to locate silencing
mutations in RHAG, a gene located on chromo"
N N
some 6 that encodes the Rh-associatedglycopro- Population genetics is the study of the dlstn-
tein RhAG. Red cell membrane expression of
bution patterns of genes and of the factors that
RhAG is required for Rh antigen expression.
maintain or change gene (or allele) frequencies.
RHA G has also been found to carry variants as"
A basic understanding of population genetics,
sociated with the Rhmod phenotype."
probability, and the application of simple alge-
Several red cell antigens require the interac-
tion of the products of two or more independent braic calculations is important in transfusion
genes for their expression. Amino acids 75 to 99 medicine, where the knowledge can be applied
of GPA, the glycoprotein that carries the M and to clinical situations such as predicting the likeli-
N antigens, must be present in the red cell hood of finding compatible blood for a patient
membrane for expression of Wrb, an antigen in who has made antibody(ies) to red cell antigens.
the Diego blood group system carried on band It may be helpful to define three commonly
3. An absence of RhD and RhCEprotein (Rhnull) used. words so that their appropriate use is
results in red cells that lack LW antigens and understood. Frequency is used to describe
have reduced or no expression of V, S, and san" prevalence at the genetic level=-that is, the oc-
tigens encoded by GPB, again demonstrating currence of an allele (gene) in a population.
the interaction in the membrane of the products Prevalence is used to describe the occurrence
of two or more independent blood group genes. of a permanent inherited characteristic at the
The sequential interaction of gene products phenotypic level-for example, a blood group
from several loci is required for the expression antigen-in any given population. Incidence is
ofABO, H, LE, and I antigens on red cells and in used when describing the rate of occurrence in
secretions. These antigens are carbohydrate de" a population of a condition that changes over
terminants carried on glycoproteins or glycolip- time, such as a disease.
Phenotype Prevalence The prevalence of donors of European ances-

try who are: K- = 91 %; S- = 48%; Jk(a-) =
The prevalence of a blood group antigen or phe-
24%.
notype can be determined by testing red cells The percentage of such donors negative for
from a large random sample of people of the each antigen is expressed as a decimal and
same race or ethnicity with a specific antibody multiplied: 0.91 (K-) x 0.48 (S-) x 0.24
and calculating the percentage of positive and Uk(a-)] = 0.1048.
negative reactions. The larger the cohort being 0.1048 expressed as a % = 0.1048 x 100% =
tested, the more statistically significant is the re- 10.48%.
sult. For antithetical antigens, the sum of the 10.48% expressed as occurrence = 10.481
percentages for the prevalence of the pheno- 100 = approximately 1110.
types should equal 100%. For example, in the Thus, approximately 1 in 10 ABO-compati-
FY blood group system, the prevalence in a ran- ble Red Blood Cell (RBC) units are expected
dom population of African ancestry for the to be K-, S-, Jk(a-).
The patient in question requires 3 units, so
Fyta-b-), Fy(a-b+J, Fy(a+b+J, and Fy(a-b-)
on average, 30 units would need to be test-
phenotypes is 9%, 22%, 1%, and 68%, respec-
ed. This can be calculated as follows:
tively; together, these percentages total 100%. If
0.1 of X =3.
the red cells from 1000 donors of European an- X= 3/0.1.
cestry are tested with anti-c, and 800 of the X = 30 units.
samples are positive and 200 are negative for
the Rh antigen c, the prevalence of the c+ phe- The prevalence of a particular antigen (or
notype is 80% and that of the c- phenotype is phenotype) can vary with race,' and the preva-
20%. Thus, in this donor population, approxi- lence for a combined antigen-negative pheno-
mately 20% of ABO-compatible units of blood, type calculation should be selected on the basis
or 1 in 5, should be compatible with seruml of the predominant race found in the donor pop-
plasma from a patient who has made anti-c. ulation.
Calculations for Antigen-Negative Allele (Gene) Frequency

Phenotypes The allele frequency is the proportion of one al-
When blood is provided for a patient with anti- lele relative to all alleles at a particular gene lo-
cus in a given population at a given time. This
bodies directed at one or more red cell antigens,
frequency can be calculated from the prevalence
a simple calculation can be used to estimate the
of each phenotype observed in a population.
number of units that need to be tested to find
The sum of allele frequencies at any given locus
the desired antigen combination. To calculate must equal 100% (or 1 in an algebraic calcula-
the prevalence of the combined antigen- tion) in the population sample tested. The geno-
negative phenotype, the prevalence of each of type frequency is the number of individuals with
the individual antigens are multiplied together if a given genotype divided by the total number of
the antigens are inherited independently of each individuals sampled in a population.
other. This approach is not valid when the anti-
gens are encoded by alleles that are closely The Hardy-Weinberg Equilibrium
linked and are inherited as haplotypes (M, N, S,
Gene frequencies tend to remain constant from
s) or reside on the same carrier protein (C, c, E, generation to generation in any relatively large
e). If a patient with antibodies to K, S, and Jka population unless they are influenced by factors
antigens requires 3 units of blood, for example, such as selection, mutation, migration, or non-
prevalence of the antigen-negative phenotype random mating, any of which would have to be
and the number of units that need to be tested significant to have a discernible effect. Accord-
to find it can be calculated as follows: ing to the principles proposed by the British
mathematician Hardy and the German physi- q2 = 1 - 0.09

cian Weinberg, gene frequencies reach equilibri-
q = ./0.91
um. This equilibrium can be expressed in alge-
braic terms by the Hardy-Weinberg formula or q = 0.95 = the frequency of KEL *02
equation: Because the sum of the frequencies of both
alleles must equal 1.00:
p+q=l
If two alleles, classicallyreferred to as A and
p=l-q
a, have gene frequencies of p and q, the homo-
zygotes and heterozygotes are present in the P = 1 - 0.95
population in the followingproportions: P = 0.05 = the frequency of KEL *01
Having calculated the allele frequencies for

KEL *01 and KEL *02, it is possible to calculate
In such a two-allele system, if the gene fre- the percentage of k+ (both K+k+ and K-k+)
quency for one allele, say p, is known, q can be and K+ (both Ksk- and K+k+) people:
calculated by p + q = 1.
The Hardy-Weinberg equation permits the Prevalence of k+ 2pq + q2
estimation of genotype frequencies from .the 2(0.05 xO.95) + (0.95)2
phenotype prevalence in a sampled population = 0.9975><100
and, reciprocally, allows the determination of a calculated prevalence
genotype frequency and phenotype prevalence of 99.75% (the observed
from the gene frequency. The equation has a prevalence of the k+
number of applications in blood group genetics, phenotype is 99.8%)
and its use is demonstrated below. Prevalence of K+ 2pq + p2
In a population of European ancestry, the fre- = 2(0.05 xO.95) + (0.05)2
quencies of the two alleles encoding the anti- 0.0975
thetical antigens K (KEL *01) or k (KEL *02) can = 0.0975 x 100 = a calcu-
be determined as follows: lated prevalence for K+
of 9.75% (the observed
Frequency of the KEL *01 allele = p prevalence of the K+
Frequency of the KEL *02 allele = q phenotype is 9%)
Frequency of the KEL *01 genotype = p2 The Hardy-Weinbergequation also can be ap-
Frequency of the KEL *011KEL *02 genotype = plied to calculate the frequencies of the. three
2pq possible genotypes KEL *01 I KEL *01, KEL *011
KEL *02, and KEL *021KEL *02 from the gene
Frequency of the KEL *021 KEL *02 genotype = frequencies KEL*01 (p) = 0.05 and KEL*02 (q)
q2 =0.95:
The K antigen is expressed on the red cells of p2 + 2pq + q2 = 1

9% of people of European ancestry; therefore: Frequency of KEL *01 lKEL *01 = p2 = 0.0025
p2 + 2pq = the frequency of people who carry Frequency of KEL *OllKEL *02= 2pq = 0.095
KEL*01 and are K+
Frequency of KEL *021 KEL *02 = q2= 0.9025
Thus, p2+ 2pq = 0.09
q2 = 1 (p2+ 2pq) = the frequency of people
-e-- If antibodies are availableto test for the prod-
who carry KEL *02102 and are K- ucts of the alleles of interest (in this example,
anti-K and antl-k), the allele frequencies also can may explain many of the differences in allele fre-
be obtained by direct counting as demonstrated quencies between populations.
in Table 9-3. The allele frequencies obtained by
direct testing are the observed frequencies for
the population being sampled, whereas those L ION HI T
obtained by gene frequency calculations (above)
are the expected frequencies. The various Polymorphisms are inherited characteristics or
calculations above, when applied to a two-allele genetic markers that can distinguish between
people. The blood groups with the greatest
situation, are relatively simple; the calculations
number of alleles (greatest polymorphism) have
for three or more alleles are much more com-
the highest power of discrimination. Blood is a
plex and beyond the scope of this chapter.
rich source of inherited characteristics that can
For a given population, if the prevalence of
be detected, including red cell, HLA, and plate-
one genetic trait, such as a red cell antigen, is
let antigens. Red cell and HLAantigens are easi-
known, the Hardy-Weinberg equation can be ly identifiable, are polymorphic, and follow
applied to calculate allele and genotype frequen- Mendelian laws of inheritance. The greater the
cies. The Hardy-Weinberg equilibrium principle level of genetic variation in a system, the less
is valid when the population is sufficiently large chance there is of finding two people who are
that chance alone cannot alter an allele frequen- identical. The extensive polymorphism of the
cy and when the mating is random. A selective HLAsystem alone allows the exclusion of >90%
advantage or disadvantage of a particular trait of falsely accused men in cases of disputed pa-
and other influencing factors, such as mutation ternity. However, serologic methods of identity
or migration in or out of the population, are as- testing were surpassed and replaced by DNA-
sumed to be absent when the Hardy-Weinberg based assays" (referred to as DNA fingerprinting
equilibrium principle is applied. When all of or DNA profiling) that were pioneered by Jef-
these conditions are met, the gene pool is said to freys and colleagues.P:" Tandemly repeated se-
be in equilibrium and allele frequencies do not quences of DNA of varying lengths occur pre-
change from one generation to the next. If the dominantly in the noncoding genomic DNA,
conditions are not met, changes in allele fre- and they are classified into groups depending on
quencies may occur over a few generations and the size of the repeat region. The extensive vari-
TABLE 9·3. Allele Frequencies of K (KEL* 01) and k (KEL* 02) Calculated Using Direct Counting
(assuming the absence of null alleles)
K+k- 2 4 4 o
88 176 88 88
K-k+ 910 1820 o 1820
Totals 1000 2000 92 1908
Allele frequency 0,046 0.954
--
A random sample of 1000 peopletested for K and k antigens has a total of 2000 alleles at the KEL locus becauseeach
person inherits two alleles, one from each parent. Therefore,the two persons with a K+k- phenotype(each with two
alleles)contribute a total of four alleles,To this are added88 K (KEL*01) allelesfrom the K+k+group, for a total of 92 K
(KEL *01) alleles, or an allele frequency of 0,046 (92 + 2000), The frequency of the k (KEL *02) allele is 0,954 (1908 +
2000).
ation of these tandemly repeated sequences be- OD GR U E E

tween individuals makes it unlikely for the same
number of repeats to be shared by two individu-
als, even if these individuals are related. Mini- Gene mapping is the process through which a
satellite [also referred to as variable number of gene locus is assigned to a location on a chromo-
tandem repeats (VNTR)] loci have tandem- some. The initial mapping of blood group genes
repeat units of 9 to 80. base pairs, whereas mi- was accomplished by testing many families for
crosatellite [alsoreferred to as short tandem re- selected red cell antigens. Pedigrees were ana-
peat (STR)]loci consist of two to five base-pair lyzed for evidence of recombination between
tandem repeats." the genes ofinterest to rule out or establish link-
Commonly used assays for VNTR and STR age of a blood group with another marker hav-
sequences involve the electrophoretic separa- ing a known chromosomal location.
tion of DNA fragments according tosize. DNA The gene encoding the antigens of the FY
profiling involves amplification of selected, blood group system was the first to be assigned
informative VNTR and STR loci using locus- to a chromosome, by showing that the gene is
specific oligonucleotide primers, and subse- linked to an inherited deformity of chromo-
quently measuring the size of the polymerase some l.so Subsequently, recombinant DNA
chain reaction {peR} products.Hundredsof STR methods were used to establish the physical lo-
loci have been mapped throughout the human cations of genes. The Human Genome Project,"
genome, and many have been applied to identi- completed in 2003, resulted in construction of a
ty testing. Analysisof different STRloci (usually physical gene map indicating the position of
at least 12) is used to generate a person's DNA gene loci, and the distance between loci is ex-
profile that is virtually guaranteed to be unique pressed by the number of base pairs of DNA.
to that person (or to two identical twins). DNA The 1000 Genomes Project,S2 completed in
fingerprinting is a powerful tool not only for 2015, created the largest public catalogue of hu-
identity testing and population genetics but also man genetic variation.
for monitoring chimerism after marrow trans- Currently, 39 blood group systems are recog-
plantatlon." STRanalysis also has been used to nized by the ISBT.1OThe genes for all of them
monitor patients for graft-vs-host disease after have been cloned and assigned to their respec-
organ transplantation, particularly after a liver tive chromosomes (Table 9-1). For the more re-
transplant." With the advent of next-genera- cently mapped blood group antigen genes, a
tion sequencing (NGS), SNVpanels are being variety of methods was used, including peptide
developed for use in genetic Identification." sequencing, massively parallel sequencing, and
In a case of disputed paternity, if an alleged biolnformatlcs approaches. Mapping of the VEL
father cannot be excluded from paternity, the blood group system was performed by two
probability of his paternity can be calculated. different groups. Storry et alused SNV·mapping
The calculation compares the probability that of 20 Vel-negativeindividuals in Sweden, which
the alleged father transmitted the paternal oblig- identified a region on chromosome 1 that
atory genes with the probability that any other tracked with the Vel-negative phenotype. 18
randomly selected man from the same racial or Sanger sequencing of several candidate genes
ethnic group transmitted the genes. The result is expressed in red cells identified a 17-base-pair
expressed as a likelihood ratio (paternity index) deletion inSMIMI in the Vel-negativeindividu-
or as a percentage. MBB has developed stan- als. Details on procedures for gene mapping are
dards and guidance documents for laboratories beyond the scope of this chapter, but reviews
that perform relationship testing." are available.12
GEN D each blood group system is genetically indepen-

GROU dent from every other blood group system and
represents a single gene or a cluster of two or
Gene and protein nomenclature is dictated by more homologous genes.
the Human Genome Organization (HUGO) The failure of an antibody to be reactive with
Gene Nomenclature Committee.f HUGO ap- red cells of a particular null phenotype is not suf-
proves both abbreviated gene symbols and lon- ficient for assignment of the corresponding anti-
ger descriptive names. Gene symbols are writ- gen to a system. Some null phenotypes are the
ten in italics (eg, !(EL). Genes may have one or result of inhibitor or modifying genes that may
more aliases, such as ACKRJ, which encodes suppress the expression of antigens from more
the FY system antigens, with aliases DARC, than one system leg, the Rhnull phenotype lacks
CD234, and FY, which is the ISBTgene name. A not only Rh antigens but also LW system anti-
Locus Reference Genomic (LRG)record, includ- gens, Fy5 antigen (FYsystem), and sometimes U
ing curated genomic, transcript, and protein ref- antigen (MNSsystem)]. Similarly,a blood group
erence sequences, is being developed for the antigen must be shown to be inherited through
blood group antigen genes." Coordinates of nu- family studies, or the expression of the antigen
cleotides within the coding region of a gene are must be demonstrated to be associated with a
numbered starting with the A of the translation- variation in the nucleotide sequence of the gene
al start codon ATG. Thus, the location of the controlling the system, to be assigned antigen
SNV in the FY gene that determines the Ft or status by the ISBTterminologyworking party. A
Fyb antigen expression is expressed FY c.125, blood group antigen must be defined serological-
such that a heterozygote can be written FY 1yby an antibody; a genetic variant that is de-
c.125A1G, with the nucleotide found in the ref- tectable only by DNA analysis and for which
erence sequence (in this case A) written first. there is no corresponding antibody cannot be
Coordinates of polypeptides are numbered start- called a blood group antigen.
ing with the first amino acid of the mature pro- The working party established a terminology
tein. Thus, the location of the amino acid deter- consisting of uppercase letters and Arabic nu-
minant of the Ft or Fybantigen is expressed FY merals to represent blood group systems and an-
p.Asp42Gly, with the amino acid found in the tigens.10,55 Each system can be identified by a set
reference protein (in this case Asp)written first. of numbers (eg, ABO system = 001; RH system
Blood group antigens were originallynamed = 004). Similarly,each antigen in the system is
using an alphabetical (eg, AlB, C/c) notation, assigned a number (eg, A antigen = 001; B anti-
or they were named after the proband whose gen = 002; D antigen = 001). ThUS,001001
red cells carried the antigen or who made the identifies the A antigen and 004001 identifies
first known antibody (eg, Duclos). A symbol the D antigen. Alternatively, the sinistral zeros
with a superscript letter (eg, Lu', Lub;Ik", Ikb) may be omitted so that the A antigen becomes
was used, and a numerical terminology (eg, Fy3, 1.1 and the D antigen becomes 4.1. Each sys-
Ik3, Rh32) was introduced. In blood group system also has an alphabetical abbreviation (Table
tems, antigens are named using more than one 9-1 gives the abbreviated system names, which
scheme (eg, the KELblood group system: K, k, are often analogous to the gene names); thus,
Is', Isb, K11, K17, TOU). KELis the ISBTsymbol for the Kellsystem, the
In 1980, the ISBT established its Working Rh ISBTsystem symbol is RH,and an alternative
Party on Terminology for Red Cell SurfaceAnti- name for the D antigen is RHl. This alphanu-
gens. 10 The working party was charged to devel- meric terminology, which was designed primari-
op a uniform nomenclature that would be "both ly for computer use, is not ideal for everyday
eye and machine readable" and "in keeping communication. To achieve uniformity, a
with the genetic basis of blood groups." A recommended list of user-friendly alternative
blood group system consists of one or more names was compiled."
antigens under the control of a single gene locus The ISBTworking party meets periodicallyto
or of two or more homologous genes. Thus, assign names and numbers to newly discovered
antigens. The working party is also charged to such as patient/donor blood transfusion incom-
develop, maintain, and monitor a terminology patibility or maternal-fetal incompatibility, and it
for blood group genes and their alleles; this is reo is the reason why antigen-negative blood is reo
fleeted byits current name, the Red Cell Immu- quired for safe transfusion in these patients.
nogenetics and Blood Group Terminology Work" Hemagglutination is simple, quick, and relative"
ing Party." The terminology takes into account ly inexpensive. When carried out correctly, it
the guidelines for human gene nomenclature has a specificityand sensitivity that is appropri-
published by HUGO, which is responsible for ate for most testing. However, hemagglutination
naming genes based on the International System has limitations;for example, it is difficult and of"
for Human Gene Nomenclature." For antigen ten impossible to obtain an accurate phenotype
terminology criteria;' tables listing the systems, for a recent transfusion recipient or to type red
antigens, and phenotypes; and information reo cells that are coated with tmmunoglobulla.G
garding the current status of gene and allele (IgG), and some typing reagents are in short sup"
terminology, see the ISBT Red Cell Immunoge- ply or not available. Because the genes encoding
netics and Blood Group Terminology web reo the 39 known blood group systems have been
sources." An example of ISBTterminology as it cloned and sequenced and the genetic bases of
applies to blood group names, alleles; pheno- most blood group antigens and phenotypes are
types, and antigens is shown in Table 9"4. Clini- known, severai of the more recent blood group
cal significanceof alloantibodies to specific anti" antigen gene discoveries involved the use of ge-
gens varies widely." nomic approaches,such as SNVanalysis forJR,IS
and NGSot Vel-negativeindividuals.19 Addition"
ally, targeted exome sequencing has been em"
B OD G UP GEN Res ployedto resolve problematic serologic cases."
DNA"based methods (genotyping) are in"
As discussed in earlier sections of this chapter, creasingly being used as an indirect method to
the antigens expressed on red cells are the prod" predict a blood group phenotype. This approach
ucts of genes and can be detected directly by has intrcducedblcod group genomics, .often.re-
hemagglutlnatlon techniques (as long as rele- ferred tq as.molesWar.i111111\l!l()h~matology, into
vant antisera are available).Their detection is an the practice of transfusion medicine. Prediction
important aspect of the practice of transfusion of a blood group antigen by testing DNA is
medicine because an antigen can,if itis intro- simple and reliable for the majority of antigens
duced into the circulation of an individual who because most result from SNVsthat are inherit"
lacks that antigen, elicit an immune response. ed in a straightforward Mendelian manner. For
It is the antibody from such an immune reo example, the antithetical antigens Sand s arise
sponse that causes problems in clinical practice, from' GYPB alleles that differby one nucleotide,
TABLE9·4. Examples of ISBT Terminology in Blood Group Systems
KEL K K+ KEL:1 KEL *01.01 Anti-K

Duffy FY Fy" FYi, or Fy(a+) FY:1 FY*01or Anti-Fy"
008001, or FY*A
8.1
l43T for S and l43C for s, and the resulting nucleotide level. Any PCR-basedtesting method
proteins differby one amino acid, methionine at can result in a false-negativeprediction due to
amino acid residue p.48 for S and threonine for failureto detect an allele resulting from the pres-
s (designated c.143T>C p.Met48Thr). ence of variation in the gene that inhibits ampli-
Detailed serologic and genetic studies, in- fication or detection of the variant being interro-
cluding whole genome sequencing, have shown gated. This is a phenomenon called allele
that there are far more alleles than phenotypes, dropout.
and this is especiallyrelevant clinicallyfor ABO Low-resolution molecular methods typi-
and Rh. Hundreds of allelesencoding the glycos- callyinterrogate an SNVor other geneticvariant
yltransferasesresponsible for the four ABOtypes in a DNA sample. This includes sequence-
have been identified, and a single nucleotide specific primer PCR (SSP-PCR), also known
variant in an A or B allele can result in an inac- as allele-specificPCR, where each DNA sample
tive transferase and a group 0 phenotype (see is amplified using two sets of primers, each of
Chapter 10). Testing for the common Rh anti- which amplifiesonly one allele, when the alleles
gens D, C/c, and E/e is uncomplicated for most differ by a single nucleotide. Another low-
populations, but antigen expression is more resolution method follows PCR amplification
complex in some ethnic groups. There are >500 with digestion of the resulting peR product us-
RHD alleles, including those encoding weak D ing a restriction enzyme [PCR-RFLP (restric-
or partial D phenotypes, and > 150 RHCE al- tion fragment length polymorphism)]. Both
leles, including those encoding altered, or novel, of these methods typicallyuse agarose gel elec-
hybrid Rh proteins, some of which result in trophoresis to separate the DNA fragments by
weakened antigen expression (see Chapter 11). size, and imaging to visualize the fragments.
RH genotyplng, particularly in minority popula- These methods are considered low throughput
tions, requires sampling of multiple regions of and are not easily automated. Real-time PCR
the genets) and algorithms for interpretation. uses fluorescent probes with quantitative or
qualitative readout. This method is less labor-
Molecular Methods for Predicting intensive and can be automated, yet each reac-
Blood Group Antigen Phenotypes tion typically interrogates only one SNV.Bead-
based or slide-based DNA arrays or single-base
Genomic DNA for molecular genotyping can be primer extension followed by matrix-assisted
isolated from any nucleated cell source. Periph- laser desorption/ionization time-of-flight
eral blood mononuclear cells (MNC) are by far (MALDI-TOF) mass spectrometry allows for
the most common specimen type used for pre- higher throughput genotypinginvolvinga multi-
dicting a red cell phenotype. If a patient has a plex PCR reaction with multiple gene targets
low MNC count that hampers DNA isolation amplified simultaneously and the ability to test
from peripheral blood, buccal swabs are an alter- multiple samples (typically48, 96, or 384) si-
nate source; this source is also useful for com- multaneously. These approaches allow for the
parison in marrow transplant recipients if anti- determination of numerous antigens in a single
body formation suggests a loss of engraftment. assay.
Complementary DNA (cDNA) can be made Medium-resolution molecular methods
from messenger RNA(mRNA)from red cells for include SNV genotyping panels that simultane-
investigation of splicingor for cloning of distinct ously interrogate multiple variants in a single
transcripts to determine if the multiple DNA gene. For blood group antigen genotyping,array-
variants identified in a blood group antigen gene based and mass-spectrometry-based assays are
are in cis or trans. used to interrogate many variants in a single
Most DNA-basedassaysinvolve amplification gene (eg, RHD) to obtain a more accurate pre-
of a target gene sequence through PCR, fol- diction, including weak and partial antigens as
lowed by one of a variety of downstream analy- well as low- and high-prevalence antigens. It is
ses. (See also Chapter 8.) Genotyping methods important to note that both low- and medium-
can be grouped by their level of resolution at the resolution genotyping approaches can result in
false-positive antigen predictions, because they further the understanding of the impact of ge-
would typically not detect a null mutation that netic variation on antigen expression as well as
would silence expression of an antigen. the frequency of such variation.
High-resolution molecular methods in-
volve interrogating, at the very least, the coding qi...
ical Applic~ti9n of the .Predidion of
region of a gene or genes, thus obtaining the nu- Blood Groups by lIIIolecularTes,~ing
cleotide sequence that can be used to predict
the amino acid sequence and splice sites. A A major use of DNAbased molecular testing is
commonly used high-resolution method is to predict the red cell phenotype of a fetus or of
Sanger sequencing, which can be performed us- a transfusion recipient, or when red cells are
ing DNA fragments generated by gene-specific coated with IgG.Additionalapplicationsinclude
PCR of genomic DNA or cDNA. When cDNA is the resolution of discrepancies in the ABO and
used as template, splice variants can be detect- RH systems and identification of the genetic ba-
ed. Massively parallel sequencing (MPS), sis of unusual serologicresults. Molecular DNA
which includes NGS, is a high-resolution meth- analysis also provides information to aid in dis-
od that can be used to sequence the whole ge- tinguishing alloantibodies from autoantibodies.
nome, the whole exome (covering all protein- This section gives an overview of some of the
coding regions), or a set of selected or targeted major applications of molecular testing that are
genomic regions (such as the genes critical for currently employed in patient and donor testing.
blood group antigen. expression). MPS is being These and additional. clinical applications are
employed in research and clinical settings, summarized in Table 9-5.
where it has the potential to revolutionize and
personalize diagnostics, prognostics, and selec- rvrolecular Testing to Predict the Bed Cell
tion of optimal therapies, including blood com- Phenotype
ponents." High-resolution approaches are more
accurate than low- and medium-resolution ap-
Recent Recipients of Transfusion. In patients
proaches but often involve a more complex receiving chronic or massive transfusions, the
analysis process and interpretation. These ap- presence of donor.red cells makes typing by
proaches can, also identify variants of un- hemagglutination inaccurate. Time-consuming
known significance (VUS)that can complicate and cumbersome cell separation methods that
a clinical interpretation. are often unsuccessful in isolating the patient's
Molecular immunohematology laboratories reticulocytes for typing can be avoided when
use methods that are similar to those used for DNA typingis used. PCR-basedassaysprimarily
HLA typing, that is, low-resolution typing of one use DNA extracted from MNCs isolated from a
or a few SNVs as well as high-resolution se- sample of peripheral blood. Interference from
quencing of entire gene regions. High-resolution donor-derived DNA is avoided by targeting and
methods are often used to investigate new al- amplifyinga region of the gene that is common
leles and resolve discordant findings comparing to all allelesso that the minute quantity of donor
serologic and molecular results. The application j)t'-Ji\is not detes.ted. This approach makes pos-
of these methods has been reviewed by several sible reliable blood group determination with
groupS.61.64 DNA prepared from a blood sample collected af-
Public databases of blood group antigen vari- ter transfusion. DNA isolated from a buccal
ation are critical to the sharing of information swab or saliva is also suitable for testing. In
for both research and clinical purposes. Besides transfusion-dependent patients who produce al-
the blood .group allele tables managed by the loanjjbodies, an extended antigen profile Is im-
ISBT working party that include phenotype in- portant to determine additional blood group an-
formation, when available, there are otherre- tigen~ to which the patient can become
sources, either specific to blood group a11ti- sensitized.
gens65.67 or more general," that are ;useful In the past, when a patient with autoim-
catalogues of human variation. These resources mune hemolytic anemia received transfusion
TABLE9-5. Applications of Molecular Testing to Predict Red Cell Antigens in Patients and Donors
To predict a patient's red cell phenotype:
After a recenttransfusion
- Aid in antibody identification and RSCunit selection
- Select red cells for adsorption
fj When antibody typing reagent is not available(eg, anti-Do", _DOb,-Js", -V, -VS)
III Distinguish an alloantibody from an autoantibody (eg, anti-e. antl-Kp'')
(il Help identify alloantibody when a patient's type is antigen-positive and a variant phenotype is possible (eg,
antl-D in a D-positive patient, antl-e in an e-positive patient)
® When the patient's red cells are coated with immunoglobulin (DAT+)
- When direct-agglutinating antibodies are not available
- When the antigen is sensitive to the IgG removal treatment (eg, antigens in the Kell system are denatured
by EDTA-glycine-acidelution)
- When testing requires the indirect antiglobulin test and IgG removal techniques are not effective at remov-
ing cell-bound immunoglobulin
- When antisera are weakly reactiveand reaction is difficult to interpret (eg, antl-Do', anti-Do'; anti-Fyb)
€I Obtain a patient's phenotype before or after administration of monoclonal antibody therapeutic drugs
(eg, anti-CD38, anti-CD47, daratumumab) that cause interferences with serologic testing methods
e After allogeneic stem cell transplantation [If an antibody problem arises, test stored DNAsamples (or buccal
swab) from the patient and the donor(s) to guide selection of units for transfusion]
• To detect weakly expressedantigens (eg, Fybwith the Fyxphenotype)
® Identify genetic basis of unusual serologic results, especially Rh variants
III Resolveblood group discrepancies (eg, A, S, and Rh)
11) Aid in the resolution of complex serologic investigations, especiallythose involving high-prevalenceantigens
when reagentsare not available
Ij) Identify if a fetus is at risk for HDFN(mother with antl-D; father homozygous or heterozygousfor RHD)
To predict a donor's red cell phenotype:
Screenfor antigen-negative donors

.. When antibody is weak or not available (eg, anfi-Do', _DOb;-Js', -Js"; -VNS)
.. Mass screening to increase antigen-negative inventory
<» Find donors whose red cells lack a high-prevalenceantigen
® Identify donors with RHvariant alleles that express partial Rh antigens (for Rh allele-matching to patients)
8 Resolve blood group discrepancies (eg, A, S, and Rh)
0& Resolveantigen-typing discrepancies caused by alleles encoding weak or partial antigens
II; Type donors for reagentred cells for antibody screening cells and antibody identification panels (eg, Doa,Do",
Jsa,V, VS)
<Ii Determinezygosity of donors on antibody detection/identification reagent panels, especially D, S, Fya,and Fyb
RBCs= Red Blood Cells; DAT = direct antiglobulin test; HDFN= hemolytic diseaseof the fetus and newborn.
before the patient's red cell phenotype for amounts of CD47 expression on the red cell. Al-
minor antigens was established, time- and so, in some cases a patient who is sensitized
resource-consuming differential allogeneic ad- may cause an initial positive DAT that may be
sorptions were required to determine the pres- negative upon subsequent testing. The detec-
ence or absence of alloantibodies underlying tion of MoAbs is dependent on when the last
the autoantibody. Establishing the patient's dose was received in relation to pretransfusion
most probable phenotype through molecular testing, on the patient disease, and on serologic
testing makes it possible to match the antigen method used."
profile of the adsorbing red cells to that of the Many laboratories use thiol-treated red cells
patient, thereby reducing the number of cell to test the patient's serum/plasma after anti-
types required for adsorption. This approach CD38 administration; however, they are unable
also allows matching of the antigen profile of to rule out antibodies to the KEL,LU, KN, LW,
the donor to that of the. patient for the most and YTblood group system antigens. Currently,
clinically significant, common antigens (eg, Rh; several anti·CD47 clones are in clinical trial.
Ika, Ikb, S, s) when transfusion is required. The interference due to HU5F9-G4 can be
Transfusion of units that are antigen-matched
avoidedusing a monoclonal anti·lgGthat is non-
for clinically significant blood group antigens
reactive with IgG4 or by allogeneic platelet or
prevents delayed transfusion reactions and
red cell adsorption." The development of meth-
avoids additional alloimmunization.
When Red Cells Are Coated with IgG. In ods to negate interference by different clones of
patients with or without autoimmune hemolytic anti·CD47is under way as the trial continues.
anemia, the presence of immunoglobulin bound Molecular testing may help with mitigating
to the red cells [positive result on direct anti- the MoAb interferences with serologic testing
globulin testing (OAT)]often makes antigen typ- by obtaining a baseline genotype before the pa-
ing results by serologic methods invalid. Cer- tient begins drug therapy. Molecular testing may
tain methods, such as treatment of the red cells also be used to predict a patient's phenotype
with chloroquine diphosphate or EDTA-glycine after drug therapy is already started if whole
acid (EGA), may be employed to remove the blood or buffy coat is used for DNA extraction.
red-cell-bound IgG. These methods are not al- The patient's extended phenotype can be used
ways successful or accurate"; the antigen of in- to predict the potential antibodies that the pa-
terest may be denatured by the treatment (eg, tient might make. Phenotype-similar red cells
EGA destroys antigens of the Kell blood group can be provided for transfusion while the pa-
system), and direct agglutinating antibodies for tient is receivingMoAbdrug treatments."
the antigen of interest may not be available.Mo-
lecular testing allows determination of an ex- DNA-Based Assays to Distinguish
tended antigen profile to select antigen-negative Alloantibody from Autoantibody
RBCunits for transfusion that match the antigen
profile. When an antibody specificityis found in a pa-
When Red Cells Are Coated with Thera- tient whose red cells express the corresponding
peutic Drugs. As a result of the increasing suc- antigen, it is essential to know whether the anti-
cess of treating patients with monoclonal antibody is an allo-or autoantibody, and molecular
body (MoAb)therapeutic drugs (eg, anti·CD38, testing is helpful for transfusion management. If
anti·CD47), there has been an increase in inter- molecular testing predicts the red cells to be an-
feringsubstance detection, causingseveral serol- tigen positive, further investigation of the blood
oW! methods to become invalid. For example, group antigen coding regions by a higher-
anti-CD38 is broadly reactive when performing resolution molecular method such as Sanger se·
antibody detection with an indirect antiglobulin quencing should be considered, because the
test (IAT),due to the expression of CD38 on the sample may have an amino acid change from
red cell, and anti-CD47 presents with a broader the reference sequence in the protein carrying
reactivity range than anti·CD38 due to its large the antigen. Amino acid changes can result in
new epitopes or altered (weakened or partial) (HDFN).Antigen prediction by molecular meth-

expression of the conventional antigen. ods can be used to identify the fetus who is not
This is especially relevant in patients with at risk of HDFN (ie, who is predicted to be anti-
sickle cell disease (SCD)or thalassemia who re- gen negative) so that the mother need not be ag-
quire long-term transfusion support and are at gressively monitored. Testing of fetal DNA
risk of alloimmunization that is often complicat· should be considered when a mother's serum
ed by the presence of antibodies. In patients of contains an IgG alloantibody that has been asso-
African ancestry who have SCD, partial expres- ciated with HDFN and the father's status for the
sion of common Rh antigens (D, C, c, and e) is corresponding antigen is heterozygous or inde-
prevalent. Such patients frequently present with terminable, or he is not available for testing. For
a combination of anti-D, -C, and -e, and yet their more detail on perinatal issues, including HD-
red cells type serologically as D+, C+, and/or FN, see Chapter 23.
N. Although such patients may make alloanti- Molecular Testing to Identify a Fetus at
bodies to these antigens, autoantibody produc- Risk for HDFN. The first application of a DNA-
tion with Rh-related specificityis common, and based method for the prediction of blood group
distinguishing between the two is critical for phenotype occurred in the prenatal setting and
safe transfusion practice to avoid hemolytic was reported by Bennett et al," who tested fetal
transfusion reactions (and conserve rare DNA for the presence of RHD. Because of the
blood).73,74Delayed hemolytic transfusion reac- clinical significance of anti-D, RHD is probably
tions, in particular, places patients with SCD at the most frequent target gene for fetal testing,
risk for life-threatening anemia, pain crisis, acute but molecular testing can be used to predict the
chest syndrome, and/or acute renal failure. Pa- antigen type of the fetus for any antigen if the
tients may also experience hyperhemolysis, in genetic basis is known. When the implicated
which hemoglobin levels decline below pre- IgG antibody in the maternal circulation is not
transfusion levels as a result of bystander hemol- antl-D, it is prudent, when possible, to also test
ysis of the patients' own antigen-negative red the fetal DNA for RHD to preempt unnecessary
cells. RH genotyping has revealed that many of requests for D- blood for intrauterine transfu-
these patients have variant RHD and/or RHCE sion; this is particularly relevant to avoid the use
alleles that encode amino acid changes in Rh of rare rr' or rr" blood when anti-c or anti-e is
proteins, resulting in altered or partial antigens. the implicated antibody.
For details on RHD and RHCE alleles that en- PCR analyses for the prediction of fetal D
code partial antigens, refer to Chapter 11. phenotype are based on detecting the presence
Reports of autoantibodies to Jka and Jkb are or absence of specificportions of RHD. In popu-
not uncommon. With the discoveryof variant JK lations of European ancestry, the genetic basisof
alleles that encode partial Jk"and Jkbantigens, it the D- phenotype is usually associated with de-
is probable that some previously identified auto- letion of the entire RHD, but several other mo-
antibodies were alloantibodies. (The JK system lecular bases have been described. In popula-
is discussed in Chapter 12.) DNA analysisfor JK tions of Asian ancestry, 15% to 30% of D-
variants is helpful to clarify the situation. As in people have an intact but inactive RHD, while
some other blood group systems, JK system ge- others with red cells that are nonreactive with
netic diversityis higher in populations ofAfrican antt-D have the Del phenotype. Approximatelya
ancestry. quarter of D- people of African ancestry have an
inactive RHD gene (RHD'P), which does not en-
Molecular Testing in Prenatal Practice code the D antigen, and many others have a hy-
brid RHD-CE-D gene (eg, the r's phenotype).
DNA-basedtesting has affected prenatal practice (For details on the RH system, see Chapter 11.)
in the areas that are discussed below. Hemagglu- Predicting the D type by molecular methods re-
tination, including antibody titers, gives only an quires probing for multiple nucleotide changes.
indirect indication of the risk and severity of he- The choice of assays depends on the patient's
molytic disease of the fetus and newborn ethnicity and the degree of discrimination de-
sired." Establishing the fetal KEL genotype is whose red cells lack some epitopes of D (partial
also of great clinical value in determining D) and are at risk for D immunization, from
whether a fetus is at risk for severe anemia, be- those with a weak D phenotype who are not at
cause the strength of the mother's Kantibody of· risk for D immunization. Red cells with partial
ten does not correlate with the severity of the D may type as D+, some by direct tests and oth-
infant's anemia. The same is true for anti-Ges." ers by indirect tests. These women might bene-
The risk of HDFN can be predicted by deter- fit from receiving RhIGprophylaxis if they carry
mining whether the fetus carries the antigen a D+ fetus. RHD genotyping can distinguish D
that corresponds to the mother's antibody. If the variants to guide RhIG prophylaxis and blood
father demonstrates heterozygous expression of transfusion recommendations." RHD genotyp-
the antigen or if antigen expression is indeter- ing is recommended in females of childbearing
minable, testing of fetal DNA should be consid· potential, to avoid treatment of females with
ered. If the fetus is predicted not to carry an an- weak D types 1, 2, or 3 who have little or no
tigen against the mother's antibody, the invasive risk of RhD alloimmunization, as well as to iden-
and expensive monitoring of the mother may tify women with other D variants, including
not be necessary. women of Africanancestry, in whom partial D is
Amniocytes, harvested from amniotic fluid, common and who may benefit from Rh immu-
are the most common source of fetal DNA. Cho- noprophylaxts.v-"
rionic villus sampling and cordocentesis are not Molecular Testing of Paternal Samples.
favored because of their more invasive nature The father's red cells should be tested for the an-
and associated risk to the fetus. A noninvasive tigen corresponding to the antibody in the ma-
sample source is the cell-free fetal DNA that is ternal plasma. If the red cells are negative, the
present in maternal plasma as early as 5 weeks fetus is not at risk. If the father is positive for the
of gestation; the amount of DNA increases with antigen, zygosity testing can be performed. Zy-
gestational age, and reliable results in DNA- gosity describes the number and similarity of
based assays are obtained starting at about 15 two alleles; whereas homozygous or heterozy-
weeks of gestation (sometimes earlier, depend- gous individuals carry two copies, in scenarios
ing on the gene of interest]." Becausefetal DNA where there is copy number variation, individu-
is generally composed of shorter fragments, the als can be hemizygous, or only carry a single
testing capability may be limited compared to gene copy.
cellular DNA.79 However, these assays are par- Zygosity testing of paternal samples is most
ticularly successful for D typing because the D- often performed when testing for possible
phenotype in the majority of samples is due to HDFN due to anti-D or anti-K. If the paternal
the absence of the RHD gene. red cells are K+ and the mother has anti-K,they
Testing for the presence or absence of a gene can be tested serologicallyfor expression of the
is less demanding than testing for a single gene allelic k antigen. However, many centers do not
polymorphism or SNV to predict, for example, have a licensed reagent available, such that ge-
the K/k antigen status. Cell-freefetal DNA from netic counselors often request DNA testing of
the maternal plasma is routinely used in some the father to predict K antigen status. If the pa-
European countnes=" to test for the presence ternal red cells are K-, the maternal antl-K is
of a fetal RHD gene to eliminate the unnecessary most likely the result of immunization through
administration of antepartum Rh Immune Glob- transfusion, or pregnancy by another partner.
ulin (RhIG)to the - 40% of D- women who are When molecular methods are used to deter-
carrying a D- fetus." In a multiethnic popula- mine paternal RHD zygosity,several different ge-
tion, accuracy of noninvasive fetal RHD geno- netic events cause a D- phenotype, and multiple
typing requires distinguishingnormal RHD from assays are often used to accurately determine
hybrid or inactive alleles." RHD zygosity,especiallyin non-European ethnic
RHD Genotyping to Determine D Anti- groups. If the father is homozygous (carries two
gen Status in Pregnant Women. Serologic functional RHD alleles), all of his children will
typing for D cannot easily distinguish women be D+, and any pregnancy in his partner needs
to be monitored. If the father is heterozygous, ods are typically focused on ruling out variant
the fetus has a 50% chance of being at risk. De- antigens that are known to be associated with
termining the D type of the fetus can prevent variability in antigen typing. Examples of dis-
unnecessary monitoring of the pregnancy and crepancies that have been resolved using molec-
use of immune-modulating agents. ular methods are listed in Table 9-6. A detailed
discussion of causes of both ABO and RhD typ-
Using Molecular Methods to Screen for ing discrepancies is beyond the scope of this
Antigen-Negative Blood Donors chapter; several publications describe approach-
es to handling such findings.86.88
DNA-based typing to predict donor antigen pro-
files in the search for antigen-negative units is
Molecular Testing to Confirm the D Type of
routine for most blood centers, especially when
Donors
suitable antibodies are not avallable. Because
red cell typing for DO antigens is notoriously dif- Donor centers must serologically test donors for
ficult, one of the most frequent approaches is to weak D to avoid labeling a product as D- that
type for Do"and DOb.Many other antibody spec- might result in anti-D in response to transfused
ificities are unavailable for mass donor screen- RBCs. Some donor red cells with very weak D
ing. These speciflclties include anti-Hy, -Io", expression (weak D type 2, and especially those
-Jsa, -Jsb,-Cw, -V,and -VS. with the Del phenotype) are not typed as D+ us-
Red cell genotyping panels are used by many ing current methods and are labeled as D-. The
blood donor centers to screen for multiple mi- prevalence of weak D red cells not detected by
nor antigens in a single assay format and have serologic reagents is approximately 0.1 % (but
been used for mass screening of blood donors." may vary depending on the test method and
Several platforms licensed by the Food and Drug population). Although the clinical Significance
Administration (FDA)are now available. The re- has not been established, donor red cells with
sults can be used for the labeling of donor units weak D expression have been associated with
for extended antigen profiles. This practice not alloimmunization. RHD genotyping could be
only increases the antigen-negative inventory by used to confirm blood donors who type D_,89
expanding combinations of the minor antigens but a high-throughput and cost-effective plat-
and some high-prevalence antigens, but also al- form is not yet avallable.
lows provision of donor components that are
molecularly matched to the patient's type. li- Molecular Testing to Match Red Cell Donors
censed genotyping tests that predict ABO and to Rh-Alloimmunized Patients Carrying Rh
RhD are not avallable for the labeling of donor Variants
units.
Rh variant antigen expression is common in in-
dividuals of African ancestry.P:" Antibodies to
Using Molecular Methods to Resolve
variant Rh antigens, including anti-hr" and anti-
Serologic Typing Discrepancies
hr", have been associated with insufficient de-
It is not uncommon in either the donor center crease in sickle hemoglobin (HbS) after ex-
or hospital blood bank setting to have a discrep- change transfusion in patients with SCD, as well
ancy between two serologic typings of a donor as with transfusion reactions and/or monocyte
or patient. The discrepancy may involve the cur- monolayer assay incompatibility. In some cases,
rent vs a historical type, or two typings on a cur- patients have benefited from transfusion with
rent sample. In the latter scenario, the discrep- RBC units matched to their RH alleles. RH al-
ancy may involve the use of two different lele-matching describes the process of selecting
methods (eg, solid-phase vs tube testing) or two blood donors based on the RHCE and RHD al-
different reagents (eg, monoclonal vs polyclonal leles of the patient, using a tiered system." This
antisera). Molecular methods can be employed approach has yet to be implemented on a large
as part of discrepancy investigations. Such meth- scale but has been modeled.93,94
TABLE9·6. Examples of Serologic Discrepancies Resolved Using Molecular Methods
D- prevl- RHDarray
ously, currently types and variants
weak D+ RHD cDNAanalysis cDNAanalysis identified
RHDc.19T (p.7W) associ-
ated with weak D type 18
RH Sampletyped C+ previ- RBCpanel, RHC£ RHC£ array identified
ously, currently types C- and RHC£*ceT/; RHD array
RHDarrays identified RHD*DIV; this
haplotype is associated
with anti-C reactivity
FY Sampletyped Fy(b-) RBCpanel FY*OlIFY*02W.Ol
previously; now types Fy(a+b+w)
Fy(b+w)with one antisera,
Fy(b-) with another, and
Fy(b+) with a third
JK Sampletyped Jk(a+) pre- SSP-PCRfor JK JK*Ol W.OlIJK*02
viously, now types Jk(a-) c.130G/A (Jka+wb+)
ssp-peR = sequence-specific primer polymerase chain reaction.
ABO Genotyping ofSo/id-Organ Donors verse ethnicities are tested. Causes of discrepan-
cies include recent transfusions, stem cell trans-
Due to insufficient ABO-compatibledonors for
plantation, and natural chimerism. Stem cell
kidney transplantation, ABO-incompatibletrans-
transplantation and natural chimerism also may
plantations have become routine. A recent cause differences between the results of testing
meta-analysis showed good, albeit inferior, out- DNA from somatic cells and results of testing
comes of such transplantations." Transplanta- DNA extracted from peripheral MNCs. Thus,
tion of other ABO-incompatible solid organs is when using molecular methods, it is important
on the rise." Living-donor registries that use to obtain an accurate medical history. Many ge-
buccal swabs (instead of peripheral blood) for netic events can cause apparent discrepant re-
initial DNA testing can use ABO genotyping sults between hemaggiutination and molecular
strategies to predict the blood type. ABO geno- test results (Table9-7). Weak antigen expression
typing is also useful for predicting the blood type may not be detected by hemagglutination, and
of recently deceased individuals of unknown the genotype may not always predict the pheno-
ABOtype who received massive transfusion. type." Table 9-8 lists some important consider-
ations when resolving discrepancies between se-
Discrepancies between Serologic (Phenotype) rologic and molecular testing.
and Molecular (Genotype) Testing
Differences between serologic and DNA testing Challenges in Using SNV Genotyping to
results do occur and should be investigated. Of- Predict Red Cell Phenotypes
ten, these discrepancies lead to interesting dis- Molecular testing interrogates a single SNV or a
coveries such as the presence of a novel allele or few SNVs associated wlth antigen expression
genetic variant, particularly when people of dl- and cannot sample every nucleotide in the gene.
TABLE 9-7. Examples of Some Molecular Events Where Analysis of Gene and Phenotype May Not
Agree
Alternative splicing Nt changein splice site: partial/ S-s-; Gy(a-)

complete skipping of exon
Prematurestop codon Deletionof nt(s) -7 frameshift Fy(a-b-); D-; c-E-; RhnulI; Gy(a-);
GE:-2,-3,-4; Ko;McLeod
Insertion of nt(s) -7 frameshift D-; Co(a-b-)
Nt change Fy(a-b-); r': Gy(a-); Ko;McLeod
Amino acid variant Missensenucleotidevariant
Reducedamount of Missensenucleotidevariant FyX;Co(a-b-)
protein
Hybrid genes Crossing over GP.Vw;GP.Hil;GP.TSEN
Geneconversion GP.Mur; GP.Hop;D--
Interacting protein Absenceof RhAG

Absenceof Kx Weak expression of KELantigens
Modifying gene KLF1 mutations In(Lu) phenotype
~~~."~~~- .~~~~-~~- ~-~~--~~~~--~~~~~~-
Nt = nucleotide.
Although a gene may be detected by DNA test- To avoid misinterpretation, routine assays
ing, there are times when the gene product is must include appropriate tests to detect variants
not expressed on the red cells, because of a mu- that silence gene expression if prevalent in the
tation that silences the gene or reduces expres- population tested. Silenced alleles can be specif-
sion levels, and it is not detected by routine ic to a particular ethnic group. For example, in
hemagglutination testing. Such changes result in the FYblood group system, an SNV(c.-67T>C)
discrepancies in the typing of patients and do- within the promoter region (GATAbox) of FY
nors. Homozygosity (or compound heterozygos- prevents transcription of FY*A and/or FY*B in
ity) for a silenced gene results in a null pheno- red cells but not in other tissues. Although si-
lencing of FY*A is rare, silencing of FY*B is fre-
type, and most null phenotypes have more than
quent in persons of African ancestry where ho-
one genetic basis."
mozygosityfor the -67T>C SNVin FY*B results
With donor typing, the presence of a grossly
in the Fy(a-b-) phenotype, which has a preva-
normal gene whose product is not expressed on
lence of 60% or higher. To ensure accuracy, test-
the red cell surface results in the donor being
ing for the GATAbox variant must be included
falsely typed as antigen positive. Although this in typing for FYin persons of Africanancestry.
situation means loss of an antigen-negative do- When an assay is used to predict the pres-
nor, it does not jeopardize the safety of blood ence or absence of D antigen, particularly in
transfusion. However, if a grosslynormal gene is populations of African ancestry, it is essential to
detected in a patient but the gene is not ex- include a test for the complete but inactive RHD
pressed, the patient remains at risk for the corre- gene RHD'F. If an assay is designed to predict S
sponding antibody if he or she receives a trans- and s antigen expression in persons of African
fusion of antigen-positiveblood. ancestry, it should include interrogation of a
TABLE 9·8. Common Methods for Resolving Discrepancies between Serologic (Phenotype) and
Molecular (Genotype) Testing
'" Some common factors that can leadto discordant results:

- The number of epltopes present on the red cell at testing
- The use of human, polyclonal, or monoclonal reagents on historical typing
- Process errors during testing
- Clerical errors during computer entry
Review of serologic and molecular results entry to verify no clerical errors:

- Data entry errors
- Sample labeled with the incorrect patient/donor information
- Sample tubes switched during testing and/or DNAextraction
'" Clerical error detected:

- Invalidate the test result(s); order and result the corrected interpretation
- Obtain a new properly labeled sample with the correct patient/donor information
- Repeatsample testing and/or DNA extraction if a switched sample tube(s) is suspected
Repeatserologic testing:
- If the sample is available, repeatserologic testing on the current sample
- Repeattesting of the current sample with a different source of serologic reagent(s)
- If the sample is not available for repeat testing, try to obtain another sample from the patient/donor for
additional testing upon subsequent donation or next encounter
'" Repeat molecular testing:

- If the sample is available, repeat molecular testing on the current sample and indicate if a new DNA
extraction was used for repeat sample testing
- Repeattesting of the current sample with a different molecular platform, if available
- Sample may needto be sent to an outside laboratory for further testing (eg, sequencing)
Determination of final results:

- If the repeattesting resolved the discrepancy, no further action is required
- If repeat testing does not resolve the discrepancy, notify management team, including ellA laboratory
director or designeefor further instructions
C>T SNVat nucleotide c.230 in GYP*B exon 5 ed with the loss of JK antigen expression occur
or an SNV in intron 5 (+5g>t) because both more often in people of Asian ancestry, whereas
SNVsprevent expression of S antigen and result nucleotide variants encoding amino acid vari-
in expression of UVAR• ant that weaken JK expression occur in people
Other common causes of discrepancies in- of African ancestry.
clude the presence in the sample of an altered
FY*B allele that encodes an amino acid variant
causing the FyXphenotype with greatly reduced u
expression of Fybantigen. The red cells type as
Fy(b-) with most serologic reagents. The preva- Blood group genetics has become an essential
lence of the allele encoding the Fyxphenotype component of the practice of transfusion medi-
in people of European ancestry is as high as 2%, cine. The field of molecular immunohematology
and the allele has been found in persons of Afri- has provided a greater understanding of genetic
can ancestry also. Silencing mutations associat- blood group variants, including the complexity
of Rh variants and the associated partial Rh anti- cipients, especially those with hemoglobinopa-
gens. Molecular methods used to predict blood thies, is growing. Increasingly, RHD genotyping
group antigen expression vary in their through- is being used to resolve discrepancies and assess
put capabilities and resolution. Genotyping as- candidacy for RhIG. Blood group genomics, in-
says that can predict many antigens in multiple cluding the use of massively parallel sequencing,
samples simultaneously are being used routinely is poised to provide even richer information
by blood centers to identify red cell donors lack- about patient and donor red cell phenotypes
ing antigens of clinical significance. The use of that has the potential to make transfusion medi-
these red cell panels in chronic transfusion re- cine even more personalized.
KEY POINTS
1. Genetics is the study of heredity; that is, the mechanisms by which a trait, such as expression
of a blood group antigen, is passed from parents to offspring.
2. A gene is a segment of DNA and is the basic unit of inheritance; it occupies a specific location
on a chromosome (the gene locus).
3. Alleles are alternative forms of a gene (eg, alleles JK*A and JK*B are different forms of
SLC14Al that encode the Jka and Jkbantigens, respectively).
4. A human somatic cell is diploid, containing 46 chromosomes in 23 pairs: 22 palrs are auto-
somes, and the remalning pair are the sex chromosomes, with males carrying X and Y, and fe-
males, two X chromosomes.
5. Somatic cells divide by mitosis, where the chromosomes replicate and are divided into two
new diploid cells that have all the genetic information of the parent cell.
6. Germ cells divide by meiosis,where after chromosomalreplication and two diviSions,four haploid
gametes are formed, each having half the chromosomalcomplement of the parent somatic cell.
7. The term "genotype" traditionally refers to the complement of genes inherited by each person
from his or her parents; the term is also used to refer to the set of alleles at a single gene locus.
Whereas the genotype of a person is his or her genetic constitution, the phenotype is the observ-
able expression of the genes and reflects the biologicactivity of the gene(s). Thus, the presence or
absence of antigens on the red cells, as determined by serologictesting, represents the phenotype.
8. When identical alleles for a given locus are present on both chromosomes, a person is homozy-
gous; when nonidentical alleles are present at a particular locus, the person is heterozygous;
and a person carrying only one allele copy is hemizygous.
9. The expression of blood group antigens on the red cell may be modified or affected by gene in-
teractions. This can involve structural interactions (eg, RHAG gene mutations associated with
Rhnull phenotype) or transcriptional interactions leg, KLFl mutations associated with In(Lu)
phenotype].
10. There are multiple types of genetic variation, including single nucleotide, insertion/deletion,
and structural variants. SNVs can result in changes to amino acids, cause premature stop co-
dons, and alter messenger RNAsplicing, all of which can alter blood group antigen expression.
11. A blood group system consists of one or more antigens under the control of a single gene locus
(eg, KEL encodes the Kellblood group antigens) or of two or more homologous genes (eg, RHD
and RHCE encode the Rh blood group antigens). ThUS,each blood group system is genetically
independent. Currently, 39 blood group systems are recognized.
12. The genes encoding the blood group systems have been identified, and the genetic bases of
most antigens and phenotypes are known.
13. DNA-based assays ofvarying resolUtion and throughput are being used by molecular immuno-
hematology laboratories to predict antigen status in both donors and patients.
REFERENCES
1. Brown TA. Introduction to genetics:A molecular 13. Uigdberg L, Reid ME, ZelinskiT. Human blood
approach. London, UK:Garland Science, 2011. group genes 2010: Chromosomal locations and
2. Lesk A. Introduction to genomics. 3rd ed. New cloning strategies revisited. Transfus Med Rev
York:Oxford University Press, 2017. 2011;25:36-46.
3. Reid ME, Shine 1. The discovery and Signifi- 14. Svensson L, Hult AK, Stamps R, et al. Forssman
cance of the blood groups. Cambridge, MA: SBB expression on human erythrocytes: Biochemical
Books,2012. and genetic evidence of a new histo-blood group
4. Schurr TG. The peopling of the New World: system. Blood2013; 121:1459-68.
Perspectives from molecular anthropology. 15. Zelinski T, Coghlan G, Liu XO, et al. ABCG2
Annu RevAnthropoI2004;33:551·83. null alleles define the Jr(a-) blood group pheno-
5,.Pierron D, Heiske M, Razafindrazaka H, et al. type. Nat Genet 2012;44:131-2.
-Strong selection during the last millennium for 16. Saison C, HeliasV, BallifBA,et al. Null allelesof
"African ancestry in the admixed population of ABCG2 encoding the breast cancer resistance
Madagascar. Nat Commun 2018;9(1):932. protein define the new blood group system Ju-
6. Clark DP, Russell LD.Molecular.biology: Made nior. Nat Genet 2012;44: 174-7.
simple and fun. St. Louis, MO: Cache River 17. HeliasV, Saison C, BallifBA,et al. ABCB6is dis-
Press, 2010. pensable for erythropoiesis and specifies the
7. Reid ME, Denomme GA..DNA-basedmethods new blood group system Langereis. Nat Genet
in the immunohematology reference laboratory. 2012;44:170-3.
TransfusApher Sci 2011 ;44:65-7. 18. Storry JR, loud M, Christophersen MK, et al.
8. Nussbaum RL, McInnes RR, Willard HF. Homozygosityfor a null allele of SMIMI defines
Thompson & Thompson genetics in medicine. the Vel-negative blood group phenotype. Nat
8th ed. Philadelphia: Elsevier/Saunders, 2016. Genet 2013;45:537-41.
9. Reid ME, Lomas-Francis C, Olsson ML. The 19. Cvejic A, Haer-Wigman L, Stephens JC, et al.
blood group antigen factsbook. 3rd ed. San Di- SMIM1 underlies the Vel blood group and influ-
ego, CA: Academic Press, 2012. ences red cell traits. Nat Genet 2013;45:542-5.
10. International Society of Blood Transfusion. Red 20. BallifBA,HeliasV, PeyrardT, et al. Disruption of
Cell Immunogenetics and Blood Group Termi- SMIM1 causes the Vel- blood type. EMBOMol
nology (working group). Blood group terminolo- Med 2013;5:751-61.
gy. Amsterdam, the Netherlands: ISBT, 2019. 21. Anliker, M, von Zabern I, Hochsmann B, et al. A
[Availableat http://www.isbtweb.orglworking- new blood group antigen is defined by anti-
parties/red -cell-immunogenetics-and -blood- CD59, detected in a CD59 deficient patient.
group-terminology/ (accessed October 4, Transfusion 2014;54: 1817-22.
2019)·1 22. Daniels G, BallifBA, HeliasV, et al. Lack of the
11. An international system for human cytogenetic nucleoside transporter ENTl results in the
nomenclature (1978) ISCN (1978). Report of Augustine-null blood type and ectopic mineral-
the Standing Committee on Human Cytogenet- ization. Blood2015;125:3651-4.
ic Nomenclature. Cytogenet Cell Genet 1978; 23. LyonMF. Xchromosome inactivation. Curr Btol
21:309-404. 1999;9:R235-R237.
12. Lewis M, Zelinski T. Linkage relationships and 24. Daniels G. Human blood groups. 3rd ed. Ox-
gene mapping of human blood group loci. In: ford, UK:BlackwellSCience,2013.
Cartron l-P, Rouger P, eds. Molecular basis of 25. Clemson CM, Hall LL, Byron M, et al. The X
major human blood group antigens. New York: chromosome is organized into a gene-rich outer
Plenum Press, 1995:445-75. rim and an internal core containing silenced
non genic sequences. Proc Natl Acad Sci USA 39. Cho D, LeeJS, YazerMH, et al. Chimerism and
2006; 103:7688-93. mosalcism are important causes of ABOpheno-
26. Redman CM, ReidME. The McLeod syndrome: type and genotype discrepancies. Immunohema-
An example of the value of integrating clinical tology 2006;22: 183-7.
and molecular studies. Transfusion 2002;42: 40. SingletonBK,Burton NM, Green C, et al. Muta-
284-6. tions in EKLF/KLF1form the molecular basis of
27. Russo DCW, Lee S, Reid ME, Redman CM. the rare blood group In(Lu) phenotype. Blood
Point mutations causing the McLeod pheno- 2008; 112:2081-8.
type. Transfusion2002;42:287-93. 41. Singleton BK,RoxbyD, StirlingJ, et al. A novel
28. Garratty G. Blood groups and disease:A histori- GATA-1mutation (Ter414Arg)in a familywith
cal perspective. Transfus Med Rev 2000; 14: the rare X-linkedblood group Lu(a-b-) pheno-
291-301. type (abstract).Blood2009; 114:783.
29. Thorisson GA, Stein LD. The SNP consortium 42. Pena SDJ, Chakraborty R. Paternity testing in
website: Past, present and future. Nucleic Acids the DNAera. Trends Genet 1994;10:204-9.
Res2003;31:124-7. 43. JeffreysAJ, Wilson V, Thein SL. Hypervariable
30. Blumenfeld 00, Patnaik SK. Alielic genes of 'minisatellite' regions in human DNA. Nature
blood group antigens: A source of human muta- 1985;314:67-73.
tions and cSNPs documented in the Blood 44. Jeffreys AJ, Wilson V, Thein SL. Individual-
Group Antigen Gene Mutation Database. Hum specific 'fingerprints' of human DNA. Nature
Mutat 2004;23:8-16. 1985;316:76-9.
31. US Department of Energy and National Insti- 45. Butler JM, Reeder D]. Short tandem repeat
tutes of Health. Human Genome Project. Wash- DNA internet database. NIST standard refer-
ington, DC: US Department of Energy Genome ence database SRD 130. Galthersburg, MD: Na-
Programs, Office of Biologicaland Environmen- tional Institute of Standards and Technology,
tal Research, 2010. [AvaIlableat http://web.or- 2017. [Aval1ableat http://www.cstl. nist.gov/
nl. gov/ sci/ techresources/H uman_ Genom e/ div831/strbase/index.htm (accessedMarch 20,
index.shtml.] 2017).]
32. Reid ME. Molecular basis for blood groups and 46. ClarkJR, Scott SD,Jack AL, et al. Monitoring of
function of carrier proteins. In: Silberstein LE, chimerism following allogeneic haematopoietic
ed. Molecular and functional aspects of blood stem cell transplantation (HSCT):Technical rec-
group antigens. Bethesda, MD: AABB,1995:75- ommendations for the use of short tandem re-
125. peat (STR)based techniques, on behalf of the
33. Storry JR, Olsson ML. Genetic basis of blood United Kingdom National External Ouality As-
group dlversity.BrJ HaematoI2004;126:759-71. sessment Service for Leucocyte Immunopheno-
34. Moller M, Lee YO, Vidovic K, et al. Disruption typing Chimerism Working Group. BrJ Haema-
of a GATA1-binding motif upstream of XG/ to12015; 168(1):26-37.
PBDXabolishes Xg(a) expression and resolves 47. Dorniati-Saad R, Klintmalm GB, Netto G, et al.
the Xg blood group system. Blood 2018; Acute graft versus host disease after liver trans-
132(3):334-8. plantation: Patterns of lymphocyte chimerism.
35. Roulis E, Hyland C, Flower R, et al. Molecular AmJ Transplant 2005;5:2968-73.
basis and clinical overview of McLeod syn- 48. Alvarez-CuberoMJ, Salz M, Martinez-Garda B,
drome compared with other neuroacanthocyto- et al. Next generation sequencing: An applica-
sis syndromes: A review. JAMA Neuro12018; tion in forensic sciences? Ann Hum Blol 2017;
75:1554-62. 44(7):581-92.
36. Rice JP, Saccone NL, Corbett J. The lod score 49. Maha GC, ed. Standards for relationship testing
method (review).Adv Genet 2001;42:99-113. laboratories. 14th ed. Bethesda, MD: AABB,
37. Bluth MH, Reid ME, Manny N. Chimerism in 2020.
the immunohematology laboratory in the mo- 50. Donahue RP, Bias WB, Renwick JH, McKusick
lecular biologyera. Transfus Med Rev 2007;21: VA. Probable assignment of the Duffy blood
134-46. group locus to chromosome 1 in man. Proc Natl
38. Yu N, KruskallMS, Yunis J], et al. Disputed ma- Acad SciUSA 1968;61(3):949-55.
ternity leading to identification of tetragametic 51. National Human Genome Research Institute.
chimerism. N EnglJ Med 2002;346(20):1545- The Human Genome Project. Bethesda, MD:
52. National Institutes of Health, 2019. [Avallableat
h ttps:/ /www.genome.gov/human-genome- 65. The Human RhesusBase.Version 2.4. [Avallable

project.] at http://www.rhesusbase.info/(accessed Oc-
52. International Genome Sample Resource. Ge- tober 4,2019).]
nome browsers: 1000 Genomes data in Ensem- 66. Moller M,]oud M, StorryJR, Olsson ML. Eryth-
bl. Hinxton, UK: European Bioinformat ics rogene: A database for in-depth analysis of the
Institute, 2018. [Available at http://www. extensive variation in 36 blood group systems in
internationalgenome.org/1000-genomes-brows the 1000 Genomes Project. Blood Adv 2016;
ers/ (accessed October 4,2019).] 1(3):240-9.
53. Yates B, Braschi B, Gray K, et al. Gene 67. Lane WI, Westhoff CM, G1eadallNS, et al. Au-
names.org: The HGNC and VGNC resources in tomated typing of red blood cell and platelet an-
2017. Nucleic Acids Res 2017;45(D1):D619-25. tigens: A whole-genome sequencing study. Lan-
54. MacArthur JA, Morales J, Tully RE, et al. Locus cet HaematoI2018;5(6):e241-e251. [Avallable
Reference Genomic: Reference sequences for at https://b1oodantigens.com/cgi-bin/a/a.fpl
the reporting of clinically relevant sequence (accessedOctober 4,2019).]
variants. Nucleic Acids Res 2014;42(Database 68. Singie Nucleotide Polymorphism Database
issue):D873-8. (dbSNP). Bethesda, MD: National Center for
55. Daniels GL, Anstee DJ, Cartron l-P, et al. Blood BiotechnologyInformation, US National Library
group terminology 1995. ISBT Working Party of Medicine, 2019. [Avallab1eat https://www.
on Terminology for Red Cell Surface Antigens. ncbLnlm.nih.gov/projects/SNP / (accessed Oc-
Vox Sang 1995;69:265-79.
tober 4,2019).]
56. Garratty G, Dzlk WH, Issitt PD, et al. Terminol-
69. Horn T, Hamilton], Kosanke], et al. Assessment
of common red blood cell pretreatments to yield
ogy for blood group antigens and genes: Histori-
an accurate serologic antigen phenotype com-
cal origins and guidelines in the new millenni-
pared with genotype-predicted phenotype. Im-
um. Transfusion 2000;40:477-89.
munohematology 2017;33(4):147-51.
57. HGNC guidelines. Cambridge, UK: HUGO
70. Dizon MF. The challenges of daratumumab in
Gene Nomenclature Committee, 2002. [Avail-
transfusion medicine. Lab Med 2017;48(1 ):6-9.
able at https://www.genenames.org/about!
71. LiuI, Wang L,Zhao F, et al. Pre-clinicaldevelop-
guidelines/ (accessed October 4,2019).] ment ofa humanized anti-CD47 antibody with
58. Reid ME, Lomas-Francis C, Bloodgroup antigens
anti-cancer therapeutic potential. PLoS One
and antibodies:A guide to clinicalrelevance and 2015;10(9):e0137345.
technical tips. New York:SBBBooks,2007. 72. Mitigating the anti-CD38 interference with se-
59. Schoeman EM, RoulisEV, Liew YW, et al. Tar- rologic testing. Association bulletin #16-02.
geted exome sequencing defines novel and rare Bethesda,MD: AABB,2016. [Avallableat http://
variants in complex blood group serology cases www.aabb.org/programs/publications/bulle
for a red blood cell reference laboratory setting. tins/Documents/ab 16-02.pdf (accessed Octo-
Transfusion2018;58(2):284-93. ber 4,2019).]
60. Yohe S, Thyagarajan B. Review of clinical next- 73. Chou ST, Westhoff CM. The role of molecular
generation sequencing. Arch Pathol Lab Med immunohematology in sickie cell disease. Trans-
2017;141(11):1544-57. fus Apher Sci 2011 ;44:73-9.
61. Avent ND. Largescale blood group genotyping. 74. Noizatt-Pirenne F, Tournamille C. Relevance of
TransfusClin Bioi2007;14:1 0-15. RHvariants in transfusion of sickie cell patients.
62. Monteiro F, Tavares G, Ferreira M, et al. Tech- Transfus Clin Bioi2011;18:527-35.
nologies involved in molecular blood group ge- 75. Pate LL,Myers J, Palma], et al. Anti-Ge3 causes
notyping. ISBTSciSer 2011;6:1-6. late-onset hemolytic disease of the newborn:
63. Gassner C, Meyer S, Frey BM, et al. Matrix- The fourth case in three Hispanic families.
assisted laser desorption/ionization, time of Transfusion2013;53:2152-7.
flight mass spectrometry-based blood group ge- 76. Hyland CA,Millard GM, O'Brien H, et al. Non-
notyping - the alternative approach. Transfus invasive fetal RHD genotypingfor RhD negative
Med Rev2013;27:2-9. women stratified into RHD gene deletion or
64. Keller MA. The role of red cell genotyping in variant groups: Comparativeaccuracy using two
transfusion medicine. Immunohematology blood collection tube types. Pathology 2017;
2015;31 (2):49-52. 49(7):757-64.
77. Daniels G, Hnnlng K, Martin P, SoothillP. Fetal 85. Denomme GA, Schanen MJ. Mass-scaledonor
blood group genotyping from DNA from mater- red cell genotyping using real-time array tech-
nal plasma: An important advance in the man- nology (review). Immunohematology 2015;
agement and prevention of haemolytic disease 31(2):69-74.
of the fetus and newborn. Vox Sang 2004;87: 86. Luo X, Keller MA, James I, et al. Strategies to
225-32. identify candidates for D variant genotyping.
78. Clausen FB, Christiansen M, Steffensen R, et al. BloodTransfus 2018;16(3):293-301.
Report of the first nationally implemented clini- 87. Meny GM. Recognizingand resolvingABOdis-
cal routine screening for fetal RHD in D- preg- crepancies. Immunohematology 2017;33(2):76-
nant women to ascertain the requirement for 81.
antenatal RhD prophylaxis. Transfusion 2012; 88. Vege S. Resolution of discordances between se-
52:752-8. rology and DNA. In: Vege S, Gannett M, Del-
79. Li Y, Zimmermann B, Rusterholz C, et al. Size aney M. Introduction to moleclar immunohe-
separation of circulatory DNA in maternal plas- matology. Bethesda, MD: AABBPress, 2020:91-
ma permits ready detection of fetal DNA poly- 114.
morphisms. Clin Chern 2004;50(6):1002-11. 89. Wagner FF. RHD PCR of D-negative blood do-
80. Haimila K, Sulin K, KuosmanenM, et al. Target- nors. TransfusMed Hemother 2013;40: 172-81.
ed antenatal anti-D prophylaxis program for 90. Pham B-N,Peyrard T, Tourret S, et al. Anti-HrB
RhD-negativepregnant women - outcome of the and anti-hrB revisited. Transfusion 2009;49:
first two years of a national program in Finland. 2400-5.
Acta Obstet Gynecol Scand 2017;96(10): 1228- 91. ReidME, HipskyCH, Velliquette RW,et al. Mo-
33. lecular background of RH in Bastiaan, the RH:
81. de Haas M, Thurik FF, van der Ploeg CP, et al.
-31,-34 index case, and two novel RHD alleles.
Sensitivity of fetal RHD screening for safe guid-
Immunohematology 2012;28:97-103.
ance of targeted anti-D immunoglobulin prophy-
92. KellerMA, Horn T, Crowley J, et al. Genotype
laxis: Prospective cohort study of a nationwide
compatibility tables for matching patients and
programme in the Netherlands. BMJ2016;355:
i5789. donors for RH variants. Transfusion 2013;
82. Clausen FB. Lessons learned from the imple- 53(2S):174A.
mentation of non-invasive fetal RHD screening. 93. GaspardiAC, Sippert EA,De Macedo MD, et al.
Expert RevMol Diagn 2018;18(5):423-31. Clinicallyrelevant RHD-CEgenotypesin patients
83. Sandler S, Regel W, Westhoff CM, et al. It's with sickle cell disease and in African Brazilian
time to phase in RHD genotyping for patients donors. BloodTransfus2016; 14(5):449-54.
with a serologicweak D phenotype. Transfusion 94. Chou ST, Evans P, Vege S, et al. RH genotype
2015:55:680-9. matching for transfusion support in sickle cell
.84. Johnson ST, Katz L, Queenan IT, et al for the disease. Blood2018;132(11): 1198-207.
Interorganizational Work Group on RHD Geno- 95. de Weerd AE, Betjes MGH. ABO-incompatible
typing.Joint statement on phasing-inRHD geno- kidney transplant outcomes: A meta-analysis.
typing for pregnant women and other females of CHnJ Am Soc NephroI2018;13(8):1234-43.
childbearing potential with a serologic weak D 96. Dean CL, Sullivan HC, Stowell SR, et al. Cur-
phenotype. Bethesda, MD: AABB.[AVailableat rent state of transfusion practices for ABO-
http://www.aabb.org/advocacy/statements/ incompatible pediatric heart transplant patients
Pages/statement150722.aspx (accessed Octo- in the United States and Canada. Transfusion
ber4,2019).J 2018;58(9):2243-9.
rate Blood
MD, Westman, PhD
HE (eg, type 1 chain vs type 2 chain) are responsible

the ABO, PIPK, LE (Lewis), H, I, GLOB for the tissue-specificdistribution of many blood
(Globoside), FORS, and SID blood group group antigens.v' Studies have shown that these
systems are defined by immunodominant carbo- blood groups have roles in development, cell ad-
hydrate epitopes on glycoproteins and glycolip- hesion, malignancy, and infectious disease, al-
ids. The synthesis of these antigens requires the though many of the exact mechanisms underly-
action of a series of enzymes known as glycosyl- ing these roles are still unknown.v"
transferases [Fig10-1 (A)]. These enzymes reside
mainly in the Golgi apparatus and are responsi-
ble for adding specific sugars, in a particular se- A 0 (0
quence and steric or anomeric linkage (a-linked
or ~-linked), to growing oligosaccharide chains The ABO system was originally described by
on glycolipidsand/or glycoproteins.f Most, but Karl Landsteiner in 1900 and remains the most
not all, carbohydrate blood group antigens are important blood group system in transfusion
located at the ends of these chains. Because of medlcine," In blood, ABO antigens are found in
their wide tissue distribution, the carbohydrate- substantial amounts on red cells and also to a
based systems are often referred to as hlsto- lesser extent on platelets. In individuals who
blood groups.' have the "secretor" phenotype, antigens are
Previously, the dogma was that each glyco- present in body fluids as well. ABO antigens are
syltransferase typically uses one specific donor also expressed on many other tissues, including
substrate molecule and one specific acceptor those of the endothelium, kidney, heart, bowel,
substrate molecule, but many examples of pancreas, and lung.' This is the reason why
broader, more "promiscuous" use of acceptor these antigens also constitute a relative barrier
substrates have come to light, including those against ABO·incompatible organ transplanta-
involving carbohydrate-based blood groups. tion."
Transcriptional regulation together with the Transfusion of ABO-incompatible blood can
specificityof these enzymes for both their nucle- be associated with acute intravascular hemolysis
otide sugar donor substrates leg, uridine diphos- and renal failure, and can be fata1.9,10 Similarly,
phate (UDP)-galactose]and acceptor substrates transplanted ABO-incompatiblesolid organs can
Martin L. Olsson, MD, PhD, Professor of Transfusion Medicine, Department of Laboratory Medicine, Vice Dean,
Faculty of Medicine, Lund University, and Medical Director, Nordic Reference Laboratory for Genetic Blood
Group Typing, Office of Medical Services, Region Skcille,Lund, Sweden; and Julia S. Westman, PhD, Postdoctoral
Fellow, Sanford Burnham Prebys Medical Discovery Institute, Center for Nanomedicine, University of Califor-
nia-Santa Barbara, Santa Barbara, California
297
Globular
enzymatically
active domain
(A) (8)
FIGURE 10-1. Model of a glycosyltransferase anchored in the Golgi membrane (A), and three-
dimensional surface model of the human ABO glycosyltransferase (B). The arrow at the top shows
the catalytic cleft, and the dark surfaces highlighted with black labels correspond to the amino acid
positlons that determine A vs 8 specificity.
undergo hyperacute humoral rejection if the pa- have been shown to possessABO-likestructures
tient has not been pretreated to remove natural- on their lipopolysaccharidecoats.!':"
ly occurring anti-A and/or anti-B from plasma.
Because of the serious clinical consequences as- Biochemistry
sociated with ABO incompatibilities, ABO typ-
The A and B antigens are defined by three-sugar
ing and ABO compatibility testing remain the
terminal epitopes on glycolipids and glycopro-
foundation of safe pretransfusion testing and a
teins." As shown in Fig 10-2, the H antigen is
crucial part of a pretransplantation workup.
The ABO system contains four major ABO characterized by a terminal a 1,2 fucose, which
groups: A, B, 0, and AB. The four phenotypes is the immediate and required biosynthetic pre-
are determined by the presence or absence of cursor for expression of either the A or B anti-
two antigens (A and B) on red cells. (See Table gen. The presence of this fucose is required for
10-1.) The ABO system is also characterized by the A and B glycosyltransferasesto be able to
the presence or absence of naturally occurring use the oligosaccharide chain as their acceptor
antibodies, termed isohemagglutinins, directed substrate. In group A individuals, an Nacetylga-
against the missing A and B antigens. As shown lactosamine is added in an a 1-3 linkage to the
in Table 10-1, an inverse relationship exists be- subterminal galactose of the H antigen to form
tween the presence of A and/or B antigens on the A antigen. In group B individuals, an a 1,3
red cells and the presence of anti-A, antt-B, or galactose is added to the same subterminal ga-
both, in sera, a phenomenon often referred to as lactose to form the B antigen. In group ABindi-
Landsteiner's rule. For example, group 0 individuals, both A and B structures are synthesized.
viduals, who lack A and B antigens on red cells, In group 0 individuals, neither A nor B antigens
possess both anti-Aand anti-B. It is believed that are synthesized as a result of alterations in the
the immunizing sources for such naturally oc- ABO genes.7,13 Consequently, group 0 individu-
curring antibodies are gut and environmental als express only H antigen. A and B antigens are
bacteria, such as the Enterobacteriaceae, which also absent in the very rare Bombay phenotype
C HAP T E R 1 0 ABOand Other Carbohydrate Blood Groups 299
TABLE 10-1. Routine ABO Grouping
o + +
+ o o + o A 40 27
o + + o o B 11 20
+ + o o o AB 4 4
o o + + + Bombay* Rare Rare
*H-negative phenotype(see section on H antigen).
+ = agglutination; 0 = no agglutination.
because of the absence of the H-antig~n precur- mon and second most common A subgroup, re-
sor (see "The H System" below). spectively. The Al phenotype, compared to A2,
As terminal epitopes, the A and B. antigens has approximately five times the number of A
can be displayed on a number of oligosaccharide antigens on red cells as a result of a more active
scaffolds that differ in their size, composition, A transferase." There are also antigen differenc-
linkages, and tissue distribution. On red cells, es between the two. For instance, the Al trans-
ABH antigens are present mainly as the N-linked ferase is more prone to make type 3 (repetitive
portions of glycoproteins and as glycosphingolip- A) and type 4 (globe-A)A antigens than the A2
ids, but also to a lesser degree as an O-linked .trapsferase.15,16 In .;addition to the above, ABH
part of certain glycoproteins (Fig 10-3). ABH an- an.tigens expressed on type 1 chain. substrates
tigens are subclassified by the carbohydrate se- can form additional Le'vassociated antigens de-
quence immediately next to the ABH-defining pending on secretor and LE status.P'" (See the
sugars. In humans, ABH is expressed predomi- H and LEsystem sections.)
nantly on four different oligosaccharide periph-
eral core structures (see Table 10-2). On human ABOin Development and Aging
red cells, the majority of endogenous ABH anti-
gen synthesized is present on type 2 chain struc- ABO antigens can be detected on red cells of
tures. In addition, ABH-active type 1 chain embryos as early as 5 to 6 weeks of gestation."
structures can be adsorbed onto the red cell, es- The quantity of ABO antigens on umbilical cord
pecially in secretor individuals. 14 red cells is less than that for adults, as a result of
The ability to synthesize and use different the immaturity of type 2 chain precursors on
carbohydrate chains is genetically determined. cord red cells." (See section on the I blood
In addition to the four lIlain ABO groups men- group system below.) With increasing age, pre-
tioned above, subgroup phenotypes of A and B cursor chains become increasingly branched,
can be identified based on the quantity of A or B thereby allowing more A and B antigen to be ex-
antigen expression and which types of carbohy- pressed." Adult levels of ABO expression are
drate chains contain the A or B antigen (see sec- generally present by age 2 to 4 years.F:"
tion on ABO subgroups). For example, the A Anti-A and anti-B are not present at birth or,
phenotype can be subdivided into a number of if present, they are of maternal origin. Endoge-
subgroups, with Al and A2being the most com- nous synthesis of anti-A and antl-B can develop
A antigen
R
B antigen
R
H antigen
R
Fucose (Fuc) N-acetylglucosamine (GlcNAc)
o Galactose (Gal) D N-acetylgalactosamine (GaINAc)
FIGURE 10-2. Schematic representation of the ABH antigens. Standard symbols for glycan annotation
are used.Type 2 ABH antigens, the most common type on red cells, are shown. (Seealso Table 10-2.)
R = upstream carbohydrate sequence.
as early as age 3 to 6 months, with nearly all titer ABO antibodies can be present in group 0
children displayingthe appropriate isohemagglu- multiparous women and in patients taking cer-
tinins in their sera at 1 year of age.17,20 Titers of tain bacteria-basednutritional supplements.o'v"
anti-A and anti-B continue to increase during Although early reports indicated a decline in iso-
early childhood and achieve adult levels within hemagglutinin titers in the elderly, subsequent
5 to 10 years. studies have disputed these flndings." In indus-
Among healthy adults, ABO titers can natu- trialized countries, isohemagglutinin titers have
rally vary from 4 to 2048 or higher.17,20,21 High- generally decreased, and some studies suggest
~ N-;Jlycan V O-glycan i Glycospninqolipid
FIGURE 10-3.Schematic representationof the red cell membranewith selectedcarbohydrate-carrying

blood group molecules representing different kinds of glycans.
TABLE 10-2. The Most Important Peripheral CoreChain Variants of A Antigen in Humans
A epitope GaINAca1-3(Fuca 1-2)Galp1-R

A type 1 GalNAca 1-3(Fuca 1-2)Galp1-3G IcNAcp 1-3-R
A type 2t GaINAca1-3(Fuca 1-2)Galp1-4GlcNAcp1-3-R
A type 3 GaINAca1-3(Fuca1-2)Galp1-3GaINAca1-3(Fuca1-2)Galp1-4GIcNAcP1-3-R
(repetitive A)
A type 4 GaINAca1-3(Fuca 1-2)Galp1-3GaINAcp1-3Gala1-4Galp1-4Glc-Cer
*Underlined sequencesdenotethe critical differences betweentype 1,2, and 4 chains. Linkagesand anomery (a- or ~-linked)
otthe galactose in these A antigen variants are shown in bold. Bracketedsequencesdenote the repetitive sequence charac-
teristic. of type 3 chain A antigen. Note: There is also an alternative type 3 chain that is denoted the O-linked mucin type,
which has the characteristic Gal~1-3GaINAcbinding but not the repetitive A sequence.
tBy far the predominant type on human red cells.
Cer = ceramide; Fuc = fucose; Gal = galactose; GalNAc = N-acetylgalactosamine; Glc = glucose; GlcNAc= N-acetylglucos-
amine; R = upstream carbohydrate sequence.
that increasing consumption of processed foods gens, respectively [Fig 10-1 (B)].7,13The rare
is a factor." cisABphenotype has a chimeric enzyme with a
mix of A- and B-specificamino acids at or near-
Genetics by those amino acid positions." A plethora of
mutations associated with weak A and B sub-
The ABO gene is located on chromosome 9q34 groups has been described. As an example,
and consists of seven coding exons spread over group Az (the second most common A subgroup
-19 kb.' The largest portion of the open reading after All is commonly the result of a nucleotide
frame is located in exons 6 and 7. The gene is deletion (c.106 1delC) resulting in a frameshift
transcriptionally regulated by several mecha- and an enzyme with an additional 21 amino ac-
nisms, including promoter methylation, antisense ids at the C-terminus of the molecule.F" Most
RNA, tissue-specific transcription-factor-binding of the weak A or B subgroups described below
motifs, and a minisatellite enhancer region and in Table 10-3 depend on single nonsynony-
-4 kb upstream of exon 1.7In addition, recent mous changes compared to A or B consensus,
studies have shown the importance of an erythro- resulting in the substitution of a conserved ami-
specific GATA-bindingmotif in intron 122and no acid that is important for the activity, specific-
the possibilitythat micro-RNAsmay be involved ity, or localization of the enzyme.
by binding to the 3' end of the sequence." ABO An 0 allele encodes a gene product without
expression is also regulated by the H(FUTl / enzymatic function or no protein at all. The
FUr2) genes, which are responsible for the syn- blood group 0 phenotype, therefore, is an auto-
thesis of H antigen, the precursor of A and B an- somal recessive tralt representing inheritance of
tigens. The H genes are in turn tightly regulated two nonfunctional ABO genes. Overall, more
in a tissue-specificmanner through transcription than 100 0 alleles have been identified.7,26The
factors and promoters. In the total absence of H, two most common 0 alleles are ABO*O.Ol.Ol
no A or B antigen can be expressed regardless of (formerly known as 01 or 001) and
ABO genotype. This results in the Bombay or 0h ABO*0.01_02 (known as o=: or 002).
phenotype.I'" These alleles contain the same nucleotide dele-
Following the purification of A glycosyltrans- tion, c.261delG, which leads to a frameshift and
ferase from lung tissue" and the subsequent premature truncation of the protein that lacks
cloning of the ABO gene," a series of studies the enzymatically active domain. A principally
has identified the molecular basis for A, B, 0, different but infrequent 0 allele is ABO*0.02
cisAB,and weak ABO subgroupS.7,13,26 Although (originally described as (j2 but later also called
hundreds of ABO alleles have been found and 003), representing a group of nondeletional 0
characterized, the vast majority of individuals alleles that contains a nonsynonymous polymor-
have alleles givingrise to Aj) A2, B, or 0 expres- phism (c.802G>A) encoding amino acid 268
sion. The AI and B consensus alleles are written (p.Gly268Arg),which is otherwise a critical resi-
as ABO*Aj_Ol and ABO*B.Ol, respectively, in due for donor substrate binding. These alleles
the blood group allele terminology developed by were also found to be involved in cases of sus-
the International Society of Blood Transfusion pected A subgroups.F'" A subsequent study
(ISBT)(see Table 9-4 in Chapter 9 for examples found that alleles with this alteration were re-
of ISBT terminology applied to blood groups). sponsible for 25% of all serologic ABO typing
These alleles are codominantly expressed, and discrepancies caused by reverse-grouping prob-
their coding regions differ by only seven nucleo- lems in healthy donors, despite the frequency of
tides, of which four alter amino acids in the this type of 0allele being 1%to 3% in people of
resulting glycosyltransferases.Ys" Three amino European ancestry but less common or virtually
acid substitutions (A vs B; p.Gly235Ser, absent in other populations.F'" It was speculat-
p.Leu266Met, and p.Gly268Ala) are important ed that the weak anti-A observed in plasma
in determining whether the glycosyltransferase could reflect weak residual glycosyltransferase
uses UDP-Nacetylgalactosamine or UDP-galac- activity. However, a later study was unable to
tose donor substrate to synthesize A or B anti- demonstrate A antigen or enzyme activity in
TABLE10-3. Serologic Reactions Observed in Selected A and B Subgroups
A1 4+ 0 4+ 0 0 4+ 0 A, H
A2 4+ 0 4+ 2+ 0/2+* 4+ 0 A,H
A3 3+mft 0 3+mft 3+ 0/2+* 4+ 0 A, H
Ax O/± 0 1-2+ 4+ 0/2+* 4+ 0 H
Ael 0* 0 0 4+ 0/2+* 4+ 0 H
Am O/± 0 O/± 4+ 0 4+ 0 A,H
B 0 4+ 4+ 0 4+ 0 0 B,H
B3 0 3+mft 3+mft 4+ 4+ 0 0 B,H
Bweak 0 ±/2+ ±I2+ 4+ 4+ 0 0 H
Bel 0 0 0 4+ 4+ 0 0 H
Bm 0 O/± O/± 4+ 4+ 0 0 B,H
*The occurrenceof anti-At is variablein these phenotypes.
tThis reaction can be read as 2+ or 3+ mixed field but typically looks like one or a few large agglutinatesamong a large
number of free cells.
tpositive adsorption/elution test with anti-A.
1+ to 4+ = agglutination of increasingstrength; ± = weak agglutination; mf = mixed-field agglutination;0 = no agglutination.
group 0 donors with the ABO *0.02 allele but the second most common subgroup (20%). It is
confirmed that the anti-A titers appear to be difficult to estimate the absolute number of A
lower." The clinical significance, if any, is un- antigen sites per red cell. Some investigators
clear and the units are labeled group O. have suggested approximately 1 million for Al
and .220,000 for Az/z but others have suggested
ABOSubgroups two to three times as many." Both Ai and Az
subgroups are strongly agglutinated by reagent
ABO subgroups are phenotypes that differin the anti-A in routine direct testing. Ai can be distin-
amount of A and B antigen carried on red cells guished from Azby the lectin Dolichos bijlorus,
and present in secretions (for individuals who which agglutinates Ai red cells but is diluted to
have the secretor phenotype). Clinically, the a level that should not agglutinate Az red cells.
two most common subgroups encountered are Because the Azphenotype reflects the inefficient
Ai and Az.Ai represents the majority of group A conversion of H-to-Aantigen, Az red cells have
donors (-80% among people of European ances- increased reactivity with anti-H lectin Ulex eu-
try) and, as previously mentioned, is character- ropaeus. Enzyme studies comparing Ai and Az
ized by approximately fivetimes more A antigen glycosyltransferaseactivity show that the Ai en-
epitopes per red cell compared to Az, which is zyme is 5 to 10 times more active than the Az
enzyme, resulting in quantitative and qualitative 1. Degree of red cell agglutination by monoclo-
differences in A antigen expression.i-" The lat- nal (and possibly polyclonal) anti-A and anti-
ter includes the synthesis of type 3 and type 4 Al (in the case of the latter, Dolichos btfiorus
chain A antigens on Al red cells that are either lectin can also be used).
not present or expressed at much lower levels 2. Degree of red cell agglutination by human
on A2and other weaker A subgroups.131516 ' , polyclonal and some monoclonal anti-A,B.
In addition to A2, several weaker A sub-
3. Degree of H antigen expression (reactivity
groups have been described (eg, A3'~' An, and
Ae1).Similarly,multiple weak B subgroups have with monoclonal anti-H and/or Ulex euro-
been described (eg, B3, s, Bm, and Bea. The paeus lectin).
weak A and B subgroups are infrequently en- 4. Presence or absence of antl-Al in serum
countered and are usually recognized by appar- (Method 2-9).
ent discrepancies between red cell (forward) 5. Presence of A and H in saliva (an analysis
and serum or plasma (reverse) grouping. Most now seldom performed).
weak A and B subgroups were originally de- 6. Adsorption and elution studies with poly-
scribed before the advent of monoclonal typing clonal anti-A,
reagents, and the hemagglutination patterns re- 7. Family (pedigree)studies.
ported were based on reactivity with human 8. Molecular testing (genotyping).
polyclonal anti-A, anti-B. and anti-A,Breagents.
Weak A subgroups are frequently nome active
In the case of suspected weak B subgroups,
with human polyclonal anti-A (see Table 10-3)
the investigation is similar to the above but
and can show variable reactivity with human
anti-Breplaces anti-A (and ann-Al). Presence of
polyclonal anti-Al and anti-A,B and murine
B and H can be investigated in saliva.
monoclonal antibodies (not shown).13,15,26 The
Presently, many reference laboratories also
degree of reactivity with commercial murine
use genetic typing of the ABO gene as a comple-
monoclonal reagents is clone dependent, and
ment to establish the underlying reason for an
clones may be used together as monoclonal
ABO discrepancy." In addition to various geno-
blends as an anti-A,Breagent to allow for the ag-
typing methods available, Sanger sequencing of
glutination of ~ red cells. This is a requirement
the ABO gene or next-generation sequencing
of the European In-Vitro Diagnostic Directive
may be used. Furthermore, some ABO sub-
(IVDD), although not required in the United
groups exhibit characteristic patterns when test-
States. Because of the reciprocal relationship be-
ed by flow cytometry with selected monoclonal
tween H and synthesis of A and B antigens,
ABO reagents." This method is very useful for
most weak A and B subgroups have H expres-
differentiatinga low-grade chimera from a weak
sion similar to group 0 cells? In clinical prac-
subgroup, or the inherited A3 subgroup from a
tice, it is seldom necessary to identify a patient's
mixed-field pattern after transfusion. It should
specificA or B subgroup except where identify-
be noted that some of the above-mentioned
ing a group A2 kidney donor allows for trans-
methods to characterize ABO subgroups have
plantation of the kidney to a group 0 or Brecipi-
not yet obtained regulatory approval and that
ent. To avoid unnecessary use of group 0 Red this varies depending on the geographicregion.
Blood Cell (RBC) units, however, it can be
worthwhile to define the ABOblood group care-
B(A},A(B},and cisAB Phenotypes
fully in chronic transfusion recipients. Great
care should be exercised to understand the un- The B(A)phenotype is an autosomal dominant
derlying reason for any ABO discrepancy in a phenotype characterized by weak A expression
blood donor. For instance, a chimera should be on group Bred cells.17,36Serologically,red cells
differentiated from an A3subgroup, even if both from B(A)-phenotype individuals are strongly
may exhibit a mixed-fieldagglutination pattern. reactive with anti-B and weakly reactive with
When performed, classification of weak A certain monoclonal anti-A «2+), and they may
subgroups is typicallybased on the following: possess a strong anti-Athat is reactive with both
C HAP T E R 1 0 ABO and Other Carbohydrate Blood Groups 305
Al and Az red cells in their sera. B(A)red cells with anti-A, typically shows weak agglutination
can show varying reactivity with monoclonal (2+ or less) with certain monoclonal and most
anti-Areagents. Testing the sample with a panel polyclonal anti-B, and contains a strong anti-B in
of polyclonal and monoclonal anti-A may re- serum. Despite reactivity of the patient's red
solve the discrepancy, but genetic testing is the cells with antl-B, the patient's serum is not reac-
most accurate. Absence of the B-characteristic tive with autologous red cells.
c.703G>A polymorphism (p.Gly235Ser) in a B Acquired B is the result of deacetylation of
allele will make it a B(A) allele, but also other the A antigen's Nacetylgalactosamine, yielding
genetic alterations in B alleles will result in this a B-like galactosamine sugar.39,40 In patients'
phenotype." The basis of the phenotype is samples, acquired B is often present in the set-
that the B-like glycosyltransferase in these in- ting of infection by gastrointestinal bacteria.
dividuals has an increased capacity to use Many enteric bacteria possess a deacetylase en-
UDP-Nacetylgalactosamine in addition to UDP- zyme capable of converting A antigen to a B-like
galactose, resulting in detectable A antigen syn- analog." Identification of the acquired B pheno-
thesis. type can also be influenced by reagent pH and
An A(B) phenotype has also been described, specificrnonoclonal anti-B typing reagents." In
using monoclonal anti-B. The A(B) phenotype the past, anti-B reagents containing the ES-4
was associated with elevated H antigen and clone were associated with an increased detec-
plasma H-transferase activttv." It has been hy- tion of acquired B..
pothesized that the increased H precursor on To resolve a patient's true red cell type and
these cells may permit the synthesis of some B confirm the presence of acquired B, red cells
antigen by the A glycosyltransferase. should be retyped using a different monoclonal
The cisABphenotype can occur when an in- anti-Bor acidified (pH 6.0) human anti-B.Acidi-
dividual has inherited an ABO gene encoding an fied human anti-B does not react with acquired
ABO glycosyltransferase that can use both A- B antigen. ABO genotyping can also be useful.
and B-specificnucleotide sugars in a more equal Monoclonal anti-B, which recognizes acquired
way than in the B(A)or A(B)phenotypes." If a B, should not be used in clinicalpractice.
cisAB allele is inherited together in trans with
an 0 allele, an unusualphenotype with weak Antibodies to ABO Blood Group System
expression of both A and B is observed (eg, Antigens
AzB3). Anti-B is often present in serum. There
are different variants of cisAB, but the most Anti-A and Anti-B
common one (ABO*cisAB.Ol) is relatively prev- Immunoglobulin M (IgM) is the predominant
alent in some parts of East Asia, and conse- isotype found in group A and group B individu-
quently in individuals with origin from these als, although small quantities of IgG antibody
regions. In this variant, an Al allele exhibits can be detected. In group 0 serum, IgG is a ma-
the presence of a B-specific polymorphism, jor isotype of anti-A and anti-B. As a conse-
c.803G>C (p.Gly268A1a),which alters the en- quence, the incidence of ABOhemolytic disease
zyme's specificityfor donor substrate. of the fetus and newborn (ABQ-as~ociated
HDFN)ishigher among the offspringof group 0
Acquired B Phenotype
mothers than of mothers with other blood types,
The acquired B phenotype phenomenon is a because IgG can cross the placenta but IgM can-
transient serologic discrepancy in group A indi- not. However, ABO-associatedHDFN is less of a
viduals that causes red cell grouping discrepan- clinicalproblem than RhD-associatedHDFN.
cies." Acquired B should be suspected when a Both IgM and IgG anti-Aand anti-Bpref~r~IJ-
patient or donor who has historically typed as tially agglutinate red cells at room temperature
group A now presents with weak B expression (20 to 24 C) or cooler, and both can efficiently
on forward red cell typing. Serologically,the ac- activate complement at 37 C. The complement-
quired B phenotype shows strong agglutination mediated lytic capability of these antibodies
becomes apparent if serum testing includes an (confirmatory typing). The latter is not always
incubation phase at 37 C. Hemolysis caused by practiced outside the United States. Recipient
ABO antibodies should be suspected when ei- samples are typed before transfusion. ABO
ther the supernatant serum is pink to red or the grouping requires both antigen typing of red
cell button is smaller or absent. Hemolysis is in- cells for A and B antigen (red cell grouping or
terpreted as a positive result. The use of plasma forward type) and screening of serum or plasma
for testing or of reagent red cells suspended in for the presence of anti-A and anti-B isohemag-
solutions that contain EDTA prevents comple- glutinins (serum/plasma grouping or reverse
ment activation and hemolysis. type). Both red cell and serum/plasma grouping
are required for donors and patients because
An ti-A,8 each grouping serves as a control for the other.
Reverse or serum grouping is not required in
Sera from group 0 individuals contain an anti-
two circumstances: I) for confirmation testing
body speclflclty known as "anti-A,B" because it
of labeled, previously typed donor red cells and
is reactive with both A and B red cells. Such
2) in infants younger than 4 months of age. As
anti-A and anti-B reactivity cannot be separated
previously discussed, isohemagglutinins are not
by differential adsorption, suggesting that the
present at birth and develop only after 3 to 6
antibody recognizes a common epitope shared
months of age.
by the A and B antigens.v" This is the reason
Commercially available anti-A and anti-B for
why ISBT has acknowledged A,B as the third
red cell typing are extremely potent and aggluti-
antigen of the ABO system. Saliva containing se-
nate most antigen-positive red cells directly,
creted A or B substance can inhibit the activity
even without centrifugation. Most monoclonal
of anti-A,Bagainst both A and Bred cells.
typing reagents have been formulated to detect
many weak ABOsubgroups. (Seemanufacturers'
Anti-A 1
inserts for specific reagent characteristics.) Addi-
Ann-Al is present as an alloantibody in the se- tional reagents [antl-Al and anti-A,B) and spe-
rum of 1% to 8% of Az individuals and 22% to cial techniques to detect weak ABO subgroups
35% of AzBindividuals, and is sometimes pres- are not necessary for routine testing but are
ent in the sera of individuals with other weak A helpful for resolving ABO typing discrepancies.
subgroups. Group 0 serum contains a mixture In contrast to commercial ABO typing re-
of anti-A and anti-At." Because of the presence agents, human anti-A and anti-B in the sera of
of the antibody, ISBThas recognized the Al an- patients and donors can be relatively weak, re-
tigen as the fourth antigen in the ABO system. quiring incubation and centrifugation. Tests for
Anti-Al can cause ABO discrepancies during serum grouping, therefore, should be performed
routine testing and lead to incompatible cross- using a method that adequately detects human
matches with Al and AlB red cells. Anti-Al is anti-A and anti-B. Several methods are available
usually of IgM isotype, reacting best at room for determining ABO group, including slide,
temperature or below, and is usually considered tube, microplate, and column agglutination (gel)
clinically msigniflcant. Anti-Al is considered techniques.
clinically significant if reactivity is observed at
37 C.40 Group Azpatients with an anti-Al that is ABO Discrepancies
reactive at 37 C should receive group Az or 0
Table 10-1 shows the results and interpretations
red cells for transfusion; group AzB patients
of routine red cell and serum tests for ABO. A
should receive group Az,AzB,B, or 0 red cells.
discrepancy exists when the results of red cell
tests do not agree with those of serum tests,
Routine Testing for ABO
usually due to unexpected negative or positive
Donor blood samples are routinely typed for results in either the forward or reverse typing.
ABO at the time of donation and on receipt of (See Table 10-3.) ABO discrepancies may arise
red cell units in the hospital transfusion service from intrinsic problems with either red cells or
serum or from technical errors in performing the 6. False-positive reactions that are caused by a
test. (See Table 10-4 and section on resolving pll-dependent autoantibody, a reagent-
ABO discrepancies.) dependent antibody (eg, EDTA or paraben),
When a discrepancy is encountered, the dis- or rouleaux.
crepant results must be recorded, but interpreta- 7. Anomalous red cell grouping resulting from
tion of the ABO group must be delayed until the
acquired B, B(A),cisAB,or A(B)phenotypes.
discrepancy has been resolved. If the specimen
is from a donor unit, the unit must be quaran- 8. Polyagglutination (eg, T activation) resulting
tined and cannot be released for transfusion. from inherited or acquired abnormalities of
When an ABO discrepancy is identified in a pa- the red cell membrane, with exposure of
tient, it may be necessary to transfuse group 0 "cryptic autoantigens."" Because all human
red cells pending an investigation. It is import- sera contain naturally occurring antibodies to
ant to obtain a sufficient pretransfusion blood such cryptic antigens, those abnormal red cells
sample from the patient to complete any addi- are agglutinated also by ABO-compatible
tional studies that may be required. human sera. Monoclonal anti-A and anti-B
reagents do not detect polyagglutination.
Red Cell Testing Problems
ABO testing of red cells may give unexpected re- Problems with Serum or Plasma Testing
sults for many reasons, including the following:
Problems may arise during ABO testing of se-
1. Weak ABO expression that results from rum or plasma, including the following:
inheritance of a weak ABO subgroup. Some
1. Small fibrin clots in plasma or incompletely
patients with leukemia and other malignan-
clotted serum that can be mistaken for red
cies, as well as during pregnancy, can also
cell agglutinates.
show weakened ABO expression.13,35,42
2. Lack of detectable isoagglutinins in infants
2. Mixed-field agglutination with circulating
younger than 4 to 6 months. Children do
red cells of more than one ABO group follow-
not develop isoagglutinins until 3 to 6
ing out-of-group red cell transfusion or
months of age. ABO antibodies present at
hematopoietic progenitor cell (HPC) trans-
birth are passivelyacquired from the mother.
plantation (eg, group 0 to group A). Mixed-
Children between 6 months and 1 year of
field agglutination is also present in some
age may also have lower levels of anti-Ar-B.
ABO subgroups (eg, A3), blood group chime-
3. Unexpected absence of ABO agglutinins
rism in fraternal twins, and rare cases of
caused by a weak A or B subgroup. (See
mosaicism arising from dispermy.
Table 10-3.)
3. Neutralization of anti-A and anti-B typing
4. Unexpected absence of antt-B in children
reagents by high concentrations of A or B
receiving long-term parenteral and enteral
blood group substance in serum, resulting in
nutrition and who are in a sterile environ-
unexpected negative reactions with serum-
or plasma-suspended red cells. ment, free of bacteria."
4. Spontaneous agglutination or autoagglutina- 5. Unexpected absence of anti-A agglutinins in
tion of serum or plasma-suspended red cells patients receiving equine-derived immuno-
caused by heavy coating of red cells by globulins.44
potent autoagglutinins. 6. ABO-incompatibleHPC transplantation with
5. Nonspeclfic aggregation of serum- or plasma- induction of tolerance. For example, a group
suspended red cells caused by abnormal con- A patient receiving a group 0 marrow trans-
centrations of serum proteins or infused mac- plant will have circulating group 0 red cells
romolecular solutions. but will produce only anti-Bin serum."
TABLE 10-4. Possible Causesof ABO Typing Discrepancies
Weak/missing red cell reactivity ABOsubgroup

Leukemia/malignancy
Transfusion
Pregnancy
Intrauterine fetal transfusion
Transplantation
Excessivesoluble blood group substance
Extra red cell reactivity Autoagglutinins/excess protein coating red cells
Unwashed red cells: plasma proteins
Unwashed red cells: antibody in patient's serum to reagent constituent
Transplantation
Acquired B antigen or other polyagglutinable conditions
cisAB or B(A) phenomenon
Out-of-group transfusion
Mixed-field red cell reactivity ABOsubgroup
Recenttransfusion
Transplantation
Fetomaternalhemorrhage
Twin or dispermic (tetragametic) chimerism
Weak/missing serum reactivity Age related «4-6 months old or elderly)
ABO subgroup
Hypogammaglobulinemia
Transplantation
Excessiveanti-A or anti-B (prozone effect)
Hemodilution, egoby excessive infusion of intravenous fluids
Extra serum reactivity Cold autoantibody
Cold-reactive alloantibody
Serum antibody to reagentconstituent
Excessserum protein
Transfusion of plasma components
Transplantation
Infusion of intravenous immune globulin
(See Chapter 27 for more information on 6. Under" or overcentrifugation of tests"

ABO"mismatchedtransplantation.) 7. Incubating forward" or reverse-typing reac-
7. Severe hypogammaglobulinemiasecondary tions at a suboptimal temperature.
to inherited immunodeficiency or disease 8. Incorrect interpretation or recording of test
therapy. Hypogammaglobulinemiawith dllu- results.
tion of isoagglutinins can also occur after
several courses of plasma exchange with Resolving ABO Discrepancies
albumin replacement.
The first step in resolving an apparent serologic
8. Cold alloantibodies(eg, anti-M) or autoanti- testing discrepancy should be to repeat the test
bodies (eg, anti-I) reactive with correspond" with the same sample to exclude the possibility
ing antigen-positivereverse-groupingcells. of a technical error during testing. Additional
9. Antibodies directed against constituents in studies may include testing a new sample to
the diluents used to preserve reagent Aj and avoid mix-up: testing washed red cells; testing
Bred cells." for unexpected red cell alloantibodies; and reo
10. Nonspecific aggregation or agglutination viewing the patient's medical record for condi-
caused by high-molecular-weight plasma tions, medications, or recent .transfusions that
may have contributed to the conflicting test reo
expanders, rouleaux, high serum-protein
sults (Method 2"4). Samples with apparent
concentrations, or altered serum-protein weak or missing ABO antigens and/or antibod-
ratios. ies may require tests using methods that en"
11. Recent transfusion of out-of-group plasma" hance antlgen-antlbody binding, including incu-
containing components (eg, a group A bating red cells at 4 C (Method 2"5), using
patient provided with platelets from a group enzyme-treated red cells (Method 2"6), and con"
o donor, causing unexpected passively ducting adsorption and elution studies (Method
acquired anti-Ain the patient's plasma)" 2"7), as well as flow cytometry and molecular
12. Recent infusion of intravenous Immuncglob- testing when warranted. In some instances, it
ulin, which can contain ABOisoagglutinins. may be necessary to test for the secretion of
13. Anti"CD47, a monoclonal antibody therapy ABH antigens in saliva (Method 2"8). Patients
in US clinical trials, will give positive reac- with suspected B(A),acquired B, or A(B)pheno-
types should be retested using different mono"
tions with all test red cells, as they possess
clonal and human polyclonalreagents.
the CD47 protein. ABO discrepancies caused by unexpected se-
14. The test method, which may impact the rum reactions are not uncommon. Commonly
ability to detect anti-Ar-B. Column agglutt- encountered causes of serum-grouping discrep-
nation testing may show weaker reactivity ancies include cold autoantibodies, rouleaux,
than test tube methods. cold-reacting alloantibodies (eg, anti-M), and
weak A subgroups with an anti-Al. In addition,
Technical Errors the presence of certain nondeletional a alleles
(ABO *0. 02) is a common reason for lower anti"
Technical problems with a sample or during
A titers in group 0 individuals, as discussed
testing can also lead to problems in ABO group" above.29,46 To resolve an ABO discrepancy
ing, including:
caused by an anti-Al in a group A individual,
red cells should be tested with Dolichos biflorus
1. Specimen mix-up. lectin, which agglutinates group Aj but not Az
2. Too-heavyor too-light red cell suspensions. and weaker A subgroups. The presence of an
3. Failure to add reagents. anti-Al should be confirmed by testing serum
4. Missed observation of hemolysis. against All A2, and 0 red cells (Method 2"9). Re-
5. Failure to follow the manufacturer's instruc- verse-grouping problems resulting from either a
tions. cold alloantibody(Method 2"I0) or autoantibody
can be identified with a room-temperature anti- sues.5,47,48 H antigen has been implicated in cell
body detection test and an autocontrol at room adhesion, normal hematopoietic differentiation,
temperature. Techniques to identify ABO anti- and several malignanctes.vr"
bodies in the presence of cold autoantibodies in-
clude testing at 37 C without centrifugation Biochemistry and Genetics
(Method 2-11) and cold autoadsorption (Meth-
od 4-5). Serum or plasma properties can induce H antigen is defined by the terminal disaccha-
rouleaux formation that resembles agglutina- ride fucose(a 1,2)galactose. Two different fucos-
yltransferase (Fuc-T) enzymes are capable of
tion with A1and B red cells. Saline replacement
or saline dilution (Method 3-7) can be used to synthesizing H antigen: a2Fuc-Tl (encoded by
FUT1, alsoknown as the Hgene) and a2Fuc-T2
distinguish rouleaux from agglutination and
identify ABO antibodies. Incubating or diluting (encoded by FUT2, the secretor gene). The
FUTl enzyme preferentially fucosylates type 2
the agglutination reaction with a citrate solution
chain oligosaccharides on red cell glycoproteins
has a similar effect.
Cold autoantibodies can cause autoagglutina- and glycolipidsto form H type 2. In contrast, the
FUT2 enzyme prefers type 1 chain precursors to
tion of red cells and unexpected reactions
during red cell typing. Red cells heavily coated form H type 1 antigens in secretions (Fig 10-
with autoantibodies can spontaneously aggluti- 4).13Secretion of type 1 chain ABH antigens in
nate and cause false-positivereactions in tests saliva and other fluids requires a functional
FUT2 (secretor) gene. FUT2 is not expressed in
with anti-Aand anti-B. Usually,false-positivere-
actions caused by cold autoantibodies can be red cells but is expressed in salivary glands, gas-
trointestinal tissues, and genitourinary tissues.13
eliminated by washing red cells with warm sa-
line (Method 2-17). Autoagglutination caused Type 1 chain ABH antigens present on red cells
by IgM can also be inhibited or dispersed by in- are passively adsorbed from circulating glycolip-
cubating red cells in the presence of either dlth- id antigens present in plasma." (See "The LE
iothreitol (Method 3-16) or 2-aminoethyliso- System.") Several inactivating and weakening
thiouronium bromide. These reagents reduce gene variants have been described in both the
FUTl and FUT2 genes." Many of the variants
the disulfide bonds on IgM molecules, decreas-
ing their polyvalency and ability to agglutinate are geographicallyand ethnically restricted. For
instance, approximately 20% of people of Euro-
red cells directly.
pean ancestry are nonsecretors, and this is main-
ly due to homozygosity for the common
TH H YS {01 FUT2*OlN02 allele with c,428G>A, which
leads to a premature stop codon (p.Trp143Stop)
H antigen is expressed on red cells from all indi- and a nonfunctional enzyme.
viduals except those of the rare Bombay pheno-
type. Because H antigen serves as the precursor Null Phenotypes
to both A and B antigens, the amount of H anti- Bombay (Oh) Phenotype
gen on red cells depends on an individual's ABO
type. H antigen is highly expressed on group 0 Originallydescribed in Bombay,India, the Ohor
red cells because group 0 individuals lack a Bombay phenotype is a rare, autosomal reces-
functional ABO glycosyltransferase.In group A sive phenotype characterized by the absence of
and B individuals, the amount of H antigen is H, A, and B antigens on red cells and in secre-
considerably less because H is converted to the tions. Genetically, O, individuals are homozy-
A and B antigens, respectively. The amount of gous (or compound heterozygous) for nonfunc-
H antigen on red cells, based on agglutination tional FUTl and FUT2 genes, resulting in a
with the anti-H lectin Ulex europaeus, differs complete absence of H type 1 and H type 2
between the ABO phenotypes as follows: chain, and consequentiy also loss of A and B in-
O>Az>B>AzB>A1>A1B.H antigen is present on dependent of ABO genotype. The original
HPCs, red cells, megakaryocytes, and other tis- Bombay phenotype is actually due to the
!
o
(f)
3:
UJ
....J
01!
I
CO
-c
Z
«
I
o
UJ
o,
~
a::
o
I
'"
c:
'"
<= E
"'- 0)
'E '"
.~i ~ '"'"
0
~ ~ ~ u
'" co
'" N
'" .5:!'
'"
~ ,,0 8. >..
Q;
>-
~ '-' ~«s
I
CO o ,,;,'" ,,;,
«
Z O~
liD
~ ~
o :r:
o
~ ea, 0)
~u
0
C\I .~i ~ e~
UJ ;[ .... ~ '" 0'"
u..
o,
~ :r:
~~ ~ .. ~a. ~O
Soo ~
FUTl*OlN.09 allele with a critical missense occur in group 0 individuals, as evidenced by

mutation (c.72ST>G, p.Leu242Arg), while the trace type 1 chain H on their red cells and in
entire FUT2 gene is deleted. Ohred cells type as their secretions.
H negative with anti-H lectin Ulex europaeus, In laboratory testing, red cells from para-
monoclonal anti-H, and human polyclonal anti- Bombay individuals may (or may not) have
H from other Ohindividuals. Because these indi- weak reactions with anti-A and anti-B reagents.
viduals lack a functional FUT2 (secretor) gene In some cases, A and B antigens may be detect-
necessary for Leb synthesis, Oh individuals also ed only after adsorption and elution. Ah and Bh
type as Le(b-). (See "The LESystem.") Genotyp- para-Bombayred cells are nonreactive with anti-
ing studies have described a wide range of inac- H lectin, monoclonal anti-H, and human anti-H
tivating mutations in both the FUTl and FUT2 from Oh individuals. The sera of para-Bombay
genes in Ohindividuals. 13,26 The Ohphenotype is individuals contain antl-H, anti-HI, or both and,
also present in leukocyte adhesion deficiency 2 depending on their ABO type, anti-A and anti-
(LAD2)because of mutations in the GDP-fucose B.17,40The anti-H is typically weaker and less
transporter gene.50
clinically significant in para-Bombay than Bom-
Because they lack all ABH antigens, Oh indi-
bay individuals.
viduals possess isoagglutinins to A, B, and H.
The para-Bombay phenotype can also occur
(See Table 10-1.) In routine ABO typing, these
in nonsecretors (ie, without a functional FUT2
individuals initially type as group O. The 0h phe-
notype becomes apparent during antibody gene). In these rare cases, both FUTl alleles car-
screening tests with group 0 red cells, which ry mutations that diminish, but do not abolish,
are rich in H antigen. The anti-H present in O, the enzyme activity. Therefore, these individuals
individuals strongly agglutinates all group 0 red express small amounts of H type 2 antigen on
cells and sometimes demonstrates in-vitro he- red cells but lack H type 1 in secretions (and on
molysis. The Ohphenotype can be confirmed by red cells). If A or B alleles are inherited at the
demonstrating an absence of H antigen on red ABO locus, the phenotypes can be described as
cells and the presence of a strong antl-H in se- Ah and Bh, respectively. In principle, it is also
rum that is reactive with group 0 red cells but possible to have a weak FUTl-encoded enzyme
not with 0h red cells from other individuals. in combination with a fully functional FUT2-
encoded enzyme. The resulting phenotype is
Para-Bombay Phenotype para-Bombay.
Individuals with the para-Bombay phenotype
Antibodies to H Blood Group System
can be secretors whose red cells are apparentiy
deficient of H antigen.7,13 Genetically, these indi- Antigens
viduals are homozygous for a nonfunctional Alloanti-H (Bombay and Para-Bombay)
FUTl gene, but they have inherited at least one
functional FUT2 (secretor) gene. The red cells The anti-H found in Bombay (Oh)individuals is
from these H-deficient secretors lack serological- clinically significant and associated with acute
ly detectable H antigen but can carry small hemolytic transfusion reactions. These antibod-
amounts of H, A, and/or B antigen because, un- ies are predominantly of IgM isotype and exhibit
like persons with classic Bombay phenotype, a broad thermal range (4 to 37 C) with all red
para-Bombaypersons express type 1 chain ABH cells except O, red cells. As with anti-A and
antigens in their secretions and plasma (Method antl-B, alloanti-His capable of activating comple-
2_8).40Type 1 chain A and/or B antigens in plas- ment and causing red cell hemolysis intravascu-
ma are then passively adsorbed onto red cells, larly. The anti-H found in para-Bombayindividu-
resulting in weak A or B antigen expression. Red als may show lower titers and be less prone to
cells from para-Bombayindividuals are designat- cause direct lysisin vitro but is potentially signif-
ed "Ah'" "Bh,"and "ABh'"Para-Bombaycan also icant, especiallyin nonsecretors.
C HAP T E R 1 0 ABO and Other Carbohydrate Blood Groups 313
Au toanti-H and Autoanti-H! LE antigens are not synthesized by the eryth-

roid cells but are passively adsorbed onto red
Autoantibodies to H and HI antigens can be en-
cell membranes froma pool of soluble LE glyco-
countered in healthy individuals. When present,
lipid present in plasma/? The gastrointestinal
these autoantibodies are most common in Al in- tract, which is rich in. LE-active glycolipid and
dividuals, who have low levels of H antigen on glycoprotein, is thought to be the primary
their red cells. Autoanti-H and autoanti-HI are source of LE glycolipid in plasma. Because LE
usually of IgM isotype and are reactive at room antigens are passively adsorbed onto red cell
temperature. membranes, they can be eluted from red cells af-
ter transfusion or by increases in plasma volume
Transfusion Practice and increased circulating lipoproteins, which
Alloanti-H is highly clinically significant and is also adsorb LE glycolipid. For example, LE anti-
capable of fixing complement and causing he- gen is often decreased on red cells during preg-
molytic transfusion reactions. As a result, pa- nancy, and some women transiently type as
tients with alloanti-H caused by the Bombay Le(a-b-), which is attributed to an increase in
phenotype should be provided with H-negative circulating plasma volume and a fourfold in-
(Oh) red cells for transfusion. The same is typi- crease in lipoproteln." The levels of LE anti-
cally not necessary for para-Bombay secretor pa- gens also decrease on stored red cells; therefore,
tients, but evaluation of the clinical significance LE phenotyping should be done sooner than lat-
and thermal range of the antl-H in the individual er to avoid false-negative results. Lebexpression
para-Bombay case is worthwhile, particularly in and immunoreactivity are also influenced by
nonsecretors. ABH type as a result of the synthesis of struc-
In contrast, autoantibodies against H and HI tures carrying both LE and ABH activity (Fig 10-
4).5,13,48
are generally clinically insignificant. In most pa-
tients, transfused group-specific or group 0 red
cells should have normal in-vivo survival. Rare- Biochemistry and Synthesis
ly, autoanti-HI can result in decreased red cell LE antigen synthesis depends on the interaction
survival and hemolytic transfusion reactions af- of fucosyltransferases encoded by two distinct
ter transfusion of group 0 red cells.":" Hemoly- genes (Fig 10-4): FUT3 (the IE gene) and FUT2
sis may follow transfusion of group 0 red cells (the secretor gene).50,51Unlike. the enzyme en-
to a group Au B, or AlB patient with an unusual- coded by FUTl, which is relatively specific for
ly potent high-titer anti-HI that is reactive at type 2 chain substrates, the enzymes encoded
37 C.40 In such patients, transfusion of group- by FUT2 and FUT3 preferentially fucosylate
specific (AI, B, or AB) red cells is advised. type 1 chain substrates. The FUT2 gene there-
fore controls addition of a terminal a 1,2 fucose
to type 1 chain precursors to form H type 1 anti-
TH T gen. FUT3, the IE gene, encodes an a1,3/4-
fucosyltransferase that transfers a fucose, in an
The LE blood group system consists of two main a 1-4 linkage, to the penultimate Nacetylglucos-
antigens, Lea (LE1) and Le'' (LE2), and three amine of type 1 chain precursor (also known as
common phenotypes, Le(a+b-), Le(a-b+), and Lewis c) to form Lea antigen. The LE enzyme
Le(a-b-). Four additional LE antigens represent can also add a second fucose to H type 1 antigen
composite reactivity between Le", Leb,and ABO to form Leb antigen. Note that Leb cannot be
antigens: Leab(LE3), LebH(LE4),ALeb(LES),and formed from Leabecause the presence of a sub-
BLeb(LE6).26,50In addition to being present on terminal fucose on Leasterically inhibits binding
red cells, LE antigens are widely expressed on by the secretor enzyme.'
platelets, endothelial cells, and kidney, as well In individuals with both LE and secretor en-
as on genitourinary and gastrointestinal epitheli- zymes, H type 1 chain is favored over Lea syn-
um. thesis. As a result, most of the LE antigen
synthesized is Leb [Le(a-b+) phenotype]. In cause secretor activity increases with develop-
group Al and B individuals, Leb and H type 1 mental age. An Le(a+b+) phenotype is also pres-
chain can be further modified by ABO glycosyl- ent in 10% to 40% of individuals of Asian
transferases to form LE5 and LE6, type 1 anti- ancestry (eg, 16% of Japanese individuals) as a
gens that express a combination of Leb activity result of inheritance of a very common weak
and A or B, respecttvely.v" In group A1individu- secretor gene in East Asian populations
als, the majority of LE antigen in plasma is actu- (FUT2*OlW.02, previously SeW).26,30 In the ab-
allyALeb•52 sence of a functional FUT3 gene (Ielle), neither
Leanor Leb can be synthesized, leading to the
Genetics and LE Phenotypes Le(a-b-), or LE null, phenotype. Type 1 chain
The three LE phenotypes commonly encoun- ABH antigens may still be synthesized and se-
tered represent the presence or absence of LE creted in individualswho have inherited at least
and secretor enzymes (see Table 10-5). Le(a+b-) one functional FUT2 allele (Method 2-8). The
individuals have inherited at least one function- Le(a-b-) phenotype is significantly more com-
al FUT3 gene (traditionally denoted Le) but are mon in persons of African ancestry. Although
homozygous for nonfunctional FUT2 alleles (de- rare, an Le(a-b-) phenotype is also present in
noted selse). As a result, such individuals syn- patients with LAD2 due to defects in fucose
thesize and secrete Leaantigen but lack Leb and transport." This leads to lack of slalyl-LewisX
type 1 chain ABHantigens. (sl.e'), which is normally involved in leukocyte
The Le(a-b+) phenotype reflects inheritance extravasation.
of both functional FUT3 (Le) and FUT2 (Se) al- Several inactivating mutations have been
leles, leading to the synthesis of Lea, Leb, and identified in both the FUT3 (LE)and FUT2 (se-
type 1 chain ABH. Because most type 1 chain cretor) genes." Many of the mutations are geo-
precursor is converted to Leb in those individu- graphicallyand ethnically restricted, with many
als, they appear to type as Le(a-). An Le(a+b+) populations displaying a few predominant al-
phenotype is transiently observed in infants be- leles.
TABLE 10-5. Adult Phenotypes and Prevalence Rates in the LE System
+ 0 Le(a+b-) 22 23 Le se/se Lea
0 + Le(a-b+) 72 55 Le Se Lea, Leb, ABH
0 0 Le(a-b-) 6 22 Ie/Ie se/se Type 1 precursor

Ie/Ie Se Type 1 ABH
+ + Le(a+b+)* Rare Rare Le Sew Lea, Leb
'Probable genotype at the LE (FUT3) and secretor (FUT2) loci.

tType 1 chain antigens present in saliva and other secretions.
tLe(a+b+) is present in 10-40% of people of East Asian ancestry and is also transiently observed in infants.
Le = at least one FUT3gene encoding functional LE enzyme (represents the genotypes LelLe or Lei/e); /el/e = homozygous
for FUT3 gene encoding an inactive enzyme; Se = at least one FUT2gene encoding active secretor enzyme (represents the
genotypes SelSe or Selse); selse = homozygous for FUT2 gene encoding an inactive enzyme; Sew= FUT2 gene encoding
weak secretor enzyme.
C HAP T E R 1 a ABOand Other Carbohydrate Blood Groups 315
LE Expression in Children reagent is used). LE antibodies. can sometimes

cause hemolysis in vitro, especially when fresh
Table 10-5 shows the distrib.ution of LE types
serum and enzyme-treated red cells are used.
in adults. In contrast, most newborns type as
Le(a-b-). Approximately 50% of newborns sub-
sequently type as Lela-] after ficin or papain Transfusion Practice
treatment. The prevalence of Le~antigen, how- In .ge.Btral, LE antibodies are not considered
ever, is low in newborns compared to adults be- clinlcallyslgmficant. Red cells that are compati-
cause of developmental delays in secretor ble in tests at 37 C, regardless of LEphenotype,
(FUT2-encoded fucosyltransferase) activity. An are expected to have normal in-vivo survival. It
Lela-b-] phenotype can beYttansientlypresent is not necessary to transfuse antigen-negative
in children as the level of secretor activity ap- red cells in most patients. Unlike ABO antigens,
proaches adult levels. A valiciLE phenotype is LEantigens are extrinsic glycolipidantigens that
not developed until age 5 or 6.17 are readily eluted and shed from transfused red
cells within a few days of transfusion." Further-
Antibodies to LE Blood Group System more, LEantigens in transfused plasma can neu-
Antigens tralize LE antibodies in the recipient. For these
reasons, hemolysis in vivo is very rare following
Antibodies against LE antigens are generally of
transfusion of either Le(a+) or Le(b+) red cells,
IgM isotype and occur naturally. Clinically, LE
but exceptions occur."
antibodies are most often encountered in the
LEantibodies are not a cause of HDFN.17LE
sera of Le(a-b-) individuals.and may contain a
antibodies are predominantly of IgM isotype and
mixture of anti-Le", anti-Le", and antl-Le", an
antibody capable of recognizing both Le(a+) and do not cross the placenta. In addition, LE anti-
Lelb-) red cells. Because small amounts of Lea
gens are poorly expressed on neonatal red cells,
are synthesized in the Le(a-b+) phenotype, with many newborns typing as Le(a-b-). How-
Le(a-b+) individuals do not make anti-Lea.Anti- ever, anti-Leahas surprisingly been associated
Le" is present infrequently in the Le(a+b-) phe- with stillbirth."
notype. A transient L~(a-b-) phenotype, accom-
panied by LE antibodies, is commonly observed
I AND i ANTIGENS Of THE ~
during pregnancy. Hnally, anti-Lebcan demon-
strate ABO specificity (anti-Le'", anti-Ale", and BLOOD GROUP SYSTEM (027)
anti-BLeb),and is preferentially reactive with ND i BLO D U
Le(b+) red cells of a specific ABO groUp.40,48 LLECTION
Anti-Le'", the most common specificity,is more
strongly reactive with Le(b+) group a and A2 The I and i antigens are ubiquitous, structurally
red cells than with group Al and B red cells, related antigens present on all cell membranes.
which have low H antigen levels. Anti-Le'" is The I antigen is the only antigen in the I blood
strongly reactive with all Le(b+) red cells, re- group system, while .i antigen is still genetically
gardlessofABOgroup. unsolved and remains in the.Ii blood group col-
Most examples of LEantibodies are saline ag- lection. The minimum epitope common to both
glutinins that are reactive at room temperature. antigens is a repeating lactosamine (Gal~1-4Glc-
Unlike ABO, the agglutination is relatively frag- NAc) or type 2 chain precursor. The minimum i
ile and easily dispersed, requiring gentle resus- antigen epitope is a linear, nonbranched struc-
pension after centrifugation. Agglutination is ture containing at least two successive lac-
sometimes observed after 37 C incubation, but tosamine motifs." The I antigen is a polyvalent,
the reaction is typically weaker than that at branched glycan derived from the i antigen (Fig
room temperature. On occasion, LE antibodies 10-5). Both i and I serve as substrates ~? sSClf-
can be detected in the antihuman globulin folds for the synthesis of ABH, LEX[GaW1-
(AHG)phase. Such detection may reflect either 4(Fuca 1-3)GlcNAc],and other type 2 chain an-
IgG or bound complement (if polyspecificAHG tigens.4,5,18On red cells, i and I antigens are
GCNT2 chr 6p24.2 i antigen
~< I ,~3' ~R
ATG
~ 1/ TGA I
POIYlact~samine
I
6_~_N_aCetylgIUCOS_~
.. aminyltransferase
~ ~3 3R I antigen
~4
~
~4 3R
I antigen
3
~4
FUT1 chr 19q13.33
1 2 3 4
5'-fl-r'~~ I ~3'
z-c-tucosyt-
transferase
ATG~
HI antigen
A. Fucose • N-acetylglucosamine
o Galactose
FIGURE 10-5. GCNT2 and its erythroid transcript that gives rise to the enzyme synthesizing the I
antigen from i antigen. Further elongation by FUTt-encoded fucosyltransferase results in the HI
antigen. The linkages created by each enzyme are highlighted in bold. ATG and TGA represent the
start and stop codons of the gene, respectively.
R = upstream carbohydrate sequence.
present on Nlinked glycoproteins and glyco- The i phenotype is characteristic of neonatal red
sphingolipids. cells as well as other maturing erythroid cells,
whereas 1+ is the common phenotype for ma-
Phenotypes ture red cells as seen in an adult's peripheral cir-
Two phenotypes are recognized according to the culation. With increasing age, there is a gradual
presence or absence of I antigen: I and i (1-). increase in I antigen accompanied by a recipro-
cal decrease in i antigen as glycan chains are exon 1Aor exon lB. In the iadult phenotype with
branched; most children develop an adult I+ cataracts, there is a loss of I antigen synthesis in
phenotype by age 2.18An increase in i antigen all tissues, caused by either gene deletion or mu-
can occur in people with chronic hemolytic dis- tations in exons 2 and 3.
orders and is a sign of stressed erythropoiesis.55
Certain genetic disorders are associated with Antibodies to I and i Antigens of the I
an increase in i antigen." The iadultphenotype Blood Group System and i Blood Group
(I-i+) is a rare autosomal recessive phenotype Collection
caused by mutations in the GCNT2 (previously
known as the lor IGnT gene). In patients of Anti-/
Asian ancestry, the iadultphenotype can be associ- Anti-I is common in the serum of healthy indi-
ated with congenital cataract. Increased i anti- viduals and is typically autoreactive. Anti-I is
gen levels are also present in people with usually of IgM isotype and is strongly reactive at
Diamond-Blackfan anemia and congenital 4 C with titers of <64. Samples with higher ti-
dyserythropoietic anemia type II (alsoknown as ters may also be detectable at room tempera-
hereditary erythroblastic multinuclearity with ture. Anti-Iis identified by strong reactions with
positive acidifiedserum lysistest, or HEMPAS). adult red cells but weak or no agglutination
with cord red cel1s(see Table 10-6). Anti-I can
Genetics be enhanced by 4 C incubation, the presence of
The GCNT2 gene encodes a ~1-6-Nacetylglu- albumin, or use of enzyme-treated red cel1s.An
cosaminyltransferase that converts the linear i alloanti-Ican be seen in the iadult
phenotype.
antigen into the branched I antigen.I8,26The Some examples of anti-I can demonstrate
gene resides on chromosome 6p24 and contains complex reactivity and are more strongly reac-
five exons, including three tissue-specificexons tive with red cells of specific ABO, PI' or LE
(exons lA, 1B, and 1C). As a result, three differ- phenotypes. Many of those antibodies appear to
ent messenger RNA(mRNA)transcripts are syn- recognize branched oligosaccharides that have
thesized, depending on which exon 1 is used. been further modified to express additional
In the iadultphenotype without cataracts, blood group antigens. Anti-HIis commonly pres-
there are mutations in exon 1C, which is specif- ent in the serum of Al individuals. Anti-HI is
ic for I antigen synthesis in red cells. As a conse-
quence, I antigen is missing on red cells but is
°
more strongly reactive with group and group
A2red cel1s,which are rich in H antigen, than
still synthesized in other tissues that use either with group Al red cel1s. Anti-HI is suspected
TABLE 10-6. Comparative Typical Serologic Behavior of Antibodies to Iii Blood Group System
Antigens with Saline Red Cell Suspensions
4C I adult 4+ 0-1+
i cord 0-2+ 3+
i adult 0-1+ 4+
22 C I adult 2+ 0
i cord 0 2-3+
i adult 0 3+
when serum from a group A individual directly oratory testing, these antibodies can be reactive
agglutinates all group 0 red cells but is compati- in the AHG phase of testing, particularly when
ble with most group A donor blood tested. polyspecificAHG is used. Such reactions rarely
Other examples of complex reactivity include indicate antibody activity at 37 C but are the
anti-lA,-IPl, -IBH,and _ILehH•40 consequence of antibody binding, followed by
complement binding, at low temperatures. Usu-
Anti-i ally, avoiding room-temperature testing and us-
Autoanti-iis a relativelyuncommon cold aggluti- ing anti-IgG-specificAHG prevents detection of
nin in sera from healthy individuals. Like anti-I, nuisance cold autoantibodies. For stronger anti-
anti-i is primarily of IgM isotype but is weakly body samples, autoantibody can be removed
reactive at 4 to 10 C. Anti-i is most strongly re- from serum by cold autoadsorption techniques
active with cord and i.dult red cells and more (seeMethod 4-5). Cold autoadsorbed serum can
weakly reactive with 1+ adult red cells (Table also be used forABOtesting.
10-6). Patients with infectious mononucleosis
often have transient but potent anti-i, As with
antt-I,complex reactivity can sometimes occur. 8)
Cold Agglutinin Syndrome

The first antigen of (what was previouslyknown
Autoanti-I and autoanti-i are pathologically sig-
as) the P blood group system was discovered by
nificant in cold agglutinin syndrome (CAS)and
Landsteinerand Levinein 1927 in a series of ex-
mixed-type autoimmune hemolytic anemia. In
periments that also led to the discovery of M
those disorders, autoantl-I (or anti-l) behaves as
and N antigens. Because this was actually what
a complement-bindingantibody with a high titer
and wide thermal range. Primary CAS occurs is now called the PI antigen (and because the P
with lymphoproliferativedisorders (eg, Walden- antigen belongs to another system, GLOB), it
strom macroglobulinemia, lymphoma, and was decided in 2010 to change the name of this
chronic lymphocytic leukemia). A potent auto- system to PIPK. Severalrelated glycosphingolip-
anti-I can also occur in the setting of infection. id antigens belong to the PIPK system (PI, pk,
Mycoplasma pneumoniae infections are a com- NOR) and the GLOB system (P, PX2).26,56 In
mon cause of autoanti-I and can be accompa- 2018, the GLOBcollection was made obsolete,
nied by a transient intravascular hemolysis and and the LKEantigen moved to the 901 series of
subsequent hemoglobinuria. (See Chapter 14 antigens.57 P\ P, PX2, and LKE are high-
for additional information on CAS_) prevalence antigens expressed on the red cells
The specificity of the autoantibody in CAS of nearly all individuals except in rare null phe-
may not be apparent when undiluted samples notypes, which lack P, PX2, and LKEantigens
are tested. Titration and thermal amplitude stud- (Pk phenotype) or P, pk, and LKE antigens (p
ies may be required to discern the specificityof phenotype) (see Table 10-7), although PX2 is
the autoantibodies and their potential clinical particularly strongly expressed on red cells of p
significance. Table 10-6 illustrates the serologic phenotype.58Red cells are particularly rich in P
behavior of antl-I and anti-i at 4 C and 22 C. antigen (alsoknown as globoside),which is the
(SeeChapters 13 and 14 and Method 4·7 for ad- most abundant neutral red cell glycolipid.59,60 pk
ditional information regarding titration and ther- and P antigens are also widely expressed on
mal amplitude studies.) nonerythroid cells, including lymphocytes,
platelets, kidney, lung, heart, endothelium, pla-
Transfusion Practice centa, and synovium cells.61,62In contrast, PI
Autoanti-I can interfere with ABO typing, anti- antigen seems to be mainly expressed on red
body detection, and compatibilitytesting. In lab- cells.?'
TABLE10-7. Phenotypes and Prevalences in the Pi PK and GLOB Group Systems
+ + 0/+ + None P1 79 94
0 + o/± + Anti-Pi * P2 21 6 80
0 ot 0 0 Anti-PP1 pk p Rare Rare Rare

(Tja)
+ 0 + + Anti-P, P1k Rare Rare Rare

-PX2
0 0 + + Antl-P, P2 k Rare Rare Rare

-Pi, -PX2
* An anti-Pi is approximately 25% P2 individuals.

tUsually negative. examples of antl-P may be weakly positive as a result of cross-reactivity of antl-P
with PX2 on p red cells.
Phenotypes acetylgalactosaminyltransferase (encoded by

B3GALNTI), which adds a I3I-3-linked N
More than 99.9% of donors have the Pj (Pl+) acetylgalactosamineto the terminal galactose of
or Pz (Pl-) phenotypes. (See Table 10-7.) Both pk (Gb3) to form P antigen (Gb4). In some cells,
phenotypes express pk and P antigens and .differ including red cells, the P antigen is further elon-
only in the expression of PI antigen. Three rare, gated to form additional, globo-familyantigens,
autosomal recessive phenotypes have been iden- such as Luke (LKE),type 4 chain ABH antigens
tified (p, Pj\ p/), as well as some rare weak (globo-H,-A,and -B),and NOR. NOR+ is a rare
variants.63,64 In analogy with the ABO system, polyagglutinablered cell phenotype and the re-
the rare p and pk phenotypes are associated with sult of unusual globe-familyantigens character-
the presence of naturally occurring antibodies ized by the addition of an a1-4 galactose to the
against missing antigens (anti-PI, anti-P, anti-P\ terminus of P and related long-chain globo-
and anti-PX2). glycolipids(Table 10-8).65
Unlike pk and P antigens, the PI antigen is
Biochemistry not a globe-seriesglycosphingolipidbut a mem-
ber of the neolacto family (type 2 chain glyco-
The synthesis of the P\ P, and PI antigens pro- sphingolipids). In P, individuals, A4GALT adds
ceeds through the stepwise addition of sugars to an a 1-4 galactose to the terminus of paraglobo-
lactosylceramide, a ceramide dihexose (CDH) side. Whether PI antigen is present on red cell
(Fig 10-6 and Table 10-8). The first step in this glycoproteins or not has remained unclear.66,67
process is the synthesis of the pk antigen, the However, the PI determinant is found on glyco-
precursor of all globo-series glycosphingolipids. proteins in other species, and paragloboside is
To make the pk antigen, aI,4-galactosyltransfer- also the precursor of H antigen, which together
ase (encoded by A4GALT) adds a terminal with A and B determinants are mostly found on
galactose, in an a1-4 linkage, to CDH. The pk Nlinked glycans.68,69A recent study shows that,
antigen can then serve as a substrate for PI ,3-N in fact, the PI epitope is present on glycopro-
III
III c
<: 'E
'E m
VI g!
m
VI
0
o
~ ';:
::J
C, '"rn
'iii '0
G)
UJ
z- '0
o
l III
o
m
m
,2
~
.c
o
'"
:;\: :;\: (fJ (!j
.0+
III
VI
<1l 0
VI
8 11 ili
::J
'"
'iii 8
::J
(!j o LL
.0<l1li
CHAPTER 10 ABO and Other Carbohydrate Blood Groups 321
TABLE10-0. Structures of Blood-Group-Carrying Molecules in the Pi PK and GLOB Systems, and

Related Glycosphingolipids
Globo (Gb) Gb3, pk Galai-4Gall3i-4Glcl3i-i Cer
Gb4,P GaINAcl3i-3Galai-4Gall3i-4Glcl3i-i Cer
Gb5 Gall3i-3GaINAcl3i-3Gala i-4Gall3i-4Glcl3i-i Cer
NORi Galai-4GaINAcl3i-3Gala1-4Gall3i-4GlcI31-i Cer
FORSi GaINAca1-3GaINAcl3i-3Galai-4Gall3i-4Glcl3i-i Cer
Globo-H Fucai-2Gall3i-3GaINAcl3i-3Galai-4Gall3i-4Glcl3i-1 Cer
LKE NeuAca2-3Gall3i-3GaINAcl3i-3Galai-4Gall3i-4Glcl3i-i Cer
NOR2 Galai-4GaINAcl3i-3Galai-4GaINAcl3i-3Galai-4Gall3i-4Glcl3i-1 Cer
Neolacto (nLc) GlcNAcl3i-3Gall3i-4GlcI31,i Cer
Gall3i-4GlcNAcl3i-3GaI131-4Glcl3i-1Cer
Pi Galai-4Gall3i-4GlcNAcl3i-3Gall3i-4Glcl3i-i Cer
SPG Neu5Aca2-3Gall3i-4GlcNAcI31-3GaI131-4Glcl3i-1Cer
GaINAcl3i-3GaI131-4GlcNAcI31-3Gall3i-4Glcl3i-iCer
Sialyl-x2 Neu5Aca2-3GaINAcI31-3Gall3i-4GlcNAcl3i-3Gall3i-4GlcI31-iCer
family. Neolactoare type 2 glycosphingolipids.
CDH = ceramide dihexose or lactosylceramide; Cer = ceramide; Gal = galactose; GalNAc = N-acetylgalactosamine;Glc =
glucose, GlcNAc = N-acetylglucosamine; Neu5Ac = N-acetylneuraminic acid (sialic acid); PG = paragloboside; SPG =
sialylparagloboside.
teins on human red cells." The weak Hike ac- the protein-coding sequence of A4GALTbut can
tivity on p red cells is conferred by x2 (renamed also be due to deletions in noncoding upstream
PX2 in 2010), a related type 2 chain glycolipid. exons." In the absence of A4GALT-encoded ac-
B3GALNTI is capable of synthesizing PX2, and tivity, there is a loss of all globo-familyand PI
this explains why PX2 is lacking on red cells of antigens. These individuals have a compensato-
the pk phenotype. 58 ry increase in type 2 chain glycolipid synthesis,
as evidenced by increased paragloboside, sialo-
Genetics paragloboside, and pX2.58 Mutations in B3GAL-
NTl give rise to the pk phenotype, which is
Several inactivating mutations have been identi- characterized by a loss of P, LKE,and PX2 anti-
fied in both A4GALTand B3GALNT1.26,71,72 The gens and by increased pk expression." The
p phenotype is the consequence of mutations in mechanism underlying the PI vs P2 phenotypes
has long been an enigma. Interestingly, individu- stance derived from pigeon eggs. Inhibiting anti-
als with weak PI expression are heterozygous PI activity may be helpful when testing sera
for pI p2 alleles and have fewer transcripts of containing multiple antibodies.
A4GALT as compared to homozygotes for pI.63
A recent study identified a transcription-factor Alloanti-PPl pk and Alloantt-P
binding site encompassing the intronic pl/ p2_
associated single nucleotide polymorphism Anti-PPlpk (historically known as antl-Tj') is a
(SNP)rs5751348 in A4GALT, where binding of separable mixture of anti-P, anti-PI, and anti-P"
the hematopoietic transcription factor runt- in the sera of p individuals. Alloanti-P (and also
the recently described anti-PX2)58is present in
related transcription factor 1 (RUNXl) was ab-
the sera of PIk and P2 k individuals, occurs natu-
sent on p2 alleles." In addition, another study
rally, and is predominantly of IgM isotype or a
showed binding of the transcription factor early
mixture ofIgM and IgG. (See Table 10-7.) The
growth response 1 (EGRl) to pI alleles on a mo-
antibodies are potent hemolysins and are associ-
tif including rs5751348_76To further complicate
ated with hemolytic transfusion reactions and,
the P/P 2 story, haploinsufficiency for Kruppel-
occasionally,HDFN. There is an association be-
like factor 1 (KLFl), as seen in the In(Lu)pheno-
tween anti-PPlpk and early, recurrent sponta-
type, is connected to lowered PI antigen expres- neous abortions. The placenta, which is of fetal
sion, suggesting KLFI involvement in A4GALT origin, is rich in pk and P antigen and is a target
regulation and PI antigen expression." for maternal cytotoxic IgG antibodies."
Antibodies to Antigens of the P1 PK and
Autoanti-P (Donath-Landsteiner)
GLOB Blood Group Systems
An autoantibody with P specificityis present in
Anti-Pl patients with paroxysmal cold hemoglobinuria
Anti-PI is present in the sera of one-quarter to (PCH),a clinicalsyndrome that most commonly
two-thirds of P2 donors." Anti-PI is a naturally occurs in children following viral infection. In
occurring antibody of IgM isotype and is often PCH, autoantl-P is an IgG biphasic hemolysin
detected as a weak, room-temperature aggluti- capable of binding red cells at colder tempera-
nin. In rare cases, anti-PI is reactive at 37 C or tures, followed by intravascular hemolysis at
shows in-vitro hemolysis. Because anti-PI is body temperature. This characteristic can be
nearly always IgM, anti-PI does not cross the demonstrated in vitro in the Donath-Landsteiner
placenta and has not been reported to cause test (see Chapter 14 and Method 4-11).
HDFN. Anti-PI has only rarely been reported to
cause in-vivohemolysis. Anti-PI titers are often Transfusion Practice
elevated in patients with hydatid cyst disease or Alloanti-PPlpk and alloanti-Pare clinicallysignif-
fascioliasis(liverfluke) and in bird handlers. It is icant antibodies associated with acute hemolytic
believed that PI-like substance in bird excre- transfusion reactions and spontaneous abortion.
ment can stimulate anti-PI levels. Some people For transfusion, rare individuals of p and pk
with anti-PI also have I blood group specificity phenotypes should be provided with antigen-
(anti-IPI).40 negative, crossmatch-compatiblered cells of the
PI expression varies in strength among indi- p and pk phenotypes, respectively. Because pk in-
viduals according to genotype" and has been re- dividuals have both anti-P and anti-PX2in their
ported to decrease during in-vitro storage." As a sera, the provision of red cell units of p pheno-
consequence, anti-PI may not be reactive with type should be avoided even if they are P nega-
all PI + red cells tested. Anti-PI can be en- tive. The p phenotype exhibits the highest ex-
hanced by incubation at low temperatures (eg, pression of PX2 of all phenotypes.58However,
4 C) or by testing serum against enzyme-treated the clinical significance of anti-PX2 is not
red cells. Anti-PI reactivity can be inhibited in known, and p units could be considered if pk
the presence of hydatid cyst fluid or PI sub- blood is not available.
In general, anti-PI is a clinicallyinsignificant, with FORSl-posltive red cells of group O. Thus,

room-temperature agglutinin. Patients with anti- the FORSl-posltrve phenotype was originallyre-
PI, which is reactive only at room temperature ported in 1987 as a new ABO subgroup, Apae'
or below, can safely receive N+ red cells for found in three English familtes." These red cells
transfusion, which results in normal red cell sur- reacted wealdy with some anti-A but were
vival.. It is not necessary to provide antigen- strongly positive with Helix pomatia lectin and
negative units to these patients. Very rarely, anti- negative with Dolichos biflorus. When ABO ge-
PI can cause decreased red cell survival.and he- notyping showed homozygosity for common 0
molytic transfusion reactions. alleles, it was revealed that the reactive antigen
Anti-PI that is capable of fixing complement was not A but FORS1.8o The responsible gene is
at 37 C and is strongly reactive in the AHG GBGTl, which encodes Forssman synthase, a
phase of testing is considered potentially clini-
glycosyltransferase able to create this specific
cally significant. In such rare instances, units se-
linkage in many mammals. This gene had previ-
lected for transfusion should be nonreactive at
ously been considered a pseudogene in humans
37 C and in an indirect antiglobulin test with ei-
ther polyspecificAHG or anti-Cd." but was shown to be reactivated by c.887G>A
(p.Arg296Gln) in FORSI-positive Indtvtduals."
Interestingly, most people have naturally occur-
THE F R B ring anti-FORSI in plasma. These antibodies
T (0 1) may cause hemolysis in vitro, but their clinical
relevance is not yet known. A summary of the
Another addition to the carbohydrate blood FORS blood group system with more informa-
group systems came in 2012, when the FORS tion was recently published."
system was acknowledged by ISBT.This system
harbors a single low-prevalence antigen, FORSI,
a glycosphingolipid synthesized by addition of u
Nacetylgalactosamine in an ex, 1-3 linkage to the
P antigen (Fig 10-6). Because this antigen bears
a certain resemblance to the A antigen, which In 2019, a new carbohydrate blood group sys-
terminates with the same ex, 1-3-linkedsugar resi- tem including the Sd" antigen was acknowl-
due, some polyclonal anti-A reagents may react edged and designated SID (see Chapter 12).
4. H antigen is ubiquitously expressed on all red celis, except in the rare Bombay (Oh)phenotype,
in which both H-synthesizing fucosyltransferases encoded by the FUTl and FUT2 genes are in-
active or absent.
S. H antigen is the precursor to both A and B antigens; thus, the amount of H antigen on red cells
depends on the person's ABO group. H antigen is highly expressed on group 0 red cells be-
cause group 0 persons lack a functional ABO gene. In group A[ and B persons, the amount of
H antigen is considerably less because H is converted to the A and B antigens, respectively.
6. LE antigens are not synthesized by red cells but are passively adsorbed onto red cell mem-
branes from soluble LE glycolipids present in plasma.
7. The three common LE phenotypes indicate the presence or absence of functional glycosyltrans-
ferases encoded by the FUT3 (LE)and FUT2 (secretor) genes.
8. As young children age, there is a gradual increase in I antigen accompanied by a reciprocal de-
crease in i antigen. Most children develop an adult 1+ phenotype by age 2.
9. Autoanti-I and autoanti-i are pathologically significant in cold agglutinin syndrome and mixed-
type autoimmune hemolytic anemia.
10. More than 99.9% of donors have the p[ (PI+) or P2 (PI-) phenotype. Both phenotypes synthe-
size pk and P antigens and differ mainly in the expression of the PI antigen. Other rare pheno-
types (P/, P/, and p) exist, in which naturally occurring antibodies against pk and P can give
rise to hemolytic transfusion reactions and recurrent spontaneous abortions.
REFERENCES
1. Paulson JC, Colley KJ. Glycosyltransferases. 9. Sazama K. Reports of 355 transfusion-associated

Structure, localization, and control of cell type- deaths: 1976 through 1985. Transfusion 1990;
specific glycosylation. J Biol Chern 1989;264 30(7):583-90.
(30):17615-18. 10. Linden N, Wagner K,VoytovichAE, Sheehan J.
2. Hansen SF, Bettler E, Rinnan A, et al. Exploring Transfusionerrors in New YorkState: An analy-
genomes for glycosyltransferases. Mol Biosyst sis of 10 years' experience. Transfusion 2000;
2010;6(10):1773-81. 40(10):1207-13.
3. Clausen H, Hakomori S. ABHand related histo- 11. SpringerGF.Blood-groupand Forssmanantigen-
blood group antigens; Immunochemical differ- ic determinants shared between microbes and
ences in carrier isotypes and their distribution. mammalian cells. ProgAllergy1971;15:9-77.
Vox Sang 1989;56(1):1-20. 12. Daniel-JohnsonJ, Leitman S, KleinH, et al. Pro-
4. LoweJB,Marth JD. A genetic approach to mam- biotic-associated high-titer anti-B in a group A
malian glycan function. Annu Rev Biochem platelet donor as a cause of severe hemolytic
2003;72:643-91. transfusion reactions. Transfusion 2009;49(9):
5. Marionneau S, Cailleau-ThomasA, Rocher J, et 1845-9.
al. ABHand Lewishisto-bloodgroup antigens, a 13. Daniels G. Human blood groups. 3rd ed. Ox-
model for the meaning of oligosaccharidediver- ford: Wiley-Blackwell,2013.
sity in the face of a changing world. Biochimie 14. Henry S, Oriol R, Samuelsson B. Lewis histo-
2001 ;83(7):565-73. blood group system and associated secretory
6. Anstee DJ. The relationship between blood phenotypes. Vox Sang 1995;69(3):166-82.
groups and disease. Blood 2010;115(23):4635- 15. Clausen H, Levery SB, Nudelman E, et al. Re-
43. petitive A epitope (type 3 chaln A) defined by
7. StorryJR, OlssonML. The ABOblood group sys- blood group Al-speclfic monoclonal antibody
tem revisited: A review and update. Immunohe- TH-1: Chemical basis of qualitative Al and A2
matology2009;25(2):48-59. distinction. Proc Nat! Acad Sci USA 1985;
8. Rydberg L. ABO-incompatibilityin solid organ 82(4):1199-203.
transplantation. Transfus Med 2001;11(4):325- 16. Svensson L, Rydberg L, de Mattos LC, Henry
42. SM. Bloodgroup A(1) and A(2) revisited:An im-
munochemical analysis. Vox Sang 2009;96(1): typing problems. Transfusion 2005;45(8): 1331-
56-61. 4.
17. Klein HG, Anstee DJ. ABO, H, LE, PIPK, 30. Moller M, loud M, StorryJR, Olsson ML. Eryth-
GLOB, I and FORS blood group systems. In: rogene: Adatabase for in-depth analysis of the
Mollison's blood transfusion in clinical medi- extensive variation in 36 blood group systemsin
cine. 12th ed. Oxford: Wiley-Blackwell,2014: the 1000 Genomes Project. Blood Adv 2016;
118-66. 1(3):240-9.
18. Cooling L. Polylactosamines,there's more than 31. YazerMH, Hult AK,Hellberg A, et al. Investiga-
meets the "Ii'': A review of the I system. Immu- tion into A antigen expression on 02 heterozy-
nohematology2010;26(4):133-55. gous group O-Iabeledred blood cell units. Trans-
19. TwuYC, Hsieh CY, Lin M, et al. Phosphoryla- fusion 2008;48(8):1650-7.
tion status of transcription factor C/EBPalpha 32. Cartron JP. [Quantitative.and thermodynamic
determines cell-surfacepoly,LacNAcbranching study of weak A erythrocyte phenotypes]. Rev
(I antigen) formation in erythropoiesisand gran- Fr Transfus Immunohematol1976; 19(1):35-54.
ulopoiesis. Blood2010;115(12):2491-9. 33. Berneman ZN, Van Bockstaele DR, Uyttenbro-
20. Auf der Maur C, Hodel M, Nydegger UE, Rie- eck WM, et al. Flow-cytometric analysis of
ben R. Age dependency of ABO histo-blood erythrocytic blood group A antigen density pro-
group antibodies: Reexamination of an old dog- file.Vox Sang 1991;61(4):265-74.
ma. Transfusion 1993;33(11):915-18. 34. Hosselni-MaafB, Hellberg A, Chester MA, Ols-
21. Mazda T, YabeR, NaThalang0, et al. Differenc- son ML.An extensive PCR-ASPstrategy for clin-
es in ABOantibody levels among blood donors: icalABOblood group genotypingthat avoids po-
A comparison between past and present Japa- tential errors caused by null, subgroup and
nese, Laotian, and Thai populations. Immuno- hybrid alleles. Transfusion 2007;47(11 ):2110-
hematology 2007 ;23(1):38-41. 25.
22. Sano R, Nakajima T, Takahashi K, et al. Expres- 35. Hult AK, Olsson ML. Many genetically defined
sion of ABO blood-group genes is dependent ABOsubgroups exhibit characteristic flow cyto-
upon an erythroid cell-specific regulatory ele-
metric patterns. Transfusion 2010;50(2):308-
ment that is deleted in persons with the B(m)
23.
phenotype. Blood2012;119(22):5301-10.
36. Beck ML, Yates AD, Hardman J, KowalskiMA.
23. Kronstein-WiedemannR, Nowakowska P, Mila-
Identificationof a subset of group B donors reac-
nov P, et al. miRNA regulation of blood group
tive with monoclonal anti-Areagent. Am J Clin
ABOgenes (abstract). Blood2015;126:158.
24. Clausen H, White T,Takio K, et al. Isolation to PathoI1989;92(5):625-9.
37. Yazer MH, Olsson ML, Palcic MM. The cis-AB
homogeneity and partial characterization of a
histo-blood group A defined Fuc alpha 1~2Gal blood group phenotype: Fundamental lessons in
alpha 1-3- N-acetylgalactosaminyltransferase glycobiology. Transfus Med Rev 2006;20(3):
from human lung tissue. J Bioi Chern 1990; 207-17.
265(2):1139-45. 38. Garratty G, Arndt P, Co A, et al. Fatalhemolytic
25. YamamotoF, Clausen H, White T, et al. Molecu- transfusion reaction resulting from ABOmistyp-
lar genetic basis of the histo-blood group ABO ing of a patient with acquired B antigen detect-
system. Nature 1990;345(6272):229-33. able only by some monoclonal anti-B reagents.
26. Reid ME, Lomas-Francis C, Olsson ML. The Transfusion 1996;36(4):35 h7.
blood group antigen factsbook. 3rd ed. London: 39. Okubo Y,SenoT, TanakaM, et al. Conversion of
AcademicPress, 2012. group A red cells by deacetylation to ones that
27. Hosseini-MaafB, Irshaid NM, Hellberg A, et al. react with monoclonal antibodies specific for
New and unusual 0 allelesat the ABOlocus are the acquired B phenotype. Transfusion 1994;
implicated in unexpected blood group pheno- 34(5):456-7.
types. Transfusion 2005;45( 1):70-81. 40. Issitt PD, Anstee DJ.Appliedblood group serolo-
28. SeltsamA, Das Gupta C,Wagner FF,BlasczykR. gy. 4th ed. Miami, FL: Montgomery Scientific
Nondeletional ABO*O alleles express weak Publications, 1998.
blood group A phenotypes. Transfusion 2005; 41. Obukhova P, KorchaginaE, Henry S, Bovin N.
45(3):359-65. Natural anti-A and anti-B of the ABO system:
29. Wagner FF,BlasczykR,Seltsam A. Nondeletion- Allo-and autoantibodies have different epitope
al ABO*O alleles frequently cause blood donor specificity.Transfusion2012;52(4):860-9.
42. Olsson ML, Irshaid NM, Hosseini-MaafB, et al. by anti-Lela) active at 37 degrees C. Blood
Genomic analysis of clinical samples with sero- Transfus 2013;11(3):456-9.
logic ABOblood grouping discrepancies: Identi- 54. Fan J, Lee BK,Wikman AT, et al. Associationsof
fication of 15 novel A and B subgroup aileles. Rhesus and non-Rhesus maternal red blood cell
Blood2001;98(5):1585-93. alloimmunization with stillbirth and preterm
43. CoolingLW,SitwalaK, Dake LR,et al. ABOtyp- birth. Int I Epidemiol2014;43(4):1 123-31.
ing discrepancies in children requiring long- 55. Navenot JM, Muller JY, Blanchard D. Expres-
term nutritional support: It is the gut after all! sion of blood group i antigen and fetal hemoglo-
Transfusion 2007;47(Suppl I): IOA. bin in paroxysmal nocturnal hemoglobinuria.
44. Shastry S, Bhat SS, Singh K.A rare case of miss- Transfusion 1997;37(3):291-7.
ing antibody due to anti-snake venom. Transfu- 56. StorryJR, Castilho L, Chen 0, et al. Internation-
sion 2009;49(12):2777-8. al Society of Blood Transfusion Working Party
45. Hult AK,DykesJH, StorryJR, OlssonML. A and on Red Cell Immunogenetics and Terminology:
B antigen levels acquired by group 0 donor- Report of the Seoul and London meetings. ISBT
derived erythrocytes following ABO-non- Sci Ser 2016;11(2):118-22.
identical transfusion or minor ABO-incompatible 57. StorryJR, Clausen FB, Castilho L, et al. Interna-
haematopoietic stem cell transplantation. Trans- tional Societyof BloodTransfusionWorking Par-
fus Med 2017;27(3):181-91. ty on Red Cell Immunogenetics and Blood
46. Yazer MH, Hosseini-MaafB, Olsson ML. Blood Group Terminology: Report of the Dubai, Co-
grouping discrepancies between ABO genotype penhagen and Toronto meetings. Vox Sang
and phenotype caused by 0 alleles. Curr Opin 2019;114(1 ):95-102.
HematoI2008;15(6):618-24. 58. Westman JS, Benktander J, StorryJR, et al. Iden-
47. Molne J, Bjorquist P, Andersson K, et al. Blood tification of the molecular and genetic basis of
group ABOantigen expressionin human embry- PX2, a glycosphingolipid blood group antigen
onic stem cells and in differentiatedhepatocyte- lacking on globoside-deficient erythrocytes.
and cardiomyocyte-like cells. Transplantation J BioiChern 2015;290(30):18505-18.
2008;86(10):1407-13. 59. Fletcher KS, Bremer EG, Schwarting GA. P
48. Larson G, SvenssonL, HynsjoL,et al. Typingfor blood group regulation of glycosphingolipidlev-
the human Lewisblood group systemby quanti- els in human erythrocytes. J Bioi Chern 1979;
tative fluorescence-activated flow cytometry: 254(22):11196-8.
Large differences in antigen presentation on 60. Suzuki A, Kundu SK,Marcus DM. An improved
erythrocytes between AI, A2, B, 0 phenotypes. technique for separation of neutral glycosphin-
Vox Sang 1999;77(4):227-36. golipidsby high-performance liquid chromatog-
49. Hosoi E, Hirose M, Hamano S. Expressionlevels raphy. J LipidRes 1980;21(4):473-7.
of H-typealpha(1,2)-fucosyltransferasegene and 61. Cooling L, Downs T. Immunohematology. In:
histo-blood group ABO gene corresponding to McPherson RA,Pincus MR, eds. Henry's clinical
hematopoietic cell differentiation. Transfusion diagnosis and management by laboratory meth-
2003;43(1):65-71. ods. Philadelphia:Saunders, 2007:618-68.
50. Combs MR. Lewis blood group system review. 62. Dunstan RA. Status of major red cell blood
Immunohematology 2009;25 (3):112-18. group antigens on neutrophils, lymphocytes and
51. Cooling L. Carbohydrate blood group antigens monocytes. BrJ HaematoI1986;62(2):301-9.
and collections. In: Cooling L, Davenport R, 63. Thuresson B, Westman IS, Olsson ML. Identifi-
Schwartz J. eds. Transfusionmedicine: Key con- cation of a novel A4CALT exon reveals the ge-
cepts in a changing world. Bethesda, MD: netic basis of the P/P 2 histo-blood groups.
AABBPress, 2020 (in press). Blood2011; 117(2):678-87.
52. Lindstrom K, BreimerME, IovallPA, et al. Non- 64. Cooling L, Dake LR, Haverty D, et al. A hemo-
acid glycosphingolipid expression in plasma of lytic anti-LKE associated with a rare LKE-
an Al Le(a-b+) secretor human individual: Iden- negative, "weak P" red blood cell phenotype:
tification of an ALeb heptaglycosylceramide as Alloanti-LKEand alloanti-Precognize galactosyl-
major blood group component. J Biochem 1992; globoside and monosialogalactosylgloboside
111(3):337-45. (LKE)antigens. Transfusion2015;55(1):115-28.
53. Hoglund P, Rosengren-LindquistR,WikmanAT. 65. Duk M, Singh S, ReinholdVN, et al. Structures
A severe haemolytic transfusion reaction caused of unique globosideelongation products present
in erythrocytes with a rare NOR phenotype. PI PK-nullalleles.Transfusion2014;54(7): 1831-

Glycobiology2007; 17(3):304-12. 5.
66. Haselberger CG, Schenkel-Brunner H. Evidence 74. Hellberg A, Poole], Olsson ML. Molecular basis
for erythrocyte membrane glycoproteins being of the globoside-deficientpk blood group pheno-
carriers of blood-group PI determinants. FEBS type. Identification of four inactivating muta-
Lett 1982;149(1):126-8. tions in the UDP-N-acetylgalactosamine:
67. YangZ, Bergstrom], KarlssonKA.Glycoproteins globotriaosylceramide3-P-Nacetylgalactosamin-
with Gala 4Gal are absent from human erythro- yltransferase gene. ] Bioi Chem 2002;277:
cyte membranes, indicating that glycoJipidsare 29455-9.
the sole carriers of blood group P activities. 75. Westman ]S, Stenfelt L, Vidovic K, et al. Allele-
] BioIChern 1994;269(20): 14620-4. selective RUNXI binding regulates PI blood
68. KhooKH, Nieto A, Morris HR, Dell A. Structur- group status by transcriptional control of
al characterization of the N-glycans from Echi- A4GALT. Blood2018;131 (14):1611-16.
nococcus granulosus hydatid cyst membrane 76. Yeh CC, Chang C], Twu YC, et al. The differen-
and protoscoleces. Mol Biochem Parasitol1997; tial expression of the blood group pl-A4GALT
86(2):237-48. and p2-A4GALTalleles is stimulated by the tran-
69. Suzuki N, Yamamoto K.Molecular cloning of pi-
scription factor early growth response 1. Trans-
geon UDP-galactose: P-D-galactoside a 1,4-
fusion 2018;58(4):1054-64.
galactosyltransferase and UDP-galactose:p-D-
77. Eernstman N, Heshusius S, Philipsen M, et al.
galactoside PI ,4-galactosyltransferase, two
KLFI regulates PI expression through transcrip-
novel enzymes catalyzing the formation of
Gala 1-4Galp 1-4Galp 1-4GlcNAc sequence. tional control of A4GALT(abstract). Vox Sang
] BioiChem 2010;285(8):5178-87. 2017;112(Sl):25.
70. Stenfelt L, Westman ]S, Hellberg A, Olsson ML. 78. Lindstrom K, van dem Borne AE, Breimer ME,
The PI histo-blood group antigen is present on et al. Glycosphingolipid expression in sponta-
human red blood cell glycoproteins.Transfusion neously aborted fetuses and placenta from blood
2019;59(3): 1108-17. group p women. Evidence for placenta being
71. HellbergA, RingressiA,YahalomV, et al. Genet- the primary target for anti-Tj-antlbodies. Glyco-
ic heterogeneity at the glycosyltransferase loci coni] 1992;9(6):325-9.
underlying the GLOB blood group system and 79. Stamps R, Sokol R], Leach M, et al. A new vari-
collection. Br] HaematoI2004;125(4):528-36. ant of blood group A: Apae. Transfusion 1987;
72. Ricci Hagman ], Hult AK, Westman ]S, et al. 27(4):315-18.
Multiple miscarriages in two sisters of Thai ori- 80. Svensson L, Hult AK, Stamps R, et al. Forssman
gin with the rare pk phenotype caused by a nov- expression on human erythrocytes: Biochemical
el nonsense mutation at the B3GALNTJ locus. and genetic evidence of a new histo-blood group
TransfusMed 2019;29(3):202-8. system. Blood2013;121(8):1459-68.
73. Westman ]S, Hellberg A, Peyrard T, et al. Large 81. Hult AK, Olsson ML. The FORS awakens: Re-
deletions involving the regulatory upstream review of a blood group system reborn. Immuno-
gions of A4GALT give rise to principally novel hematology 2017;33(2):64-72.
1
The Rh System
Thierry Peyrard, PharmD, PhD, EurSpLM, and Franz F. Wagner, MD
HE RH SYSTEM IS COMPOSED OF of childbearing potential, and the significantrisk

two genes, each encoding a polypeptide, of harm to a D-positive fetus make D antigen-
that together are responsible for the ex- matching a routine practice in transfusion medi-
pression of 55 antigens (Table 11-1). The blood cine.
group system name is Rh, and the international
symbol is RH. The attention to red cell allolm-
munization relative to this system stems from RI A R
the D antigen, which is the most immunogenic
of all minor blood group antigens. The transfu- The clinical impact of the D antigen dates to
sion of D-positive Red Blood Cells (RBCs)to D- 1939, when Levine and Stetson made the key
negative individuals results in an immunization observation that the serum of a pregnant wom-
rate of 80% to 90% in healthy volunteers,' and an agglutinated some 80% of ABO-compatible
20% to 50% in patients." samples. The authors proposed that "products of
Anti-D remains a major cause of severe he- the disintegrating fetus" and an adverse transfu-
molytic disease of the fetus and newborn sion reaction in the mother to a blood transfu-
(HDFN). A true success story in transfusion sion from her husband were related to the hem-
medicine therapy in the mid-1960s, the devel- agglutinin found in her serum." Landsteiner and
opment of RhImmune Globulin (RhIG)prophy- Wiener used an antiserum from guinea pigs im-
laxis, arose partly from the observation that munized with red cells from Rhesus macaques
ABO incompatibility between a mother and fe- to distinguish "Rh-posltive" and "Rh-negative"
tus had a partial protective effect against immu- red cells. The fact that this serum probably rep-
nization to D.s The administration of immuno- resented an anti-LW (LWantigen is increased in
globulin G (IgG) anti-D obtained from human D-positivered cells) and the failure of Levine to
plasma was effective in the prevention of HD- name their antigen triggered a fierce debate
FN.6 With the use of RhIG, alloimmunization to about who really discovered Rh. A good histori-
D in pregnancy has been reduced to occur in cal account of the confusion of the D antigen
about 1 in 4000 live births," Once a woman has with the LW system has been described by Ros-
become antl-D immunized, RhIG does not pre- enfield.?
vent strengthening of the anti-D during pregnan- "Rh-positive"and "Rh-negative" refer to the
cy. The high antl-D immunization risk, the im- D antigen status of red cells.The D and ABO.an-
pact of alloimmunization in D-negative females tigens are the principal antigens matched for
Thierry Peyrard, PharmD, PhD, EurSpLM, Head of Department, National Immunohematology Reference Labora-
tory, and Assistant Director, Medical and Scientific Director, National Institute of Blood Transfusion, Paris,
France; and Franz F. Wagner, MD, Priv-Doz, Head of Laboratory, Red Cross Blood Service NSTOB, Springe, Ger-
many, and Medical Director, ambulatory health-care center Clementinenkrankenhaus, Springe, Germany
T. Peyrard has disclosed no conflicts of interest. F. Wagner receives royalties from patents on the molecular struc-
ture ofRh.
329
TABLE 11-1. Rh Antigens by Common Name, ISBT Terminology, and Prevalence
ISBT Terminology
0 004.001 RH1 Common 85%/92% Whites/Blacks
C 004.002 RH2 Common 68%/27% Whites/Blacks
E 004.003 RH3 Common 29%/22% Whites/Blacks
c 004.004 RH4 Common 80%/96% Whites/Blacks
e 004.005 RH5 98%
ce or f 004.006 RH6 Common 65%/92% Whites/Blacks
Ce or rh; 004.007 RH7 Common 68%/27% Whites/Blacks
CW 004.008 RH8 2% Whites
CX 004.009 RH9 N2% Finns
V 004.010 RH10 30% Blacks
EW 004.011 RH11 Low
G 004.012 RH12* Common 84%/92% Whites/Blacks
RH13-RH16 Obsolete
Hro 004.017 RH17t High
Hr or Hrs 004.018 RH18* High Hrs- in Blacks
hrs 004.019 RH19§ 98% hrs- in Blacks
VS 004.020 RH20 Low 32% Blacks
CG 004.021 RH21 Common 68% Whites
CE 004.022 RH22 Low <1%
OW 004.023 RH23° Low on OVa
RH24/RH25 Obsolete
c-like 004.026 RH26 High
cE 004.027 RH27 Common 28%/22% Whites/Blacks
hr" 004.028 RH28 Low
total Rh 004.029 RH29~ High 100% except Rhnull

C HAP TE R 1 1 The Rh System 331
TABLE 11·1. Rh Antigens by Common Name, IS8T Terminology, and Prevalence(Continued)
Goa 004.030 RH300 Low Blacks
hr" 004.031 RH31§ 98% hrB- in Blacks
Rh32 004.032 RH32# Low Blacks, on DBT
RoHar,DHAR 004.033 RH33 Low <1%, Germans
HrB 004.034 RH34** High HrB- in Blacks
Rh35 004.035 RH35 Low
Bea 004.036 RH36 Low
Evans 004.037 RH37 Low on DICE hybrids
RH38 Obsolete
C-like 004.039 RH39 High
Tar 004.040 RH40 Low on DVII
Ce-like 004.041 RH41 High 70% Whites
Ces 004.042 RH42 Low 2% Blacks
Crawford 004.043 RH43 Low 0.1% Blacks
Nou 004.044 RH44 High
Riv 004.045 RH45 Low
Sec 004.046 RH46 High Sec- in Blacks
Dav 004.047 RH47 High
JAL 004.048 RH48 Low
STEM 004.049 RH49tt Low 6% Blacks
FPTT 004.050 RH50 Low on DFR, RoHar
MAR 004.051 RH51 High Finns
BARC 004.052 RH52° Low on DVI
JAHK 004.053 RH53 Low
(Continued)
TABLE 11-1. Rh Antigens by Common Name, IS8T Terminology, and Prevalence (Continued)
OAK 004.054 RH54° Low Blacks (on Dina, DOL, RN)
LORC 004.055 RH55 Low
CENR 004.056 RH56 Low
CEST 004.057 RH57 High Antithetical to JAL,

CEST- in Blacks
CELO 004.058 RH58 High Antithetical to RH43,

CELO- in Blacks
CEAG 004.059 RH59 High CEAG- in Blacks
PARG 004.060 RH60 Low
CEVF 004.061 RH61 High CEFV- in Blacks
CEWA 004.062 RH62* High
*Present on red cells expressingCor D antigen.

tAntibody made by individualswith D-deletionphenotypesD--, Dc-, and DC"-.
*Antibody made by individuals with alterede and/or D phenotypesprevalentin groups of African ancestry.
§Absentfrom red cells with DcE/DcE(R2R2) phenotypeor variant e found in groups of African ancestry.
°Low-prevalenceantigenassociatedwith the partial D indicated.
iAntibody made by individualswith Rhnull red cells.
#Low-prevalenceantigen expressedby red cells with RNor the partial DBTantlqen.'
** Antibody made by individualswith alteredC, e, and/or D phenotypesprevalentin groups of African ancestry.
ttAssociatedwith 65% of hrs-Hrs- and 30% of hrB-HrB- red cells.
transfusion. Along with the D antigen, four Rh 32,000 kDa.13,14 Nterminal amino acid se-
antigens-antithetical C/c and E/e, named by quencing of Rh was accomplished in the late
Fisher using the next available letters of the 1980s.15 The findings led to the cloning of the
alphabet-are responsible for the majority of RHCE gene in 199016 and of the RHD gene in
clinicallysignificantRh antibodies. In sickle cell 1992.17,18 The genetic basis of four different
patients, antibodies to high-prevalence Rh anti- RHCE alleleswas identified in 199319 and con-
gens such as Hrs and HrBmay be a major obsta- firmed in 1994.20
cle to transfusion support."
Rh proteins, unlike most membrane proteins,
are neither glycosylated nor phosphorylat- IN y
ed.11,12 The use of immunoprecipitation fol-
lowed by sodium dodecylsulfatepolyacrylamide Early Rh nomenclature reflects the differences
gel electrophoresis led to the discoverythat Rh in opinion concerning the number of genes that
proteins have a molecular weight of 30,000 to encode DCEce antigens. The Hsher-Racetermi-
C HAP TE R 1 1 The RhSystem 333
nology was based on the premise that three and the remaining three diglts refer to the anti-
closely linked genes, CI c, Ele, and D, were re- genic specificity; the Rh system was assigned
sponsible. In contrast, the Wiener nomenclature number 004, with RH as the international sym-
(Rh-Hr) was based on the belief that a single bol. The Rh system has recorded 62 antigens,
gene encoded several blood group factors. How- with 7 antigens deemed obsolete. The true anti-
ever, the Rh system is composed of two genes, genic variabilityis even higher, as epitopes miss-
as firstproposed by Tippett." ing in partial D and some RhCEantigens are not
The Fisher-Race DCE terminology is often glven separate antigen numbers. (See Table 9-4
preferred for written communication, but a in Chapter 9 for examples of ISBTterminology
modified version of Wiener's nomenclature for blood group antigens, alleles, and pheno-
makes it possible to identify the Rh antigens types.)
present on one chromosome using a single term,
that is, using a haplotype (Table 11-2). In the
modified Wiener's nomenclature, "R" indicates u
that D is present, and a number or letter indi-
cates the C/c and E/e antigens: R, for Ce, Rzfor Chromosomal Structure
cE, Ro for ce, and R. for CEoThe lowercase "r"
indicates haplotypes lacking D, with the C/c The Rh antigens are located on two proteins,
and E/e antigens indicated using symbols: r' for RhD and RhCE.,encoded by two genes, RHD
Ce, r" for cE, and rY for CE (Table 11-2). and RHCE, closely linked near the 3' end of
The International Society of Blood Transfu- chromosome 1p36.11. RHD and RHCE are ori-
sion (ISBT)Working Party on Red Cell Immuno- ented in a tail-to-tailarrangement: telomere - 5'-
genetics and Blood Group Terminology adopted RHD-3' - 3'-RHCE-S' - centromere. A blood
six-digit numbers to indicate red cell antigens. group irrelevant gene, TMEM50A, is located be-
The first three numbers represent the system, tween RHD and RHCE and partially overlapsthe
TABLE11·2. Prevalence of the Principal Rh Haplotypes
Rh positive
DCe Rl 42 17 70
DeE R2 14 11 21
Dee Ro 4 44 3
DCE Rz <0.01 <0.01
Rh negative
ee 37 26 3
Ce r' 2 2 2
eE r" <0.01 <0,01
CE rY <0.01 <0.01 <0.01

3' end of RHCE. In addition, another gene, differs from the next most similar protein,
RSRPI, completely overlaps RHD but has oppo- RhCE,in several amino actds, allowing for many
site orientation [Fig 11-1 (A)]. RHD likely arose possibleT-cellstimulating peptides and exposing
from RHCEin a duplication event." Both genes many immunogenic epitopes.
have 10 exons, and overall they share 97% se- RHCE is found in all but rare D- - individu-
quence identity in the coding region. als (the dashes represent missing C/c and E/e
Segmental nucleotide exchanges are com- antigens) and encodes both C/c and E/e anti-
mon between RHD and RHCE and, conversely, gens on a single protein (Ce, cE, ce, or CE). The
and are thought to be facilitated by the opposite E and e antigens differby one amino acid: a pro-
orientation of RHD and RHCE.23 One gene acts line or alanine at position 226 (p.Pr0226Ala),lo-
as a donor template during replication but re- cated on the fourth extracellular loop of the pro-
mains unchanged in the process (so-calledgene tein. Most C-positivehaplotypes derive from the
conversion mechanism). The donated region c-positivehaplotypes by a gene conversion lead-
can be one nucleotide or span several base pairs ing to an RHD-like segment surrounding exon 2
(bp), single exons, or multiple exons, typically in RHCE; this mechanism explains both the
leading to RHD-CE-D or RHCE-D-CE hybrid al- large number of amino acid differencesbetween
leles. C and c (p.TrpI6Cys, p.Leu60Ile, p.Asn68Ser,
p.Serl03Pro) and the molecular basis of the G
Gene Products (Rh Proteins) antigen (p.SerI30), expressed both by RhD and
C from RhCE.
RHD encodes the D antigen, and RHCE encodes The five principal antigens are responsible for
the CcEe antigens in four combinations (ce, cE, the majority of Rh incompatibilities, although
Ce, or CE). Both genes encode 417 amino acids. the Rh system as a whole is more complex (Ta-
The two polypeptides encoded by RHD vs ble 11-1). New antigens may result from single
RHCE differby 32 to 35 amino acids, depending nucleotide polymorphisms (SNPs)or major gene
on whether RhD is compared to RhC or Rhc. rearrangements. For example, the genetic ex-
The last decade has witnessed the development changes between RHD and RHCE can create hy-
of an abundance of information on the genetic brid proteins that express an RhDprotein with a
diversity of the RH locus, and antigen variants portion of RhCE,or vice versa.
identified by DNA-based testing have far ex-
ceeded the number identified by serology. More
than 500 RHD and 150 RHCE alleles with an GENOTY
impact on the Rh phenotype have been docu-
mented. A directory of RHD alleles is main- Inheritance studies of the five principal Rh anti-
tained by the RhesusBasedatabase," and on the gens have been used to determine Rh haplo-
ISBTwebsite, where the Working Party on Red types (Table 11-3) and to predict RHD zygosity:
Cell Immunogenetics and Blood Group Termi- in many populations, the D-negative haplotype
nology maintains, names, and catalogs new al- is associated with ce, while the D-positivehaplo-
leles." types carry C or E. However, alternate haplo-
Most D-negative (Rh-negative) phenotypes types like Ce and Dee exist and prevent a mean-
are the result of complete deletion of the RHD ingful prediction in some populations (eg, the
gene, likely through a nonsister chromatid ex- frequencies of RoRovs Rorare nearly identical in
change involving regions termed the "Rhesus persons of African ancestry). Moreover, the use
boxes," which flank RHD [Fig 11-1 (B)].23The of inferred haplotype frequencies in multiethnic
result is a "hybrid Rhesus box" that may be used societies makes prediction of RHD zygosity un-
for direct detection of the RHD deleted allele. certain. The strength of anti-Dhemagglutination
The absence of the whole RHD gene encoding cannot reliably show a difference between a sin-
the RhD protein explains why exposure of D- gle or a double dose of the D antigen: less D an-
negative individuals to D-positive red cells re- tigen is expressed when the C antigen is pres-
sults in a robust immune response. Indeed, RhD ent, a phenomenon called the "Ceppellini
C HAP TE R 1 1 TheRhSystem 335
A.
RSRPl
Rhesus boxes y
"'30 kbo
B. Chromatid crossing over
p-tel
Rhesus boxes
RHD-deleted haplotype
p-tel RHCE
FIGURE 11·1. The RH locus. (A) Organizationof RHD and RHCE in the short arm (p) region of chromosome
1p36.11. The two genes are each approximately 55,000 base pairs (bp) in size and are separated by
approximately 30,000 bp (N30 kpb). RHD is flanked by two long homologous regions (Rhesus boxes) of
approximately9000 bp. The orientation of the RH locus is: p-telornere(p-tel) - RHD - RHCE. Other genesare
in the region but are irrelevantto the expressionof Rh. (8) The origin of the RHD-deletedhaplotype. During
meiosis, a chromatid crossing-over misalignment occurs between the upstream Rhesus box (5') of one
chromatid and downstream Rhesus box (3') of another (upper figure). An RHD-deletedhaplotype (lower
figure) results from the resolution of the chromatid exchange(solid arrows), with the formation of a hybrid
Rhesusbox (5'/3'). The alternatehaplotype,two RHD in tandem (hatcharrow), has not beenobserved.
TABLE 11-3. Results of Tests with Five Principal Rh Antisera with Phenotype and Predicted RH
Genotype
~~."""",,"-~"-~~~-" ~~- ~~-~" =-""""-~"~~-~~~--'-~"
~~~-~ .. ==--="'~ .. ~~~~~-
Antisera
Predicted Alternative
Anti-D AnU-C Anti-E Anti-c Anti-e Phenotype GenotYpe* Genotype
Rh positivet
+ + 0 + + D, C, C, e R1r R1 RO
DCe/ee DCe/Dee
RO r
Dee/Ce
+ + 0 0 + D, C, e R1 R1 R1 t'
DCe/DCe DCe/Ce
+ + + + + D,C,c, E,e R1 R2 R1 r"
DCe/DeE DCe/eE
R2r'
DeE/Ce
Rzr
DCE/ee
RORz
Dee/DCE
+ 0 0 + + D,c, e ROr RORO
Dee/ee Dee/Dee
+ 0 + + + D,c, E,e R2r R2RO
DeE/ee DeE/Dee
ROr"
Dee/eE
+ 0 + + 0 D,c, E R2R2 R2r"
DeE/DeE DeE/CE
+ + + 0 + D, C, E,e R1 Rz Rzr'
DCe/DCE DCE/Ce
+ + + + 0 D,C,c,E R2Rz Rz t'
DeE/DCE DCE/eE
+ + + 0 0 D,C,E RzRz Rz rY
DCE/DCE DCE/CE
TABLE11·3. Results of Tests with Five Principal Rh Antisera with Phenotype and Predicted RH
Genotype (Continued)
Rh negative*
0 0 0 + + C, e ((
ce/ce
0 + 0 + + C, C, e t' (
Ce/ce
0 0 + + + C, E,e (" (
cE/ce
0 + + + + C, C, E,e t' r
Ce/CE
*Each genotypeis shown in both Wiener and nomenclature.
tRare genotypes (RO rY, RI rY, and R2 rY) not shown (prevalenceof <0.01%).
*Raregenotypes(rrY, rr, r'r, and rYrY) not shown (prevalenceof <0.01%).
effect."26This effect is seen both in cis [less D and patients. Testing for the common CcEe anti-
antigen is expressed in DCe/DCe (R)R))individ- gens is performed primarily during antibody in-
uals than in DcE/DcE (R2R2)individuals] and in vestigations or to provide antigen-matched
trans [less D antigen is expressed in DCe/Ce blood for certain chronic transfusion recipients,
(R)r? )individuals than in DCe/ ce (R)r)individu- such as patients with sickle cell disease (SCD)
als]. RHD homozygous. (RHD/ RHD) DCe/DCe and thalassemia, to minimize allolmmunlza-
(R)R))red cells express about as much D antigen tton." In many European countries, CcEe typing
as hemizygous (RHD/-) DcE/ce (R2r)red cells. of donors is standard, allowingfor widespread
For this reason, it is important to choose red use of Celie-matched transfusion strategies (eg,
cells with the same Rh phenotype when per- in women of childbearing potential).
forming serial anti-D titrations in the antenatal
setting because significantly different titers can D Antigen
be obtained if the red cells differ in their under-
lying zygosity. RHD zygosity (see below) can be Tests with partial D red cells revealed that
determined by DNA-basedtesting. However, de- monoclonal antl-D reagents bind to numerous
pending on the population, in addition to the different epitopes. The main epitopes are desig-
RHD deletion, many different nonfunctional nated eplil to epD9. Each epitope has addition-
RHD alleles have to be detected.27,28 al subdivisions (eg, epD6.l), leading to at least
30 different D epitopes. Most D epitopes are
highly conformational and consist of more than
simple linear amino acid residues.
AN N
D'-Positive (Rh-Positive) Phenotypes
Manufactured licensed reagents are available to
detect the expression of the principal Rh anti- Most individuals with a D-positivered cell phe-
gens-D, C, c, E, and e (Table 11-3). D antigen notype express a conventional RhD protein.
phenotyping is routinely performed on donors However, >500 RHD alleles have been reported
that encode amino acid changes. These alleles lar or transmembrane region of the protein,
can cause numerous variations in the expression rather than on the exofacialdomain of the RhD
of D antigen, and red cellswith some form of al- protem." (See Fig 11-2.) Wagner and Flegel
tered D expression are encountered in routine (et al)32proposed a system to classifyaltered D
transfusion practice. An estimated 1%of individ- red cells on the basis of their nucleotide substi-
uals of European ancestry carry only RHD alleles tutions (reviewed in Flegel and Denomme"].
that encode altered D antigens, and the inci- Not included in the definition is whether a per-
dence in individuals of African ancestry is much son with a weak D type can or cannot make
higher (up to 30% in some populations"]. Al- alloanti-D.
tered D is organized into four groups: weak D, Generally, intracellular and transmembrane-
partial D (including category D), Del, and non- ous amino acid changes are thought to affect
functional RHD.31 the insertion of the polypeptide into the
Weak D Types. Traditionally, the weak D membrane and thus result in a reduced number
phenotype was defined as red cells with a re- of D-antigen sites on the red cells. More than
duced amount of D antigen that required an in- 150 weak D types are distinguished." Other
direct antiglobulin test (IAT)for detection (for- mechanisms leading to diminished D antigen
merly called D"]. However, the number of expression without major antigenic changes are
samples identified as having weak D expression mutations interfering with splicing" and dele-
depends on the typing reagent and method tions" or duplications" of RHD exons. In per-
used, which have changed over the years. The sons of European ancestry, the most common is
vast majority of such samples carry RHD alleles weak D type 1, which has a valine-to-glycine
that encode for proteins with amino acid chang- amino acid substitution at position 270
es predicted to be located within the intracellu- (p.Va1270Gly).Types 1, 2, and 3 represent
FIGURE 11-2. Structural models of weak D (according to RhesusBase24)and partial D (according to the ISBT
listing of normal and partial D25). The locations of amino acid changes in alleles with single amino acid
substitutions are indicated by gray disks for weak D and black rings for partial D. Black disks indicate alleles with
unknown phenotype, normal D phenotype, or disputed partial D phenotype. The position of exofacialloops 3, 4,
and 6 (encoded by RHD exons 4, 5, and 7) that harbor exofacial differences between RhD and RhCE are
indicated by thick arrows. Weak D types 1, 2, and 3 (position indicated by thin arrows) are found in
approximately 90% of people of European ancestry with weak D phenotypes. Partial D types are encoded by
single amino acid changes that are generally present on the exterior (erythrocyte surface) of the cell. (Adapted
from Flegel33and Wagner.34)
approximately 90% of the weak D types.32 Weak RHCE exon 5, DIVb by RHCE exon 7, DVI by
D types can be further weakened when C is RHCE exons 4 and 5, and DBTby RHCE exons
presentin transto a weak D type (Ceppellinief- 5 and 7]. The novel sequences of the hybrid pro-
fect); for example, r.in transwithweak D type tein resulting from regions of RhD joined to
2 (R2r,).39 RhCE also explain the expression of new low-
Partial D Types. Individuals with partial D frequency antigens (FPTTin DFR, DW in DVa,
type express D antigen on their red cells. but BARCin DVI).
may form alloanti-D. Initially, partial D types Second, amino acid substitutions in exofa-
were grouped into different "category D" types cial RhD protein segments have minor impact
(categoryII to VII)based on the mutual reactivi- and may lead to phenotypes difficultto discrimi-
ty with their respective alloanti-Dr'?Nowadays, nate from standard or weak D, like DNB and
mostly monoclonal anti-D is used to investigate DHMi. In rare cases, in-frame deletions or inser-
D epitope expression, and the molecular hetero- tions of a single amino acid have a similar im-
geneity has been even greater than anticipated. pact.
Several molecular mechanisms underlie this Third, often, several amino acid substitutions
phenotype. are dispersed along the protein, as in DIna,
First, many partial D types initially recog- DNa, and DAR. Such partial D are especially
nized as D categories were due to RHD-CE-D frequent in individuals of African ancestry. In
hybrid allelesthat code for RhD proteins lacking contrast to weak D types, partial D changes are
RhD-specificityamino acids in certain protein predicted to be located on the exterior mem-
parts. The phenotype is determined mainly by brane surface" or, alternatively, can be internal
the origin of exons 4, 5, and 7 that encode for but alter extracellular epitopes.
RhD-specificexofacial protein segments (see Fig The use of a panel of monoclonal antibodies
11-3). DFR-like phenotypes were caused by to assigna D variant to a specificpartial type may
RHCElike exon 4 [category D Va (DVa) by not be reliable." Many partial D phenotypes
FPTI BARC
DFR DVI
@Rh32 @
DBT CE
@ Evans @ Rh33
DIV DHAR
FIGURE11·3. Schematic model of RhDas viewedfrom the red cell surface. RhD·like loops are indicated by gray
disks, and RhCE·likeloops by black disks. Loops 1, 2, and 5 show no constant differences betweenRhDand all
RhCEproteins (loop 2 of RhD is similar to RhCe but differs from Rhce). The antigenic character of hybrid
proteins depends largely on the replacementof RhD·like loops by RhCE-likeloops. Eachpossible combination
is associatedwith a distinct partial D phenotypeandthe presenceof a low-prevalenceantigen.
have reduced D antigen expression, making a found in individuals of African ancestry. These
serologic characterization difficult,and evidence two examples are notable because the red cells
of the partial D character may be difficult to as- show strong reactivity with some monoclonal
sess unless an alloanti-D is observed and fully anti-D reagents but are nonreactive with others,
proven as an alloantibodyby ad-hoc methods, and are thus a source of D typing discrepancies
Del Types. Red cells that express extremely (Table 11-4). Other changes in the RhCE poly-
low levels of D antigen that cannot be detected peptide may mimic aD epitope. They are encod-
by routine serologicmethods (includingIAT)are ed by alleles designated by the amino acid
designated as "D-elution," or Del, types, be- changes as ceRrand ceSL43,44The red cells are
cause their D antigens can be detected by more or less reactive with some, but not all,
adsorption/elution studies only, Del cells are monoclonal anti-D. A new RHCE allele express-
found in 10% to 30% of seemingly D-negative ing D epitopes, RHCE*ceRG, was recently de-
populations of Asian ancestry and are less com- scribed in a patient of European ancestry, show-
mon in individuals of European ancestry ing a strong reactivity with several anti-D clones
(0.027%). In Asia, the predominating allele is (MS26, HMlO, ESDI, and HMI6).45 Most im-
RHD(1227G>A),42 but the allelic background is portant, individuals with DHAR,Crawford, and
more heterogeneous in Buropeans." The major RHCE*ceRG lack the expression of a conven-
mechanisms leading to Del phenotypes are mu- tional RhD protein and can be sensitized to D if
tations at splice sites and missense mutations. no functional RHD gene is present in trans.45.47
The discrimination of Del from D negative may Elevated D. Several rare deletion pheno-
be difficult, because adsorption/elution tests types, designated as D--, Dc-, and DCw-,have
have a substantial rate of falsepositives and false an enhanced expression of D antigen and no or
negatives, and some alleleswith seeminglyinac- weak or altered C/c and E/e antigens." These
tivating mutations leg, RHD(97dupT) with a variants are the converse of partial D and result
frameshift in exon 1] may express a Del pheno- from the replacement of portions of RHCE by
type. RHD. The additional RHD sequences in RHCE
Nonfunctional RHD Alleles. RHD genes result in the additional expression of (hybrid) D
that do not encode a full-length polypeptide are antigen along with a normal RHD often in trans,
nonfunctional and have been given the ISBTal- which explains the enhanced D expression and
lele designation RHD*OIN (with "N" indicat- reduced or missing C/c and E/e antigens.
ing "null") to indicate that they are not ex-
pressed." In D-negative individuals of African O-Negative (Rh-Neqative) Phenotype
ancestry, a nonfunctional allele is prevalent,
which especially contains a 37-bp insertion and The D-negative phenotype is more common
results in a premature stop codon rendering the in people of European ancestry (15-17%), is
gene nonfunctional. It has been designated less common in people of African ancestry
RHD'P.27 In addition, hybrid alleles lacking (approximately8% in African-Americans),and is
RHCE, like exons 4 to 7, and alleleswith inacti- rare in people of Asian ancestry «0.1%).49 The
vating mutations may be the molecular basis of D-negative phenotype has arisen multiple times
such alleles. in human history, as evidenced by the different
D Epitopes on RhCE. Expression of D epi- nonfunctional alleles responsible for the lack of
topes by the protein product of the RHCE gene, D expression in various ethnic groups.
in the absence of RHD, further complicates sero- Worldwide, the D-negative phenotype most
logic determination of D status. Several RhCE frequently results from a deletion of the entire
proteins have D-specific amino acids and epi- RHD gene." In people of European ancestry,
topes that are reactive with some monoclonal other alleles are rare and usually associatedwith
anti-D. These are more often found in a specific uncommon haplotypes [r' (Ce) or r" (cE)].28In
population. Examplesinclude DHAR(Rhce'Har'), people of African ancestry, RHD"Pand a hybrid
also named R,Har, which is found in individuals allele derived from RHD*DIIJa included in the
of European ancestry, and Crawford (ceCF), so-called (Clce' type 1 haplotype are almost as
CHAPTER 11 The Rh System 341
0) Ol Ol
Q) Q) Q)
;;::: ;;::: ;;:::
+- 0) 0) 0) 0)
0)
Q) Q) Q) Q) Q)
;;::: Ol Ol Ol Ol
~o t:;, ~ E:
0)
Q)
2:
Q)
2:
Q)
2:
Q)
;;:::
0)
Q)
;;:::
0)
Q)
;;::: ~
Q) Q) Q) Q)
0... ;;::: ;;::: ;;::: ;;::: .~
1;;
0>
Q.)
c::
II
0>
Q.)
c::
0)
Q)
0)
Q) g
..c:;;
(/)
o o
(/) en
o
;;:::
...._ en
o
(/)
o +en en
o
en
o
en
o
(/)
o E .."
o
0... 0... 0... Ol
Q)
0... 0... 0... 0... 0... 0... 0... 0)
o
Q)
;;::: 0-
;;::: Il
en
o
0.
,s;
en (/)
::;
o o .c
(/) (/) (/) (/) o
0... c;,
o o o
~ o t:;,
0... 0... 0... Q)
0...
Q)
c::
;;::: ;;::: '":::>
E
..c::
~
(/) (/) en en en (/) en
'"
II
o o o o o o o C!l
0... ...._
0... 0... en Ol Ol Ol Ol :I:
Q) <0::
t:;,
Q)
0)
Q)
'<,
0)
Q)
~Q)
Q)
2: ;;::: ~Q)
Q)
2:
Q)
;;:::
Q)
2: ~Q)
~Q) <=
;;::: ;;::: ;;::: 2: 2: ;;::: ;;::: .is..
en
Q.)
1;;
:c
Q.)
E
c;:; c;:; c;:; .~
c:: c:: c::
CO 0 <0 0 II
o
Cle:; Ne:; Ci ~
~ go CI)
~ §
0
~
(5
CI)
u.J ~-
E E 0... .5
::;
.c
o
c;,
o
c::
:::>
E
co C') .~
N
si:
~ II
I- Cl :2
....J
!?
<:t Lr:> "0 ~

.Q
(/) en c:: "0
Q) Q) Q) Q) Q) c::
.;::: .;::: c:: iii <.:> Q)
Q)
c:: Q) Q) o c::
Q) zs
o CI)
.....
CI)
.....
(3
o
.&i Q)
c::
~ o o CO o o
E
<.:>
:::l
<.:>
:::l o co ~
E E E .r::: co
1::: '"
.Q
'"
(.!J E E a « ....J
<:(
common." In people of East Asian ancestry, Typing Donors for D

large hybrid alleles of the Ce haplotype are an-
The goal of D typing of donors, including the
other frequent mechanism. In people of Asian identification of units with weak D or partial D
ancestry, 10% to 30% of those who serologically types, is to prevent antl-D immunization of
type as D negative are actually De1.42 transfusion recipients. The AABB Standardsfor
Blood Banks and Transfusion Services (Stan-
Testing for 0 dards) requires donor blood to be tested using a
Monoclonal antibody production technology in- method that is designed to detect weak expres-
sion of D. There is no requirement that the typ-
troduced in the 1980s freed manufacturers from
ing should be done using an IAT, and some auto-
reliance on human source material to manufac- mated systems use enzymes to enhance
ture antl-D reagents. These antibodies are specif- detection of weak D. If the test results are posi-
ic for a single D epitope and do not detect all D- tive, the unit is labeled "Rh positive."S2(P31)
Most
positive red cells. Bythe 1990s it became appar- weak- or partial-D-antigen units are detected,
ent that monoclonal antibodies could be used in but infrequently some very weak D red cells or
a "blended" fashion to avoid deeming a preg- unusual partial D types are not detected; Del red
nant woman or transfusion recipient D-positive cells will be nome active with anti-D. Red cells
if she or he expressed the category DVIvariant. with weak D antigen are less immunogenic than
It had been known for many years that category normal D-positivered cells, but even Del donor
DVI types could make anti-D and cause signifi- units may stimulate anti_D.S3-S7 Once shipped to
cant HDFN.so,SlReagents for D phenotyping an institution, a unit labeled "Rh negative" must
be confirmed D negative by testing an integrally
were selected to circumvent this problem. The
attached segment before transfusion, but testing
immediate-spin (IS)phase uses an IgM monoclo- by IAT is not required. Units labeled "Rh posi-
nal antl-D that fails to react with red cells with tive" do not require a confirmatory test.S2(p3S)
partial DVI. This antibody is blended with a
monoclonal or polyclonal IgG that requires an Typing Patients for D
antiglobulin test (IAT)for the detection of D. In
this way, typing partial DVI as D positive could When the D type of a patient is determined, a
weak D test is not recommended except to as-
be avoided in pregnancy and transfusion by per-
sess the red cells of a newborn to determine ma-
forming only the IS phase of testing. Cord blood ternal risk for D immunization. Today, monoclo-
is tested in both the IS phase and the IAT phase nal IgM reagents type many samples as D-
to assign a D-positivestatus to most D variants. positive by IS that would have previously been
Since their development, blended anti-D re- detected only by lAT.
agents from various manufacturers have used DVI is one of the most common partial D
different monoclonal anti-D. Most Food and types found in people of European ancestry, and
Drug Administration (FDA)-approvedanti-D re- anti-D produced by women with partial DVIhas
agents combine a monoclonal IgM, which caus- resulted in fatal HDFN.sl Current FDA-licensed
es direct agglutination at room temperature, monoclonal IgM reagents are selected to be
with a monoclonal or polyclonal IgG that is re- nonreactive with red cellswith partial DVIin di-
active by IAT, for the determination of weak ex- rect tests (Table 11-4). Therefore, performing
only the direct test on red cells from female chil-
pression of D. Antl-D for column agglutination
dren and women of childbearing potential
testing may contain only monoclonal IgM or a avoids the risk of sensitization by classifyingfe-
blend of IgM and IgG. FDA-licensed reagents males with DVI as D-negative for transfusion
contain unique IgM clones, and these may ex- and RhIG prophylaxiS. However, the results of
hibit different reactivity with red cells that have positive rosetting tests (to detect fetomaternal
certain weak D, partial D, or D-like epitopes, in- hemorrhage) must be carefully evaluated; ma-
cluding DHARand Crawford (Table 11-4). ternal weak D types that are reactive only in the
IAT phase have a false-positive rosette test re- Unfortunately, licensed anti-D reagents can-
sult not distinguish individuals with partial D from
those expressing a normal D antigen. Many par-
D Typing Discrepancies tial D red cells such as DIna or DAR,two of the
most common partial D types in people of Afri-
D typing discrepancies should always be investi-
can ancestry, type strongly D positive in the IS
gated and resolved. (See "Resolving D Typing
phase and, in the absence of RHD genotyping,
Discrepancies.") D-negative blood is an appro-
are recognized as a D variant only after the pa-
priate option for female patients needing imme-
tients produce anti-D.
diate transfusion, but a thorough clerical and se-
Policies regarding D typing procedures and
rologic investigation should be performed. RHD
selection of blood components for transfusion
genotyping is also useful to resolve D typing dis-
should be based on the patient population, risk
crepancies." (See "Clinical Considerations.")
Because donor centers use test methods to of immunization to D, and supply of D-negative
detect weak D phenotypes, and generally hospi- blood. Policies should address procedures when
tals do not, a donor who is correctly classifiedas an unexpected D phenotype is encountered. Al-
D-positive may be classified as D-negative as a though it is important to prevent D immuniza-
transfusion recipient This discrepancy should tion in females of childbearing potential to avoid
not be considered problematic but, rather, HDFN, for other patients the complications of
should be communicated to the patient and anti-D are less serious, and the decision to trans-
health-care staff and be noted in the patient's fuse D-positive or D-negative blood should take
medical record. into consideration the D-negative blood sup-
ply.65
Clinical Considerations As previously stated, not all D-negative
patients make anti-D when they are exposed
The long history of providing recipients who to D-positive red cells. The incidence in
have weak D phenotype cells with D-positive D-negative hospitalized patients receiving
RBCs has suggested that some weak D pheno- D-positive blood components is variable but ap-
types are unlikely to make anti-D. In 2015, a proximates 30%.2-4MBB Standards requires
work group evaluated the scientificliterature on that transfusion services have policies that ad-
anti-D alloimmunization among Indtviduals dress the administration of D-positive red cells
whose red cells have a weak D phenotype .and to D-negative patients and the use of RhIG,
concluded that weak D types 1, 2, and 3 can be which is a human blood product that is not en-
safely treated as D-positive in pregnancy." The tirely without risk.52(pp38,47)
recommendations have been adopted by MBB,
the College of American Pathologists, the Amer- G Antigen
ican College of Obstetricians and Gynecologists,
and the Armed Services Blood Program. RHD The G antigen is found on red.cells possessing C
genotyping can be performed with reasonable or D and maps to the shared exon 2 and the 103
cost recovery that is in line with the costs associ- serine residue on RhD, RhCe, and RhCE pro-
ated with unnecessary administration of RhIG_60 teins. Antibodies to G appear as anti-D plus anti-
ThUS,implementing the committee recommen- C that cannot be separated. However, the anti-
dations can help to avoid exposing pregnant body can be adsorbed by either D-C+ or D+C-
women to RhIG unnecessarily. Recent data sug- red cells. The presence of anti-C can explain
gest that antt-D immunization in weak D type why a D-negative person who received D-C+
4.0 and 4.1 is very rare, too, but opinions on the blood, or a D-negative woman who delivered a
transfusion strategy in these genotypes remain D-C+ child, can subsequently appear to have
controversial.P'P" Other weak D types such as made anti-D. Anti-D, -C, and -G can usually be
11 and 15 have been reported to make anti-D." distinguished by adsorption and elution stud-
and information on risk for alloanti-D for other ies.66 The analyses are not often necessary in the
weak D types is not yet available. pretransfusion setting. However, it is important
to provide RhIG prophylaxis to pregnant among people of African ancestry receiving C+

women who have anti-G only and are at risk for blood, and anti-Hr" immunization may become
anti-D. an obstacle to transfusion support.
Transfusion recipients who express partial e
(/c and Ele Antigens antigens frequently appear to make antibodies
with e-like specificity, such as antl-hr" or anti-
The RHCE alleles encode the principal C/c and hrs. The red cells may lack the high-prevalence
E/e antigens. More than 150 different RHCEal- HrBor Hrs antigens.10,68,69 Partial e expression is
1e1esare known, and many are associated with associated with several RHCE*ce alleles.10
altered or weak expression of the principal anti- These alleles are found primarily in people of Af-
gens and, in some cases, loss of high-prevalence rican ancestry; some examples are shown in Fig
antigens." Partial C and many partial e antigens 11-4. The molecular variability suggests that
are well recognized, with the majority reported anti-hr'Vantt-Hr" and anti-hrvanti-Hr- may not
among individuals of African ancestry. represent a single entity; in fact, red cells desig-
nated as hrB-/HrB- or hrs-/Hrs- by serologic
Altered or Variant C and e Antigens testing may not be compatible with anti-hr"/
Nucleotide changes in RHCE result in quantita- -HrBor -hrv'-Hr" produced by patients with oth-
tive and qualitative changes in C/c or E/e anti- er RHCE alle1es.7o,71
gen expression; altered or partial C and e anti- An additional complication is that altered
gens are encountered most frequently. In RHCE*ce is often inherited with a partial D (eg,
persons of European ancestry, altered C is asso- DIll, DAU, or DAR).72As discussed above, pa-
ciated with amino acid changes on the first ex- tients with partia1-Dred cells are at risk of pro-
tracellular loop of RhCe and the expression of ducing anti-D.
CW (G1n41Arg)or CX (Ala36Thr) antigens. In
people of African ancestry, variability of RHCE is CE,Ce, eE,and ee Compound Antigens
much greater than in those of European ances- Compound antigens define epitopes that depend
try, and altered or partial C and e antigens are on conformational changes resulting from ami-
frequent. no acids associated with both C/c and E/e.
Altered or partial C is associated with chang- These antigens were referred to previously as cis
es that result in the expression of the products to indicate that the antigens were ex-
novel antigens JAHK (p.SerI22Leu) and JAL pressed from the same haplotype; that is, on a
(p.ArgI14Trp), but partial C expression most single Rhce polypeptide protein. These antigens
often results from the inheritance of an are shown in Table 11-5 and include ce (fl, Ce
RHD*DIIla-CE(4-7)-Dhybrid, and less often of (rh}, CE (Rh22), and cE (Rh27).
an RHD-CE(4-7)-Dhybrid." These two hybrids
are located in the RHD gene but do not encode
Clinical Considerations
the D antigen; rather, they encode a C reactivity
on a hybrid background that differs from the It has long been recognized that allotmmunlza-
normal background (Fig 11-4). The allele has an tion represents a significant problem in patients
incidence of approximately 20% in people of Af- with SCD because 25% to 30% or more of those
rican ancestry. It is inherited with an RHCE al- who are chronic transfusion recipients develop
lele, designated as RHCE*ctf (capital S), that red cell antibodies in the absence of minor blood
encodes partial e antigen and a V-VS+ pheno- group antigen-matching." To address the prob-
type." The expressed product of the hybrid lem, many treatment programs determine the
RHD*Dllla-CE(4-7)-Dlinked to RHCE*ctf is re- pretransfusion red cell phenotype in patients
ferred to as the (C)ces type 1 or rls haplotype. with SCD and transfuse RBCsthat are C, E, and
Red cells with the rlshap10type express a partial K antigen matched (ie, antigen negative if the
C and lack the high-prevalence antigen HrBbut patient lacks the antigen), because these anti-
type as moderately to strongly C positive with gens are considered to be the most immunogenic.
monoclonal reagents. Anti-C is not uncommon In addition, some programs attempt to supply
CHAPTER 11 The Rh System 345
TABLE11-5. Compound Rh Antigens on Rh RHD Zygosity Testing

Proteins
RHD zygosity can be determined by two ap-
Presellfon Red proaches: assaying RHD dosage or confirming
c°rri~.~~nd the presence of a hybrid Rhesus box.24,78 In pre-
Antige~ Rh Cens With These
Designation Protein Haplotypes natal practice, paternal RHD zygosity testing is
important to predict fetal D status when the
ee or f Rhee Dee(Ra)or ee (r) mother has anti-D. The management of HDFN
can vary depending on whether the father is ho-
Ceor rh;,Rh7 RhCe DCe(R,)or Ce(r') mozygous or hemizygous RHD positive. Care
must be taken in the interpretation of testing re-
eEor Rh27 RheE DeE(R2) or eE(r") sults using either approach. For RHD dosage,
testing of at least two target exons is needed to
CEor Rh22 RhCE DCE(Rz) or CE(rY) accurately determine zygosity, and nucleotide
polymorphisms in hybrid Rhesus boxes can con-
found the analysis, especially in ethnic minori-
ties.78,79 The presence of the nonfunctional
RBCsfrom donors of African ancestry whenever RHD'Ppseudogene should be included in zygos-
possible. Determining the donor and patient ity analysis as a routine practice because it is
genotypes may improve matching. Antigen- common among persons of African ancestry."
matching reduces alloimmunization significant-
ly, although there is not complete international Fetal RHD Typing
consensus on antigen-matching for all patients
with SCD.14,75 To determine the D-antigen status of a fetus, fe-
Despite matching for D, C/c, and E/e, some tal DNA can be isolated from cells obtained by
patients become sensitized because they express amniocentesis or chorionic villus sampling. An
Rh variants." It is not possible to predict who alternative, noninvasive approach is to test the
will become alloimmunized, and prophylactic maternal plasma, which contains cell-free, fetal-
antigen-matching of blood for these patients is derived DNA beyond 5 weeks' gestation.80,8! As
not feasible because of the low prevalence of is the case now in several countries, determina-
antigen-negative blood. tion of fetal RHD status using this noninvasive
procedure will likely become routine in clinical
practice to eliminate the unnecessary adminis-
RH G N PIN tration of antepartum RhIG to women who are
carrying a D·negative fetus.8o
RH genotyping is a powerful adjunct to serologic
testing for the typing of transfusion recipients, Confirming D Status
RHD zygosity determination, fetal RHD typing,
confirmation of D status, and identification of RHD genotyping is useful to distinguish partial
antigen-matched blood for patients with SeD. D from weak D or to resolve serologic D typing
discrepancies. Although patients with an uncer-
tain D status can be treated as D-negative for
Typing Transfusion Recipients
transfusion and RhIG administration, this ap-
In patients receiving chronic or massive transfu- proach may be unsatisfactory for females of
sions, the presence of donor red cells in the pe- childbearing potential who face unnecessary
ripheral blood makes red cell phenotyping by ag- RhIG injections, and it puts a strain on the limit-
glutination inaccurate. Genotyping overcomes ed D-negative blood supply. It is also important
this limitation because blood grouping can be to confirm the D status in weak D patients with
determined with DNA prepared from a blood a c- or e- phenotype, to know whether they do
sample, even if the sample was collected after or do not need to be provided with rare blood
transfusion .17 for transfusion. RHD genotyping in pregnancy
C HAP TE R 1 1 The RhSystem 347
allows informed decisions to be made on the ad- RhAG variants may lead to reduced expression
ministration of antenatal RhIG. (See "D Typing of all Rh proteins" or of RhD only.87
Discrepancies" and "Clinical Considerations" in Red cells lacking all Rh antigens are designat-
the "Testing for D" section above.) ed as RhnulJ• In the "amorph" type, both RHD
For donors, D typing discrepancies must be and RHCE are inactive, usually due to the com-
resolved because errors in determining D status monly observed deletion of RHD in D-negative
may be reportable to the FDA and result in the people combined with molecular alterations in
recall of blood components. D-negative, first- RHCE. In the more frequent "regulatory" type,
time donors are screened for RHD to detect red molecular alterations in RHAGprevent Rh anti-
cells with very weak D in some centers." gen expression.
RhnulJ red cells are stomatocytic and associat-
RHGenotyping for Patients with SCD ed with mild anemia, suggesting that the Rh
proteins have an important structural role in the
Currently, extensive RH genotyping is time-
erythrocyte membrane. TheRh complex is asso-
consuming and is used primarily for patients
ciated with the membrane skeleton through
with complex Rh antibody reactivity and to find
CD47 protein 4.2, ankyrin, band 3, Duffy, and
compatible donors in the American Rare Donor
glycophorin B and C interactlons.P'" Absence
Program for patients with antibodies to high-
of RhCE, as in D- -, leads to diminished CD44
prevalence Rh antigens." The availability of
high-throughput RH genotyping platforms en- and CD47 expression," whereas absence of
ables donors to be identified by genotyping. RhD diminishes LW expression.
However, RH genotype-matching for patients
with SCD who have rare Rh variant types may
not be possible for chronic prophylactic .transfu-
sions." Patients with SCD who cannot be sup-
ported with crossmatch-compatible transfusions
may be candidates for stem cell transplanta- Most. Rh an.tibodies are IgG but may have an
IgM component. Typically, Rh antibodies do not
tion."
activate complement, although rare exceptions
have been reported. As a result, in a transfusion
reaction involving Rh antibodies, hemolysis is
primarily extravascular rather than intravascular.
Rh antibodies have the potential to cause
clinicallySignificantHDFN. Anti-c may cause se-
vere HDFN, but anti-C, -E, and -e do not com-
In the red cell membrane, the two Rh proteins monly cause HDFN, and when they do, it is
RhD and RhCE are organized as trimeric "Rh usually mild to moderate. For antibody investi-
complexes" with a third protein, RhAG, which gations, Rh antibodies are enhanced by enzyme
shares 38% of its identity with RhD/RhCE, has treatment of red cells, and most are optimally re-
the same membrane topology, and is encoded active at 37 C.
by a single gene on chromosome 6. Amino acid
substitutions in the RhAG protein lead to the Concomitant Rh Antibodies
three antigens of the RHAGblood group system,
recognized as the 30th system by the ISBTjn Some Rh antibodies are often found together.
2008: Duclos (RHAGl), 010.. (RHAG2), (il1d For eXamPle, a. DCe/DCe (R[Rd patient with
DSLK(RHAG3).85Of note, RHAG4has been de- anti-E most certainly has been exposed to the
clared recently obsolete. c antigen as well. Anti-c may be present in addi-
Although the RhD/RhCE presence in the Rh tion to anti-B, but the anti-c may be weak and
complex is believed to be stochastic, RhAG is a undetectable at the time of testing. When seem-
critical component, and lack of functional RhAG ingly compatible E-negative blood is transfused,
prevents Rh antigen expression. Furthermore, it is most likely to be c-positive and may elicit an
immediate or delayed transfusion reaction. with high-protein reagents. A negative result

Therefore, some experts advocate for avoiding from a test that was performed concurrently
the transfusion of c-positive blood in this situa- with a similar reagent serves as a control. For
tion. In contrast, testing for antt-Ein serum con- example, for ABOand Rh typing, the absence of
taining anti-c is not warranted because the pa- agglutination by anti-Aor antl-Bserves as a neg-
tient has probably been exposed to c without ative control for spontaneous hemagglutination.
being exposed to E. In addition, the vast majori- For red cells that show agglutinationwith all re-
ty of c-negative donor blood is E-negative. (See agents (eg, group ABor D+), a control performed
Table 11-3.) In many European countries, con- as described by the reagent manufacturer is re-
sideration of the full Rh phenotype (C, E, c, e) is quired (with the exception of donor retyping).
standard practice once the patient is immunized In most cases, a suitable control is a suspen-
to one Rh antigen. sion of the patient's red cells with autologous se-
rum or 6% to 10% albumin. Indirect antiglobu-
Antibodies to High-Prevalence Rh lin testing is not valid for red cells with a
Antigens positive direct antiglobulin test (DAT)result un-
less a method is used to remove the IgG anti-
A1loantibodies to high-prevalence Rh antigens
body. Antigen-positive and -negative controls
include anti-Rh29, made by some Rh"ull individ-
should be tested, and the positive control cells
uals who lack Rh antigens, and others [anti-Hr",
should have a single dose of the antigen or be
-Hrs,-Sec, etc) that are most often encountered
known to demonstrate weak reactivity.
in transfusion recipients with SCD.
Rh Testing Considerations in HDFN
N I ER Red cells from an infant with HDFN are coated
IN with immunoglobulin, and a low-protein re-
agent is usually necessary to test these cells. Oc-
High-Protein Reagents and Controls casionally,red cellswith a stronglypositive OAT
Some Rh reagents for use in slide, rapid tube, or result may be so heavily coated that they are not
microplate tests contain high concentrations of agglutinated by a reagent with the same specific-
protein (20-24%)and other macromolecular ad- ity as the bound antibody. This "blocking"
ditives. These reagents are prepared from pools phenomenon probably results from steric
of human sera and give reliable results; howev- hindrance, or the epitope targeted by the mono-
er, high protein levels and macromolecular addi- clonal antiserum is occupied by maternal antl-D,
tives may cause false-positive reactions. (See causing a false-negative result. Heat elution of
"Causes of False-Positiveand False-NegativeRh the antibody performed at 45 C permits red cell
Typing Results" below.) These reagents must be typing, but elution must be performed with ap-
used according to the manufacturers' instruc- propriate controls to check for antigen denatur-
tions and with the appropriate controls. False- ation. Detection of the antibody in an eluate
positive results could cause a D-negativepatient confirms the presence of the antigen on the red
to receive D-positiveblood and become immu- cells, and RHD genotyping can be used for con-
nized. If red cells exhibit aggregationin the confirmation of 0 typing.
trol test, the results of the test are not valid.
Causes of False-Positive and False-
Low-Protein Reagents and Controls Negative Rh Typing Results
Most Rh antisera in routine use are low-protein False-positive typing results can be caused by
reagents formulated predominantly with IgM any of the following:
monoclonal antibodies. Spontaneous agglutina-
tion causing a false-positiveresult can occur, al- 1. Immunoglobulin-coating of the cells as a
though this happens much less frequently than result of warm or cold autoagglutinins. The
red cells should be washed several times and Resolving D Typing Discrepancies
retested with low-protein reagents by direct
To investigate D typing discrepancies, errors in
methods. If an IAT is required, IgG coating
SamPle identification or of. a clerical nature
the red cells can be removed by treating the should be eliminated by obtaining and testing a
cells with glycine/EDTA (Method 2-21) or new sample. Beyond clerical errors, ..multiple
chloroquine (Method 2-20) and retesting. variables contribute to D typing discrepancies.
2. Induction of rouleaux by serum factors .that These. variables include the use of different
can be eliminated by thoroughly washing the methods (eg, slide, tube, microplate, gel, and
red cells.and retesting. automated analyzers using enzyme-treated red
3. Use of the wrong reagent. cells), the phase of testing (DAT or IAT), differ-
4. Contaminationwith reagent from another vial. ent IgM clones in manufacturers' reagents, and
5. Nonspecificaggregationof the red cells due to the large number of RHD gene variations that af-
some component of the reagent other than the fect the level of expression and epitopes of the D
antibody (ie, a preservative, antibiotic, or dye). antigen.
6. Testing of polyagglutinablered cells aggluti- It is important to know the characteristics of
nated with reagents that contain human serum. the D typing reagent used and to always consult
7. Reactivity of the antiserum with a rare RhCE and follow the manufacturer's instructions
variant. during D typing. The FDA has drafted recom-
mendations that require manufacturers to speci-
False-negative typing results can be caused fy the reactivity of their reagents with partial
by any of the following: DN, DV, and DVIred cells."
The IgM anti-D in all .of the tube reagents
1. Failure to add the reagent. It is good practice currentlyJicensed by the FDAis reactive by di-
to add typing reagent to all test tubes or wells rect testing (initial spin) with DN and .DV red
before adding the red cells. cells but has been selected to be nonreactive
2. Use of the wrong reagent. with partial DVI red cells in direct testing. limit-
3. A red cell suspension that is too heavy for a ed studies have been performed to characterize
tube test or too weak for a slide test. the reactivity of anti-D reagents with other par-
4. Failure to detect a weak D reaction with tial D and weak D red cells. These studies have
direct testing (immediate centrifugation). shown that the anti-D reagents cannot reliably
5. Nomeactivity of a reagent with a weak or predict whether a D variant is a weak or partial
partial form of the antigen. D antigen.92,93 Table 11-4 shows the reactivity of
6. Aggressive resuspension of the red cell but- important D variant red cells that have predict-
ton, dispersing the agglutination. able patterns among the different anti-D re-
7. Contamination, improper storage, or outdat- agents. In general, females of childbearing po-
ing of the reagent. tential with partial D should be considered to be
8. Red cells with a strongly positive DAT result D positive when they are blood donors, but D
and antigen sites blocked because of a large negative when they are transfusion recipients
amount of bound antibody (most common in and with regard to antenatal prophylaxis with
severe HDFNcaused by anti-D). anti-D immunoglobulin.
350 A A B B TE C H N I CAL MAN U A L
2. "Rh positive" and "Rh negative" refer to the presence or absence, respectively, of the D anti-
gen.
3. Contemporary Rh terminology distinguishes between antigens (such as D and C), genes (such
as RHD and RHCEj, alleles (such as RHCE*ce and RHCE*Ce), and proteins (such as RhD and
RhCE).
4. Most D-negative (Rh-negative)phenotypes result from complete deletion of the RHD gene. Ex-
posure of D-negative individuals to RhD often results in the development of anti-D.
5. RHCE encodes both C/c and E/e antigens on a single protein. C and c differ by four amino ac-
ids, whereas E and e differ by one amino acid.
6. Routine donor and patient Rh typing procedures test only for D. Testing for other common Rh
antigens is used to resolve or confirm antibody identification and, for many SCD transfusion
programs or for other patients receiving chronic transfusions, to match patients and donors for
D, C, and E. .
7. Weak D phenotypes are defined as having a reduced amount of D antigen and may require an
IAT for detection. Weak D usually results from amino acid changes that impair the insertion of
the protein in the membrane. Many different mutations cause weak expression of D.
8. RHD genotyping can identify those pregnant females and blood transfusion recipients with a
serologic weak D phenotype who can be managed safely as D positive.
9. Most antl-D reagents approved by the FDA combine a monoclonal IgM (that is reactive at
room temperature for routine testing) and a monoclonal or polyclonal IgG (that is reactive by
IAT for the determination of weak D). Artti-D for column agglutination testing may contain
only IgM. These reagents may show different reactivity with red cells that have weak D, partial
D, or D-like epitopes.
10. When determining the D type of a patient, an IATfor weak expression of D is not recommend-
ed except when testing the red cells of an infant born to a mother at risk of D immunization.
D-negative donors must be tested by a method that detects weak D.
11. Most Rh antibodies are IgG, although some may have an IgM component. With rare excep-
tions, Rh antibodies do not activate complement and, thus, cause primarily extravascular rath-
er than intravascular hemolysis. Antibodies almost always result from red cell immunization
through pregnancy or transfusion.
REFERENCES
1. Klein HG, Anstee DJ. The Rh blood group sys- 5. Mollison PL, Hughes-Jones NC, Lindsay M,
tem (including LW and RHAG).In: Mollison's Wessely J. Suppression of primary RH immuni-
blood transfusion in clinical medicine. 12th ed. zation by passively-administered antibody: Ex-
Hoboken, NJ:Wiley-Blackwell,2014:167-213. periments in volunteers. Vox Sang 1969; 16:
2. Selleng K, Ienichen G, Denker K, et al. Emer- 421-39.
gency transfusion of patients with unknown 6. Freda V, Gorman J, Pollack W. Rh factor: Pre-
blood type with blood group 0 Rhesus D posi- vention of isoimmunization and clinicaltrials in
tive red blood cell concentrates: A prospective, mothers. Science 1966;151:828-30.
single-centre, observational study. Lancet Hae- 7. Zwingerman R, Jain V, Hannon J, et al, Alloim-
mato12017;4:e218-24. mune red blood cell antibodies: Prevalence and
3. FlommersfeldS, Mand C, KUhneCA, et al. Un- pathogenicity in a Canadian prenatal popula-
matched type 0 RhD+ red blood cells in multi- tion. J Obstet GynaecolCan 2015;37:784-90.
ple injured patients. Transfus Med Hemother 8. Levine P, Stetson RE.An unusual case of intra-
2018;45(3):158-61. group agglutination.JAMA1939;113:126-7.
4. Frohn C, Dumbgen L, BrandJ-M, et al. Probabil- 9. Rosenfield R. Who discovered Rh? A personal
ity of antl-D development in D- patients receiv- glimpse of the Levine-Wiener argument. Trans-
ing D+ RBCs.Transfusion2003;43:893-8. fusion 1989;29:355-7.
10. Noizat-Pirenne F, Lee K, Pennec PY, et al. Rare 25. International Society of Blood Transfusion
RHCE phenotypes in black individuals of Afro- Working Group on Red Cell Immunogenetics
Caribbean origin: Identification and transfusion and Blood Group Terminology. Blood group ter-
safety. Blood 2002; 100:4223-31. minology: Blood group allele tables, Amster-
11. Green FA Phospholipid requirement for Rh an- dam: ISBT,2019. [Availableat http://www.isbt
tigenic activity. J Biol Chem 1968;243:5519. web.org/working-parties/red-cell-immunoge
12. Gahmberg CG. Molecular characterization of netics-and-blood-group-terminologyl (accessed
the human red cell Rho{D) antigen. EMBO J October 8,2019).]
1983;2:223-7. 26. Ceppellini R, Dunn LC, Turri M_An interaction
13. B10y C, Blanchard D, Lambin P, et al. Human between alleles at the RH locus in rnan which
monoclonal antibody against Rh{D) antigen: weakens thereactivity of the Rh(O)Factor (D).
Partial characterization of the Rh(D) polypeptide Proc Nat! Acad SciUSA 1955;41 :283-8.
from human erythrocytes. Blood 1987;69: 1491- 27. Singleton BK, Green CA, Avent ND, et al. The
7. presence of an RHD pseudogene containing a
14. Moore S, Woodrow CF, McClelland DB. Isola- 37 base pair duplication and a nonsense muta-
tion of membrane components associated with tion in Africans with the Rh D-negative blood
human red cell antigens Rh(DJ, (cl, (E) and Fy. group phenotype. Blood 2000;95: 12-18_
Nature 1982;295:529-31. 28. Wagner FF, Frohmajer A, Flegel WA RHD posi-
15. SabooriAM, Smith BL, Agre P. Polymorphism in tive haplotypes in D negative Europeans. BMC
the Mr 32,000 Rh protein purified from Rh(D)- Genet 2001 ;2:1O.
positive and -negative erythrocytes. Proc Natl 29_ Compernolle V, Chou ST,Tanael S, et al. Inter-
Acad Sci USA 1988;85:4042-5. national Collaboration for Transfusion Medicine
16. Cherif-Zahar B, Bloy C, Le Van Kim C, et al. Mo- Guidelines. Red blood cell specifications for pa-
lecular cloning and protein structure of a human tients with hemoglobinopathies: A systematic
blood group Rh polypeptide. Proc Nat! Acad Sci review and guideline. Transfusion 2018;58(6):
USA 1990;87:6243-7. 1555-66.
17_ Le Van Kim C, Mouro I, Cherif-Zahar B, et al. 30. Granier T, Beley S, Chiaroni J, et at A compre-
Molecular cloning and primary structure of the hensive survey of both RHD and RHCE allele
human blood group RhD polypeptide. Proc Nat! frequencies in sub-Saharan Africa. Transfusion
Acad Sci U SA 1992;89:10925-9. 2013;53{Suppl2):3009-17_
18. Arce MA, Thompson ES, Wagner S, et al. Mo- 31. FlegelWA Molecular genetics and clinical appli-
lecular cloning of RhD cDNA derived from a cations for RH. Transfus Apher Sci 2011 ;44{I):
gene present in RhD-positive, but not RhD- 81-91.
negative individuals. Blood 1993;82:651-5. 32. Wagner FF, Gassner C, Muller TH, et al. Molec-
19. Mouro I, Colin Y, Cherif-Zahar B, et al. Molecu- ular basis of weak D phenotypes. Blood 1999;
lar genetic basis of the human Rhesus blood 93:385-93.
group system. Nat Genet 1993;5{1 ):62-5_ 33. Flegel WA Molecular genetics of RH and its
20. Simsek S, de Iong CAM, Cuijpers HTM, et al. clinical application. Transfus Clin Biol 2006; 13:
Sequence analysis of cDNA derived from reticu- 4-12.
locyte mRNAs coding for Rh polypeptides and 34. Wagner FF.Molecular genetics: The two Rhesus
demonstration of E/e and C/c polymorphism. genes and their Rhesus boxes (presentation).
Vox Sang 1994;67:203-9_ Ulm, Germany: DRK BlutspendedienstBaden-
21. Tippett P. A speculative model for the Rh blood Wlirttemberg-Hessen,2004. [Availableat http://
groups. Ann Hum Genet 1986;50(Pt 3):241-7. www.uni-ulm.de/-wflegel/RH/SympDG-
22. Wagner FF, FlegelWA RHCErepresents the an- TI20041 4WagnerDGT12004MA. pdf (accessed
cestral RH position, while RHDis the duplicated October 8, 2019).]
gene, Blood 2002;99:2272-3. 35. FlegelWA, Denomme GA AlIo-and autoanti-D
23. Wagner FF, Flegel WA RHD gene deletion oc- in weak D types and in partial D. Transfusion
curred in the Rhesus box. Blood 2000;95:3662- 2012;52:2067-9.
8. 36. Ogasawara K, Sasaki K, Isa K, et al. Weak D al-
24. Wagner FF, FlegelWA The human RhesusBase. leles in Japanese: A c.960G>A silent mutation
Version 2.3. [Available at http://www.rhesus in exon 7 of the RHD gene that affects D expres-
base.info (accessedAugust 8, 2018).] sion. Vox Sang 2016; 110(2):179-84_
37. Fichou Y, ChenJM, Le Marechal C, et al. Weak in an Rh-positiveDu variant mother. Transfusion
D caused by a founder deletion in the RHD 1983;23:91-4.
gene. Transfusion 2012;52(11):2348-55. 52. Gammon R, ed. Standards for blood banks and
38. Fichou Y, Parchure D, Gogri H, et al. Molecular transfusion services. 32nd ed. Bethesda, MD:
basis of weak D expression in the Indian popula- AABB,2020.
tion and report of a novel, predominant variant 53. Schmidt PJ, Morrison EC, Shohl J. The antige-
RHD allele.Transfusion 2018;58(6): 1540-9. nicity of the Rho (DU)blood factor. Blood 1962;
39. Wagner FF,Frohmajer A, Ladewig B, et al. Weak 20: 196-202.
D alleles express distinct phenotypes. Blood 54. Wagner T, Kormoczi GF, Buchta C, et al. Anti-D
2000;95:2699-708. immunization by DELred blood cells. Transfu-
40. Tippett P, Sanger R. Observations on subdivi- sion 2005;45:520-6.
sions of the Rh antigen D. Vox Sang 1962;7:9- 55. Yasuda H, Ohto H, Sakuma S, Ishikawa Y. Sec-
13. ondary anti-D immunization by Delred blood
41. Denomme GA, Dake LR, Vilensky D, et al. Rh cells. Transfusion 2005;45: 1581-4.
discrepancies caused by variable reactivity of 56. FlegelWA, Khull SR,Wagner FF. Primary anti-D
partial and weak D types with different serologic immunization by weak D type 2 RBCs.Transfu-
techniques. Transfusion 2008;48:473-8. sion 2000;40:428-34.
42. Shao CP, Maas JH, Su YO, et al. Molecular back- 57. Mota M, Fonseca NL, RodriguesA, et al. Anti-D
ground of Rh D-positive, D-negative, D(el) and alloimmunization by weak D type 1 red blood
weak D phenotypes in Chinese. Vox Sang 2002; cells with a very low antigen density. Vox Sang
83: 156-61. 2005;88: 130-5.
43. Wagner FF, Ladewig B, Flegel WA. The RHCE 58. Flegel WA, Denomme GA, Yazer MH. On the
allele ceRT: D epitope 6 expression does not re- complexity of D antigen typing: A handy deci-
quire D-specificamino acids. Transfusion 2003; sion tree in the age of molecular blood group di-
43: 1248-54. agnostics. J Obstet Gynaecol Can 2007;29:746-
44. Chen 0, Hustinx H, Flegel WA. The RHCE al- 52.
lele ceSL:The second example for D antigen ex- 59. Sandler SG, Flegel WA, Westhoff CM, et al. It's
pression without D-specific amino acids. Trans- time to phase in RHD genotyping for patients
fusion 2006;46:766-72. with a serologic weak D phenotype. College of
45. Vrignaud V, Ramelet S, Gien D, et al. A novel American Pathologists Transfusion Medicine Re-
RHCE allele expressing RHD epitopes responsi- source Committee Work Group. Transfusion
ble for a false-positive D typing and post- 2015;55:680-9.
transfusion anti-D alloimmunization in a patient 60. Kacker S, Vassallo R, Keller MA, et al. Financial
of Western European descent (abstract). Trans- implications of RHD genotyping of pregnant
fusion 2018;58(Suppl S2):44A. women with a serologic weak D phenotype.
46. Beckers EA, Porcelijn L, Ligthart P, et al. The Transfusion 2015;55:2095-103.
ROHAR antigenic complex is associated with a 61. Ouchari M, Srivastava K, Romdhane H, et al.
limited number of D epitopes and alloanti-D Transfusion strategy for weak D Type 4.0 based
production: A study of three unrelated persons on RHD alleles and RH haplotypes in Tunisia.
and their families.Transfusion 1996;36: 104-8. Transfusion 2018;58(2):306-12.
47. Westhoff CM. Review: The Rh blood group D 62. FlegelWA, Peyrard T, Chiaroni J, et al.A propos-
antigen: Dominant, diverse, and difficult. Im- al for a rational transfusion strategy in patients
munohematol2005;21 :155-63. of European and North African descent with
48. Daniels G. Human blood groups. 2nd ed. Cam- weak D type 4.0 and 4.1 phenotypes. Blood
bridge, MA: BlackwellScience, 2002. Transfus 2019; 17(2):89-90.
49. Race RR, Sanger R. Blood groups in man. 6th 63. Westhoff CM, Nance S, Lomas-FrancisC, et al.
ed. Oxford: Blackwell, 1975. Experience with RHD*weak D type 4.0 in the
50. Colin Y, Cherif-Zahar B, Le Van Kim C, et al. USA.BloodTransfus 2019;17(2):91-3.
Genetic basis of the RhD-positive and RhD- 64. Flegel WA. Homing in on D antigen immunoge-
negative blood group polymorphism as deter- nicity. Transfusion 2005;45:466-8.
mined by Southern analysis. Blood 1991 ;78: 65. Schonewille H, van de Watering LM, Brand A.
2747-52. Additional red blood cell alloantibodies after
51. LaceyPA, CaskeyCR, Werner DJ,MouldsJI. Fa- blood transfusions in a nonhematologic alloim-
tal hemolytic disease of a newborn due to anti-D munized patient cohort: Is it time to take pre-
cautionary measures? Transfusion 2006;46:630- related to anti-D. Prenat Diagn 2010; 12-13:
5. 1207-12.
66. Issitt PD, Anstee DJ. Applied blood group serolo- 79. Matheson KA, Denomme GA. Novel 3' rhesus
gy. 4th ed. Durham, NC: Montgomery Scientific box sequences confound RHD zygosity assign-
Publications, 1998. ment. Transfusion2002;42:645-50.
67. Daniels GL, Faas BH, Green CA, et al. The VS 80. LoYM,Corbetta N, Chamberlain PF, et al. Pres-
and V blood group polymorphisms in Africans: ence of fetal DNA in maternal plasma and se-
A serologic and molecular analysis. Transfusion rum. Lancet 1997;350:485-7.
1998;38:951-8. 81. Van der Schoot CE, SoussanAA, KoelewijnJ, et
68. Reid ME, Storry JR, Issitt PD, et al. Rh haplo- al. Non-invasiveantenatal RHDtyping. Transfus
types that make e but not hrBusually make VS. Clin BioI2006jI3:53-7.
Vox Sang 1997;72:41-4. 82. Wagner FF. RHD PCR of D-negative blood do-
69. Vege S, Westhoff CM. Molecular characteriza- nors. TransfusMed Hemother 2013;40: 172-81.
tion of GYPB and RH in donors in the American 83. Chou St, Westhoff CM. The role of molecular
Rare Donor Program. Immunohematol 2006; immunohematology in sicklecell disease. Trans-
22:143-7. fus Apher Sci 2011 j44:73-9.
70. Pham BN, Peyrard T, Tourret S, et al. Anti-Hrls 84. Fasano RM, Monaco A, Meier ER,et al. RH ge-
and anti-hrb revisited. Transfusion 2009;49: notyping in a sickle cell disease patient contrib-
2400-5. uting to hematopoietic stem cell transplantation
71. Pham BN, Peyrard T, [uszczak G, et al. Analysis donor selection and management. Blood 2010j
of RhCE variants among 806 individuals in 116:2836-8.
France: Considerations for transfusion safety, 85. Tilley L, Green C, Poole J, et al. A new blood
with emphasis on patients with sickle cell dis- group system, RHAG:Three antigens resulting
ease. Transfusion 2011;51:1249-60. fromamino acidsubstitutionsin the Rh-associated
72. Westhoff CM, Vege S, Halter-Hipsky C, et al. glycoprotein.Vox Sang2010;98: 151-9.
DIIla and DIll Type 5 are encoded by the same
86. Cherif-ZaharB, RaynalV, Gane P, et al. Candi-
date gene acting as a suppressor ofthe RH locus
allele and are associated with altered RHCE*ce
in most cases of Rh-detlciency.Nat Genet 1996;
alleles: Clinical implications. Transfusion 2010;
12(2):168-73.
50: 1303-11.
87. Mu S, Cui Y,Wang W, et al. A RHAGpoint mu-
73. Vichinsky EP, Earles A, Johnson RA, et al. Allo-
tation selectively disrupts Rh antigen expres-
immunization in sickle cell anemia and transfu-
sion. TransfusMed 2019;29(2): 121-7.
sion of racially unmatched blood. N Eng! J Med
88. Dahl KN, Parthasarathy R, Westhoff CM, et al.
1990;322: 1617-21.
Protein 4.2 is critical to CD47-membrane skele-
74. Ness PM. To match or not to match: The ques-
ton attachment in human red cells.Blood 2004;
tion for chronically transfused patients with 103:1131-6.
sickle cell anemia. Transfusion 1994;34:558-60. 89. Nicolas V, Le Van Kim C, Gane P, et al.
75. Vichinsky EP, Luban NL, Wright E, et al. Pro- RhRhAG/ankyrin-R, a new interaction site be-
spective RBC phenotype matching in a stroke tween the membrane bilayer and the red cell
prevention trial in sickle cell anemia: A multi- skeleton, is impaired by Rhlnulll-associatedmu-
center transfusion trial. Transfusion 2001 ;41: tation. J BioIChem 2003;278:25526-33.
1086-92. 90. Flatt JF, Musa RH, Ayob Y, et al. Study of the
76. Chou ST, Jackson T, Vege S, et al. High preva- D--phenotype reveals erythrocyte membrane al-
lence of red blood cell alloimmunization in sick- terations in the absence of RHCE.BrJ Haematol
le cell disease despite transfusion from Rh- 2012;158(2):262-73.
matched minority donors. Blood 2013; 122: 91. Food and Drug Administration. Draft guidance:
1062-71. Recommended methods for blood grouping re-
77. Reid ME, Rios M, Powell VI, et al. DNA from agents evaluation. (March 1992) Silver Spring,
blood samples can be used to genotype patients MD: CBER Office of Communication, Out-
who have recently received a transfusion. Trans- reach, and Development, 1992. [Available at
fusion 2000;40:48-53. https://www.fda.gov/downloads/Biologics
78. Pirelli KJ, Pietz BC,Johnson ST,et al. Molecular BloodVaccines/ GuidanceComplianceRegula to
determination of RHD zygosity: Predicting risk rylnformation/Guidances/Blood/UCM080926.
of hemolytic disease of the fetus and newborn pdf.]
92. Judd WJ, Moulds M, Schlanser G. Reactivityof 93. Denomme GA, Dake LR,Vilensky D, et al. Rh
FDA-approved anti-D reagents with partial D discrepancies caused by variable reactivity of
red blood cells. Immunohematol 2005;21: 146- partial and weak D types with different serologic
8. techniques. Transfusion2008;48:473-8.
1
Other Blood Group
Systems and Antigens
Cami Melland, MLS(ASCPYCMSBB,and Sandra Nance, MS, MT(ASCP)SBB
HIS CHAP T E R DESCRIBES31 OF 901 series, respectively.1 Several references re-

the 39 blood group systemsrecognized by lating to these antigens and their corresponding
the International Society of Blood Trans- antibodies are availablefor guidance."
fusion (ISBT).A blood group system is defined Blood group antigens may be glycoproteins,
by one or more antigens controlled at a single polypeptides, or glycolipids. The structure of
gene locus, or by two or more very closely blood group antigens provides information
linked homologous genes With little or no ob- about function, which can be helpful not only
servable recombination between them.':' The for antibody identification but also for assessing
blood group systems are listed in ISBTorder in the potential immunogenicity of an antigen and
Table 12-1. The full ISBT classificationcan be for applications related to the emerging interest
found on the ISBTwebsite(http://www.isbtweb. in immunotherapy. (See Tables 12-1 and 12-2.)
org/working-parties/red-cell-immunogenetics- However, the most important aspect of blood
and-blood-group-terminology/],and Appendix 6 group antigens in transfusion medicine is
lists all antigens assignedto systems. ISBTabbre- whether their corresponding antibodies are clin-
viations (symbols)for blood group systems-for ically significant and therefore have the poten-
example, JK instead of Kidd-will be used in tial to cause hemolytic transfusion reactions
this chapter. See Table 9-4 in Chapter 9 for ex- (HTRs)and hemolytic disease of the fetus and
amples of ISBT terminology applied to blood newborn (HDFN). Identifying antibodies to
group system antigens, alleles, and phenotypes. these antigens and determining their clinicalsig-
Many more references to blood group systems nificance or lack thereof can determine the clini-
and antigens than can be. provided here are cal course of action for .a patient who has
availablein various textbooks and reviews. 3-5 produced these antibodies. Emerging drug regi-
The other groups of antigens described at the mens that interfere With blood bank testing (eg,
end of this chapter are not yet assigned to a sys- anti-CD38 therapy) and autoantibodies add to
tem. Those that are serologically,biochemically, the difficulty in not only determining alloanti-
or genetically related to a blood group system body specificity but also finding compatible
but do not meet all the criteria are grouped into blood." Fortunately, advances in molecular red
"collections." Others are classified together in cell genotyping have helped predict the patient
the low- or high-prevalence groups from most phenotype for some of the antigens discussed in
major populations and make up the 700 and this chapter."
Cami Melland, MLS(ASCP)cMSBB,Immunohematology Reference Lab Manager, Vitalant Mountain Division,

Denver, Colorado; and Sandra Nance, MS, MT(ASCP)SBB,Senior Director, National Laboratories, American Red
Cross, Senior Director, American Rare Donor Program, and Adjunct Assistant Professor, University of Pennsylva-
nia, Philadelphia, Pennsylvania
355
eCD
a;
en
"C
CD
en
::::l
ro
'-'
en
ro
..c::
~ I
2
E
ro
;2
u..
u;
o
o I ;2
2 I CD u;
u;
o "C en
::::l
o
CD I
I en ~ "C
::::l
M
N
CD
en
::::l
ro
s
CD
2-
ca::::l
CD
en
::::l
ro
"C '-' ~ en '-'
c::: c::: ..c:: ::::l CD
ro
o ~ "5 ~
..c::
en c:::
.!E ~ en en
CD
"C CD
o
.c
"C
c:::
ro
o
"C
:a
o
:;::; ..c
u) c::: ~ :1:: :;::;
I :2'
u;
ro2 I 7 c:::
ro
'-i=l .....JLl... :;::; '-i=l
c::: o o o UJO o c::: c::: o o (.)
« I 2 2 ~ I 2 « « 2 2 CI)
"C
c:::
ro
en
a:
en
a:
l- l-
I I
« o
CD "C
en c:::
::::l ro
ro
'-' en
(.) a:
l-
r-, I
en C'")
«
s::
Cl:l
Cl
1<1 e
CD
:;::::;
s::
~ >
CD
« t5
ro
en
CD
c.. N en en
::::s CD ::::l
e
CD
N
"C
c:::
C.
E
ro
'-'
c:::
ro
"'C o ~ ~ en
o CD
en a:
o
C5
e CD
:a l-
- ~ I
o
o
..c o
en 1i5. :;::;
c:::
.~ >~ ro
>-1-
"'C
o "C:I
00
.....J
UJ
~
o 2o
2
..0
:;::::;
s::
«
<+-
o
Cl:l
t.;)
s::
C"<:l
t.;) l!) l!) N
to N
'+=
'i:;
Cl
U5
cti
.52
s::
CI)
("_) o 2 I
.....J
(.)
CCl UJ UJ
« 2 a: ~ .....J CI)
,.- N C'") eo co r-... 00 O"l

,...
Cl
,... N ,...
C'")
o
o
o
o
o
Cl
""'"
o
o
o
o
Cl
Cl
Cl
Cl
Cl
Cl
o
Cl o Cl
-r-'
Cl Cl
CHAPTER 12 Other Blood Group Systems and Antigens 357
Cl)
~
.<::
C"")
o
(.)
I
-0
c::
ro::s
~o
y
""§
ro::s
2:
LL
o
I
e
Cl)
>
Cl)
(/J
-0
ffi2:
:::lLL
ro::sO
<':>I
:=~~'E
0-0
Y 55
~ ~ ~
o
2: 2:
o o
2:
o
2:
o 2:
2:
o o
2:
o
2:
o
2:
-0
.@ -0
Cl)
o (/J
Cl. :::l
e ro::s
<.:> <Ii
0:::
c:: Cl)
l-
Cl)
Cl)
..0
~
x: I
Cl)
.s ~
::;
~
sz: -0 <.:>
.@
~
-
o
I Cl. ~
'<-
o
~ .5
c::
en Cl)
Cl)
-0
Cl)
Cl)
(/J ..0 (/J
ro::s (/J
:::l
<.:> ro::s ro::s
o .<:: <.:>
c:: Cl)
-0
± ~
c:: 2: x:
ro::s -;> en
e ~ro::s
Cl)
e ::0
o
.c
2:' '0
Cl)
>
Cl) ~ro::s
Ci.
.!:Q
:1::
Eo::: Cl)
-0
'en
aI ~I- o
Cl)I
.c
o 2:
o o o o ac::
E 2: Cl)«
c:: o o
2: 2: 2: « ro::s (5 2:
(!l
2: co
--
0::: I
a c, I a
a a $: I ~ u.J 0::: 2: ~ « 2: ...J
...J
0 (.) ...J (.) I X (!l (.) ~ 2: a 0::: -, (!l (!l
'<t co CD r..... co 0).

,_ 0 T- N C"") '<t l[) CD r..... co 0)
T-
0 o T-
0 o 0 0
N
0
N
0
N
0
N
0
N
0
N
0
N
.0
N
0
N
0
N
0
c::
..
a:>
a:>
.e
Z a:>
>
;;;::
Q
~
LL
Cl
'"
.c::
....J
...= :c
a:>
:::2:
0
..CI
:=:
Q)
>
a:>
-ci
a:>
!f<=
CI) (/) -ci t:: c::
Z
"CJ
;;;::
LL
"C
a:>
a:>
t:: 0
c. '"c::
"C
= Cl
(/) 0
e
C.
'"
.. '"
:::I
"'
tn
:c
'"a:> ::: c::
=
a;
'+-
0
a:>
U
>
c::
a:>
a:>
a:>
a:>
.e
~
<=
c::
CI)
(/)
u '"
sz: .e
a:>
a:>
>
'"
>-
:S a:>
"L
7
> '"
x: .e
'5 c::
0 ".;::;
c::
'"
x:
2:
2:
LL
2:
LL
CI) "C Cl Cl <Ii
tn
"'tn
a:>
'"
'+-
LL
Cl :c :c Q)
'"
(/)
CI) :::I 0
(/)
:c e e -0
c.
Ci '"
u a:>
i5.
:2 a:>
> -ci a:>
> -ci -ci -ci :::J
,_
U
(/)
"E a:> a:> a:> a:> a:> a:>
~ '"
.s:
'<t
E
'"><
'+-
0 b
(/)
t::
0
c.
(/)
'+-
0 -ci
a:>
t::
0
c.
t::
0
c.
t::
0
c.
.E
!!l
Ci
E
C.!J
<:(
a:>
0
(/)
a:>
(/)
(/)
a:>
(/)
e
.....
(/)
a:>
(/)
t::
0 ..... ::: .....e
e ..... "en
..c
Q)
:c 0 c. ;s:
'" e
CI) 0 0 0 0
:c a: 2: ~ '"
(.) '"
(.) 2: (.) 2: 2: 2: I-
eD
!::Q
a:> Q)
.e a:> Q)
~ en
.....
0
.s::; -ci
"C Q)
a:> -ci Q5
~
t::
0
a:>
t:: > I a;
:0- c. 0 <=
c::
-0
Q)
:::s
e c.
e .....'"0
(f)
'"
;s:
e:: c::
:;::; a:> c:: O"l
a:> a:> a:> C')
e:: .e a:>
.e
:::I 0
0 a:> "C a.
2- ii: (/)
a: -ci :::J
I- ~
x: '"
x: I-
a:>
t:: e
en :c :c Cl
e:: -::.~ a: c:: Cl 0
C. -0
Q)
01 =0:
01-
I-
:c '" :::
....J e a
a
zs
:;::;
e:: =:c
UQ
<:(
'+-
<=
I
c::
a:>
>
a:>
c::
a:> (f)
<X: '" a:>
c..
«S-
&g~
O
a:> ..... 0
0
(/) .e '"
...J
t;
:::s (/)
..... (/)
e 'w= ~>- 0 '"

U
a:>
a:>
:::I
"C
::s!
·E
'"
.c:: Q)
>
CD CI)
Q c:: a: ~ ea.
"C ~.
ii;,;,;.
0 'L I- "C
<= C.
0 "C :c c::
0 =0 c::1 c::
'"
'" c:: ::: a:'" -ci '" .s
C'a~
iii "'0: a:> -ci .s -ci -ci -ci
_
,","l- (/) a:> Cl
0
...... .S::! :c a: '"
>-
t::
>
(J;>
l-
:c 0
a:>
t:: a:>
(J;>
t::
a:>
t::
a:>
t::
c::
:;:::;
l- .e
>.< .._,. :c "C (/) <:( c. 0 :::I 0 0 0 Q)
::: :::
(/) ..
c. "C C. C. c. Q)
"~ 0 a.I Cl a:> .....

0
a:> a: ::: ::: ::: E
"C E :; :2
(/)
:::I ::s!
(J;>
>
c::
(J;>
'-
..... I- ..... ..... ..... €'
0
.c CI) U 0 0 a:> (J;> 0 :c 0 0 0
:;::; :c< 2: 2: ~ '" ~
u en .e 2: <:( 2: 2: 2: 8!. c:
e:: Cl
c:: 0
<X: :g ·00
..... :::J
0
Q)
(.;)
tn
= :s:a Ii;c::
e:: '5 CI)=
l-
eD ~
Cd !::Q -0
(.;) c::i ~ C') '<t N 0
:.;::::: z< Q) 0
"c
Cl
:E CO
- .....0
(f)
U5 (f)
Q) p
c;;
,_
c.· u
Q)
(.;)
·c .s en
0
ECi a 1:: e;;
C3 CI).c en
C.!J CTl 2: N Q)
;s: c::
<:(a: C.!J 2:
.,....
i.~
en en
:c a
a: LL a:
J
2:
<:(
....J
....J
w
>
I.!)
Cl
(.)
:::::>
<:(
<:(
~
Cl
U5
....J
l-
(.)
e;;
:::J ~
0
c:: E
N
.,.... '" -=
E !!l
_.
IoU
•
Q)
:E II
= Iii
c::i 0
C')
,-
C')
N
C')
C')
C')
'<t
C')
I.!)
C')
(.0
C')
I"-
C')
00
C')
CTl
C')
(f)
<C
I-
eD
~ ~ z 0 0 0 0 0 0 0 0 0 0 • !::Q
C HAP T E R 1 2 Other Blood Group Systemsand Antigens 359
TABLE 12-2. Blood Group Function"
Receptors and Adhesion
FY Proinflammatory chemokine receptor and Plasmodium vivax and knowlesi receptor

KN/CROM Receptor to remove complement-coated immune complexes
MNS Plasmodium falciparum receptor and association with the anion transporter band 3
IN (C044) Adhesion of leukocytes to endothelial and stromal cells, stimulating T- and s-een activation
Lu/LW Ig superfamily adhesion moleculars, receptors, and signal transducers
OK (C0147) Receptor for cyclophilin A; facilitation of dermoblasts to increase matrix metalloproteinases
for healing and/or development
CH/RG Complement activation; null types associated with lupus
Transporters and Channels
01 Anion exchange; maintenance of the red cell Shape/structure
JK Urea transport
CO Water channel
Enzymatic Reactivity
YT Acetylcholinesterase essential in neurotransmission

Ig = immunoglobulin;ATP= triphosphate.
toskeleton.. GPAis abundant, with about 106

copies per red cell, whereas GPBhas only about
MNS is a complex blood group system consist- 200,000 copiesp.ercell. GPAforms an associa-
ing of 49 antigens on two glycoproteins. Much tion in-the membrane with band 3 (Dr blood
of its complexity arises from recombination be- group system), and. botll GRA. and GPB appear
tween closely 'linked homologous genes. Al- to be Part of the band 3/Rh ankyrin macrocorn-
though there are many antigens in the MNS sys- plex (Fig 12-1).9 The relatively Jew copies of
tem, the most well known are M, N, 5, and s. GPB on the red cell.surface could be linked to
The M and N antigens are located on glyco- the somewhat Iowincidence' of identification of
phorin A (GPA,CD235A), and the S and s anti- antibodies to S and s antigens.'?
gens are located on glycophorin B (GPB, GYPkand GYPB, the genes encoding GPA
CD235B). and GPB, are located on chromosome 4q31.21
Both GPA and GPB cross the membrane and .include seven and five exons, respectively.
once and have an external N-terminal domain Most genetic recombination between GYPA and
and a C-terminal cytosolic domain. The extra- GYPBgenes occurs.in a 2-kb stretch from exons
cellular domains of both molecules have many 2 to 4, resulting in many polymorphisms and di-
sialic-acid-richO-glycans. GPAis N-glycosylated versity of antigen expression within this system
at asparagine-45 (position 26 in the mature pro- (Fig 12-2).9A third gene in this GPA gene fami-
tein), whereas GPB is not N-glycosylated.The ly, GYPE, probably produces a third glycopro-
long cytosolic tail of GPAinteracts with the cy- tein, glycophorin E, but this plays little or no
<C
c..
(!)
co
c..
(!)
<C
c..
o
C HAP T E R 1 2 OtherBloodGroupSystems and Antigens 361
8283848586
'¥
GYP(B-A-B)
Mur.
82/ \848586
83'¥ A3
Extracellular Cytosol
FIGURE 12-2. GYPA, GYP8, and the hybrid GYP(8-A-8) gene responsible for GP.Mur, and a representation of
the proteins they encode, showing the regions of proteins encoded by the various exons.
\jf = pseudoexon not representedin the mRNA or the encoded protein.
part in MNS antigen expression and is not de- fifth proteins are also known as positions 20 and
tectable by routine methods. 25 because GYPA creates 19 amino acids that
GPA is restricted to blood cells of erythroid are removed when the GPAprotein binds to the
origin and is often used .as anerythroid marker. red cell membrane and becomes a mature pro-
A GPAlike molecule has been detected on renal tein. The N+ GPAhas leucine and glutamic acid
endothelium. Both GPA and GPB are exploited at those positions. The amino-terminal 26 ami-
by the malaria parasite Plasmodiumfslciperum no acids of the GPB mature protein are usually
as receptors for binding to red cells and may be identical to those of the N form of GPA,includ-
critical to the invasion process. 11 As a result of ing the cleaved amino acids 1 to 19. Thus, in al-
this phenomenon, indlvlduals from ethnic most all people of European ancestry and most
groups in areas where P. fslciperum is endemic people of other ethnicities, GPB expresses 'N.'
are more likely to be negative for the S and s an- However, because GPB is much less abundant
tigens than other populations. This fact can be than GPA, most anti-N reagents do not detect
helpful information during the course of anti- the 'N' antigen on GPB.
body identification. Sand s are another pair of polymorphic anti-
thetical antigens of the MNS system, carried on
M (MNS1), N (MNS2), S (MNS3), and GPB. Familystudies show linkage between M/
s (MNS4) Nand Sis. S+ GPB has methionine at the 29th
position of the mature protein, whereas s+ GPB
M and N are antithetical antigens and polymor- has threonine.
phic in all populations tested (see Table 12-3 for The N-termin.al re~ol1 qf GPA j~. cleaved
phenotype frequencies). M and N are located at from intact red .Cyllsby trypsin,.wher~as that of
the N-terminus, or free-amine group (-NHz)' at GPBis not. Consequently, M and Na_nqgel1son
the end of GPAin M+ and/or N+ red cells. M+ GPA are trypsin sensitive, and S, s, and 'N' on
GPAhas serine and glycine at the first and fifth GPB are trypsin resistant. In contrast, with
positions of the mature protein. The first and a-chymotrypsin treatment of red cells, M and N
TABLE 12·3. Prevalence of Some Phenotypes are not active at 37 C, are not clinicallysignifi-
of the MNS System cant, and can generallybe ignored in transfusion
practice. If room-temperature incubation is elim-
inated from compatibility testing and screening
for antibodies, these antibodies are often not de-
tected. When M or N antibodies active at 37 C
are encountered, antigen-negative red cells or
those that are compatible by an indirect anti-
M+N- 30 25
globulin test (IAT)should be provided. Very oc-
M+ N+ 49 49 casionally, anti-M has been implicated as the
cause of acute and delayed HTRs, and anti-M
M-N+ 21 26 has very rarely been responsible for severe
S+s- 10 6 HDFN.12 Anti-N is not generally associated with
HTR or HDFN. A few cases of warm autoim-
S+s+ 42 24 mune hemolytic anemia (AIHA)caused by auto-
S- s+ 48 68 antl-N have been described, one of which had a
fatal outcome. However, in the absence of he-
S-s- 0 2 molysis, the specificity of an autoantibody is
generally not significant. (See Chapter 14 for
more information on autoantibodies with alloan-
tibody spedftcity.)
activity is only partially reduced, whereas S, s, Anti-S and -s are usually IgG antibodies that
and 'N' expression is completely destroyed. M, are active at 37 C. They have been implicated in
N, S, s, and 'N' are all destroyed by treatment of HTRs and have caused severe and fatal HDFN.
the red cells with papain, ficin, bromelin, or pro- Autoanti-S has caused AIHA. If immunized, in-
nase, although this effect with S and s may be dividuals with S-s-U- red cells may produce
variable. antl-U. Anti-U has been responsible for severe
and fatal HTRsand HDFN. Autoanti-U has been
S-s-U- Phenotype implicated in AIHA. Because U- types are rare
and are most often encountered in people of
The red cells of about 2% of Americans of Afri- African ancestry, it can be helpful to screen
can ancestry and a higher proportion of Africans those donors for S and S. If S-s- is determined
are S-s- and lack the high-prevalence antigen U serologically,it is helpful to submit samples for
(MNS5). The S-s-U- phenotype often results molecular testing to detect U+VAR, which is diffi-
from homozygosity for a deletion of the coding cult to determine serologicallydue to the unreli-
region of GYPB,but other, more complex molec- abilityof antisera.
ular phenomena involving hybrid genes may
also give rise to an S-s- phenotype with expres- Other MNSSystem Antigens and
sion of a variant U antigen (U+VAR). U is general- Antibodies
ly resistant to denaturation by proteases-
papain, ficin, trypsin, and a-chymotrypsin. The other MNS system antigens are of either
However, antt-U is not reactive with papain- high or low prevalence in most populations. The
treated red cells in rare cases. similarity of sequence between certain regions
of GYPA and GYPB may occasionally lead to
Antibodies to M, N, S, 5, and GYPA pairing with GYPB during meiosis. If re-
UAntigens and Their Clinical combination then occurs, either by crossing
Significance over or by gene conversion, a hybrid gene can
be formed consisting partly of GYPA and partly
Anu-M is a relatively common antibody, where- of GYPB. A large variety of these rare hybrid
as anti-N is less common. Most anti-M and -N genes exist, and they give rise to low-prevalence
C HAP T E R 1 2 OtherBloodGroupSystems and Antigens 363
antigens and, in the homozygous state, to phe- THE ru S T (005 )

notypes that lack high-prevalence antigens,"
One well-known example of an antigen created LU (Lutheran) is a polymorphic system consist-
by this recombination is the Mi(a+) phenotype ing of 25 antigens, the majority of Which are
created by the hybrid gene that is responsible for highly prevalent in all populations tested. There
the GP.Mur (previously Mi.Ill) phenotype. The are four antithetical patra=-Luvl.u", Lu6/Lu9,
hybrid gene is mostly GYPB, but a small region Lu8/Lu14, and Aua/Aub-ofwhich ur, Lu9,
of GYPB encompassing the 3' end of the pseudo- and Lu14 are oflow prevalence." Au" and Aub
have a prevalence of around 80% and50%, re-
exon and the 5' end of the adjacent intron has
spectively, in people .ot European ancestry. Of
been replaced by the equivalent region from GY-
most relevance in transfusion medicine, Lu'
PA. This means that the defective splice site in
(LUl) has a prevalence of about 8% in people of
GYPB is now replaced by the functional splice European or African ancestry but is rare else-
site from GYPA, and the new, composite exon is where; its antithetical antigen, .Lub (LU2), is
expressed in the messenger RNA (mRNA) and common everywhere.
represented in the protein." This provides an LU antigens are destroyed by treatment of
unusual amino acid sequence that is immuno- the red cells with trypsin or a-chymotrypsin,
genic and represents the antigens Mur and Mia. whereas papain and ficin have little effec:t.Most
Another recombination example is the amino LU antibodies are not reactive with red cells
acid sequencethatresultSfr~TI1 th~'unction of treated with the sulfhydryl reagents 2-amino-
exons B3 and A3, which gives rise to Hil and ethylisothiouronium bromide (AET) pr dithio-
MINY (Fig 12-2). threitol (DTT), which reduce the idisulfide
Mur antigen is rare in people of European bonds of the immunoglobulin superfamily (IgSF)
and African ancestry but has a prevalence of domains (Method 3-18).
about 7% in people of Chinese ancestry and The LU antigens are located on a pair of gly-
10% in people of Thai ancestry. Anti-Mur has coproteins that differ by the length of their cyto-
the potential to cause severe HTRs and HDFN. plasmic domains as a result of alternative RNA
In Hong Kong and Taiwan, anti-Mur is the most splicing. They are encoded by the BeAM allele
common blood group antibody after anti-A and located on chromosome 19q13.2, and the mole-
-B. In Southeast Asia, it is important that red cule is called CD239. The proteins span the
cells for antibody detection include a MUT+sam- membrane once and have five extracellularlgSF
pie. 14 domains. The function of IgSF proteins is
Antibodies with the generic name anti-Bn' thought to be associated with medlatton of cell-
to-cell adhesion for cell recognition and innate
may be made by very rare individuals who lack
and adaptive immune responses, and they pos-
all or part of GPA; these antibodies have caused
sess structural features shared with .immuno-
severe HTRs and HDFN. There are some ex-
globulins." The isoform or protein with the lon-
tremely rare individuals who lack both GPA and ger cytoplasmic domain interacts with spectrin
GPB as a result of being homozygous for the Mk of the red cell membrane skeleton. The location
silencing allele. The red cells of these individu- of the LU antigens on the IgSF domains is
als type M-, N-, S-, s-, U-, and En(a-).1s An shown in Fig12-3. The LU glycoproteins are ad'
antibody made by an individual Withthis pheno- hesion molecules that bind isoforms of laminin
type did cause severe HDFN requiring intrauter· that contain 0,-5 chains. Laminin.is a glycopro-
ine transfusion with blood donated. by family tein of the extracellular matrix, and Lll-lamlnin
members. Three types of ann-Err have been interactions may playa role in the migration of
recognized serologically [anti-Bn'Pi, -EnaFR,and mature erythroid cells from the marrow to the
-EnaTS), based on reactivity with ficin- and peripheral blood at the latest stages of erythro-
trypsin-treated red cells. poiesis. Upregulation of LU glycoproteins on red
reacting with all red cells except those from

Lua/Lub LURC Lu(a-b-) individuals. The In(Lu) phenotype is
Lu21 another rare group with extremely weak expres-
Lu5 sion of LV antigens detectable only by adsorp-
Lu17 tion/elution or predicted by molecular tech-
Lu12 niques. In(Lu) is the result of mutations in the
erythroid transcription factor gene KLFI. Muta-
Lu4
tions in KLFI also affect other blood group genes
Lu8/Lu14
Lu16
and cause weakened expression of several other
LUGA antigens, including PI, In", and AnWj and can
LUAC be associated with hematologic abnormall-
ties.20,21 This is a rare, dominant suppressor of
Lu6/Lu9 LV antigens. The In(Lu) phenotype has a preva-
Lu20
lence of around 0.03%. In one family, hemizy-
gosityfor a mutation in the Xllnked gene for the
major erythroid transcription factor GATA-1re-
sulted in a Lu(a-b-) phenotype with an Xlinked
mode of inheritance."
Clinical Significance of Antibodies to

LU Blood Group System Antigens
LV antibodies are most often IgG and demon-
strate reactivity best by rAT;they have generally
been implicated only in mild delayed HTRs.
Antl-Lu" may be "naturally occurring" or im-
mune, and it is often IgM but may also be IgG
and IgA. These antibodies are usually reactive
by direct agglutination of Lu(a+) red cellsbut of-
eOOH ten alsoreactive by an lAT.Anti-Lu''may show a
"mixed-field't-like agglutination. Lu(a-b-) peo-
ple may form anti-Lu3,which appears to be anti-
FIGURE 12-3. Diagramof thetwo isoforms of the LU Lu"plus anti-Lu".
glycoprotein, showing the five extracellular
immunoglobulin superfamily domains and the
location of the LU antigens on these domains,the ND XK (01 )
single membrane-spanning domain, and the
cytoplasmicdomains.
The antigen often referred to as "Kell," but cor-
rectlynamed "K" or "KELl," is the originalanti-
cells of patients with sickle cell disease could
gen of the KELsystem and the first blood group
playa part in adhesion of these cells to the vas-
antigen to be identified following the discovery
cular endothelium and the resultant crises of
of the antiglobulin test in 1946. Its antithetical
vascular occlusion."
antigen, k or KEL2,was identified 3 years later.
The KEL system consists of 36 antigens num-
Rare LU Phenotypes
bered from KELl to KEL39,of which three are
The extremely rare Lunull phenotype arises due obsolete.f The KELsystem includes seven pairs
to homozygous inheritance of the inactive LU (K/k, Jsa/Jsb, K111K17, K14/K24, VLAN/
gene." Red cells from these individuals lack ex- VONG, KYO/KYOR, and KHVLlKEAL) and
pression of LV antigens, may produce anti-Lux, one triplet (Kp"/Kpb/KpC)of KELantithetical an-
tigens. Initially, most antigens joined the KEL Kx blood group antigen (XKl). Absence of Xk
system through genetic associations observed in protein from the red cell results in reduced ex-
family studies. These associations have now pression ot the KELglycoprotein and weakened
been conflrmed by DNA sequencing of the KEL KELantigens (McLeo.dphenotype, see below).
gene. The KEL gene is located on chromosome
7q33.Itisorganized iIlto.19 exons: exon 1en-
The KELGlycoprotein and the KEL Gene codes the probable translatio.n-initiatingmethl-
onine; exon 2, the cytosolic domain; exon 3,
The KELantigens are located on a red cell mem-
the membrane-spanning domain; and exons 4
brane glycoprotein (CD238). KEL is a type II
through 19, the large extracellular domain.
membrane glycoprotein; it spans the membrane
once and has a short N-terminal domain in the
KELAntigens
cytosol and a large C-terminal domain outside
the membrane.23,24 The K antigen has a prevalence of about 9% in
The extracellular domain has 15 cysteine res- people ot European ancestry, about 2%in people
idues and is extensively folded by disulfide or African ancestry, and is rare in East Asia (Ta-
bonding, although crystallographic studies are ble 12-4). The k antigen is highly prevalent in all
required to. determine the molecule's three- populations. K and k result from a single nucleo-
dlmensional structure. KELsystem antigens de- tide polymorphism (SNP)in exon 6, which en-
pend on the conformation of the glycoprotein codes Met193 in K and Thr193 in k.
and are sensitive to. disulfide-bond-reducing Kp"(KEL3)is found in about 2% of people of
agents, such as O.2M DTT and AET (Fig 12-4). European ancestry and is not present in people
The KELglycoprotein is linked through a sin- of African DrJapanese ancestry (Table 12-4); Kph
gle disulfide bond to. the Xk protein (Fig 12-4), (KEL4)has high prevalence in all populations.
an integral membrane protein that expresses the Kp" (KEL21), an antigen with very low preva-
Disulfide bond
FIGURE 12·4. Diagramof the KELand XK proteins, whose cysteine residuesare linked by disulfide bonds. The
remaining 14 cysteine resldues.on the KEL protein link with one another with more disulfide bonds. This
configuration makesthe KELsystem antigenssensitiveto dithiothreitol (OTT),which destroys disulfide bonds.
TABLE 12-4. Prevalence of Some KEL cells (Fig 12-5). If the second allele inherited by
Phenotypes an individual carries an SNP producing the K an-
tigen, instead of the k antigen for example, the
person will express the K antigen in addition to
the k antigen."
The remaining five antithetical antigen pairs
(KI11KI7, K14/K24, VLANIVONG, KYOI
K- k+ 91.0 KYOR,KHULIKEAL);the low-prevalence anti-
gens Ul" and K23; and the high-prevalence anti-
K+ k+ 8.8 2 gens K12, K13, K18, K19, K22, TOU, RAZ,
KALT,KTIM, KUCI, KANT, KASH,KELP,and
Kt k- 0.2 Rare
KETl all result from single amino acid substitu-
Kp(a-bt) 97.7 100 tions in the KELglycoprotein.
KEL antigens are resistant to papain, ficin,
Kp(atbt) 2.3 Rare trypsin, and a-chymotrypsin but are destroyed
by a mixture of trypsin and a-chymotrypsin.
Kp(a+b-) Rare 0 They are also destroyed by 0.2M DTT and AET
and by EDTA-glycine.
Js(a-bt) 100.0 80
Clinical Significance of Antibodies to
Jsta-b«) Rare 19
KEL Blood Group System Antigens
Js(atb-) 0 KELsystem antibodies are usually IgG, and pre-
-~~~- dominantly IgG1. They should be considered
*K, Kp', and Js' are extremely rare in populations of potentially clinically significant from the per-
Asianancestry.
spective of causing severe HDFN and HTRs. Pa-
tients with KEL system antibodies should re-
ceive antigen-negative blood whenever possible.
Anti-Kis the most common immune red cell
lence, is the product of another allele at the same
antibody outside the ABO and RH systems; one-
locus as Kp" and Kpb, and the antigen results
third of all non-Rh red cell immune antibodies
from different single nucleotide substitutions investigated are anti-K. An antiglobulin test is
within codon 281. The mutation associated usually the method of choice for detecting anti-K,
with Kp' expression reduces the quantity of KEL although occasional samples may agglutinate
glycoprotein in the red cell membranes, giving red cells directly. Most anti-K appears to be in-
rise to a slight reduction in expression of KEL duced by blood transfusion. Anti-K can cause se-
antigens in Kp' 1Kp" homozygotes but a more vere HDFN, and in some countries, it is the
obvious weakening of KELantigens in individu- practice for girls and women of childbearing po-
als who are heterozygous for Kp' and the null al- tential to receive only K- red cells. Antibodies
lele F!. to K, k, Kpa, Kpb,Isa, Isb, Ku, ur, K11, K19,
Js" (KEL6)is almost completely confined to K22, and KEALare all reported to have caused
people of African ancestry. The prevalence ofIsa severe HDFN and many have been implicated in
in African Americans is about 20% (Table 12-4). acute or delayed HTRs.
Isb (KEL7)is highly prevalent in all populations, The pathogenesis of HDFN caused by anti-K
and Is(a+b-) has not been found in persons of differs from that resulting from anti-D. Anti-K
non-African ancestry. HDFN is associated with lower concentrations
The expression of antigen is based largely on of amniotic fluid bilirubin than antl-D HDFN of
the inheritance of two alleles. The inheritance of comparable severity. Postnatal hyperbilirubin-
one normal KEL allele generally results in the emia is not prominent in infants with anemia
expression of the k, Isb, and Kpbantigens on red caused by anti-K. There is also an unexpected
FIGURE 12-5. Diagram of a KELprotein resulting from inheritance of one mutated allele K, which expressesthe
K antigen. If the second allele inherited is the normal KEL allele, the individual will also express the k, Kpb,and
JSbantigens.
low reticulocyte count in the presence of pro- not received transfusion; in another, microbial
found anemia and erythroblastosis in HDFN infection was implicated as an immunizing
caused by anti-K compared with anti-D. These agent. 26
symptoms suggest that anti-K HDFN is associat-
ed with a lower degree of hemolysis and that fe- KEL Null (Ko) and KmodPhenotypes
tal anemia in anti-K HDFN results predominant-
ly from a suppression of erythropoiesis." The Like most blood group systems, KELhas a null
KELglycoprotein appears on erythroid progeni- phenotype (Ko),in which no KEL antigens are
tors .at a much earlier stage of erythropoiesis expressed and the KEL glycoprotein cannot be
than do RH system antigens. Consequently, anti- detected in the membrane. Immunized K, indi-
K probably facilitates phagocytosis of K+ eryth- viduals may produce anti-Ku (anti-KELS),an an-
roid progenitors at an early stage of develop- tibody reactive with all cells except those of the
ment by macrophages in the fetal liver, before K, phenotype. Homozygosity for a variety of
the erythroid cells produce hemoglobin. nonsense, missense, and splice-site mutations
Antibodies mimicking KEL system specifici- has been associated with K, phenotype."
ties have been responsible for severe AIHA. Kmodred cells have only very weak expres-
Presence of the autoantibody is often associated sion of KELantigens, and individuals with this
with apparent depression of all KELantigens. Al- phenotype are homozygous (or doubly heterozy-
though most examples of anti-K are stimulated gous) for missense mutations, resulting in single-
by pregnancy or transfusion, a few cases of ap- amino-acid substitutions within the KEL glyco-
parently non-red-cell immune anti-K have been protein. Some ~Od individuals make an anti-
described. In some cases, the antibodies were body that resembles anti-Ku but differs in being
found in healthy, male blood donors who had nonreactive with Kmodred cells. Other pheno-
types in which KELantigens have substantially of the XK gene." McLeod syndrome is associat-
depressed expression result from Ktr/ Kohetero- ed with the McLeod phenotype, in which Kell
zygosity, absence of Xk protein, and absence of antigens are expressed weakly, and Km (KEL20)
the GE system antigens Ge2 and Ge3, which as well as Kx are absent. When given transfu-
are located on the glycophorins C and D (GPC, sion, people with the McLeod phenotype with-
GPD). The reason for this phenotypic associa- out chronic granulomatous disease (CGD) pro-
tion between KEL and GE is not well defined, duce anti-Km only, which is compatible with
although there is biochemical evidence to show both McLeod and Kophenotype red cells.
that KELglycoprotein, Xk, GPC, and GPD are Deletion of part of the X chromosome that in-
all located within the 4.1R membrane protein cludes XK may also include CYBB, absence of
complex (Fig12_1).28,29 which is responsible for Xlinked CGD. When
given transfusion, CGD patients with McLeod
Functional Aspects syndrome usually produce anti-Kxplus anti-Km,
making it almost impossible to find compatible
The KEL protein has structural and sequence donors, because the would-be donors are most
homology with a family of zinc-dependent endo- often patients themselves. It is recommended
peptidases that process a variety of peptide hor- that transfusion for males with CGD and Me-
mones. Although the physiologicfunction of the Leod syndrome be avoided when possible.
KELglycoprotein is not known, it is enzymati-
cally active and can cleave the biologicallyinac-
tive peptide big-endothelin-3 to create the bio- THE FY SYSTE 8}
logically active vasoconstrictor endothelin-3.
Consequently, KELmight playa role in regulat- The FY (Duffy) system includes five antigens
ing vascular tone, but there is no direct evi- that reside on the FYglycoprotein, which is also
dence for this." No obvious pathogenesis is as- known as the atypical chemokine receptor 1
sociated with the Kophenotype. (ACKR1, previously known as DARC). The
In addition to erythroid cells, KELantigens ACKRI gene consists of two exons, with exon 1
may be present on myeloid progenitor cells, and encoding only the first seven amino acids of the
KELglycoproteinhas been detected in testis and FY glycoprotein." ACKRI is on chromosome
lymphoid tissues, and with Xk protein in skele- 1q21-q22.
tal muscle.
Fya(FY1) and Fyb(FY2)
Kx Antigen (XK1), McLeod Syndrome,
The antigens Fya and Fyi'differby a single amino
and McLeod Phenotype
acid change in the Nterminus of the FYglyco-
Kx is the only antigen of the XKblood group sys- protein (Gly42 and Asp42, respectively; see Fig
tem. It is located on a polytopic protein that 12-6). They are polymorphic in people of Euro-
spans the red cell membrane 10 times and is pean ancestry, giving rise to three phenotypes:
linked to the Kell glycoprotein by a single disul- Fy(a+b-), Fy(a+b+J, and Fy(a-b+) (Table 12-5).
fide bond (Fig 12-4). Xk protein is encoded by In ASia,Fy"is a high-prevalence antigen, and the
the XK gene on chromosome Xp21.1. The func- phenotype Fy(a-b+) is rarely encountered. In in-
tion of the Xk-Kellcomplex is not known, but dividuals of African ancestry, the Fy(a-b-) phe-
Xk has structural resemblance to a family of notype is most common, caused by homozygosi-
neurotransmitter transporters. ty for a silenced FY*B allele (FY*02N.Ol). Fy"
McLeod syndrome is a very rare Xlinked and Fybare very sensitive to most proteolytic en-
condition that develops almost exclusively in zymes, including bromelin, a-chymotrypsin,
males and is associated with acanthocytosis and ficin, papain, and pronase, but are not destroyed
a variety of late-onset muscular, neurologic, and by trypsin.
psychiatric symptoms." It results from hemizy- The FY*02N.Ol allele in people of African
gosity for inactivating mutations and deletions ancestry encodes the Fybantigen but is silenced
A weak form of Fybantigen known as FyXoc-

curs rarely. The allele, FY*02W.Ol, encodes an
amino acid substitution, Arg89Cys, in the cyto-
solie domain of the glycoprotein. Fyb antigen
may be undetected by.some sources of anti-Py"
(and is usually noted in the manufacturer's in-
sert). This weak Fybantigen can be detected by
adsorption/elution or predicted by molecular
methods.
eOOH Fy3, FyS, and Fy6

Very rarely people of non-African ancestry with
FIGURE 12-6. Diagram of the FY glycoprotein Fy(a-b-) red cells are homozygous for inactivat-
(previously DARC but renamed ACKR1), with a ing mutations in ACKRI. These individuals
glycosylated external N-terminal domain, seven make no FYglycoprotein at all and were identi-
membrane-spanning domains, and a cytoplasmic fied through the presence in their sera of anti-
C-terminus.The position of the Fya/Fybpolymorphism Fy3, an antibody that is reactive with all red
is shown. cells except those of the Fy(a-b-) phenotype.
Fy6, like Fya and Fyb, is sensitive to protease
treatment, whereas Fy3 and Fy5 are resistant.
Fy5 is absent not only from cells of the Fy(a-b-)
by a mutation in the promoter region, 67T>C.34
phenotype but also from cells of the Rhnull phe-
This mutation disrupts the binding site for the
notype. The FYglycoprotein may belong to the
erythroid-specific GATA-I transcription factor
junctional membrane protein complex, which
and prevents expression of the gene in erythroid
also contains Rh proteins (Fig 12-1).35
tissue. FYglycoprotein is present on many cells
throughout the body; thus, Fy(a-b-) people of
African ancestry the FY glycoprotein on Clinical Significance of Antibodies to
their red cells only. explains why they do FYBlood Group System Antigens
not make anti-Py" rarely make anti-Fy3 Anti-Fy"is a relatively common antibody; anti-
or anti-Fy5 (see GATA-I binding- Fybis about 20 times less common. IgG subclass
site African has IgG1 usually predominates, and naturally occur-
been ring examples are very rare. Antl-Fy' and anti-
Fybmay cause acute or delayed HTRs.Although
generally mild, some have proven fatal. These
and antibodies have also been responsible for mild to
Fy(atbt) 48 3 15
Fy(a-bt} FIIFI or FIIFy 32 20 4
Fy(a-b-) FylFy FylFy o 67 o
severe HDFN. Anti-Fy3has been responsible for The FY Glycoprotein and Malaria
acute and delayed HTRs, and anti-Fy5, for de-
The FYglycoprotein is a receptor for merozoites
layed HTRs.Anti-Fy5has been found only in in-
of Plasmodium vivax, the parasite responsible
dividuals of African ancestry who have received
for a form of malaria that is widely distributed in
multiple transfusions. Anti-Fy6 is defined by a
Africaand Asiabut is less severe than malaria re-
monoclonal antibody only and reacts with all
sulting from P. fslcipstum infection. Red cells
red cells except Fy(a-b-) cells. It reacts with an
with the Fy(a-b-) phenotype are resistant to in-
epitope on the N-terminus of the FY glycopro-
vasion by P. vivex merozoites. Consequently,the
tein regardless of Fy"/Fybphenotype. Anti-Fy4is
FY*02NOI allele confers a selective advantage
obsolete. in geographic areas where P. vivax is endemic;
this advantage probably balances out any poten-
Functional Aspects of the FY
tial disadvantage resulting from the absence of
Glycoprotein the chemokine receptor on red cells.
The FY glycoprotein is a red cell receptor for a
variety of chemokines, including interleukin-8,
monocyte chemotactic protein-I, and melano- JK SVSTE (00)
ma growth stimulatory activity." It traverses the
The JK (Kidd) system consists of three antigens
membrane seven times, with a 63-arnino-acid
located on the urea transporter glycoprotein
extracellular N-terminal domain that contains
with 10 membrane-spanning domains, cytoplas-
two potential N-glycosylationsites and a cyto-
mic N- and C-termini, and one extracellular
plasmic C-terminal domain (Fig 12-6). This ar-
N-glycosylationsite (Fig 12_7).22,31 The JK gene
rangement is characteristic of the G-protein-
(SLCI4Al) is located on chromosome 18qll-
coupled superfamily of receptors that includes
chemokine receptors.
The function of ACKR1 on red cells is not
known. It has been suggested that it might act
as a clearance receptor for inflammatory media-
tors and that red cells function as a "sink," or as
scavengers, for the removal of unwanted
chemokines. If so, this function must be of limit-
ed importance because FY antigens are absent
on the red cells of most individuals of African
ancestry. It has been suggested that ACKR1 on
red cells reduces angiogenesis and, consequent-
ly, the progression of prostate cancer by clearing
angiogenic chemokines from the tumor micro-
environment. This potential effect of erythroid
ACKRI could provide an explanation for the
substantially higher levels of prostate cancer in
men of African ancestry compared with those of eOOH
European ancestry."
ACKR1is present in many organs, where it is
expressed on endothelial cells lining postcapil- FIGURE 12-7. Diagramof the JK glycoprotein, a urea
lary venules. FYglycoprotein on vascular endo- transporter, with cytoplasmiC N- and C-terminal
thelium may be involved in the inhibition of domains, 10 membrane-spanning domains, and an
cancer-cell metastasis and induction of cellular N-glycan on the third extracellular loop. The position
senescence." ACKR1 may also facilitate move- of the Jka/Jkb polymorphism is shown on the fourth
ment of chemokines across the endothelium. externalloop.
q12 and contains 11 exons, of which 4 through gene, named In(Jk) in analogy with the In(Lu)
11 encode the mature protein. dominant inhibitor of LU and other antigens.
Very weak expression of Jka and/or Jkb can be
Jk UK1) and Jkb (JK2)
B detected on In(Jk) red cells by adsorption/elu-
tion tests.
Ik"and Jkbare the products of antithetical alleles
and represent Asp280 and Asn280 in the fourth Clinical Significance of Antibodies to JK
external loop of the JK glycoprotein (Fig 12-7). Blood Group System Antigens
They have similar prevalence tn populations of
European and Asian ancestry, but Ik" is more Anti-lk' and -Ik''are usually IgG1 and IgG3, but
common than Jkb in people of African ancestry some are partly IgG2, IgG4, or IgM. Anti-Ik' and
(Table 12-6). The antigens are resistant to pro- -Jkbare often found with other antibodies and
teolytic enzymes, such as papain and ficin. about 50% bind complement. The Ik' antigen
compared to other blood group antigens is quite
Jk{a-b-) and Jk:-3 immunogenic, surpassed only by the K and D
antigens. 10
The null phenotype, Jk(a-b-) or Jk:-3, usually Although antt-Ik' is immunogenic, the titer of
results from homozygosity for a silent gene at not only anti-Ik' but also anti-Ik" often decreases
the JKlocus. Although very rare in most popula- to below the level of detection. Some aggluti-
tions, the null phenotype is relativelymore com- nate antigen-positivecells directly, but the reac-
mon in people of Polynesian ancestry, with a tions are usually weak. Generally, an antiglobu-
prevalence of around 1 in 400 but as high as lin test is required, and use of enzyme-treated
1.4% in those of Niuean ancestry. The Polyne- cells may be necessary to detect weaker anti-
sian null allele (lK*02N.Ol) contains a splice- bodies. These characteristics make JKantibodies
site mutation in intron 5 that results in the ab- dangerous. For example, if a patient has antt-Ik"
sence of the protein from the membrane. In peo- and later moves to .a different hospital without
ple of Finnish ancestry, where Jk(a-b-) is less the patient antibody history, the antibody may
rare than in other populations of European an- not show in testing. If anti-Ik' is not identified,
cestry, the allele responsible (lK*02N.06) en- transfusion of Jk(a+) red cells may occur, caus-
codes a Ser291Pro substitution. Immunized in- ing a potential HTR.
dividuals with no JK glycoprotein may produce JK antibodiesmay cause severe acute or de-
anti-Jk3, an antibody that is reactive with all red layed HTRs. They are a very common cause of
cells except those of the Ik]a-b-) phenotype. An delayed HTRs, probably because they may not
extremely rare form of Jk(a-b-) phenotype be detected in pretransfusion testing due to
found in .p~ople..of Japanese ancestry results their tendency to decrease to low or undetect-
from heterozygosity>for a d()minant inhibitor able levels in plasma. Despite their hemolytic
potential, JK antibodies only very rarely cause
severe HDFN.39 JK antibodies have been impli-
TABLE 12-6. JK Phenotypes in Three cated in acute renal transplant rejection, sug-
Populations gesting that JK antigens can behave as histocom-
patibility antigens."
The JK Glycoprotein in Urea

Transportation
Jk(a+b __) 26 The JK antigens are located on the red cell urea
transporter, SLC14Al (alsoknown as HUT11 or
Jk(a+b+) 50 40 50
UT-Bl). When red cells approach the renal me-
Jk(a-b+) 24 8 27 dulla, which contains a high concentration of
urea, the urea transporter permits rapid uptake
372 AA B B TEC H N I CAL MAN UA L
of urea and prevents the cells from shrinking in red cell membrane proteins, which could func-
the hypertonic environment. As the red cells tion as a gas channel for 02 and CO2, Band 3 is
leave the renal medulla, urea is transported rap- also a component of the junctional complex that
idly out of the cells, preventing the cells from links the red cell membrane to the membrane
swelling and carrying urea away from the kid- skeleton via GPC and protein 4.1 (Fig 12·1).
ney. SLC14A1has been detected on endothelial SLC4AI encodes band 3. It is located on chro-
cells of the vasa recta, the vascular supply of the mosome 17q21.31 and consists of 20 exons.
renal medulla, but it is not present in renal tu-
bules. Oi8 (011) and Oib (012); Anti-Oi8 and
Normal red cells are rapidly lysed by 2M ob
-O I
urea because urea transported into the cells
makes them hypertonic and they burst as a re- Di", the original DI antigen, is very rare in
sult of the osmotic influx of water. Because of people of European or Africanancestry but has a
the absence of the urea transporter, Jk(a-b-) prevalence of 5% in people of Chinese or
cells are not hemolyzed by 2M urea, and this Japanese ancestry and an even higher preva-
can be used as a method for screening for lence in the indigenous peoples of North and
Jk(a-b-) donors." South America, reaching 54% in the Kainganges
The Jk(a-b-) phenotype is not associated Indians of Brazil. Dib is a high-prevalence anti-
with any clinical defect, although two unrelat- gen in almost all populations. Dia and Dib repre-
ed Jk(a-b-) individuals had a mild urine- sent an amino acid substitution in the seventh
concentrating defect."
THE (010) Rba

Trs
Band 3, the Red Cell Anion Exchanger WARR
The 22 antigens of the DI (Diego)system are lo- Vga
Wda
cated on band 3, the common name for the red swa/SW1
BOW/NFLD
cell anion exchanger or solute carrier family
Wu/DISK
4Al (SLC4A1). Band 3 is a major red cell Jna/KREP
membrane glycoprotein with -106 copies per Bpa
red cell and contributes to red cell structural in-
tegrity. Band 3 has a transmembrane domain
that traverses the membrane 14 times, with an
N-glycanon the fourth extracellular loop. Band
3 also has a long, cytoplasmic N-terminal do-
main that interacts with the membrane skeleton
proteins ankyrin, 4.1R, and protein 4.2 and
functions as a binding site for hemoglobin (Figs
12-1 and 12-8). The short cytoplasmic C-termi-
COOH
nal domain binds carbonic anhydrase II. Car-
bonic anhydrase II is an enzyme involved in
CO2exchange on the red cell membrane." FIGURE 12·8. Diagram of band 3, the 01
Band 3 in red cells has at least two major glycoprotein and anion exchanger,with cytoplasmic
functions: the rapid exchange of HC03- and Nand C-terminal domains, 14 membrane-spanning
Cl ions, which are important in CO2 transport, domains,and an N-glycanon the fourth extracellular
and attachment of the red cell membrane to the loop (although the precise conformation is still
cytoskeleton." Tetramers of band 3 form the controversial). The locations of the 22 antigens of
core of the band 3/Rh ankyrin macrocomplex of the 01system on the extracellularloops are shown.
extracellular loop of band 3: Leu854 and Fr', and SWl. Anti-DISK detects a high-
Pr0854, respectively. prevalence antigen that is antithetical to Wu and
Anti-Diaand _Dibare usually IgGl plus IgG3, has caused severe HDFN. Anti-ELO and anti-
which can induce hemolysis. The antibodies BOWhave also caused severe HDFN.
typically require an antiglobulin test for detec- Antigens of the DI system are not destroyed
tion. Anti-Dl", which is present in ~3.6% of by proteolytic enzymes, such as papain, flcin, or
multiple-transfusion recipients in Brazil, can trypsin; however, the antigens carried on the
cause severe HDFN. Anti-Di" has rarely been re- third extracellular loop (Rba, rr, WARR, Vg",
sponsible for serious HDFN. Generally, neither Wda, BOW, NFLD,Wu, DISK, Ina, KREP, and
antl-Di' nor anti-Df cause HTRs; one example Bp"] are sensitive to a-chymotrypsin.
of anti-Dr causing a delayed reaction has been
reported, and anti-Di"rarely causes mild delayed
reacttons." T (0
The YT system consists of five antigens encoded
Wra (DI3) ilInd Wrb (DI4); Anti-Wr and8
by the ACHE gene on the long arm of chromo-
-Wrb some 7. Yta (YT1; His353) and Ytb (YT2;
The low-prevalence antigen Wra and its antithet- Asn353) are antithetical antigens on acetylcho-
ical antigen of extremely high prevalence, Wrb, linesterase. High-prevalence antigen YT3, or
represent an amino acid substitution in the YTEG, is the result of the nucleotide change
fourth loop of band 3: Lys658 and Glu658, re- 266G>A in exon 2 with the predicted amino
spectively. Wrb expression depends on the pres- acid change Gly89Glu. Acetylcholine-esterase
ence of GPA.Despite the presence of Glu658 on (AChE), an enzyme that is important in neuro-
band 3, Wrb is not expressed in the rare pheno- transmission, has an unknown function in red
types associated with a complete absence of the cells. Ytb has a prevalence of about 8% in people
MN glycoprotein GPAor of the part of GPAthat of European ancestry but is more common in
is close to insertion into the red cell membrane. people from the eastern Mediterranean; Yt' has
This provides strong evidence for an interaction relatively high prevalence in all populations. Yt'
between band 3 and GPA within the red cell is not affected by trypsin but is destroyed by
membrane. Band 3 has been associated with a-chymotrypsin treatment of the red cells. The
clearance of senescent and oxidatively stressed Yt' antigen is variably sensitive to papain and fi-
red cells." cin. Yt' and Ytb are sensitive to the disulfide-
Anti-Wr' is a relatively common antibody and bond-reducing agents AET and DTT.
can be naturally occurring. It is usually detected Three high-prevalence antigens~YTEG
by an antiglobulin test but sometimes by direct (YT3),YTU (YT4),and YTOT(YT5)-were add-
agglutination of red cells. Wra antibodies are ed to this system in 2018 with the help of solu-
mostly IgG1 but sometimes IgM or IgM plus ble recombinant proteins. All three origlnated
IgG. Anti-Wr' has been responsible for severe from homozygous mutations in exon 2 resulting
HDFN and HTRs. A1loanti-Wrbis rare and little in amino acid changes in AChE.45
is known about its clinicalsignificance,but auto- YT antibodies are usually IgG and require an
anti-Wr'' is a relatively common autoantibody IATfor detection. They are not generally consid-
and may be implicated in AIHA. ered to be clinicallySignificant,although anti-Yt'
may cause accelerated .destruction of Yt(a+)
transfused red cells and has been implicated in
Other DI Antigens
acute and delayed HTRs.46Rare blood may be
Over the years, many antigens of very low prev- required for a patient with anti-Yr. Often a
alence have been shown to represent amino monocyte monolayer assay is used to determine
acid substitutions in band 3 and have joined the whether the antl-Yt' is predicted to cause overt
DI system: Wda,. Rba, WARR, ELO, Wu, Bpa, destruction of transfused antigen-positive red
Moa,Hg",Vg",Sw",BOW, NFLD,Ina, KREP,Tra, cells.
HE (01 ) allyreactive by an lAT.Treatment of red cellsby

proteolytic enzymes has little effect, but
The two antigens of the XGblood group system, disulfide-bond-reducingagents (AETand DTT)
Xg' (XG1) and CD99 (XG2),are encoded by ho- substantiallyreduce reactivity.
mologous genes. The XC gene lies partly within Emerging progress in the use of monoclonal
the X chromosome pseudoautosomal region antibodies to treat disease is increasing interest
(Xp22.32), a section at the tip of the short arm in the function of various antigens. ERMAPis
that pairs with the Y chromosome. XC is one of known to inhibit T-cell function by decreasing
few genes not inactivated by lyonization, which cell proliferation and cytokine secretion. This
is the random inactivation of one of two X al- understanding has led to proposals to employ
leles in females." The CD99 gene is homolo- soluble ERMAP to help regulate the immune
gous to XC and is located on both X and Y chro- system for individuals with autoimmune dis-
mosomes (Yp11), with pairing occurring at ease."
meiosis. Xg' is polymorphic and has a preva-
lence of about 66% in males and 89% in fe-
males. Both CD99 and Xg"expression are con- TH Y T (Ol )
trolled by a common regulator gene, XC. XC
transcription is impacted by allele rs311103C, The DO (Dombrock)system consists of 10 anti-
which disrupts a GATAbinding site on the XC gens: the polymorphic antithetical antigens Do'
gene, ultimately resulting in an Xg(a-) pheno- (DOl; Asn265) and DOb(D02; Asp265) and the
type." Although Xg' antibodies occasionallyag- high-prevalence antigens Gy", Hy, Io", DOYA,
glutinate red cells directly, they are generally DOMR, DOLG, DOLC, and DODE.51 Doa and
19Gand are reactive by an lAT. They are not re- DObhave a prevalence of 66% and 82%, respec-
active with red cells treated with proteolytic en- tively, in populations of European ancestry (Ta-
zymes. Anti-Xg' is not clinically significant. ble 12-7). The prevalence of Do" is somewhat
CD99 antibodies in common use are mostly lower in populations of African ancestry and
monoclonal and of mouse origin; a few human substantially lower in people of East Asian an-
alloanti-CD99 occur, although little is known cestry. Anti-Gy"is the antibody that is character-
about their characteristics. istically produced by immunized individuals
with the DOnull [Gy(a-)]phenotype that results
from various inactivating mutations. Two un-
H common phenotypes are present in individuals
of African ancestry: Hy-, Jo(a-) (Gly108Val);
The SC (Scianna)system consists of seven anti- and Hy--", Jo(a-) (Thrl17I1e). These are usually
gens on erythrocyte membrane-associated pro- associated with weak expression of DOband
tein (ERMAP), a member of the 19SFthat has Doa, respectively (Table 12-7). The DO glyco-
one 19SFdomain." Sc1 (Gly57)and Sc2 (Arg57) protein (ART4;CD297) is an adenosine diphos-
are antithetical antigens of high and low preva- phate ribosyltransferase encoded by ART4, lo-
lence, respectively. Rd (SC4) is of low preva- cated on chromosome 12p13-p12, although its
lence; Sc3, STAR,SCER,and SCANare of high function on red cellsis not known.
prevalence. Anti-Sc3is produced by individuals DO antigens are resistant to papain and ficin
with the very rare SCnull phenotype and are pretreatment of red cells but are sensitive to tryp-
dicted to have red cell membrane integrity is- sin, a-chymotrypsin, and pronase. They are also
sues. sensitive to the disulfide-bond-reducingagents
Antibodies to SC antigens are rare and few AETand DTT.
have been implicated in an HTR, but several The Do" and DObantigens are poor immuno-
have been implicated in mild to severe HDFN. gens and therefore are uncommon antibodies.
Evidence is limited due to the scarcity of the an- When formed, the antibodies are weak or vari-
tibodies. Although directly agglutinating SC1 ably reactive, usually 19G, and reactive by an
antibodies are known, SC antibodies are gener- lAT.The antibodies are usually found in the sera
C HAP T E R 1 2 OtherBloodGroupSystemsand Antigens 375
TABLE12-7. Phenotypes of the DO System and Their Approximate Prevalence
Do(a+b-) + + + + 18 11
Do(a+b+) + 44
Do(a-b+) + 45
Gy(a-) 0
Hy- Rare
Jo(a-) +w Rare
DOYA- +w +w Rare
DOMR- +w Rare
DOLG- + +w + + Rare Rare

+W= weakened expression of antigen.
of individuals with other alloantibodies." all red cells except those of the extremely rare
Screening for DO-compatible donors, therefore, Co(a-b-) phenotype that results from .various
is best performed by molecular genotyping. inactivating mutations. Co4 (Gln47) is.a high-
Anti-Do' and -Dobhave been responsible for prevalence antigen whose presence is requlred
acute and delayed HTRs. Anti-Hy, anti-Gy', and for the expression of Co" because of the proximi-
anti-Io' have been reported to cause moderate ty of the polymorphism. 54 The CO antigens are
transfusion reactions but are not always clinical- located on the red cell's water transporter,
ly significant. DO antibodies have not been re- aquaporin-l , encoded by AOPI on chromosome
ported to cause HDFN, although some neo- 7p14 (Fig 12-9). CO antibodies are usually 19G
nates have been born with a positive direct and reactive by an lAT, although agglutinating
antiglobulin test (DAT)result. 19M antl-Co" has been found. CO antibodies
have been implicated in severe HDFN and
HTRs. CO antigens are resistant to proteolytic
H E enzymes.
The CO (Colton) system consists of 4 antigens.
Co' (CO1; Ala4S) is a hlgh-prevalence antigen; H ( 16)
COb(C02; Val4S), its antithetical antigen, has a
prevalence of about 8% in people of European The LW (Landsteiner-Wiener] system consists of
ancestry but is less common in other ethnic three antigens. LWa (LWS) and LWb (LW7;
groups; Co:-3 is the rare COnull phenotype; and GIn1OOArg)are antithetical antigens of high and
Co4 is a high-prevalence antigen originally low prevalence, respectively= Anti_Lwabis reac-
thought to be C03.53 Anti-Co3 is reactive with tive with all red cells except those of the ex-
FIGURE 12-9. A model of aquaporln-t, showing the six membrane-spanning domains as cylinders. The first
extracellular loop is glycosylated and contains the Coa/Cobpolymorphism. The third extracellular loop and first
intracellular loop contain alanine (A), proline (P), asparagine (N) motifs and form a channel in the membrane
through which water molecules pass.
tremely rare LWnull phenotype and Rhnull cells, involved in the stabilization of erythroblastic is-
which are also LW(a-b-). LW antigens are ex- lands in the marrow during the later stages of
pressed more strongly on D+ than D- red cells erythropoiesis.54 ICAM-4is also part of the band
and more strongly on umbilical cord red cells, 3/Rh ankyrin macrocomplex (Fig 12-1) of red
even D- red cells, than on those of adults, mak- cell surface antigens and might maintain close
ing D- cord red cells valuable in identification contact between the red cell surface and the
studies for anti-LW. LW antigens are unaffected vascular endothelium. Upregulation of ICAM-4
by treatment of the red cells with papain, flcin, on red cells of patients with sickle cell disease
trypsin, or a-chymotrypsin but are destroyed by has been implicated in the adhesion of these
pronase. Disulfide-bond-reducing agents (AET cells to the vascular endothelium and the resul-
and DTT) either destroy or greatly reduce LWa tant crises of vascular occlusion.56
or iw- (LW6)on red cells. Most LW antibodies are reactive by an lAT.
The LW glycoproteinis intercellular adhesion They do not require exposure to the LW anti·
molecule-4 (ICAM-4), an IgSF adhesion mole- gen, are not generally considered to be clinically
cule encoded by ICAM4 on chromosome significant, and have not been implicated in
19p13.2. ICAM-4 binds integrins on macro- HTRs or HDFN. Acquired and often temporary
phages and erythroblasts, and it is probably LW-negativephenotypes sometimes occur with
production of anti-Lw" or anti-LWab.This phe- 2q 14-q21, by initiation of translation at two

nomenon is usually associated with pregnancy different sites on the mRNA. GPD lacks the
or hematologic malignancy. The transient anti- N-terminal 21 amino acids of GPC. GPC and
bodies seem like alloantibodies because they GPD are part of the junctional complex of mem-
present when the patient's antigen is undetect- brane proteins." Their C-terminal cytoplasmic
able. If the patient recovers and the red cells are domains interact with the membrane skeleton
tested later, the LWantigensare againexpressed. through 4.lR, p55, and adducin and serve as an
important link between the membrane and its
skeleton.
THE CH G SYSTEM «(11) GPC is exploited as a receptor by some
strains .of the malaria parasite P. ja!ciparum.
The nine antigens of the CH/RG (Chido/Rodg- There are three types of "GE-negative" pheno-
ers) system, although considered blood group types (Table 12-8). Ge:-2,-3,-4 is the true null,
antigens, are not produced by erythroid cells. in which both GPC and GPD are absent from
They are located on a fragment of the fourth the red cells, and the cells are elliptocytic. In
component of complement (C4d) that attaches the other phenotypes, Ge:-2,3,4 and Ge:-2,
to the red cells from the plasma. Ch 1 to Ch6, -3,4, GPD is absent and an abnormal form of
Rgl, and Rg2 each have a prevalence >90%; GPC is present. Ge2, Ge3, and Ge4 are de-
WH has a prevalence of about 15%.A complex stroyed by trypsin treatment of red cells.
relationship exists -b~tween the. nine determi- Although Ge2 and Ge4 are also sensitive to pa-
nants and SNPs in C4A and C4B, the genes paln treatment, Ge3 is resistant. Consequently,
encoding the C4a chains. The expression of papain-treated red cells can be used for distin-
CH/RG on red cellsis destroyed by treatment of guishing anti-Ge2 from anti-Ge3 in the absence
the cells with proteolytic enzymes, such as pa- of red cells of the very rare Ge:-2,3,4 pheno-
pain or ficin. type.
No CH/RG antibodies are known to have GE antibodies may be IgM and directly ag-
caused HTRs or HDFN, and antigen-negative glutinating, but most are IgG and require an IAT
blood is not required for transfusion. However, for detection. Anti-Ge2 is not generally consid-
rare cases of anaphylactic reactions and de- ered to be clinicallysignificant,but anti-Ge3 has
creased red cell survival caused by these anti- caused mild to moderate HTRs. Anti-Ge3 has
bodies have been described." CH/RG antibod- caused HDFN that tends to manifest 2 to 4
ies are generally IgG. Detection of these weeks after birth, and with severe neonatal ane-
antibodies with native red cells usually reqllires mia associated with suppression of erythropoie-
an IAT, but they directly agglutinate red cells sis. Ge:-2,-3 blood is rare and difficult to ob-
coated artificiallywith C4d. Binding of CH/RG tain. Monocyte monolayer assays may be of
antibodies to red cells is readily inhibited by value in determining potential clinical signifi-
plasma from CH/RG-positiveindividuals; this is cance. Some autoantibodies with specificities
a useful aid to identification of these antibod-
ies (Method 3-17). TABLE 12-8. Phenotypes Lacking High-
Prevalence GE Antigens and the Antibodies That
May Be Produced
THE Ci (020)
The GE (Gerbich) system consists of six high-

prevalence antigens-Ge2, Ge3, Ge4, GEPL, Ge:-2,3,4 (Yus type) Anti-Ge2
GEAT,and GETI-and five antigens with very Ge:-2,-3,4 (Gerbich Anti-Ge2 or -Ge3
low prevalence: Wb, Is", Ana, Dh', and GElS. type)
These antigens are located on GPC, GPD, or
Ge:-:-2,-3,-:A(Lea,~h Anti-Ge2, -Ge3, or -Ge4
both. These two glycoproteinsare produced by
type)
the same gene, GYPC, located on chromosome
resembling anti-Ge2 or -Ge3 have been respon- antibodies are usually IgG and require an IATfor
siblefor AIHA detection. They may be inhibited by serum
or concentrated urine from antigen-positive in-
dividuals.
R E (0 1)
The 20 CROM(Cromer) antigens are located on KN (0

the complement-regulatory glycoprotein called
decay-accelerating factor (DAF or CD55).59 The 10 antigens of the KN (Knops)system are
They include the antithetical antigens Tca/Tcb/ located on the complement-regulatoryglycopro-
Tc' and WESa/WESb• Tc" and WESb have high tein called complement receptor 1 (CR1 or
prevalence, and Tc", Tc', and WEsa have low CD35).61All are polymorphic, although Kn',
prevalence, although both Tcb and WESa are McC",S11(Sl''),S13,and Yk' have relativelyhigh
present in approximately 0.5% of people of Afri- prevalence (Table 12-9).
can ancestry, and WESais present in 0.6% of The Helgeson phenotype, an apparent KNnu11
people of Finnish ancestry. The other antigens phenotype, has very low levels of red cell CR1
have high prevalence: o-, Dra, Es, IFC, UMC, and very weak expression of KN system anti-
GUTI, SERF, ZENA, CROV, CRAM, CROZ, gens. KN system antigens are generally resistant
CRUE,CRAG,and CROK. to papaln and ficin, although this may depend
Anti-IFCis the antibody made by individuals on the antibodies used, and are destroyed by
with the very rare CROMnullphenotype (Inab trypsin or a-chymotrypsin treatment. They are
phenotype), and it is reactive with all red cells alsoweakened or destroyed by AETand DTT.
other than those of the Inab phenotype. CROM CR1 is a receptor for P. fslciperum and ap-
antigens are readily destroyed by a-chymo- pears to be involved in the rosetting of red cells
trypsin treatment of red cells but not by papaln, that is associatedwith severe malaria. The McCb
ficin, or trypsin treatment. The disulfide-bond-
reducing agents AET and DTT reduce antigen
expression only slightly. TABLE 12-9. Approximate Prevalence of KN
CD55 helps protect the red cells from lysis Antigens in Two Populations
resulting from autologous complement by
inhibiting the action of C3-convertases. Inab-
phenotype red cells do not undergo undue
hemolysis, however, because of the activity of
another complement-regulatory glycoprotein,
CD59. Because CD55 and CD59 are both Kna KN1 99 100
linked to the red cell membrane by a glycosyl-
phosphatidylinosito1 (GPI) anchor, pathologic Knb KN2 6 0
levels of hemolysis occur in paroxysmal noctur-
nal hemoglobinuria, which is associated with a MeCa KN3 98 94
clonal defect in GPI biosynthesis and the ab- SI1(Sla) KN4 98 60
sence of both CD55 and CD59 in affected red
cells. CD55 has recently been identified as a re- Yka KN5 92 98
ceptor for P. fslciperum: 60 MeCb KN6 0 45
CROM antibodies are not usually considered
to be clinicallysignificant because there is little SI2 KN7 0 80
evidence that any have caused an HTR,and the
SI3 KN8 100 100
evidence from functional cellular assaysis equiv-
ocal. CROM antibodies have not been implicat- KCAM KN9 98 20
ed in HDFN, and they are probably sequestered
by high levels of CD55 in the placenta. CROM KDAS KN10 23 87
and SI2 alleles, present almost exclusively in in-

dividuals of African ancestry, may confer a de-
gree of protection from the parasite. This might The OK system consists of three antigens: Ok',
explain the very strong difference in the preva- OKGV,and OKVM.All three antigens have very
lence of some antigens, especially SI1, McCb, high prevalence and are located on the IgSF
S12,and KCAM,among populations of Europe- molecule CD147, or basigin, which has two
an and African ancestry (Table 12-9). IgSF domains. Ok=is resistant to proteolytic en-
KN system antibodies are not clinicallysignif- zymes and disulfide-bond-reducingagents. Very
icant and can be ignored when selecting blood few alloanti-Ok' antibodies and a single example
for transfusion. They are usually challenging to each of anti-OKGVand DKVM are known; all
work with, often making it difficult to distin- are reactive by an IAT.63In-vivo survival tests
guish antigen-negative cells from those with and cellular functional assays with one anti-Ok'
weak expression. Recombinant CR1 reagents or have suggested that it could be clinically Signifi-
recombinant blood group antigens may be use- cant, but no clinical information exists. Basigin
ful as inhibiting reagents to help in identification is another important receptor for P. fslciperum
of these antibodies. They are generally IgG and invaslon."
reactive only by an lAT.
TH H
TH IN (0
The RAPH system consists of one antigen.
The low-prevalence antigen Inaand its antitheti- MER2 (RAPH1), which is located on the tetra-
cal antigen Inb plus four other high-prevalence spanin CD151, was initially defined by mouse
antigens (INFI,INJA,INRA,and INSL)are locat- monoclonal antibodies that recognized a quanti-
ed on CD44, the predominant cell surface re- tative polymorphism, and about 8% of the
ceptor for the glycosaminoglycanhyaluronan, a population has undetectable levels of MER2 on
component of the extracellular matrix.6z AnWj their mature red cells. Alloanti-MER2was found
(901009), an antigen with very high preva- in three Israeli Jews originating from India who
lence, may also be located on or associated with had a RAPH-nullphenotype resulting from a sin-
CD44, but the evidence is incomplete. IN anti- gle nucleotide deletion that led to a premature
gens have reduced expression on red cells with stop codon. These three individuals were
the In(Lu) phenotype, and AnWj is virtually CDl St-deficient and had end-stage renal fail-
undetectable on In(Lu) cells. Inaand Inbare sen- ure, sensorineural deafness, and pretibial epider-
sitive to treatment of red cells with proteolytic molysis bullosa, suggesting that CD151 is essen-
enzymes-papain, ficin, trypsin, a-chymotryp- tial for the proper assembly of basement
sin-and are also destroyed by the disulfide- membranes in kidney, inner ear, and skin.65,66
bond-reducing agents AET and DTT. AnWj, MER2- individuals with anti-MER2 but only
however, is resistant to all these enzymes but single amino acid substitutions in CD151 do not
shows variable outcomes with reducing agents. have these symptoms.
Anti-Inaand -Inboften agglutinate red cells di- MER2 antigen is resistant to treatment of
rectiy, but the reaction is usually enhanced by red cells with papain but is destroyed by trypsin,
an lAT. IN antibodies are not generally consid- a-chymotrypsin, and pronase and by AET and
ered to be clinicallySignificant,although there is DTT.MER2antibodiesreact in an lAT.There is no
one report of anti-In" causing an HTR. Anti- evidence that anti-MER2is clinicallysignificant.
AnWj, however, has caused severe HTRs, and
In(Lu) red cells should be selected for transfu-
sion. In(b-) and In(Lu) are rare blood types, and J
use of a monocyte monolayer assay may be of
value in determining the potential clinical signif- The JMH (John Milton Hagen) system consists
icance of the corresponding antibodies. of six antigens with very high prevalence-
JMH,JMHK,JMHL,JMHG,JMHN, and JMHQ- prevalence, and absence of these antigens is as-

on the semaphoringlycoproteinCDI08 (Sema7A). sociated with an aberrant U (MNSS)antigen.
Anti-JMH is typically produced by individuals
with an acquired loss of CD108. This most often
occurs in elderly patients and is associated with T JR SY T (O
a weakly positive DAT result. The absence of
the other JMH antigens results from different The high-prevalence antigen Ir" is currently the
missense mutations in SEMA7A.67 JMH anti- only antigen in the JR blood group system, fol-
gens are destroyed by proteolytic enzymes and lowing the independent findings of two groups
disulfide-bond-reducingagents. They are not de- demonstrating that the Jr(a-) phenotype was the
tected on cord red cells. Sema7A has also been result of inactivating nucleotide changes in
shown to be a receptor for P.jalciparum. II ABCG2_70,71The gene encodes ABCG2, a multi-
JMH antibodies are usually reactive in an lAT pass membrane-protein family member of the
but they are often not detected unless the labo- adenosine triphosphate (ATP)-bindingcassette
ratory uses an anti-IgG reagent that contains transporters that is broadly distributed through-
anti-IgG4. This lack of detection is usually not out the body. Ir" has long been associated with
problematic, as they are not generally consid- drug resistance in cancer and resistance to xeno-
ered to be clinicallysignificant, although one ex- biotics, and it might be important for porphyrin
ample was implicated in an acute HTR. homeostasis."
The Jr(a-) phenotype is present predomi-
nantly in people of Japanese ancestry. Jra anti-
THE GIL Y (029) gen is resistant to proteolytic enzymes and di-
sulfide-bond-reducing agents. Anti-Ir' is
The GIL (Gill)system consists of one very high- reactive by an IAT and has caused HTRs. It is
prevalence antigen, GIL,located on aquaporin 3 not usually implicated in HDFN, although two
(AQP3). GIL is a member of the aquaporin su- fatal cases have been reported. ABCG2 trans-
perfamily of water and glycerol channels (like porter expression is stronger on cord cells than
the CO blood group systeml." AQP3 enhances adult cells, which may lead to early anti-It'
the permeability of the red cell membrane by binding to fetal cells.'?
glyceroland water.
GIL antigen is resistant to proteolytic en-
zymes and disulfide-bond-reducing agents. GIL T E NS T 3)
antibodies are reactive by an lAT. Anti-GILhas
not been implicated in HTRsor HDFN, although The LAN system consists of one antigen, Lan.
monocyte monolayer assays have suggested a Lan is a high-prevalence antigen that became a
potential to cause accelerated destruction of new blood group system followingthe discovery
GIL+red cells. that it was carried on ABCB6. The antigen is an
ATP-binding cassette transporter molecule on
the erythrocyte membrane." Unlike Ir", Lan is
THE RH s TEM «30) not associated with any single geographicor eth-
nic group, and this is mirrored by the diversity
The three antigens of the RHAG system (Ol", of mutant alleles in the Lan- individuals stud-
Duclos, DSLK)are located on the Rh-associated ied. ABCB6 is associated with porphyrin trans-
glycoprotein (RhAG), which is described in port and was thought to have an important role
more detall in Chapter 11.69RhAGis closely as- in heme synthesis; however, the existence of
sociated with the Rh protein in the membrane ABCB6-deleted individuals indicates that there
as part of the band 3/Rh ankyrin macrocomplex may be compensation by other transporters in
(Fig12-1).Ol"is very rare, and homozygosityfor the absence of ABCB6.
the allele encoding Ol' is associated with an The Lan antigen is expressed variably on red
Rhmod phenotype. Duclos and DSLKhave high cells in different individuals but is resistant
to proteolytic enzymes and disulfide-bond- a blood group system, and the antigen to which
reducing agents. Anti-Ian is reactive by an IAT the antibodywas directed was named CD59.1.2
and has been implicated in HTRsbut not gener- CD59 deficiency causes chronic inflammato-
ally in HDFN. Lan- blood is rare and use of a ry demyelinating neuropathy resulting in mus-
monocyte monolayer assay may be of value in cular weakness, lesions in the central nervous
determining potential clinicalsignificance. system, and hemolytic episodes. This deficiency
has been reported in over 10 children suffering
from severe illness. Several of these patients had
received transfusion, but only one made anti-
CD59.78
Vel is a high-prevalence blood group antigen
that has been assigned its own system. It has
been shown to depend on the presence of small H U
integral protein I (SMIMI), a protein of un-
known function newly discovered on the eryth- The AUG (Augustine)system consists of four an-
rocyte surface.":" Absence of Vel antigen in the tigens on the erythrocyte protein called equili-
majority of Vel- individuals, regardless of ethnic brative nucleoside transporter 1 (ENTl). The
background, is caused by a 17-base-pairdeletion At(a-) phenotype in individuals of African an-
in SMIMI, which results in the absence of the cestry is defined by an amino acid polymor-
protein at the cell membrane. phism on the ENTl protein, and At(a-) mem-
Vel antigen expression is generally weak on bers of a rare family affected by bone
cord red cells and differs substantially from one malformation lacked the protein because of an
individual to another. Patterns of expression are inactivating mutation in the ENTl gene
a consequence both of zygosityfor the 17-bp de- (SLC29Al).79 Based on the evidence, the blood
letion and of polymorphism in the transcription group system AUGwas created. The antigen de-
regulatory region in intron 2. Serologic expres- fined by the antibody produced by the null phe-
sion is not affected by protease treatment, al- notype was named AUG1, and the antigen de-
though sensitivity to reducing agents such as fined by the amino acid Glu391 (At") was
0.2M DTT is variable. Anti-Velis often a mix- named AUG2.
ture of IgG and IgM, readily activates comple- Anti-At"is an antibody to a high-prevalence
ment, and has been implicated in mild to severe antigen, reactive by lAT. It has been implicated
HTRs,although HDFNis rare. in an acute HTR and severe HDFN. At(a-) do-
nor units are extremely rare in the United
States, and use of a monocyte monolayer assay
HE 5 (0 may be of value in determining potential clinical
Significance.
An antibody to a high-prevalence antigen detect- The low-prevalence antigen AUG3 (named
ed in the plasma of a CD59-deficientchild recip- ATML) and the high-prevalence antigen AUG4
ient of transfusion was shown to be specific for (named ATAM)were added to the AUG blood
CD59.77 The antibody was readily inhibited group at the 2018 lSBTworking party meeting.
with soluble protein. Sequence analysis of sam- ATML is caused by a variant in SLC29AI, en-
ples from the family revealed that the parents coding p.Thr387Pro in the fifth extracellular
(first-degreecousins) were heterozygous and the loop of the ENTl protein. Anti-ATML has
child homozygous for a silencing mutation in caused severe HDFN. Anti-ATAMwas found in
CD59. The antibody was 19G,and although the 1995 in a pregnant woman of European ances-
child's red cells had been weakly DAT-positive try who had received a transfusion. ATAM is
following transfusion, incompatible blood was caused by a homozygous missense mutation in
well tolerated. Thus, CD59 has been ratified as SLC29AI encoding p.Asn81Ser.45,8o
THE KANN STEM (037) IE crt.a S T 9)
Anti-KANNO,described as a broadly reactive al- The CTL2 system was named as a system in
loantibody, was reported in some Japanese preg- 2019 and is made up of two antigens, CTL2.1
nant women." A genome-wide association and Rif. There have been no reports of transfu-
study was performed on four KANNO-negative sion reaction or HDFN.86[As the manuai went
individuais and normai KANNOindividuais. By to press, the ISBTWorking Party meeting to ap-
prove CTLas blood group 039 was delayed. See
using whole-exome sequencing, the genome
ISBTwebsite for updates.]
variation was found and confirmed. The KAN-
NO antigen was located on red-cell-specific
membrane protein by monoclonai antibody- ANTIGENS TH
specific immobilization of erythrocyte antigens NG
assay." The KANNOpolymorphismwas located
on the prion protein gene on chromosome
20p13 locus. Blood Group Collections
Although many antigens are categorized to one
of the 39 known blood group systems, five
THE SED S i'E (038)
blood groups remain uncharacterized: 205 [Cost
Sd" was described in 1967 as an antigen of
(symbol: COST)I, 207 [li (symbol: I)], 208 IEr
(symbol: ER)], 210, and 213 (symbol: MN
somewhat high prevaience with a positivityrate
CHO). Within these collections are 14 antigens,
of about 90%.83,84Some of the antigen charac-
mostlywith high or low prevaience. Someblood
teristics noted through the years include the loss group collections contain two or more antigens
of Sd" reactivity in pregnancy, possible interfer- that are related serologicaily,biochemically, or
ence with binding of Escherichia coli in the in- genetically to a blood group system but do not
testines, inhibition of the entrance of maiaria fit the criteria for system status. !
parasites into red cells, and exhibition by the an- The COST collection contains Cs' and Cs",
tibody of unique refractile agglutinates in tube antitheticai antigens with relatively high and
testing. Sd" is a carbohydrate antigen on red low prevalence, respectively. These antigens are
serologicailyrelated to those of the KNsystem
cells synthesized by enzyme ~(1,4)N-acetylga-
but do not appear to be located on CRl. COST
lactosaminyltransferase. The strength of Sd" on antibodies are not clinicallysignificant.The 207
red cellsis highly variable from one individuai to collection is named Ii and has one antigen, i.
another, and Sd' is not detected on cord red This appears with variable prevalence in serolog-
cells. Anti-Sd" agglutination of red cells has a ic testing, depending on the method of testing.
characteristic mixed-field appearance with free Era and Erbare antitheticai antigens in the ER
red cells when viewed microscopicaily.Anti-Sd" collection with very high and low prevaience,
is inhibited by urine from Sd(a+) individuais respectively.Anti-Er3 is produced by individ-
(Method 3-19) and by guinea pig urine. In 2019, uals with Er(a-b-) red cells. There is no evi-
Sd" was connected to variants of B4GALNT2 dence that Er antibodies are clinicaily signifi-
cant. The 210 collection contains Leeand Led,
gene, and a new blood group system was estab-
both lower-prevalence antigens. MN CHO con-
lished." Because Sd" is present in other tissues, tains carbohydrate antigens associated with
it is likely that only a fraction of the 10%of the MNS system antigens that are not encoded by
population who do not have detectable Sd" on GYPA or GYPB. These antigens have been
their red cells are actuaily Sd(a-), and would shown to result from altered glycosylation of
produce anti-Sd". the Olinked sugars on GPAand GPB.
The 201, 202, 203, 204,206, 209,211, and anti-Anw] is more common and is associated
212 collections have been made obsolete, as with a transiently AnWj- phenotype. The anti"
they have been assigned to a system. gen may be carried on CD44. AnWj is absent on
cord red cells .and is severely suppressed in red
High-Prevalence Antigens (901 Series) cells of the In(Lu) phenotype.
PEL is numbered as 901014. The PEL- phe-
The 901 series of the ISBT classification con"
notype and has been found in only two families,
tains six antigens (Table 12"10): all high preva-
and two examples of .antl-Plil. have been de"
lence. All are inherited, although none is eligible
scribed. A related antibody, antl-Ml'P, was non"
to join a system. All antigens are resistant to
reactive with PEL- red cells, but anti-Plil. was
papain, trypsin, «-chyrnotrypsin, and DTTI
weakly reactive with red cells of the antibody
AET treatment of the red cells, and all except
makers,
AnWj are well-expressed on cord cells.
The high-prevalence antigen ABTI, number
The first in the series is 901008, with the
901015, is serologically related to Vel. However,
symbol Emm. Little is known about the Emm
it has been excluded from SMIMI by sequenc-
antigen. Seven examples of anti-Bmm have been
ing analysis and thus has returned to the 901
described, and of those, six have occurred
series. Like Vel, ABTIexpression differs substan-
as naturally occurring antibodies, all in males
tially, and it is generally expressed only weakly
who have not received transfusion. The clinical
on cord red cells. ABTI is resistant to treatment
significance is unknown.
of red cells with proteolytic enzymes or
Anton is next (901009), with the symbol An"
disulfide-bond-reducing agents. Anti"ABTI has
Wj. AnWj is a high-prevalence antigen that
serves as the receptor for Heemophilus influen- not caused HDFN, and clinical data are limited.
For MAM (901016), there is good evidence
za on red cells. Alloanti-An Wj has been reported
that the antibodies are clinically significant. Se-
in few individuals and may cause severe HTRs,
vere HDFN and, in one case, neonatal thrombo-
although no cases are known of HDFN.87 Auto"
cytopenia have been reported to be caused by
anti"MAM.
LKE is the last antigen in this series and is
TABLE12-10. Antigens of the IS8T 901 Series
(High Prevalence) numbered 901017. Anti"LKE is rare, usually
IgM, and is generally not known to cause HTRs
or HDFN. One case of a clinically significant
anti"LKEhas been reported."
901008 No evidence of clinical
significance Low-Prevalence Antigens (700 Series)
AnWj 901009 Severe AHTRs Seventeen antigens of very low prevalence in all
of the. populations tested constitute the 700.se"
PEL 901014 No evidence of clinical
ries of.jhe ISBT classification: By, Chr', Bi, Bx',
significance
Toa, pta, Rea, W, Lia, Milne, RASM, JFV, Kg,
ABTI 901015 No evidence of clinical JONES, HJK, HOFM, and REIT.All are inherited
significance and do not fit any criteria for joining or forming
a system.
MAM 901016 Severe HDFN Antibodies to low-prevalence antigens do not
present transfusion problems because ..compati-
LKE 901017 No evidence of clinical
ble blood is readily available; however, these an"
significance
tibodies remain undetected if a serologic cross"
ISBT = International Society of Blood Transfusion; AHTR match including antiglobulin phase is not
= acute hemolytic transfusion reaction; HDFN = employed. Antibodies to JFV, Kg, JONES, HJK,
hemolytic diseaseof the fetus and newborn. and REIThave all caused HDFN.
HLA Antigens on Red Cells scribed in the LU system section above,

heterozygosity for different mutations in KLFI
"Bg" is the name given to HLAClass I antigens
has been identified in individuals with the
expressed on mature red cells. Bg' represents
In(Lu) phenotype. In these individuals, expres-
HLA-B7;Bgb,HLA-BI7 (BS7 or BS8); and Bt,
sion of antigens carried on CD44 (Inaflnb) and
HLA-A28(A68 or A69, which cross-reactswith
the AnWj and PI antigens is weak." However,
HLA-A2).Many individuals, however, do not KLFI mutations have also been shown to affect
express Bg antigens on their red cells, despite other genes, notably the ~-globingene, resulting
having the corresponding HLAantigens on their in the hereditary persistence of fetal hemoglobin
lymphocytes. syndrome." Affected individuals have elevated
There are a few reports of Bgantibodies caus- hemoglobin F levels, some >30%, and demon-
ing HTRs.87These antibodies are sometimes strate an In(Lu) phenotype. Furthermore, dis-
present as contaminants in reagents." HLAanti- crete mutations in KLFI appear to give rise to
gens on red cells are not destroyed by papain, fi- differentphenotypes; for example, the change of
cin, pronase, trypsin, a-chymotrypsin, AET, or Glu32SLysdoes not result in the In(Lu) pheno-
DTT. They can be decreased or removed from type but is associated with severe congenital
red cells with chloroquine (Method 2-20) or dyserythropoietic anemia. These red cells
acid glycinefEDTA (Method 2-21). demonstrate weakened expression of the anti-
gens in the CO (AOPl), CRaM (DAF),and LW
(ICAM-4)blood group systems."
Although mentioned above in the LUsystem
section, it is worth repeating that a mutation in
CATkl resulted in the Xlinked Lu(a-b-) phe-
notype in one family." It is likely that additional
mutations in these and other erythroid-specific
Mutations in genes encoding erythroid tran- transcription factors will be identified as the
scription factors are emerging as important mod- causes of altered blood group antigen expres-
ifiers of blood group antigen expression. As de- sion.
KEY POINTS
1. Of 363 recognized antigen specificities, 326 belong to one of 39 blood group systems repre-
senting either a single gene or two or more closely linked homologous genes. Some groups of
antigens that are not eligibleto join a system are classifiedtogether as collections. Antigens not
:~s~~eldS:i:S~~~~~~C~~~l~~ection have either low or high prevalence and make up the 700
2. M and N are antithetical, polymorphic antigens. M, N, S, s, and 'N' are generally thought to be
destroyed by treatment of the red cells with papain, ficin, bromelin, or pronase, although this
effectwith S and s is variable. M and N, but not S, s, or 'N', are destroyed by trypsin treatment.
3. Anti-M is relatively common, while anti-N is uncommon. Most anti-M and -N are reactive only
at room temperature and are not clinicallysignificant.When M or N antibodies active at 37 C
are encountered, antigen-negative or crossmatch-compatible red cells should be provided.
Anti-S, -s, and -U are generally IgG antibodies that are active at 37 C. They have been implicat-
ed in HTRsand severe and fatal HDFN.
4. Because anti-K can cause severe HDFN and HTRs, patients with anti-K should receive K-
~~~d::~~~~t~::ible. Anti-K is the most common immune red cell antibody not in the
5. The FYglycoprotein consists of five antigens. Ft and Fybpresent in four phenotypes: Fy(a+b-),
Fy(a+b+), Fy(a-b+), and Fy(a-b-).Fy3, FyS, and Fy6 are high-prevalence antigens. Ft and Fyb
REFERENCES
1. Daniels GL, Fletcher A, Garratty G, et al. Blood in extensive diversification. Hum Mutat 1995;
group terminology 2004: From the Internation- 6:199-209.
al Society of Blood Transfusion committee on 10. Stack G, Tormey CA, Estimating the immunoge-
terminology for red cell surface antigens. Vox nicity of blood group antigens: A modified calcu-
Sang 2004;87:304- 16. lation that corrects for transfusion exposures. Br
2. StorryJR, CastiIhoL, Chen 0, et al. Internation- J Haemato12016;175:154-60.
al Society of Blood Transfusion Working Party 11. Satchwell TJ. Erythrocyte invasion receptors for
on Red Cell Immunogenetics and Terminology: Plssmodiumfskipetum: New and old. Transfus
Report of the Seoul and London meetings. ISBT Med 2016;26:77-88.
Sci Ser 2016;11:118-22. 12. Wikman A, Edner A, Gryfelt G, et al. Fetal he-
3. Poole J, Daniels G. Blood group antibodies and molytic anemia and intrauterine death caused
their significancein transfusionmedicine. Trans- by antt-M immunization. Transfusion 2007;47:
fus Med Rev 2007;21 :58-71. 911-17.
4. Reid ME, Lomas-Francis C, Olsson ML. The 13. Huang CH, Blumenfeld00. Molecular genetics
of human erythrocyte MiIII and MiVI glyco-
blood group antigen factsbook. 3rd ed. London:
phorins: Use of a pseudoexon in construction of
Academic Press, 2012.
two delta-alpha-delta hybrid genes resulting in
5. Daniels G. Human blood groups. 3rd ed. Ox-
antigenic diversificatton. J BioI Chern 1991;
ford: Wiley-Blackwell,2013.
266:7248-55.
6. Daniels G. Blood group diversity and its impact
14. Heathcote DJ, Carroll TE, Flower RL. Sixty
on transfusion medicine. Transfusion science years of antibodies to MNS system hybrid glyco-
education course. Emeryville, CA: Grifols, phorins: What have we learned? Transfus Med
2018. Rev2011 ;25:111-24.
7. Anani W, Marchan M, BensingK, et al. Practical 15. Al-Iada NA. A Jordanian family with three sis-
approaches and costs for provisioning safe trans- ters apparently homozygous for MKand evi-
fusions during anti-CD38 therapy. Transfusion dence for clinical Significanceof antibodies pro-
2017;57(6):1470-9. duced by MkMkindividuals. Transfusion 2017;
8. Lux SE 4th. Anatomy of the red cell membrane 57(2):376-8.
skeleton: Unanswered questions. Blood 2016; 16. Crew VK, Green C, Daniels G. Molecular bases
127:187-99. of the antigens of the Lutheran blood group sys-
9. Blumenfeld 00, Huang CH. Molecular genetics tem. Transfusion 2003;43: 1729-37.
of the glycophorin gene family, the antigens for 17. Eng-HuiY, RoscheT, Almo S, Fiser A. Function-
MNSs blood groups: Multiple gene rearrange- al clustering of immunoglobulin superfamily
ments and modulation of splice site usage result proteins with protein-protein interaction infer-
mation calibrated Hidden Markov model se- 31. Gassner C, Bronnlmann C, Merki Y, et al. Step-
quence profiles. J Mol BioI 2014;426(4):945- wise partitioning of Xp21: A profiling method
61. for XK deletions causative of the McLeod syn-
18. Eyler CE, Telen MJ. The Lutheran glycoprotein: drome. Transfusion2017;57(9):2125-35.
A multifunctional adhesion receptor. Transfu- 32. Danek A, Rubio JP, Rampoldi L, et al. McLeod
sion 2006;46:668-77. neuroacanthocytosis: Genotype and phenotype.
19. KaramaticCrew V, Mallinson G, Green C, et al. Ann Neuro12001 ;50:755-64.
Differentinactivating mutations in the LUgenes 33. Meny GM. The Duffyblood group system: A re-
of three individuals with the Lutheran-null phe- view. Immunohematology 2010;26:51-6.
notype. Transfusion2007;47:492-8. 34. Tournamille C, Colin Y, Cartron JP, LeVan Kim
20. Singleton BK,Burton NM, Green C, et al. Muta- C. Disruption of a GATAmotif in the Duffygene
tions in EKLF/KLF1form the molecular basis of promoter abolishes erythroid gene expression in
the rare blood group In(Lu) phenotype. Blood Duffy-negativeindividuals. Nat Genet 1995;10:
2008;112:2081-8. 224-8.
21. KellerJ, Vege S, Horn T, et al. Novel mutations 35. Mohandas N, Gallagher PG. Red cell mem-
in KLF1encoding the In(Lu)phenotype reflect a brane: Past, present, and future. Blood 2008;
diversity of clinical presentations. Transfusion 112:3939-48.
2018;58(1 ):196-9. 36. Horuk R, Chitnis CE, Darbonne WC, et al. A re-
22. Westhoff CM, Reid ME. Review: The Kell, ceptor for the malarial parasite Plasmodium vi-
Duffy, and Kidd blood group systems. Immuno- vax. The erythrocyte chemokine receptor. Sci-
hematology 2004;20:37-49. ence 1993;261: 1182-4.
23. Lee S, Zambas ED, Marsh WL, Redman CM. 37. Shen H, Schuster R, StringerKF,et al. The Duffy
Molecular cloning and primary structure of Kell antigen/receptor for chemokines (DARC)regu-
blood group protein. Proc Natl Acad Sci USA lates prostate tumor growth. FASEBJ 2006;
1991;88:6353-7. 20:59-64.
24. Chaffln J. Kell kills (video). Blood Bank Guy, 38. Xu L, Ashkenazi A, Chaudhuri A. Duffy anti-
gen/receptor for chemokines (DARC)attenu-
February 2014. [Available at https://www.
ates angiogenesis by causing senescence in en-
bbguy. org/ education/videos/kellkills/ (ac-
dothelial cells.Angiogenesis2007; 10:307-18.
cessed October 10,2019).]
39. Lawicki S, Coberly E, Lee L, et al. Jk3 alloanti-
25. Daniels G, Hadley A, Green CA. Causes of fetal
bodies during pregnancy-blood bank manage-
anemia in hemolytic disease due to anti-K.
ment and hemolytic disease of the fetus and
Transfusion2003;43: 115-16.
newborn risk. Transfusion2018;58; 1157-62.
26. Marsh WL, Nichols ME, Oyen R, Thayer RS,et
40. Holt S, Donaldson H, Hazlehurst G, et al. Acute
al. Naturally occurring anti-Kellstimulated by E. transplant rejection induced by blood transfu-
coli enterocolitis in a 20-day-old child. Transfu- sion reaction to the Kidd blood group system.
sion 1978;18:149-54. Nephrol DialTransplant 2004; 19:2403-6.
27. Denomme GA. Kell and Kx blood group sys- 41. Heaton DC, McLoughlin K. Ikia-b-]red blood
tems. Immunohematology 2015;31:14-19. cells resist urea lysis.Transfusion 1982;22:70-1.
28. Salomao M, Zhang X, Yang Y, et al. Protein 42. Sands JM, Gargus J], Frohlich 0, et al. Urinary
4.1R-dependent multiprotein complex: New in- concentrating ability in patients with Jk(a-b-)
sights into the structural organization of the red blood type who lack carrier-mediatedurea trans-
blood cell membrane. Proc Natl Acad Sci USA port. J Am Soc Nephrol 1992;2: 1689-96.
2008; 105:8026-31. 43. Everaert N, Willemsen H, Hulikova A, et al.
29. Azouzi S, Collec E, Mohandas N, et al. The hu- The importance of carbonic anhydrase II in red
man Kell blood group binds the erythroid 4.1R blood cells during exposure of chicken embryos
protein: New insights into the 4.1R-dependent to C02. Respir Physiol Neurobio12010;172(3):
red cell membrane complex. Br J Haematol 154-61.
2015;171 :862-71. 44. Lutz H. Naturally occurring anti-band 3 antibod-
30. Lee S, Debnath AK, Redman CM. Active amino ies in clearance of senescent and oxidatively
acids of the Kellblood group protein and model stressed human red blood cells. Transfus Med
of the ectodomain based on the structure of Hemother 2012;39(5):321-7.
neutral endopeptidase 24.11. Blood 2003; 45. Storry J, Banch Clausen F, Castilho L, et al. In-
102:3028-34. ternational Society of Blood Transfusion Work-
ing Party on Red Cell Immunogenetics and 61. Moulds JM. The Knops blood-group system: A
Blood Group Terminology: Report of the Dubai, review. Immunohematology 2010;26:2-7.
Copenhagen and Toronto meetings. Vox Sang 62. Xu O. The Indian blood group system. Immuno-
2019;114(1):95-102. hematology 2011 ;27:89-93.
46. Byrne KM, Byrne PC. Review: Other blood 63. Smart EA, Storry JR. The OK blood group sys-
group systems-Diego,Yt, Xg, Scianna, Dom- tem: A review. Immunohernatology 2010;26:
brock, Colton, Landsteiner-Wiener, and Indian. 124-6.
Immunohematology 2004;20:50-8. 64. Crosnier C, Bustamante LY, Bartholdson SJ, et
47. Johnson NC. XG:The forgotten blood group sys- al. Basiginis a receptor essentiaIfor erythrocyte
tem. Immunohematology 2011;27:68-71. invasion by Plasmodium jalciparum. Nature
48. Moller M, Lee Y, Vidovic K, et al. Disruption of 2011 ;480:534-7.
the GATA1-binding motif upstream of XG/ 65. Karamatic Crew V, Burton N, Kagan A, et al.
PBDXabolishes Xg' expression and resolves the CD 151, the first member of the tetraspanin
Xgblood group system. Blood 2018; 132:334-8. (TM4) superfamily detected on erythrocytes, is
49. Velliquette RW.Review:The Sciannablood group essential for the correct assembly of human
system. Immunohematology 2005;21 :70-6. basement membranes in kidney and skin. Blood
50. Zang X, Ghosh K. BTNL9and ERMAPas novel 2004;104:2217-23.
inhibitors of the immune system for immuno- 66. Hayes M. Raph blood group system. Immunohe-
therapies. United States patent application publi- matology 2014;30:6-10.
cation. New York, NY:Albert Einstein College of 67. Seltsam A, Strigens S, Levene C, et al. The mo-
Medicine, 2018. lecular diversity of Sema7A, the semaphorin
51. Reid ME. Complexities of the Dombrock blood that carries the JMH blood group antigens.
group system revealed. Transfusion 2005;45: Transfusion 2007;47:133-46.
92S-99S. 68. Roudier N, Ripoche P, Gane P, et al, AOP3 defi-
52. Gubin AN, Njoroge JM, Wojda U, et al. Identifi- ciency in humans and the molecular basis of a
cation of the Dombrock blood group glycopro- novel blood group system, GIL. J Biol Chern
tein as a polymorphic member of the ADP- 2002;277 :45854-9.
ribosyltransferase gene family. Blood 2000;96: 69. Chou ST, Westhoff CM. The Rh and RhAG
2621-7. blood group systems. Immunohematology
53. Halverson GR, Peyrard T. A review of the 2010;26: 178-86.
Colton blood group system. Immunohematolo- 70. Zelinski T, Coghlan G, Liu XO, Reid ME. AB-
gy2010;26:22-6. CG2 null alleles define the Jr(a-) blood group
54. Daniels G. Functions of red cell surface proteins. phenotype. Nat Genet 2012;44: 131-2.
Vox Sang 2007 ;93:331-40. 71. Robey RW, To KK, Polgar 0, et al. ABCG2: A
55. Grandstaff Moulds MK. The LW blood group perspective. Adv Drug Deliv Rev 2009;61 :3-13.
system: A review. Immunohematology 2011; 72. Fujita S, Kashiwagi H, Tomimatsu T, et al. Ex-
27:136-42. pression levels of ABCG2 on cord red blood
56. Zennadi R, Moeller BJ, Whalen EJ, et al. cells and study of fetal anemia associated with
Epinephrine-induced activation of LW-mediated anti-Jr(a).Transfusion 2016;56:1171-81.
sickle cell adhesion and vaso-occlusion in vivo. 73. HeliasV, Saison C, BallifBA,et al. ABCB6is dis-
Blood 2007; 110:2708-17. pensable for erythropoiesis and specifies the
57. Mougey R. A review of the Chido/Rodgers new blood group system Langereis. Nat Genet
blood group. Immunohematology 201 0;26( 1): 2012;44: 170-3.
30-8. 74. Storry JR, Ioud M, Christophersen MK, et al.
58. Walker PS, Reid ME. The Gerbich blood group Homozygosityfor a null allele of SMIM1 defines
system: A review. Immunohematology 2010; the Vel-negative blood group phenotype. Nat
26:60-5. Genet 2013;45:537-41.
59. Storry JR, Reid ME, Yazer MH. The Cromer 75. BallifBA,HeliasV, Peyrard T, et al. Disruption of
blood group system: A review. Immunohemato- SMIMl causes the Vel- blood type. EMBO Mol
logy 2010;26: 109-18. Med 2013;5:751-61.
60. Egan ES, Jiang RH, Moechtar MA, et al. Mala- 76. Cvejic A, Haer-Wigman L, Stephens JC, et al.
ria. A forward genetic screen identifies erythro- SMIM1 underlies the Vel blood group and influ-
cyte CD55 as essential for Plasmodium fslci- ences red blood cell traits. Nat Genet 2013;45:
parum invasion. Science 2015;348:711-14. 542-5.
77. AnlikerM, von Zabern I, Hochsmann B, et al. A 86. Peyrard T, Vrignaud C, Mikdar M, et al. IGT4:
new blood group antigen is defined by anti- Alloantibodies directed to the SLC4412/CTL2
CD59, detected in a CD59-deficient patient. transporter define two new red cell antigens
Transfusion2014;54: 1817-22. and a novel human blood group system (ab-
78. Weinstock C, Anliker M, von Zabern I. An up- stract). Transfusion2019;59(Suppl):18A.
date on the CD59 blood group system. Immu- 87. Zhaodong X, Duffett L, Tokessy M, et al. Anti-
nohematology2019;35:7-8. AnWj causinghemolytic transfusionreactions in
79. Daniels G, BallifBA,HeliasV, et al. Lack of the a patient with aplastic anemia. Transfusion
nucleoside transporter ENTI results in the 2012;52:1476-81.
Augustine-null blood type and ectopic mineral- 88. Cooling L, Dake LR,Haverty D, et al. A hemo-
ization. Blood2015;125:3651-4. lytic anti- LKE associated with a rare LKE-
80. Daniels G. An update on the Augustine blood negative, "weak P" red blood cell phenotype:
group system. Immunohematology 2019;35: 1- Alloanti-LKEand alloanti-Precognize galactosyl-
2. globoside and monosialogalactosylgloboside
81. Kawabata K, Uchikawa M, Ohio H, et al. Anti- (LKE)antigens. Transfusion2015;55(1): 115-28.
KANNO:A novel alloantibodyagainst a red cell 89. Nance ST. Do HLAantibodies cause hemolytic
transfusion reactions or decreased RBCsurviv-
antigen of high frequency. Transfus Med Rev
al? Transfusion2003;43:687-90.
2014:28:23-8.
90. Anti-Fy"and Anti-Fy" (packageinsert). Rev 021
82. Omae Y,Ito S, Takeuchi M, et al. Integrativege-
2013. Peachtree Corners, GA:Immucor, 2013.
nome analysis identified the KANNO blood 91. BorgI, PapadopoulosP, GeorgitsiM, et al. Hap-
group antigen as prion protein. Transfusion loinsufficiency for the erythroid transcription
2019;59;2429-35. factor KLF1 causes hereditary persistence of fe-
83. Macvie SJ, Morton JA, Pickels MM. The reac- tal hemoglobin.Nat Genet 2010;42:801-5.
tions and inheritance of a new blood group anti- 92. Arnaud L, Saison C, HeliasV, et al. A dominant
gen Sd", Vox Sang 1967;13:485-92. mutation in the gene encoding the erythroid
84. Benton PH, Howell P, Ikin EW. Antl-Sd",a new transcription factor KLFl causes a congenital
blood group antibody. Vox Sang 1967;13:493- dyserythropoietic anemia. Am J Hum Genet
501. 2010;87:721-7.
85. StenfeltI, HellbergA, Moller M, et al. Missense 93. Singleton BK, Roxby DJ, Stirling JW, et al. A
mutations in the C-terminal portion of the novel GATAlmutation (Stop414Arg)in a family
B4GALNT2-encoded glycosyltransferaseunder- with the rare X-linkedblood group Lula-b-]phe-
lying the Sd(a-) phenotype. Biochem Biophys notype and mild macrothrombocytic thrombo-
Rep2019;19:100659. cytopenia.BrJ Haemato12013;161:139-42.
Identification of Antibodies to
Red Cell Antigens
Lay See Er, MSTM, SBB(ASCp;CM, CQA(ASQ), and Debra J. Bailey, MT(ASCP)SBB
ATURALLY OCCURRING ANTI-A population being studied. In chronic transfusion

and antl-B are the only red cell antibod- recipient populations, such as those with sickle
ies commonly found in human serum or cell anemia or thalassemia, as many as 14% to
plasma. All other antibodies are called "unex- 50% of individuals are reported to be alloimmu-
pected red cell antibodies." This chapter dis- nized, whereas in the general population, allo-
cusses methods for identifying unexpected red immunization is estimated to be 0.5% to 1.5%.1-4
cell antibodies once pretransfusion testing (see In some instances, no specific immunizing
Chapter 17) indicates an unexpected antibody is event due to red cell exposure can be identified.
present. Non-red-cell-stimulated or "naturally occur-
There are two types of unexpected red cell ring" antibodies have presumably resulted from
antibodies: alloantibodies and autoantibodies. exposure to environmental, bacterial, or viral
When an individual produces an antibody to an antigens that are similar to blood group anti-
antigen that he or she lacks, the antibody is gens. Antibodies detected in serologic tests may
called an alloantibody. When an individual pro- also be passively acquired from injected immu-
duces an antibody to an antigen that he or she noglobulin, donor plasma, passenger lympho-
possesses, the antibody is called an autoanti- cytes in transplanted organs, or hematopoietic
body. Therefore, by definition, alloantibodiesre- progenitor cells (HPCs) or, in the case of neo-
act only with allogeneic red cells that express nates, may be of maternal origin.
the corresponding antigens-not with the anti- After an antibody has been detected, its type
body producer's red cells. Conversely, autoanti- (auto and/or allo)-and in the case of alloanti-
bodies are reactive with the red cells of the antibody, its specificity-should be determined and
body producer. In fact, autoantibodies usually its clinical significance assessed. A clinically sig-
are reactive with most reagent red cells as well nificant red cell antibody is defined as an anti-
as with autologous red cells because. they typi- body that is associated with hemolytic disease of
cally target an antigen commonly expressed on the fetus and newborn (HDFN), hemolytic
all red cells except those of rare phenotypes. transfusion reactions, or a notable decrease in
Immunization to red cell antigens may result transfused red cell survival. Determining the
from pregnancy, transfusion, transplantation, specificity of the antibody is the most commonly
needle sharing, or injections of immunogenic used way of predicting its possible clinical signif-
material. The incidence of alloimmunization is icance. Yet, the degree of clinical Significance var-
extremely variable depending on the patient ies among antibodies with the same specificity;
Lay See Er, MSTM, SBB(ASCP)CM, COA(ASQ),Manager, Immunohematology Reference Laboratory, Bloodworks
Northwest, Seattle, Washington; and Debra J. Bailey, MT(ASCP)SBB,Lead Technologist, Immunohematology
Reference Laboratory, American Red Cross Blood Services, Detroit, Michigan
389
390 A A B B TEe H N I CAL MAN UAL
some cause destruction of incompatible red cells Variation in Adults and Neonates
within hours or even minutes, whereas others
Some antigens (eg, I, PI, Le", and Sd') show
decrease red ceu survival by only a few days,
variable expression on red cells from different
and still others do not shorten red cell survival
adults. The antigenic differences can be demon-
discernibly. Some antibodies are known to cause
strated serologically; however, the variability
HDFN, whereas others may cause a positive di-
from one antigen-positiveadult to another is un-
rect antiglobulin test (DAT)result in the fetus
without clinicalevidence of HDFN. related to zygosity.Some antigens are expressed
differently on cord/neonate red cells compared
to adult red cells. Antigen expression on cordi
BASIC S IN RED neonate red cells may be absent, weaker, or
CIELL ANTI PRESSI N stronger as compared to adult red cells. (Table
13-1provides some examples.)
Antibody identification is dependent on the reo
activity of the serum or plasma with red cells of Changes with Storage
known antigen expression. A basic understand- Blood group antibodies may be more weakly re-
ing of variables in antigen expression is critical active with stored red cells than with fresh red
to interpreting reactivity in identification stud- cells. Some antigens (eg, Fy, Fyb,M, PI, Kna,
ies. Mcca, and Bg) deteriorate more rapidly than
others during storage, and the rate varies among
Zygosity and Dosage
red cells from different lndividuals." Becausered
The reaction strength of some antibodies may cells from donors are often fresher than com-
vary because of dosage, meaning that antibodies mercial reagent red cells, some antibodies have
are more strongly reactive (or only reactive) stronger reactions with donor red cells than
with red cells that possess a "double-dose" ex- with reagent red cells. Similarly,storage of red
pression of the antigen. Dosage describes the cells in a freezer may cause antigens to deterio-
expression of an antigen on red cells, while zy- rate, thus producing misleading antibody identi-
gosity describesthe degree of similarityof alleles fication results.
(alternative forms of the same gene) present at a The pH or other characteristics of the storage
given locus. Double-dose antigen expression oc- medium can affect the rate of antigen deteriora-
curs when an individual is homozygous for the tion.4,6 For example, Fy and Fybantigens may
gene for a given allele that encodes the antigen. weaken when the red cells are stored in a medi-
Red cells from individuals who are heterozygous um with low pH and low ionic strength. Thus,
for the gene would be expected to express a certain antibodies may demonstrate differences
single-doseof the corresponding antigen(s). Red
cells with a single-doseexpression of an antigen
TABLE 13·1. Antigen Expression on Cord Red
would typically have fewer antigen sites than Cells*
those that express a double dose and, therefore,
may be weakly reactive or nonreactive with a Expressidlf ' ""Anti!lens""
weak example of the corresponding antibody.
Alloantibodiesvary in their tendency to demon- Negative Lea, Leb, Sda, Ch, Rg, and AnWj
strate dosage (see "Alleles"in Chapter 9). Many
Weak I, H, P1, Lua, Lub, W, Vel, Sg, KN,
antibodies to antigens in the RH, FY (Duffy),
and DO antigens, Yka, Csa, and
MNS, and JK (Kidd) blood group systems
Fy3
demonstrate dosage. (SeeTable 9-4 in Chapter 9
for examples of International Society of Blood Strong i, Lwa, and LWb
Transfusion terminology applied to blood group
systems.) *Modified with permission from Reidet al.5
C HAP T E R 1 3 Identification of Antibodies to RedCellAntigens 391
in reactivity with red cells from different manu- Reagents and Test Methods
facturers if the suspending media are different. Antibody Detection Red Cells
The age and nature of the specimen must be
considered when red cells are typed. Antigens °
Group red cells suitable for pretransfusion an-
tibody screening are commercially available and
on red cells from clotted samples tend to deteri-
most frequently offered as sets of either two or
orate more quickly than antigens on red cells
three (sometimesfour) samples of reagent single-
from donor units that are collected in citrate an- donor red cells. Allreagent red cell sets licensed
ticoagulants, such as acid-citrate-dextrose or by the US Food and Drug Administration (FDA)
citrate-phosphate-dextrose. Red cells in donor for this purpose must contain red cell samples
units collected in approved anticoagulants retain that collectively express the following antigens:
their antigens throughout the standard shelf life D, C, E, c, e, M, N, S, s, PI, Lea,Leb, K, k, Ft,
of the blood component. Sampleswith EDTAup Fyb, Ik', and Jkb. Three-red-cell-sampleantibody-
to 14 days old are suitable for antigen typing? detection sets often offer red cells from donors
presumed homozygous for the respective genes
However, the manufacturer's instructions per-
with double-dose expression for the following
taining to sample suitability for antigen typing common antigens: D, C, E, c, e, k, M, N, S, s,
should always be consulted when commercial Fy", Fyb, Ik",and Jkb. As mentioned above, anti-
typing reagents are used. bodies to antigens of the RH, MNS, FY, and JK
blood group systems most commonly demon-
strate dosage. Each laboratory should decide
whether to use two or three reagent single-
donor red cell samples for antibody detection
testing. When antibody detection is automated,
the instrument platform may dictate the reagent
Specimen Requirements red cell configuration. In the United States,
pooled red cells for antibody detection are ob-
Serum and plasma are interchangeable for anti- tained from two different donors and may be
body testing unless complement is required for used only when testing donor samples. Reagent
antibody detection. In such rare cases, only se- red cells should be refrigerated when not in use
rum provides complement. Throughout this and should be in date when used for antibody
detection tests.
chapter, serum can be considered interchange-
able with plasma unless the text indicates other-
Antibody Identification Red Cell Panels
wise. The use of serum or plasma may also be
dictated by the test method or the manufactur- Identification of antibodies to red cell antigens
detected in pretransfusion antibody screening
er's instructions for the reagents employed.
requires testing the serum/plasma against a pan-
Depending on the test methods used and the el of red cell samples (typically 8-16) with
hematocrit of the patient, a 5-mL to 10-mL ali- known antigenic composition. Usually, the red
quot of whole blood usually contains enough se- cell samples are obtained from commercial sup-
rum or plasma for identifying simple antibody pliers, but institutions may assemble their own
specificities; more whole blood (ie, 10 mL to panels using red cells from local sources. Except
20 mL) may be required for complex studies. in special circumstances, panel cells are group
0, thereby allowing serum/plasma of any ABO
When autologous red cells are tested, the use of
group to be tested.
samples anticoagulated with EDTAavoids prob- Each reagent red cell sample in the panel is
lems associated with the in-vitro uptake of com- from a different donor. The reagent red cells are
plement components by red cells, which may selected so a distinctive pattern of positive and
occur when clotted samples are used. negative reactions will result when the reactivity
of all the panel cells is considered. To be func- demonstrate clinically significant antibodies"
tional, a reagent red cell panel must make it pos- and "include incubation at 37 C preceding an
sible to identify with confidence those clinically antiglobulin test. "9(p36) All methods meet this
significantalloantibodies that are most common- standard, but each method offers different ad-
ly encountered, such as anti-D, -E, -K, and -Py', vantages. Tube testing offers flexibilityto test at
The phenotypes of the reagent red cells should different phases and the option to use a variety
be distributed so that single common alloanti- of additive solutions (and thus obtain varying
body specificities can be clearly identified and degrees of sensitivity). It also requlres little spe-
most others can be excluded. Ideally,patterns of cialized equlpment. Column agglutination and
reactivity for most examples of single alloantl- solid-phase technology offer stable and possibly
bodies should not overlap with any other (eg, all less subjective endpoints, workflow standardiza-
of the K+ red cells should not be the only ones tion, and the ability to be incorporated into
that are also E+). Reagent red cellswith double- semiautomated or automated systems. They
dose antigen expression are included to detect provide a sensitive detection system for most
common antibodies that frequently show dos- blood group antibodies. Column agglutination,
age. Commercial panels are accompanied by a solid-phase methods, and very sensitive tube
common antigen profile sheet that lists the phe- tests have also been shown to enhance serologic
notypes of the red cells. Table 13-2 provides an reactivity that may not be clinicallySignificantin
example of a common antigen profile sheet for a the selection of units for transfusion, including
hypothetical antibody identification panel. It is the reactivity of warm autoantibodies. The dif-
essential to use the correct panel antigen profile ferent methods offer laboratories choice in
sheet when interpreting results because the selecting a primary antibody detection and iden-
combination of red cell samples is different for tification method that, with its sensitivity, speci-
each lot of panel cells. Commercial reagent red ficity, automation capabilities (if desired), and
cells for tube testing are diluted to a 2% to 5% cost, is suitable for the patient population
suspension in a preservative solution that can be served, the laboratorysize, and its staffexpertise/
used directly from the bottle. Washing the red experience.
cells before use is usually unnecessary unless Published studies have compared the vari-
the preservative solution is suspected of interfer- ous methods for detection of wanted and un-
ing with antibody identification. wanted red cell alloantibodies as well as the po-
Panel cells beyond their expiration date tential effect of using red cell membranes vs
should not be used as the sole resource for anti- intact red cells.10·16 Laboratories should be famil-
body identification. Most laboratories use in- iar with the unique reactivity characteristics of
date reagent red cells for initial antibody identifi- their selected method. They frequently will
cation and, if necessary, use expired reagent red choose to have one or more additional methods
cells to exclude or confirm uncommon specifici- available and develop testing algorithms that in-
ties. Any laboratory that uses expired reagent volve the different methods to aid in the investi-
red cells should establish a policy and validate gation and resolution of results that are not easi-
any procedures associated with this prac- ly apparent with their primary method.
tice.8(p21)
Additive Solutions
Test Methods
Although the test method may consist solely of
All techniques for antibody detection and identi- serum or plasma and red cells (either reagent
fication in general use today are based on the red cells as provided by the manufacturer or
principle of hemagglutination (tube or column saline-suspended red cells), as in Method 3-2,
agglutination systems) or red cell adherence most serologists use some type of additive solu-
(solid phase). AABBStandards for Blood Banks tion when tube methods are used to decrease
and Transfusion Services (Standards) requires incubation time and/or increase sensitivity.Sev-
that "methods of testing shall be those that eral different additive solutions are available,
CHAPTER 13 Identification of Antibodies to RedCellAntigens 393
N <.0 co
+ + + + + + +
+ + + + +
+ + + +
+ + + + + +
+ + + + + + +
+ + + + + + + + + +
c::
o
:g +
-
(.)
E
c:: + + + + + + +
a:>
"'0
>, + + + + + + ~c:::
"'0 o
o '-'
.0
+ + + + +
.s
::>
:0 + +
c::
« '"
II
c..:>
+ + + + + + + + <C
1::
Cl)
c:n
:;::::
> + c:::
'"
'0
Cl)
+ u
c:::
Cl)
en
..0
+ + + + + + + '"en
Cl)
1i:i
'-'
+ + + + + + + + + '6
.=o
+ + + + + + + + 1::
Cl)
c:n
:;::::
c:::
+ +
'"
'0
Cl)
'-'
+ + o + o o o o o + c:::
Cl)
en
ec.
o + + + o o a o + o a + en
b++F+4-----i-----i-----i-----i-----i-----i-----i-----t-----t-----t-----t-----t-----i i§
N <.0 co m o '6
.s
+
including low-ionic-strength saline (LISS),poly- monoclonal. Anti-IgG can be either polyclonal

ethylene glycol (PEG), and 22% bovine albu- or monoclonal. Some US-licensed monoclonal
min. Additional enhancement techniques may anti-IgGdoes not detect antibodies of IgG4 sub-
be used for complex studies. Some enhance- class. This has little clinical significance, as red
ment techniques are discussed in more detail cell antibodies of pure IgG4 subclass are rare,
later in this chapter. Specifics regarding the and they are not associated with increased red
mechanism of action for 22% bovine albumin, cell destruction because monocytes do not have
LISS,and PEG can be found in the methods de- receptors for the Fc portion of IgG4 molecules.
scribing their use (Methods 3-3, 3-4, and 3-5). The property of an anti-IgG reagent not to react
Non-tube methods typicallyprescribe the use of with IgG4 subclass antibodies may be a testing
an additive solution, most often LISS, for the advantage in some cases. It has been described
same reasons they are used in tube methods. to be a valuable tool for excluding common clin-
ically significant alloantibodies of IgGI, IgG2,
Antiglobulin Reagents and IgG3 subclassesin patients who either have
made a pure IgG4 alloantibody (eg, anti-JMH)or
Most antibody detection and identification stud- have a passivelyacquired IgG4 antibody due to a
ies include an indirect antiglobulin test (IAT) biologictherapy (ie, anti-CD47, Hu5F9-G4).18.20
phase. Either antihuman globulin (AHG)specific However, a recent report has described lack of
only for human immunoglobulin G (IgG) or a reactivity of this monoclonal anti-IgG also with
polyspecific reagent that contains anti-IgG and several IgG3 isoallotypes (genetic variations
anticomplement may be used. A polyspecificre- within an IgG subclass]." These reactivity pat-
agent may detect-or may detect more readi- terns could lead to missing a clinically signifi-
ly-antibodies that bind complement, because a cant IgG3 red cell antibody if it is predominantly
single IgG molecule can deposit multiple com- of an isoallotype not detected by the reagent.
plement molecules. Therefore, presence of a Differences in anti-IgG reactivity can also be a
low-level IgG antibody may be visualized by its source of variation in detection of antibody reac-
complement activation. To detect complement tivity between laboratories testing the same
binding, serum rather than plasma must be sample. It is important to reference the manu-
used: the anticoagulant in plasma binds calcium, facturer's instructions for unique performance
making it unavailable for complement activa- characteristics of each reagent.
tion. Complement binding may be advanta-
geous in some rare instances, such as the detec-
tion of certain JK system antibodies. Because of
the sensitivity of current test methods, most se-
rologists prefer IgG-specificAHG reagents for
routine use. This avoids unwanted reactivity re- Patient History
sulting from in-vitro complement binding by
Beforeantibody identification testing begins, the
cold-reactiveantibodies."
Antiglobulin reagents may be derived from patient's medical history should be considered,
if such information can be obtalned. Multiple as-
monoclonal or polyclonal source material. Poly-
pects of the patient's medical history may influ-
clonal reagents, by nature, contain antibodies
ence antibody identification test selection as
from many B-cell clones, and collectively their
well as interpretation.
reactivity is directed at many epitopes of the tar-
get antigen. Monoclonal reagents have selec-
Prior Red Cell Exposure
tive reactivity with only specific epitopes. Re-
agents made from different clones may have Exposure to foreign red cells through blood
subtle reactivity differences: some monoclonal- transfusion or pregnancy is the usual cause of
based reagents are blended to cover a wider red cell immunization. It is uncommon for pa-
range of epitope specificities. The anticomple- tients who have never had a transfusion or been
ment reagents licensed in the United States are pregnant to produce clinicallysignificantalloan-
tibodies, although naturally occurring antibodies globulin (WIG) and Rh Immune Globulin
may be present. Women are more likely to have (RhIG) can interfere with .antibody screening
alloantibodies than men because of exposure to tests. Some lots of MG have been reported to
foreign (ie, fetal)red cells during pregnancy. In- contain unexpected antibodies, including anti-A
fants <6 months usually do not produce alloanti- and anti-B. Intravenous RhIG, which is some-
bodies, but newborns may have passive anti- times used to treat immune thrombocytopenia,
body of maternal origin. could explain the presence of anti-D in an Rh-
If the patient has received transfusion, it is positive patient.
critical to know when the most recent transfu- Monoclonal antibodies developed as immu-
sion was given. If transfusion was given during notherapeutic agents may also interfere with se-
the past 3 months, primary immunization to red rologic results. Anti-CD38 treatment [daratu-
cell antigens may be a risk, and the presence of mumab (Darzalex; Janssen Biotech)] for
circulating donor red cells affects testing. multiple myeloma and other B-cellmalignancies
Mixed-fieldresults caused by the donor red cells has been used for a number of years and is
in antigen typing tests interfere with interpreta- known to cause positive reactions in serologic
tion of an autologous phenotype. In situations tests employingthe antiglobulin phase when the
where warm autoantibodies are present in a infused monoclonal IgG anti-CD38 binds to the
sample, autologous adsorption techniques small amount of CD38 present on all normal red
would not be used because alloantibodies could cells, including reagent red cells of antibody de-
be adsorbed onto transfused donor red cells. tection and identification tests.ZZ-Z4 Anti-CD38
can also, although less often, be the cause of a
Diagnosis and Disease weakly positive DATin the treated patient. Clin-
Certain diseases have been associated with red ical trials of an anti-CD47 [humanized monoclo-
cell antibodies; depending on the methods used, nal antibody Hu5F9-G4 (Forty Seven, Inc, in
such antibodies may be detectable in antibody collaboration with Merck KGaA and Genen-
detection and identification tests. Cold aggluti- tech)] alone or with other immunotherapeutic
nin syndrome, Raynaud phenomenon, and agents and at least one additional source of
infections with Mycoplasma pneumoniae are monoclonal anti-CD38 [Isatuximab (Imrnuno-
often associated with anti-I. Infectious mononu- Gen and Sanofl-Aventisj]are under way and
cleosis is sometimes associated with anti-i, Pa- have also been reported to interfere with pre-
tients with paroxysmal cold hemoglobinuria, transfusion testing.18-20,2S-Z7
Development of oth-
which is associated with syphilis in adults and er novel immunotherapies might cause similar
viral infections in children, may demonstrate au- serologicinterferences depending on their target
toantibodies with antl-P specificityconfirmed by antigen. Communication from the health-care
the Donath-Landsteiner test. Warm autoanti- team identifying patients being treated with a
bodies often accompany diagnoses such as monoclonal immunotherapy can streamline pre-
warm autoimmune hemolytic anemia, systemic transfusion testing.
lupus erythematosus, multiple myeloma, chron-
ic lymphocytic leukemia, or lymphoma. Patients Elements of Basic Antibody
who have received solid-organ or HPC trans- Identification
plants may demonstrate passive antibodies that
originate from donor passenger lymphocytes. Autologous Control and OAT
The autologous control (autocontrol), in which
Medications and Biologic Therapies serum or plasma and autologous red cells are
Certain drugs are known to cause antibody tested under the same conditions as reagent red
identification problems. (See Chapter 14 for a cells, is an important part of antibody identifica-
discussion of drug-related mechanisms and tion and should be performed when the test
drugs that are associated with serologic prob- method allows. The autocontrol is not the same
lems.) Administration of intravenous immune as or equivalent to a DAT (Method 3-14). Incu-
bation and the presence of additive solutions most antibodies that are reactive only at lower
may cause reactivity in the autocontrol that is temperatures have little or no clinical signlfi-
an in-vitro phenomenon only. If the autocontrol cance.
is positive in the antiglobulin phase, a DAT Readings for direct agglutination taken after
should be performed. If the DAT result is nega- 37 C incubation in tube testing are influenced
tive, antibodies to an additive solution constitu- by the additive solution used. Tests employing
ent or autoantibodies that are reactive only in PEG enhancement cannot be centrifuged and
the presence of the additive solution should be read because the reagent causes nonspecific ag-
considered. Warm autoantibodies and cold auto" gregation of all red cells. LISS,albumin, and sa"
antibodies, such as anti-I, "IH,or -Pr, may be re- line (no enhancement) tests do not have this re"
active in an IAT when certain enhancement striction. A 37 C reading can detect some
techniques are used; therefore, testing should be antibodies (eg, potent anti-D, "E,or "K)that may
repeated in another medium. If the DAT result cause direct agglutination of red cells. Other an"
is positive, it must be interpreted with careful at" tibodies (eg, antl-Le' or -Ik'] may occasionallybe
tention to the transfusion history. Autoantibod- detected by the lysisof antigen-positivered cells
ies or drugs could explain a positive DATresult; during the 37 C incubation if serum is tested.
however, if the patient has an alloantibody and Omitting centrifugation and the reading at 37 C
recently received blood that expressed the cor" should lessen the detection of unwanted posi-
responding antigen, the positive DATresult may tive reactions caused by clinically insignificant
be caused by coating of the donor red cells with autoantibodies and alloantibodies. However, in
alloantibody. This situation can be associated some instances, potentially clinically significant
with a clinically significant delayed transfusion antibodies are detected only by their 37 C reac-
reaction. More information about interpreting a tivity. In 87,480 samples, 103 examples of such
positive DATresult can be found in Chapter 14. antibodies were identified (63 antl-B; 27 "K;5
"lka;4 "D; 3 "cE;and 1 "C).28 If the 37 C reading
Initio/Identification Panel is desired in a specificantibody study, an alterna-
tive strategy is to set up duplicate tests. One test
For initial antibody identification, it is common
is read after the 37 C incubation, and the other
to use a complete commercial reagent red cell
test is read only at the IATphase.
panel with the same methods and test phases as
in the antibody detection test or crossmatch.
Abbreviated Identification Panel
The gel-column and solid-phase methods in"
volve a single reading of the test at the IAT If a patient has antibodies that were identified
phase. Tube-testing protocols have greater flexi- previously, the known antibodies should be
bility for reading at different test phases (eg, im- considered when selecting reagent red cells to
mediate spin, room temperature, 37 C, and test. For example, if the patient has known anti-e,
IAT),but many serologists also use a single IAT it will not be helpful to test the patient's serum
reading because this test detects the over" with a complete commercial reagent red cell
whelming majority of clinically Significantallo- panel in which 9 of lO red cell samples are
antibodies. e-positive.Testing a selected panel of e-negative
Some serologists using tube methods may red cells is a better approach to find any newly
choose to include an immediate centrifugation formed antibodies. It is not necessary to test
reading, a room-temperature incubation that is e-positive red cells to reconfirm the previously
read before an additive solution is included, or identified anti-e because e-negative donor units
both during initial antibody identification. Such will be selected for transfusion regardless of the
an approach may enhance the detection of cer- reactivity.
tain antibodies (eg, anti-M, "N, "PI, "I, "Lea,or If the patient's red cell phenotype is known,
"Leb) and may help explain reactions detected in reagent red cells may be selected to detect only
other phases. These steps are frequently omitted those alloantibodies that the patient would po-
in initial antibody identification studies because tentially form. For example, if the patient's Rh
phenotype is C-E+c+e-, red cells selected to the genotype obtained from DNA isolated from
exclude anti-E and anti-c should not be neces- leukocytes and hematopoietic cells may differ
sary or can be limited to a single selected cell from that of other tissues in people with a histo-
sample because the patient is not expected to ry of transplantation."
form alloantibodies to these antigens. Excep-
tions include patients with weak or altered (par- Interpretation of Results
tial) Rh antigens, which are usually found in
Antibody detection results are interpreted as
populations of African ancestry and patients
positive or negative according to the presence or
whose Rh phenotype was predicted by DNA
absence of reactivity (ie, agglutination, hemoly-
testing rather than serology and who could be
sis in serum tests, or red cell effacement in solid-
carrying a silenced or altered allele. This ap-
phase tests). Interpretation of antibody identifi-
proach can minimize the amount of testing re-
cation results can be a very complex process
quired.
combining technique, knowledge, and intuitive
Autologous Red Cell Phenotype
skills. Identification panels generally include
both positive and negative results, sometimes at
Determining the phenotype of an individual's different phases of testing; each positive result
autologous red cells by serology or genotypingis should ultimately be explained. The following
an important part of antibody identification be- sections describe a systematic process for anti-
cause the antibody maker's red cells are expect- body identification interpretation.
ed to lack antigens to which they make alloanti-
bodies. This information can guide the antibody General Assessment of Positive and Negative
identification process. Reactions
Obtaining an autologous red cell phenotype
may not always be simple. Recent transfusions Both positive and negative reactions are equally
or immunoglobulins coating the patient's red important in antibody identification, and they
cells make obtaining a valid phenotype difficult. may be initially assessed to provide a general
Misleading results may occur unless techniques idea of the specificity(ies)present in the sample.
are used to circumvent these issues." Many of (See Method 1-9 for grading agglutination in
these special phenotyping techniques, such as hemagglutination assays.) The phase and
separation of autologous cells or removal of strength of positive reactions may be compared
bound immunoglobulin, are described later in to the antigen patterns of the panel red cells to
this chapter under "Selected Procedures." Red help suggest specificity. Negative reactions sup-
cell genotyping is now commonly performed to port the specificitysuggested by the positive re-
obtain phenotype information. This approach actions. A single common alloantibody usually
avoids interference from circulating donor red produces a clear pattern with antigen-positive
cells or immunoglobulin-coated patient red and antigen-negative reagent red cells. In Table
cells, Molecular testing relies on the extraction 13-2, if a sample is reactive only with red cell
of DNA from white cells. Because of nearly uni- samples 3 and 5 of the reagent red cell panel,
versal leukocyte reduction, the short life span of anti-B is very likely present. Both reactive red
white cells in vivo, and, importantly, the assay cells express the antigen, and all cells lacking
design, the presence of transfused white cells the.antigen are nonreactive. This general assess-
from donors, if present, is not a limiting factor in ment is only the first part of the interpretation
determining the patient's red cell genotype. process. The rest of the process as described be-
There are situations, however, where the geno- low must be completed even when a specificity
type of a person may not predict the red cell looks apparent at this stage. Exclusion of anti-
phenotype. Mutations that inactivate gene ex- bodies must be performed to ensure proper
pression or rare new alleles may not be identi- identification of all antibodies potentially pres-
fied by the specificassay performed. In addition, ent.
Antibody Exclusion and Initial Specificity laboratory's criteria of two examples of double-
Assessment dose common antigens or two examples of anti-
gens whose expression is not a result of zygosity
A widely used first approach to the interpreta- (eg, PI, Lea,Leb)that were nonreactive with the
tion of panel results is to exclude specificitieson patient's plasma. It should be noted that some
the basis of nonreactivity of the patient's sample specificitieswere not crossed out at the top de-
with red cells that express the antigen. Such a spite being crossed out on one or more rows.
system is sometimes referred to as a "cross-out," Anti-E was not crossed out on the top row be-
"rule-out," or "exclusion" method. Once all cause although two panel red cells (#3 and #5)
panel results have been recorded, the antigen were E+ and nonreactive with the patient's plas-
profile on the panel worksheet of the first nonre- ma, only one was a double-dose expression
active red cell is examined. If an antigen is pres- (#3), and therefore anti-B did not meet this labo-
ent on the panel red cell and the patient sample ratory's criteria for final exclusion. There are
was not reactive with it, the presence of the cor- many acceptable variations of exclusion policies;
responding antibody may tentatively be exclud- the one chosen for this example is just one alter-
ed. native.
Many laboratory scientists actually cross out Exclusion of clinicallysignificant alloantibod-
such antigens from the list at the top of the anti- ies should involve, at a minimum, those to the
gen profile sheet using a mark on the specificity followingantigens: D, C, E, c, e, K, Fy",Fyb,Jka,
to facilitate the process. Laboratories that have Jkb, S, and s. Antibodies to Lea, Leb,M, N, PI,
rule-out policies that consider whether the rule- and other antigens specific to certain patient
out was done on a single or double dose of the populations may also be added to this list. Labo-
antigen may use different notations to distin- ratories should have a policy for their antibody
guish between the two-eg, "I" and "X" re- exclusion. The policy should list alloantibodies
spectively. After all of the antigens for a red cell requiring exclusion as well as, based on the cho-
sample have been crossed out, the same process sen test method and available resources, wheth-
is performed with the other nonreactive red er the exclusion is to be performed using single-
cells for the panel; additional specificities are or double-dose antigen-positive red cells and if
then excluded. After the last nonreactive red more than one example of the antigen is need-
cell is examined, only those antigens not ed. Ideally, antibody exclusion is performed on a
"crossed out" are left for further evaluation as nonreactive double-dose antigen-positive red
the specificity(ies)responsible for the reactivity. cell sample. As antibody investigations become
Some laboratory scientists prefer a similarbut more complex, double-dose exclusion may be-
two-phased cross-out approach. In this ap- come more difficult. The laboratory policy
proach, phase one is to cross out the antigens on should also include any exceptions to the exclu-
each red-cell-sample (panel-donor) row using sion criteria.
the same or similar notation as described above The ethnicity of the donor serving as a panel
for each nonreactive panel cell. Table 13-3 illus- cell source affects antibody exclusions. Panel
trates what an antibody identification panel red cells may appear to be double dose based on
might look like with phase one rule-outs com- phenotype. Yet, for blood group systems having
plete. The second phase of this approach is to a very common silencing allele, the panel cells
apply the laboratory's exclusion policy or criteria may carry only a single dose of the antithetical
to the rule-outs and denote the final composite antigen. Most common is the Fy(a+b-) pheno-
rule-out on the top row of the panel to summa- type on the red cells of a donor ofAfrican ances-
rize the specificities excluded. Table 13-4 de- try. Because of the high frequency of the
picts an antibody identification panel with FY*02N. 01 allele with silenced Fybred cell ex-
phase-two rule-outs complete. In this example, pression in this population, these Fy(a+b-) cells
final composite cross-out (and thus antibody ex- often have only one dose of Fy" antigen. If the
clusion) was indicated with an "X" on the top- Fy(a+b-) sample is also D+C-E-, V+, or Js(a+),
row antigen list after meeting the hypothetical the donor is likely of African ancestry. Exclusion
'" ~'"
-6
b~
"'0,
_g .~
Q) '"
0, en
----_t----+_--~----_t----+_----r_--_t----+_----r_--_+----_r--__i.~ ';:
'"en:;:;go
o o + o + o o + o o o '"c..:c'"'-'
en
C') C') C')
x
Q) t::
._
EJ.?-
~ t'O
-----+----+----1-----+----+-----r----+----+-----r----+----~--__icoe '§~
CD 00 c..:>
«
-
o
CU
Ci
E
~
CD 00
I~ ~ ~ ~ ~ ~ ~ ~ ~
ClJ
w
a.. + + + +
0 0 0 0 0 0 0 0
CD
.c ~ N C") C") C")
=
I-
iii ,- N C") <::t L() co r-, 00 m 0

,-
,-
,-
0
Co) <:(
+
1\ X X \ [X
0 0 + 0 +
~
0
X1\ X \ + 0 0 0 + 0
[>( +
+
X X X X [X
0
0 0
0
+ 0
0
+ 0
0
0
0
+ 0 0
)( + 0 0 +
\ \ + 0 0 0
I~
0 0
\ 0 0 0 0 0 + 0 0
,..
>< I'" I'"
'"'" " I'"[X
a.. + 0 + + 0
>< +
0
~
0
~
0 0
+
~
0
+
+
~
0
[X
0
0
+
~
0
>< +
0
0
V< I'"
0
+
+
0
+
0
0
+
+
~
0
><
0
>< 0
~
V< 0
~
~
0
~
V< +
~
-,
+ + + +
'"
0 0 0
~ ~ ~
+
~
0 +
I'" + [X +
X
~
0 0
[X +
~
+
~ [X +
~
0
0 0
V< 0
~
0 0 0 0 0 0
+ [X 0 + 0 0 0 0 0 0
X
~
+ [X[X 0 0 0 0
~
0 0
X
iii ,- N C") <::t L() c.o r-, 00 m 0 ,- 0
Co) 'r--' -r+ <:(
of anti-Fy' on such a panel red cell would proba- sponding allele through genotyping. As ex-
bly represent only a single-doseexclusion. plained above, ethnicity influences the apparent
For any method of crossing out, final exclu- zygosity of the FY*A allele. Testing at this stage
sion should be done only after ensuring the lab- of the investigation can also reveal errors in the
oratory's policies for eliminating the presence of presumptive identification when the expected
antibodies are met. In most cases, this process positive and negative results in confirmatory
leaves one or more antibodies that have not testing are not obtained.
been excluded. The red cells that are reactive
are then evaluated. If there is an antigen pattern Probability of Accurate Identification
that matches the test reactivity exactly for an an-
tigen not excluded, this most likely identifies Accurate identification of antibody specificity
the specificityof the antibody. Additional testing greatly depends, first, on the antibody having a
may be needed to eliminate remaining specifici- sufficient titer (ie, quantity of circulating anti-
ties that were not excluded and confirm the sus- body) to provide reliably positive reactions. Sec-
pected specificity.This process of selecting red ondly, antigen strength on test cells must be ad-
cells for exclusion and confirmation is de- equate to provide a consistent target antigen. It
scribed in the next section. When additional is difficult to know exactly when both of these
testing is not needed after the initial identifica- criteria are met.Test protocols are designed to
tion panel because an initial specificity can be enhance clinically significant reactivity and di-
assigned and all other specificities can be ex- minish false negatives due to poor antibody af-
cluded, the probability of an accurate identifica- finity, if applicable, and good laboratory practic-
tion can be directly assessed (seebelow). es attempt to minimize antigen deterioration
while optimizingproper staff testing techniques.
Assuming these variables are controlled to the
Selected Red Cells for Exclusion and
greatest degree possible, it is still necessary to
Confirmation
ensure that an observed pattern of reactions is
Selected red cells, chosen for the specific anti- not the result of chance alone. Conclusive anti-
gens they carry or lack, are used to confirm or body identification requires the sample to be
rule out the presence of antibodies. For exam- tested against a sufficient number of reagent red
pIe, if a pattern of reactive red cells fits antt-Ik' cell samples that lack-and express-the anti-
exactly, but anti-K and anti-S were not exclud- gen that corresponds with the antibody's appar-
ed, the serum should be tested with selected red ent specificity. A standard approach (which is
cells. Ideally,red cellswith the followingpheno- based on Fisher's exact method) has been to re-
types should be chosen: Ik]a-), K+, S-; Jk(a-), quire that three antigen-positivered cell samples
K-, S+; and Jk(a+), K-, S-. The reaction pattern are reactive and that three antigen-negative red
with these red cells should both confirm the cell samples are not reactive for each specificity
presence of anti-Ik' and include or exclude anti- identified." When that approach is not possible,
K and anti-SoWhenever possible, selected red a more liberal approach (which is derived from
cells should have a strong expression of the anti- calculations by Harris and Hochman=) allows
gen being tested (te, from donors presumed ho- the minimum requirement for a probability (p)
mozygous for the appropriate gene or red cells value of 0.05 to be met with two reactive and
with double-dose expression). Such red cells three nonreactive red cell samples or with one
help ensure that nonreactivity with the selected reactive and seven nonreactive red cell samples
red cell sample indicates the absence of the anti- (or the reciprocal of either combination). In
body and not that the antibody was too weak to some cases, the use of two reactive and two
be reactive with a selected red cell that had a nonreactive red cell samples is also an accept-
weak expression of the antigen. It must be re- able approach for antibody confirmation.8(p25),33
membered that confirmation of double-dose ex- Additional details on calculating probability may
pression of an antigen can be accomplished only be found in the suggested readings list at the
by demonstrating homozygosity of the corre- end of this chapter.
Consistency of Antibody Identified with tioned in Figs 13-1 and 13-2 as well as others
Autologous Red Cell Phenotype are further described below.
The patient's autologous red cell phenotype is MUltiple Antibodies
used to support the presumptive antibody iden-
tification: the red cells should lack the corre- When a sample contains two or more alloanti-
sponding antigen. The phenotype as determined bodies, it may be difficultto interpret the results
by serology or genotyping may also indicate the of testing performed with a single panel of re-
need for further investigation. For example, if an agent red cells. The presence of multiple anti-
individual appears to have anti-Fy",his or her bodies may be suggested by a variety of test re-
red cells should type Fy(a-). However, if the au- sults, such as the following:
tologous red cells type as Fy(a+) (and have a
negative DAT result), the identification of an 1. The observed pattern of reactive and nonre-
antl-Fy' is in conflict with the phenotype, and active red cells does not ftt a single antibody.
additional testing should be performed. A sero- When the exclusion approach fails to indi-
logically derived antigen-positive typing should cate a specific pattern, it is helpful to deter-
be reconfirmed by testing with more than one mine whether the pattern matches two
antibody source when possible. If genotyping combined specificities. For example, if the
predicted the antigen-positive status, this dis- reactive red cells on the panel in Table 13-2
crepancy may indicate that the patient's red are numbers 3, 5, 6, 9, and 10, none of the
cells do not actually express the antigen because specificitiesremaining after crossingout fits a
of a gene-silencingmutation not targeted by the pattern exactly. However, if both E and Kare
assay. Alternatively, the patient might have an considered together, a pattern is discerned,
altered or partial antigen because of an addition-
with reagent cells 3 and 5 showing reactivity
al gene polymorphism. It is important to remem-
because of anti-B,and reagent cells 6, 9, and
ber that the antibody in the sample could be, in
fact, an alloantibody. 10 because of anti-K. If the reaction pattern
does not fit two combined speciflcittes, the
possibilitythat more than two antibodies are
present must be considered. The more anti-
bodies a sample contains, the more complex
identification and exclusion become, but the
Not all antibody identifications are straightfor- basic process remains the same.
ward. The interpretation process described 2. Reactivity occurs at different test phases.
above does not always lead directly to an an- When tube tests are performed and reactivity
swer, and additional testing and/or consultation occurs at several phases, each phase should
with an immunohematology reference laborato- be analyzed separately. The pattern at room
ry (IRL)may be required. The autologous con-
temperature may indicate a different specific-
trol that is tested with initial antibody identifica-
ity from the pattern at the rAT phase. It is
tion studies provides a starting point for
also helpful to look for variations in the
complex antibody problem resolution. If the test
method does not allow for testing of an autocon- strength of the reactions at each phase of
trol, the DAT result can be used to plan addi- testing. Table 13-5 provides information
tional testing. Figure 13-1 shows some ap- about the characteristic reactivity of many
proaches to identifying antibodies in a variety of antibodies.
situations when the autocontrol is negative, and 3. Unexpected reactions occur when attempts
Fig 13-2 shows some approaches to identifying are made to confirm the specificity of a
antibodies when the autocontrol is positive. suspected single antibody. If a sample sus-
Common types of antibody investigations men- pected to contain anti-e is reactive with some
.0'" :5
11)
ss
'"
:S
en
.='"
"0
'0;
0
0. '"~
0
0
c: '"
en c;,
c-,
c:
'"
:S '"
0>
'"e c: 0
.~
0 .=c: ~ee '"
.s ~ .g
E'" 0
'0;
:::>
'"
:S
~ ~
<5 c
$:
~
'"
:E
"5 17; .,; .s 0
c: ~
0 c: 0.
'" g S c:: 0;
'i' ~
""" .s
"2 o c::
0
'0;
en ~ .s ~ e-
.s s '"
.c:
0.
~ E 0
~
"0
c:: .!!2 en
:;_.
.!d 0.
x 2 '"
c .c
0
'"
-6'
0;
'"
c::
'" '"'"
0;
'"
en '"
0>
'"
....J
"0
~'" -0
en -0
E
0
.c ~ 0.
~ "9 0
'"'"'"e '"'"c:: '"'"c::
~ee '" Vi
'" '" ~
1: zs
:;_. ,,; '"
Qi
-0
c;
~'"
:S
0>
'"
:S
'"
+='
:::>
0 .='" ca'"
>
~
cr::
c: ~ 0.
~ '"
c: .c:;
'" $:
0.
-0
'"
E e >-
ee 0.
Co Vi '" '"
:S
'"
11) "lti
'" .c:
:S
0 f:j'
c: ~ "0
Ol
0
~ :::>
(/) c: .!!2_E t:: '" '" .2 "lti
.1: 0 0 "lti 0; ~ E
-=c: -=Ec '"
0.
_;:::.
~£
E.!!2
'" :S
0. ~ .s .c: '"
0 _;:::.:l!,l
'"'" '"
> 0
c:
'" -
0 0 0.
(/)
c c E 0'" :::> E :::>
'in
,.,0 '" '"
.c:; '@
Q. en '"
c "0
'"
.c:;
'"
e
0-
'c
(/)
0 E
- .!!2 0. X
0;
0. .c:;
c>
0. §
C
ee 0; "0
'" '" -0<.:>
.s
"0 <= 2'
.!!2
.s
c <.:> -0
c: '" '"
"0 0>
c <=
0;
<.) ~
'0 -0
e
'0; e ~a> '" '" ~
s '" ,,; '" e ,,; '" e
0> -0
:::>
(3 -0 '" <='"
en -0
'"
E
x 2' 0
c c c:: .s
'"E 0;
<.)
C:"-:"''''O
'"E.c:;
'" ~
'" '"
e
0>
'" to -'"
<.:> C'" . :f=i
0
'" ~
0.. '" 0
0.. cry '-' ~ '" I-'" '" ::l
0.. +='
0.
en 0
~::c :2; '" '"E
.,..: N .,..: N cci .,..: N cci .,..: N
<I)
,., '"
'0 1:: '"
0>
:I< '"
t::
-0 0 0>
:c: c '" '">
t5 0 '"'" .2 c
C>
+='
0 ..0
~ '" -0 C
.0
'g .so>'"
'0
'"
~ 0>
0
'" '"
«
e.....
0 -»+:
0>
.,
'" >.:g .0
.§:
c: -oc -'
'"
'"
c;, Q.
15
.0 '"
~ ~-6'0
-'" 0
.c:;
0'"
.0
::c
.s
C
'(ij
E
:::> '"
~ ._!!1<.:>
c:
'"'"
Q.
'"'"$: :g
..0
(/) E '"
'"
c >-
c::::
0
to
E
'"., >'" E to '"
>-
"0 '" '"
~~
"0
0
(..)
'"
0.
to
., '" ~
:::>
E '":::>
0. .c
0
o.~ :§ .....
0
<I) 0. 0. ~ <I)
~ "'0. c:
~
::I
~
:::>
C/:J '"
:::>
C/:J
'"
:::>
C/:J
0. 0 C/:J
'"<:;
:::>
C/:J '"
0
~
:;::;
~
0)
Q.)
c::::
.....
-
J::
a>
.~
:::>
Q)
> > 0
c::
(/)
(/)
., 't; ., a>
.0)
>
:S ~ ~ '" -0
> 0
"n e
0>
.C::
'" .c:'"
0)
> .c:
(/)
'" '" "2

Q)
0: U :;::;
~
:e
Q)
'"'" .!:!!~
11)
0:
..!!2 "0
0. 0. >,
::;::
:;;
., '"
Q)
0:
(..)
(i::
.0:
a; c::
'" '"
a>
-0
c::
'" '"
0.
0.
a;
:;::;
c::
.£l
a; 'E c..:>
'"
:S
0:
.£l .c:
'"'" '"
3: (..) Q.)
c..:> '(ij -0
0) 0> a; 0, '"
.£l '0 '" -0 :s!
-0 ",-
0: c:: c..:> c:: a; .c 0
Q)
>,
0:
'" .c:en'"
0: ;;;: ~
.,
-0
~ (..)
+=' Q) ""0
'" '" .....

a>
0: -0 c:: <= 0 0
.0 ......
E
Q.
0 Q. is!
;;;: 'E
a>
0: c:: '" 0
:;::; '03
0 E
en
'" (J)
:::>
E Q)
0
E c
>- c:::: ._
E :2! If.i <::( J::
'"
C/:J 0
en
E E
0 0
0 ......
,.. .2
en (.)
~
?""
J::
.t:
W ""0
a:: II
~
C!I
u::: 0
1=
~
~=m~
00«
1lu (1)
E 2 Ei'!1:I1l
2
Q)
'"
::J ~-g en
iii B lfi ~
C/)
i'! 'e~1--:
:.=0<:(
(5
,_
1;)
Q)
l-
'"a
a. 0'<::0
~ <II ......
c::
a. ~~~ 0
0
j~~
~.- ~
0
......
::l
<II ctl
~ ~
Cl
o OJ
OJ
> >
:;:::; 'iii
'iii ~
o
a. :2
TABLE13-5. Serologic Reactivity of Some Common Blood Group Antibodies
Anti-M IgG>lgM Most Most Rare Sensitive Resistant Rare Rare
Anti-N IgM>lgG Most Most Rare Sensitive Resistant No Rare
Anti-S IgG>lgM Most Most Variable Resistant Yes Yes
Anti-s IgG>lgM Most Variable Resistant Yes Yes
Anti-U IgG Most Resistant Resistant Yes Yes
Anti-Pi IgM Most Most Resistant Resistant No Rare

(lgG rare)
Anti-O IgG>lgM Some Some Most Resistant Resistant Yes Yes

(lgA rare)
Anti-C IgG>lgM Some Some Most Resistant Resistant Yes Yes
Anti-E IgG>lgM Some Some Most Resistant Resistant Yes Yes
Anti-c IgG>lgM Some Some Most Resistant Resistant Yes Yes
Antl-e IgG>lgM Some Some Most Resistant Resistant Yes Yes
Antl-Lu' IgM>lgG Most Most Resistantor Variable No Mild

weakened
Anti-Lu" IgG>lgM Some Most Resistantor Variable No Mild

weakened
Antl-K IgG>lgM Some Most Resistant Sensitive Yes Yes
Anti-k IgG>lgM Most Resistant Sensitive Yes Yes
Antl-Kp' IgG Most Resistant Sensitive Yes Yes
Anti-Kpb IgG>lgM Most Resistant Sensitive Yes Yes
Antl-Js' IgG>lgM Most Resistant Sensitive Yes Yes
Anti-Js" IgG Most Resistant Sensitive Yes Yes
Anti-Lea IgM>lgG Most Most Most Most Resistant Resistant No Rare
(Continued)
TABLE 13-5. Serologic Reactivity of Some Common Blood Group Antibodies (Continued)
Ami-te" IgM>lgG Most Most Most Most Resistant Resistant No No
Anti-Fy" IgG>lgM Most Sensitive Resistant Yes Yes
Anti-Fyb IgG>lgM Most Sensitive Resistant Yes Yes
Antl-Jk' IgG>lgM Most Resistant Resistant Rare Yes
Antl-Jk" IgG>lgM Most Resistant Resistant Rare Yes
Antl-Di" IgG Most Resistant Resistant Yes Rare
Antl-Dl" IgG Most Resistant Resistant Yes Rare
Antl-Yt' IgG Most Variable Sensitive No Rare

or weak-
ened
Anti-Ytb IgG Most Variable Sensitive No No

or weak-
ened
Anti-Xg" IgG>lgM Some Most Sensitive Resistant No No
Anti-Sc1 IgG Most Resistant Variable No No
Anti-Sc2 IgG Most Resistant Variable No No
Anti-Do" IgG Most Resistant Variable No Yes
Anti-Dab IgG Most Resistant Variable No Yes
Anti-Co" IgG>lgM Most Resistant Resistant Yes Yes
Anti-Cob IgG Most Resistant Resistant Yes Yes

OTT = dithiothreitol; HOFN= hemolytic diseaseof the fetus and newborn; HTR = hemolytic transfusion reaction; IAT =
indirect antiglobulintest; Ig = immunoglobulin.
e-negative red cells, another antibody may be 4. A phenotypically similar red cell is nonreec-
present or the suspected antibody may not live. When all or nearly all panel red cells are
be anti-e at all. Testing a panel of selected e- reactive, the easiest way to recognize multi-
negative red cells may help identify an addi- ple antibodies is to test a phenotypically simi-
tional specificity. lar red cell. A phenotypically similar red cell
is one that lacks the same common antigens Optimal Phase or Temperature for Antibody .
as the patient's red cells. Lack of reactivity WasNot Tested
with this. type of red cell indicates that the If weak or questionable positive results are ob-
alloantibodies are directed at common anti- tained at the IATphase, it may be helpful to per-
gens lacking from the test red cell. A selected form tube tests with readings at immediate spin,
red cell panel can then be tested to identify room temperature, or 37 C if these phases were
or exclude the common antibodies to the red not included in the original testing. This may al-
Iowan antibody that optimally reacts as a direct
cell antigens that the patient lacks. (See the
agglutinin at 37 C or below to be more clearly
discussion on selected red cells earlier in this visualized.
chapter.)
Potential Phenotype Exclusions
Reactivity without Apparent
When serum reactivity has no apparent specific-
Specificity ity, a useful approach is to type the patient's red
Zygosity(ie, copy number), variation in antigen cells byserology or genotyping for common red
expression, and other factors may contribute to cell antigens, and eliminate from initial consid-
difficultyin interpreting results of antibody iden- eration specificitiesthat correspond to antigens
tification tests. If the reactivity of the serum is on the patient's autologous red cells. This com-
very weak and/or the pattern of reactivity and bined with other techniques allows the investi-
the cross-out process have excluded all likely gation to be focused on specificitiesmore likely
to be present. Phenotyping may not be possible
specificities, alternative approaches should be
if the patient has received transfusion recently
used. Some helpful techniques and consider-
or has had a positive DATresult.
ations include those described below.
Presenceof Antigens in Common
Alternate TestMethod
Instead of excluding antibodies to antigens on
Depending on the metb.od Wigi.r!allyused, it nonreactive red .cells, dose observation may
may be necessary either to enhance antibody re- identify antigens that thereactive red cells have
activity by using a moresensitive method (eg, in common. For example, if all of the red cells
PEG, enzymes, increased incubation time, or in- reactive at room temperature are PI + .but the
creased serum-to-cellratio; see Methods 3-5 and anti-PI pattern is not complete, the antibody
3-Sthrough 3-13) or to decrease the sensitivity could be anti-PI that is not reactive with red
of the method to avoid the detection of unwant- cells with a weaker expression of the antigen.
ed and. clinically tnsignlftcant reactivity. Meth- (Such red cells are occasionally designated on
ods to inactivate certain antigens on the reagent the panel sheet as "+w.") In this case, it might
red cells may also be helpful. Enzyme treatment be helpful to use a method that enhances anti-
renders red cells negative for such antigens as PI, such as testing after incubation at lower
Ft and Fyb.(SeeTable 13-5.)Observation of the temperatures.
If all of the reactive red cells are Jk(b+) but
effect a reagent red cell treatment has on the un-
not all Jk(b+) red cells are reactive [especially
known serum reactivity can provide clues about
(Jka+Jkb+)cells],the reactive red cells might be
its possible specificity. Adsorption or elution Jlc(a-b+) with a double-dose expression of the
methods to separate antibodies (Methods 3- antigen. In this case, tube-based enhancement
20, 4,1, and 4-2) may also be useful because se- techniques, such as enzymes or PEG, or a differ-
lective adsorption can isolate unknown reactivi- ent test method, such as solid phase, might help
ty, and elution of unknown reactivity from ad- demonstrate reactivity with all of the Jk(b+) red
sorbing red cells can concentrate the antibody. cells. Typing the patient's red cells to confirm
that they lack the corresponding antigen is also ABO Group of Red Cells Tested
helpful.
The sample may be reactive with many or all of
If strongly positive results are obtained, the
the group 0 reagent red cells but not with red
exclusion method should be used with nonreac- cells of the same ABO group as the autologous
tive cells to eliminate specificities from initial red cells. Such a reaction pattern occurs most
consideration. The strongly reactive reagent frequently with anti-H, anti-IH, or anti-Le'",
cells may be examined for any antigen in com- Group 0 and A2red cells have more H antigen
mon. than Al and AlB red cells, which express very
Finally, the presence of some antigens in little H. (See Chapter 10 for more information.)
common may suppress the expression of other Thus, sera containing anti-H or anti-Ill are
antigens. This suppression can cause weak anti- strongly reactive with group 0 reagent red cells,
bodies to be missed or certain red cells to be un- whereas autologous Al or AlB red cells or donor
expectedly nonreactive when a suspected anti- red cells used for crossmatching may be weakly
body fails to show reactivity with all antigen- reactive or nonreactive. Anti-Le'" is strongly re-
positive red cells. For example, In(Lu) is known active with group 0, Le(b+)red cellsbut weakly
to suppress the expression of LV antigens, PI, reactive or nonreactive with Lefb-) red cells
Inb,and AnWj. Similarly, Kp' is known to weak- from Al or AlB individuais.
en the expression of KELantigens. (See Chapter
12 for a more detailed discussion.) Unexpected Reagent Red Cell Problems
Rarely,a pattern of reactive and nonreactive red
Inherent Variability cells cannot be interpreted because the typing
Nebulous reaction patterns that do not appear result for a reagent red cell is incorrect or the re-
to fit any particular specificity are characteris- agent red cell has a positive DAT result. If the
tic of certain antibodies, such as anti-Bg",-Kn", red cell sample is from a commercial source, the
-McCa, -sr, -Yka, -Csa,and -JMH. Antigens cor- manufacturer should be notified immediately of
responding to these antibodies vary markedly the discrepancy.
in their expression on red cells from different
individuals. For example, the expression of Warm Autoantibodies
Knops blood group antigens shows marked The presence of warm-reactive autoantibodies
differences between individuals as a result of in a patient's serum creates a special challenge
variations in the CRI copy numbers on the because the antibody is reactive with virtually
red cells." all red cells tested. The majority of warm auto-
antibodies are IgG; IgM warm autoantibodies
Unlisted Antigens are unusual, but they have caused severe (often
fatal) autoimmune hemolytic anemia." If a pa-
A sample may react with an antigen that is not
tient with warm autoantibodies requires a trans-
routinely listed on the antigen profile supplied fusion, it is important to detect any underlying
by the reagent manufacturer-Do a, Dob,and Ytb clinically significant alloantibodies. Solid-phase
are some examples. Even though serum studies and gel-column methods frequently greatly en-
yield clearly reactive and nonreactive test re- hance warm autoantibodies. PEG, enzymes,
sults, such antibodies may not be recognized. In and, to a lesser extent, LISSalso may enhance
these Circumstances,it is useful to review addi- these autoantibodies. It is often helpful to omit
tional phenotype information supplied with the the additive solutions when testing sera that
reagent panel or consult the manufacturer. If contain warm autoantibodies. If such tests are
only one cell is unexpectedly reactive, this reac- nonreactive, common alloantibody specificities
tion is most likely caused by an antibody to a can be excluded and the same procedure can be
low-prevalence antigen. These antibodies are used for compatibility testing without the need
discussed in more detail later in this chapter. for adsorptions. If such tests remain reactive, ad-
sorptions are typically required to rule out un- The use of the last two. procedures listed is
derlying alloantibodies. For more information, controversial for the purposes of circumventing
see Chapter 14 and Methods 3-20 and 4-8 cold autoantibodies. Notes on and limitations of
through 4-10. the procedures can be found within their reo
spective method descriptions or references.
Cold Autoimtibodies In some situations, the goal of testing is not
to circumvent the cold autoantibody but rather
Cold autoantibodies may be clinicallybenign or
to define its serologic characteristics (eg, speci-
pathologic. In either case, potent cold autoanti-
bodies that are reactive with all red cells at room ficity, thermal amplitude,· titer). This may be reo
temperature or below, including the patient's quested and useful if the patient's clinical
own, can create special problems-especially if situation is suggestive of a pathologic cold auto-
the reactivity persists at temperatures above agglutinin. See Chapter 14 for a more detailed
room temperature and into the IATphase of an- discussion of immune-mediated hemolysis
tibody identification. Such situations make it dif- caused by some cold autoantibodies.
ficult to detect and identify potential clinically
significant alloantibodies that are being masked Delayed Serologic/Hemolytic::
by the cold autoantibody reactivity. The detec- Transfusion Readions
tion of coldautoantibodies Can be dependent on Delayed transfusion reactions are defined by the
the test method used. Gel-columntests can give development of a new alloantibodyin a patient
a mixed-fieldappearance even though only one following transfusion that results in laboratory
cell population is present. Solid-phasetests are evidence (serologic) or laboratory and clinical
designed to minimize their detection. There are evidence [hemolytic) of the destruction of in-
different approaches to testing sera with potent compatible transfused red cells that were com-
cold autoagglutinins. Once the presence of a
patible at the time of infusion. If a patient has re-
cold autoantibody has been confirmed, the goal
ceived transfusion in the last 3 months and the
in most situations is to circumvent or remove
autocontrol is positive in the IAT phase, there
the interfering cold autoagglutinin reactivity in
may be antibody-coated donor red cells in the
order to detect underlying and potentially clini-
patient's circulation, resulting in a positive DAT
cally significant antibodies. Procedures to ac-
complish this include the following: result that can show mixed-field reactivity. An
elution should be performed, especially when
1. Omitting the room-temperature and/or tests on plasma or serum are inconclusive. For
example, a recent transfusion recipient may
immediate-spin phase of testing using test
have a positive autocontrol and show weak re-
tube methods if one was performed.
activity with most, but not all, Fy(a+} red cells.
2. The use of anti-IgG rather than polyspecific
It may be possible to confirm antl-Fy"specificity
AHG reagent for the IAT phase of antibody
in an eluate because more antibody is often
identificationwhen using test tube methods. bound to donor red cells, and importantly, the
3. Cold auto- or allogeneic adsorption of the preparation of an eluate concentrates the anti-
patient's serum or plasma to remove autoan- body. It is rare for transfused red cells to make
tibodies but not alloantibodies (Method 4-5 the autocontrol positive at a phase other than
and 3-20). IAT,but this can occur, especiallywith a newly
4. Prewarming techniques in which reagent red developing or cold-reacting alloantibody. If the
cells and patient serum or plasma are pre- DAT result does have a mixed-field appearance
warmed to 37 C separately before they are and the plasma or serum is reactive with all cells
combined (Method 3-6). tested, a transfusion reaction caused by an allo-
5. Adsorption with rabbit erythrocytes or rabbit antibodyto a high-prevalenceantigen should be
erythrocyte stroma.36,37 considered (Fig13-2).
Antibodies to High-Prevalence Ancestry of Antibody Maker

Antigens
Antibodies such as anti-U, -Mcca, -Sl",_ISb, -Hy,
If all reagent red cells are reactive in the same -Io", -Tc",-Cr",and -At' should be considered if
test phase and with uniform strength but the au- the sample is from an individual of African an-
tocontrol is nonreactive, an alloantibody to a cestry because the antigen-negative phenotypes
high-prevalence antigen should be considered. occur almost exclusivelyin this population. Indi-
Antibodies to high-prevalence antigens can be viduals with anti-Kp"are almost always of Euro-
identified by testing selected red cells of rare pean ancestry. Anti-Dib is usually found among
populations of Asian, South American Indian,
phenotypes and by typing the patient's autolo-
and Native American ancestry. Other examples
gous red cells with antisera to high-prevalence
are found in Table 13-6.
antigens. Knowing the ethnic or ancestral origin
of the antibody producer can be helpful when Serologic Clues
selectingadditionaltests to perform (Table13-6).5
Chemically modified and/or enzyme-modified Knowing the serologic characteristics of particu-
red cells (eg, DTT/ AET-treated or ficin-treated lar antibodies to high-prevalence antigens may
red cells) can give characteristic reactivity pat- help with identification.
terns that help limit possible specificities (Table
1. Reactivity in tests at room temperature
13-7). Testing rare red cellsthat lack all antigens
suggests anti-H, -I, -IH, -P, -PP1pk (-Tja),
in a blood group system leg, Ko,Rhnull, or Lu(a-
-Ena,·LW(some), -Ge (some), -sr, or -Vel.
b-) cells]can localize the reactivity to that blood
2. Lysisof reagent red cells during testing with
group system if nonreactive.
If red cells negative for a particular high- fresh serum is characteristic of anti-Vel, -P,
prevalence antigen are not available, red cells -PP1pk (-Tja),-Ik3, and some examples of
that are positive for the lower prevalence anti- anti-H and -1. Serum instead of plasma must
thetical antigen can sometimes be helpful. For be used in tests to see lysis.
example, if a sample contains anti-Co",weaker 3. Reduced or absent reactivity with enzyme-
reactions may be observed with Cola-b-r] red treated red cells occurs with anti-Ch , -Rg,
cells than with Co(a+b-) red cells because of a -En', -Inb, -JMH,-Ge2, and some examples of
dosage effect. anti-Yt',
Antibodies to high-prevalence antigens may 4. Weak nebulous reactions in the IAT phase
be accompanied by antibodies to common anti- are often associated with anti-Kn", -Mcca,
gens, which can make identification much more -Yk",and -Cs", KN system antigens are labile
difficult. In such cases, it may be necessary to during storage: antibodies may be more reac-
determine the patient's phenotype for common tive with donor red cells and fresher reagent
antigens, choose a phenotypically similar red red cells.
cell sample (ie, one that lacks the same common 5. Complement-binding autoantibodies, such as
antigens as the patient's red cells) that is incom- anti-I and -IH, or alloantibodies, such as anti-
patible with the patient's serum, and adsorb the PP1pk and -Vel, may give stronger results
antibody to the high-prevalence antigen onto when a polyspecificAHG reagent is used.
that red cell sample. This approach leaves anti-
bodies to common red cell antigens in the ad- Antibody to High-Prevalence Antigen vs
sorbed plasma or serum, where they can be Warm Autoantibody
identified with a routine selected red cell panel. When a patient produces an antibody to a high-
Because the identification of antibodies to high- prevalence antigen after transfusion, the pa-
prevalence antigens is complicated, it may be tient's posttransfusion red cell sample may have
necessary to refer such specimens to an IRL. a positive DATresult, and both serum or plasma
TABLE13·6. High-Prevalence Antigens Absent in Certain Populations*
AnWj- Transientin any populationslsraellArabs(inheritedtype)
At(a-) Blacks
Cr(a-) Blacks
Di(b-) SouthAmerlcane-NatlveAmerlcanss.lapanese
Fy(a-b-) Blacks>Arabs/Jews>Mediterraneans>Whites
Ge:-2,-3 (Gerbichphenotype) PapuaNewGulneane-Meanesiansswhltessany
Ge:-2,3 (Yus phenotype) Mexicans>lsraelis>Mediterraneans>any
Ge:-2,-3,-4 (Leachphenotype) Any
Gy(a-) EasternEuropeans(Hornanyc-Japanese
hrB- Blacks
hrs- Blacks
Hy- Blacks
In(b-) lnolane-lranlanssArabs
Jk(a-b-) Pnlyneslanssflnnsxapanese-any
Jo(a-) Blacks
Jr(a-) Japanese-Astana-turopeans-aenouln Arabs-any
Js(b-) Blacks
k- Whites>any
Kn(a-) Whites>Blacks>any
Kp(b-) Whites>Japanese
Lan- Whites>Japanese>Blacks>any
Lu(a-b-) Any
LW(a-b-) Transientin anysinherltedtype in Canadians

(Continued)
TABLE 13-6. High-Prevalence Antigens Absent in Certain Populations* (Continued)
LW(a-) Baits
O, (Bombay) Indlans-Japanese-any
Ok(a-) Japanese
P- JapanesesHnnsslsraellssany
PP1px; Swedes>Amish>lsraelis>Japanese>any
SI(a-) Blacks> Wh itessany
Tc(a-b+c-) Blacks
u- and s-s-u+var Blacks
Vel- swedes-any
WES(b-) FlnnssBlacke-any
Yk(a-) Whites>Blacks>any
Yt(a-) Arabe-Jewssany
* Adapted with permission from Reid et al.5
and the eluate may be reactive with all reagent significantlyweaker than the serum or plasma
red cells tested. Because this pattern of reactivi- reactivity would be more characteristic of an al-
ty is identical to that of many warm-reacting loantibody to a high-prevalence antigen than a
autoantibodies that appear after transfusion, the warm autoantibody, because only the transfused
two scenarios can be very difficultto differenti- cells are coated with the alloantibody.The DAT
ate. A posttransfusion DAT result that is result in a posttransfusion sample containing a
TABLE 13-7. Alterations of Antigens by Various Agents*
Antigens Usually
Proteolytic enzymes' M, N, S, Fya,Fyb,Yta, Ch, Rg, Pr, Tn, Mg, Mi"Nw, Cia,Jea, Ny", JMH,
Xga, some Ge, and Inb
Oithiothreitol (On) or Yta; JMH; Kna; McCa; Yka; LWa; LWb; all KEL, LU, ~O, CROM, and IN
2-aminoethylisothiouronium blood group antigens
bromide (AET)
* Appropriate controls should be used with modified red cells.

tsome antigens listed may be weakenedrather than completely denatured.
*Different proteolytic enzymes may havedifferent effects on certain antigens.
C HAP T E R 1 3 Identification ofAntibodies to Red CellAntigens 413
new alloantibody to a high-prevalence antigen If an antibody to a low-prevalence antigen is

would be expected to give a mixed-field appear- suspected and all common alloantibody specific-
ance (ie, some red cells agglutinated among ities have been excluded, transfusion should not
many unagglutinated red cells), again because be delayed if identification studies are per-
only the transfused red cells would be coated formed. Because antisera to type donor units for
with antibody. In practice, however, weak sensi- low-prevalence antigens are rarely available, it is
tization and mixed-field agglutination can be usually necessary to rely on the crossmatch to
difficultto differentiate. If a pretransfusion speci- avoid transfusion of antigen-positive units.
men is not available, it may be helpful to per- When the serum is reactive with only one donor
form a red cell genotype or to use red cell sepa- unit or reagent red cell sample, the most likely
ration procedures to isolate autologous red cells cause is an antibody to a low-prevalence anti-
for testing. Performing a DATon autologous red gen; however, some other possible explanations
cells, testing the posttransfusion sample or the are that the red cells may be ABO incompatible,
eluate with DAT-negative autologous red cells, have a positive DAT result, or are polyagglutin-
or both may help distinguish an autoantibody able (te, red cells that have cryptic antigens ex-
from an alloantibody. If a DATresult from autol- posed and react with all normal adult serum).
ogous red cells is negative, the reactivity is con-
sistent with an alloantibody. If the posttransfu- Antibodies to Low-Prevalence Antigens
sion serum or plasma is reactive with DAT- in Pregnancy
negative autologous red cells, the reactivity is An antibody to a low-prevalence antigen may
consistent with an autoantibody. (See Chapter also be suspected when a maternal antibody
14 and Fig 13-2.) screen is nonreactive but her ABO-compatible
newborn has a positive DATresult and/or unex-
Antibodies to Low-PrevalenceAntigens plained decrease in red cell survival. A positive
If a sample is reactive only with a single donor result when testing the mother's serum or plas-
or reagent red cell sample after alloantibody ex- ma, or an eluate from the infant's DAT+ red
clusions are complete, an antibody to a low- cells, against the father's red cells can implicate
prevalence antigen should be suspected. To an antibody to a low-prevalence antigen as the
identify such an antibody, a panel of reagent red probable cause, even if .the specificity is un-
cells that express low-prevalence antigens can known. This testing can be performed only if
be tested with the serum. Alternatively, the one the mother's sample is ABO compatible with the
reactive red cell sample can be tested with father's red cells or if the eluate from the infant's
known antibodies to low-prevalence antigens. red cells does not contain anti-A or -B that
Unfortunately, sera that contain antibodies to would react with his red cells, or if the ABO an-
low-prevalence antigens often contain multiple tibodies are removed from the serum or eluate
antibodies to low-prevalence antigens. Although by adsorption.
low-prevalence antigens are rare by definition,
naturally occurring antibodies that recognize Drug-Dependent Antibodies
some of them are much less rare. Manyantibod- Certain drugs induce the formation of antibod-
ies to low-prevalence antigens are reactive only ies in some patients. These drug-dependent anti-
at temperatures below 37 C and therefore have bodies may cause positive antibody detection/
doubtful clinical significance. To confirm the identification tests typically at the IAT phase
suspected specificities, one may need the exper- and/or a positive DAT result. Actual immune
tise and resources of an IRL.Some IRLs,howev- hemolytic anemia caused by drugs is a rare
er, do not attempt to identify antibodies to low- event, with an estimated incidence of about one
prevalence antigens because many of these anti- in a million." Prompt correlation between clini-
bodies are not clinically meaningful, and com- cal course, drug history, and serologic findings
patible units are readily available. gives opportunity for timely recognition of the
event and provision of potentially lifesaving in- a stack of coins. It may be difficultto detect anti-
formation to the patient's clinician. When antibodies by direct agglutination in a test plasma
bodies to a drug or drug/red cell membrane that contalns rouleaux-producing proteins. Rou-
complexes are detected in routine serology, ad- leaux formation itself is not observed in the IAT
ditional and sometimes complex testing may be phase of testing because the washing steps re-
needed to rule out the presence of alloantibod- move the majority of implicated plasma pro-
ies and exclude the possibility of a transfusion teins. Patient plasma samples exhibiting rou-
reaction (delayed or serologic) occurring in the leaux are, however, prone to incomplete
patient. Testing for drug-dependent antibodies washing at the IAT phase and potentially false-
and information on drug-induced immune he- negative results. Fortunately, such false-negative
molytic anemia may be found in Chapter 14. results caused by incomplete washing should
easily be recognized by the failed antihuman
Antibodies to Reagent Components globulin control cell step of the lAT. The saline
Antibodies to a variety of drugs and chemicals in replacement technique can be used to detect di-
testing reagents can cause positive results in an- rect-agglutinating antibodies in the presence of
tibody detection and identification tests. The of- rouleaux and confirm the suspected reactivity to
fending component may be found in the sus- be rouleaux if it is dispersed by the procedure
pending media of the reagent red cells or maybe (Method 3-7).
a constituent of the antibody enhancement me-
dium that is added to the test system. Most of Other Anomalous Serologic Reactions
these anomalous reactions are in-vitro phenom- Antibodies that react only with red cells freshly
ena and have no clinical significancein transfu- washed in saline, red cells that are aged (in vitro
sion therapy, other than causing laboratory or in vivo), and red cells that have been stored
problems that delay transfusions. A systematic in some plastic containers, among others, have
comparison of the sample's reactivity with cells also been described. These types of anomalous
or enhancement media sourced from different reactions are less frequently encountered but
manufacturers, with washed red cells vs cells in
are entertained as possibilitiesafter close scruti-
original diluent, or with red cells from commer-
ny of unexplained reactivity. Additional informa-
cial sources vs donor blood may identify the of-
tion can be found in the suggested reading by
fending component. For a more complete dis-
Garratty.
cussion, see the suggested reading by Garratty at
the end of this chapter as well as listed referenc-
es.39-41
T PR OUR
Rouleaux Although the same method used for antibody
Rouleaux formation is one of the most common- detection tests is routinely used for basic anti-
ly encountered anomalous serologicreactions. It body identification, alternative techniques and
is an in-vitro phenomenon that is produced by methods may be needed to resolve complex an-
abnormal patient serum/plasma protein concen- tibody identification problems. Some of the pro-
trations. Problems with rouleaux are more prev- cedures described in this section are used rou-
alent when using plasma, which has become tinely by many laboratories; others are used
the sample of choice for blood bank testing often selectivelyand may apply only in special circum-
due to automation requirements. Rouleaux for- stances. It is important to remember that no sin-
mation is caused by aggregates of red cells that gle method is optimal for detecting all antibod-
can be mistaken for agglutination upon macro- ies. When routine methods fail to indicate
scopic examination. It can occur in any test that specificity,or the presence of an antibody is sus-
contains patient plasma and reagent red cells at pected but cannot be confirmed, the use of oth-
the time of reading. If viewed microscopically, er enhancement techniques or procedures may
rouleaux red cell aggregates will often look like be helpful. Techniques involving enzyme treat-
ment of red cells, testing at lower temperatures, However, it is often possible to type antibody-
or testing with various additive solutions coated red cells with direct-agglutinating anti-
should, whenever possible, include an autocon- sera, such as IgM monoclonal reagents. With
trol to ensure proper interpretation of results. rare exceptions, most direct-agglutinating mono-
clonal reagents give valid phenotyping results
Obtaining Autologous Red Cell despite a positive DAT result." Common tech-
Phenotype niques for removing IgG antibodies, when need-
ed, include gentle heat elution (Method 2-19),
It may be difficult to determine the patient's treatment with chloroquine diphosphate (Meth-
phenotype if the individual received transfusion od 2-20), and treatment with acid glycine/
in the past 3 months. A pretransfusion speci- EDTA (Method 2-21).
men, if available, should be used to determine
the phenotype. If a pretransfusion sample is not
I..ISSandPEG
available, the patient's newly formed autologous
red cells can be separated from the transfused PEG techniques are used to enhance reactivity
red cells and then typed (Metl1qd2-?2). Separa- and reduce incubation time compared to testing
tion of young red cellsby centrifuga~oBis bas.ed in the absence of an additive solution. LISSal-
on the difference in the densities of new and lows for reduced incubation time and may be
mature red cells. Separation is most successful used to suspend test red cells for use in tube or
when 3 days have elapsed since the last transfu- column agglutination tests.or as an additive me-
sion, which will provide time for new autolo- dium for tube or solid-phasetests. Commercially
gous red cell production. New autologous red prepared LISS additives or PEG additives may
cells must be isolated from the sample while it is contain additional enhancing agents. Care
fresh. The technique is ineffectiveand can often should be taken to closely follow the instruc-
result in false-positivetyping if the sample is too tions in the manufacturer's product insert to en-
old (>24 hours) or the patient is not producing sure that the appropriate proportion of serum to
new red cells. LISSor PEG is achieved. Generic LISSand PEG
Sickle cells are quite dense, making centrifu- procedures, as well as the principles and special
gation an ineffective technique for separating considerations for each technique, can be found
the autologous red cells from the transfused do- in Methods 3-4 and 3-5. Because LISSand PEG
nor red cells in a patient with sickle cell disease. enhance .autoantibodies, their use may compli-
Autologous sickle cells may be separated from cate alloantibody identification in samples that
donor red cells using washes with hypotonic sa- also contain autoantibodies.v'"
line (Method 2-23). Sickle cells containing he-
moglobin SS are resistant to lysis by hypotonic Temperature Reduction
saline, whereas donor red cells containing he-
Some antibodies (eg, anti-M, -N, -PI, -Lea,-Leb,
moglobin M are lysed.
and -AI) react better at room temperature or be-
Cold and warm autoantibodies may also
low, and their specificitymay be apparent only
complicate antigen typing because of the immu-
at a temperature <22 C. An autocontrol is espe-
noglobulins coating the patient's red cells. It
ciallyimportant for tests at low temperatures be-
may be possible to remove cold autoantibodies
cause many sera also contain autoanti-I or other
with warm (37 C) saline washes (Method 2-17).
cold-reactiveautoantibodies.
If the cold autoantibodies are very potent, it may
be necessary to treat the red cells with O.OIM
Increased Serum-to-Cell Ratio
dithiothreitol (DTT)to dissociateIgM molecules
that cause spontaneous agglutination (Method Increasing the volume of serum incubated with
2-18). When red cells are coated with IgG auto- a standard volume of red cells may enhance the
antibodies, it is not possible to perform antigen reactivity of antibodies that are present in low
typing with reagents that require an IAT (eg, concentrations. One acceptable procedure in-
Fy",.Fyb) without first removing the bound IgG, volves mixing 4 volumes (drops) of serum or
plasma with 1 volume of a 2% to 5% saline sus- volume of 0.1 N HCI to 9 volumes of serum or
pension of red cells and incubating the mixture plasma lowers the pH to approximately 6.5. The
for 60 minutes at 37 C. Periodic mixing during acidified serum should be tested with known
the incubation promotes contact between the M-negative red cells to control for nonspecific
red cells and the antibodies. It is helpful to agglutination.
remove the serum before washing the cells for Lowering the pH, however, Significantlyde-
an IATbecause the standard three to four wash- creases the reactivity of other antibodtes." If un-
es may be insufficient to remove all of the unbuffered saline with a pH <6.0 is used to pre-
bound immunoglobulin if increased amounts of pare red cell suspensions or for washing in an
serum or plasma are used. More than four wash- IAT, antibodies in the RH, FY, JK, and MNS
es are not recommended because bound anti- blood groups may lose reactivity. Phosphate-
body molecules may dissociate. Increasing the buffered saline (Method 1-8) can be used to con-
serum-to-red-cellratio is not appropriate for tests trol the pH and enhance the detection of anti-
using LISSor commercial PEG, which may be bodies that are poorly reactive at a lower pH.47
manufactured by dissolving the PEG in LISS.
Tests performed in a low-ionic-strength medium Enzyme Modification/Destruction of
require specific proportions of serum or plasma Blood Group Antigens on Red Cells
and additive.
Ficin and papain are the most frequently used
Increased Incubation Time enzymes for complex antibody identification.
Theydestroyor weaken antigenssuch asM, N, S,
For some antibodies, the routine incubation pe- Ft, Fyb,JMH, Ch, Rg, and xga (Table 13-7).
riod (typically 10 to 15 minutes minimum for Antibodies to these antigens are nonreactive
some additive solutions, and 30 minutes for with treated red cells. Conversely, ficin-and pa-
tests with no additive solutions) may not be suf- pain-treated red cells show persistent or en-
ficient to achieve maximum antibody binding; hanced reactivity with other antibodies (eg,
therefore, the reactions may be negative or those to antigens of the RH, PIPK, I, JK, and LE
weak, particularly in saline or albumin media. systems). For this reason, enzyme techniques
Extending the incubation time to between 30 may be used to separate mixtures of antibodies.
and 60 minutes for albumin or saline tests often For example, if a serum sample contains anti-Py'
improves the reactivity and helps clarifythe pat- and anti-Ik', many of the red cell samples on the
tern of reactions. Extended incubation may be initial panel would be reactive. If a panel of en-
contraindicated when LISS or PEG is used. If zyme-treated red cells were tested, the anti-lk'
the incubation period exceeds the recommend- reactivity would persist and perhaps increase in
ed times for these methods, the reactivity may reactivity, whereas the anti-Py' reactivity would
be diminished or lost. Care must be taken to use no longer be detected because its target antigen
all reagents according to the manufacturers' di- was destroyed by the enzyme treatment. Proce-
rections. dures for the preparation and use of proteolytic
enzymes are given in Methods 3-8 to 3-13.
Alteration in pH Additional enzymes that are commonly used
Altering the pH of the test system can change in advanced IRLsinclude trypsin, a-chymotryp-
the reactivity of certain antibodies, enhancing sin, and pronase. Depending on the enzyme and
the reactivity of some and decreasing that of method used, other antigens may be altered or
others. destroyed. Antigens that are inactivated by one
Some examples of anti-M are enhanced proteolytic enzyme may not be inactivated
when the pH of the test system is lowered to by other enzymes. Trypsin treatment has been
6.5.45 If anti-M is suspected because the only reused to remove CD38 from red cells, thereby
active red cells are M+N-, a definitive pattern avoiding the interference of anti-CD38 immuno-
(ie, reactivity with M+N+ red cells also) may be therapy." The clinical significance of antibodies
seen if the serum is acidified. The addition of 1 that are reactive only with enzyme-treated cells
is questionable; such "enzyme-only" antibodies help with the identification of the specificity of
may not have clinical significance." the antibody. For example, if a suspected anti-PI
does not produce a definitive pattern of aggluti-
Chemical Modification/De$truction of nation, tlle loss 9f reactivity after .the addition of
Blood Group Antigens on Red Cells soluble PI substance strongly suggests that the
specificityiStillti-Pl if a parallel dilution control
Certain blood group antigens can be destroyed
with saline remains reactive. Inhibition results
or weakened by chemical treatment of the cells
(Table 13-7).Modified red cells can be useful for can be interpreted only when the test is nonre-
both confirming the presence of suspected anti- active and the dilution control that substitutes
bodies and detecting additional antibodies. The an equal volume of saline for the soluble sub-
use of modified red cells can be especially stance is reactive.
helpful if a sample contains an antibody to a The most commonly Used substances for in-
high-prevalence antigen because antigen- hibition include the following:
negative red cells are rare. Sulfhydryl reagents
such as 2-aminoethylisothiouronium bromide 1. LE (Lewis) substances. Lea substances, Leb
(AET), 2-mercaptoethanol (2-ME), or DTT substances, or both are present in the saliva
cleave disulfide bonds that are responsible for of individuals who possess the LE gene
the conformation of certain blood group anti- (FUT3). Leasubstance ispresent in the saliva
gens and therefore canbe used to weaken or de- of Le(a+b-) individuals, and both Leaand Leb
stroy antigens in the KELsystem and some oth- substances are present in the saliva of Le(a-
er antigens (Method 3_18).50,51 OTT treatment b+) individuals (Method 2-8). Commercially
will also destroy C038 on red cells and has prepared LEsubstance is available.
commonly been used to mitigate the interfer- 2. PI substance. Soluble PI substance is pres-
ence of anti-CD38 immunotherapy on serology
ent in hy~atid cyst fluid and the ovalbumin
testing.22,23ZZAP reagent, which contains both
a proteolytic enzyme and OTT, denatures anti- of pigeon eggs. Commercially prepared PI
gens that are sensitive to OTT (eg, all KELsys- substance is available.
tem antigens) as well as antigens that are sensi- 3. Sd' substance. Soluble Sd" blood group-
tiveto enzymes (Method 4-8).52Glycine-HClI substance is present in various body fluids,
EOTA treatment of red cells destroys Bg and but urine has the highest concentration of
KELsystem antigens as well as the Eraantigen Sda.55To confirm the presence of anti-Sd' in a
(Methods 2-21 and 4-2).53Chloroquine diphos- serum sample, urine from a known Sd(a+)
phate can be used to weaken the expression of individual (or a pool of urine specimens) can
Class I HLA antigens (Bg antigens) on red be used to inhibit the antibody reactivity
cells." Chloroquine treatrri.ent also weakens
(Method 3-19).
some other antigens, including Rh antigens 4. CHIRG [Chido and Rodgers) substances.
(Method 2-20).
CH/RG antigens are epitopes on the fourth
Inhibition Techniques component of human complement (<,:4).56,57
Most normal red cellshave a trace amount of
Soluble forms of some blood group antigens ex- C4 on their surface. Antl-Ch and anti-Rg are
ist in body fluids, such as saliva, urine, and plas- reactive with this C4 in an lAT. A useful test
ma. These substances are also present in other
to identify anti-Ch and anti-Rgis inhibition of
natural sources, or they can be prepared syn-
thetically. Soluble substance can be used to the antibodies with plasma from Ch+, Rg+
inhibit the reactivity of the corresponding individuals (Method 3-17). Although not.an
antibody that could mask the presence of under- inhibition technique, soluble Ch and Rg sub-
lying nonneutralizable antibodies. Also, inhibi- stance in plasma can also be used to coat red
tion of the reactivity by a soluble substance can cells in vitro with excess C4d. Such coated
red cells will directly agglutinate antl-Ch and Adsorption techniques are useful for the fol-
-Rg,allowing for their rapid identification/" lowing purposes:
Denaturation of Immunoglobulins 1. Separating multiple antibodies present in a

single serum.
Sulfhydrylreagents, such as DTT and 2-ME, can
2. Removing autoantibody to permit the detec-
also be used to cleave the disulfide bonds that
join the monomeric subunits of the IgM pen- tion or identification of underlying alloanti-
tamer. Intact 19S IgM molecules are cleaved bodies. (See Chapter 14 for more Informa-
into 7S immunoglobulin subunits, which have tion.)
altered serologic reactivity." The interchaln 3. Removing unwanted antibody (often anti-A,
bonds of 7S immunoglobulin monomers are rel- anti-B, or both) from serum that contains an
ativelyresistant to such cleavage. antibody suitable for reagent use.
Uses of sulfhydryl reagents to denature im- 4. Confirming the presence of specific antigens
munoglobulins include the following: on red cells by their ability to remove anti-
body of corresponding specificityfrom previ-
1. Determining the immunoglobulin class of an ously characterized serum.
antibody (Method 3-16). In a pregnant S. Confirming the specificity of an antibody by
woman's sample, IgG antibody indicates the showing that it can be adsorbed onto red
potential for HDFN. cells of only a particular blood group pheno-
2. Identifying antibodies in a mixture of IgM type.
and IgG antibodies, particularly when an
agglutinating IgM antibody masks the pres- Adsorption serves different purposes in differ-
ence of IgG antibodies. It is important to ent situations; no single procedure is satisfactory
note that when treated plasma is used for for all purposes (Methods 4-5, 4-8,4-9, and 4-
IgG antibody identification, the effect of the 10). A basic procedure for antibody adsorption
dilution caused by the treatment should be can be found in Method 3-20. The usual serum!
plasma-to-cellratio is 1 volume of serum or plas-
considered.
ma to an equal volume of washed red cell blood
3. Determining the relative amounts of IgG and component. To enhance antibody removal, a
IgM components of a given specificity (eg, larger volume of red cells increases the propor-
anti-Aor -B). tion of antigen. The incubation temperature
4. Dispersing red cell agglutinates caused by should be that at which the antibody is optimal-
IgM autoantibodies (Method 2-18). ly reactive. Pretreating red cells with a proteo-
S. RemovingIgG antibodies from red cells using lytic enzyme may enhance antibody uptake and
a mixture of DTT and proteolytic enzyme reduce the number of adsorptions required to
(llAP reagent) (Method 4-8). remove an antibody completely. Because en-
zymes destroy some antigens, antibodies direct-
Adsorption ed agalnst those antigens are not removed by
enzyme-treated red cells. To ensure that an ad-
Antibody can be removed from a serum sample sorption process is complete (ie, that no unad-
by adsorption onto red cells that express the cor- sorbed antibody remains), it is essential to con-
responding antigen. After the antibody attaches firm that the adsorbed serum is nonreactive
to the membrane-bound antigens, the antibody with a sample of the adsorbingred cells that was
remains attached to the red cells when serum/ not used for adsorption. Adsorption requires a
plasma and cells are separated. It may be possi- substantial volume of red cells, and vials of re-
ble to harvest the bound antibody by elution or agent 3% to 4% red cell suspensions are usually
examine the adsorbed serum or plasma for anti- not sufficient. Blood samples from donor units
body(ies)remaining after the adsorptionprocess. are the most convenient sources.
When separating mixtures of antibodies, the 3. Preparation of antibody-free red cells for
selection of red cells of the appropriate pheno- autologous adsorption studies (Methods 4-5
type is extremely important. If one or more anti- and 4-8).
bodies have been previously identified, red cells
that express the corresponding antigens can be Technical factors that influence the outcome
used to remove the known antibodies. For ex- of elution procedures include the following:
ample, if a person who types K+k-, Pyta-b-)
has produced antl-k, it may be necessary to ad- 1. Incomplete washing. Sensitized red cells
sorb the anti-k onto K-k+, Fy(a-b+) red cells to should be thoroughly washed before an elu-
remove the anti-k. Then, the adsorbed sample tion to prevent contamination of the eluate
can be tested with common K-k+, Fy(a+b-) red with unbound residual antibody. Six washes
cells to detect or exclude anti-Ft. with saline are usually adequate, but more
washes may be needed if the serum contains
Elution a high-titer antibody. (The considerations in
Elution dissociates antibodies from sensitized item 3 below should be kept in mind.) To
red cells. Bound antibody may be released by confirm the efficacy of the washing process,
changing the thermodynamics of antigen- supernatant fluid from the final wash should
antibody reactions, neutralizing or reversing be tested for antibody activity and found to
forces of attraction that hold antigen-antibody be nonreactive.
complexes together, or disturbing the structure 2. Binding of protein to glass surjaces. If an
of the antigen-antibody binding site. The usual eluate is prepared in the test tube that was
objective is to recover bound antibody in a us- used during the sensitization or washing
able form. phases, antibody that nonspecificallybinds to
Selected elution procedures are given in
the test tube surface may dissociate during
Methods 4-1 through 4-4. No single method is
the elution. Similar binding can also occur
best for all situations. Heat or freeze-thaw elu-
from a whole blood sample when a patient
tion techniques are usually restricted to the in-
vestigation of HDFN caused by ABO incompati- has a positive DAT result and has free anti-
bility because these elution procedures rarely body in the serum. To avoid such contamina-
work well for other antibodies. Acid or organic tion, red cells used to prepare an eluate
solvent methods are used for eluting warm- should be transferred to a clean test tube
reactive auto- and alloantibodies. Commercial before washing and then to another clean
kits are available for performing elution. (See tube before the elution procedure is initiated.
Chapter 14, Table 14-2 for a list of elution meth- 3. Dissociation of antibody before elution. IgM
ods and their uses, advantages, and disadvantag antibodies, such as anti-A or anti-M, or low-
es.) affinity IgG may spontaneously dissociate
Elution techniques are useful for the follow- from the red cells during the wash phase. To
ing: minimize the loss of bound antibody, cold
(4 C) saline or wash solution provided by the
1. Investigation of a positive DATresult (Chap- manufacturer should be used for washing.
ter 14). 4. Incorrect technique. Such factors as incom-
2. Concentration and purification of antibodies, plete removal of organic solvents or failure to
detection of weakly expressed antigens, and correct the tonicity or pH of an eluate may
identification of multiple antibody speciftci- cause the reagent red cells used to test the
ties. Such studies are used in conjunction eluate to hemolyze or appear "sticky." The
with an appropriate adsorption technique, as presence of stromal debris may interfere with
described below and in Method 2-7. the reading of test results. Careful technique
and strict adherence to procedures should 1. Prenatal studies. When the antibody has a
eliminate such problems. specificity that is known to cause HDFN,
5. Instability of eluates. Diluted protein solu- or the antibody's clinical significance is
tions, such as those obtained by elution into unknown, the results of titration studies may
saline, are unstable. Eluates should be tested contribute to the decision about performing
as soon as possible after preparation. Alter- additional procedures (eg, Doppler sonogra-
natively, bovine albumin may be added to a phy or amniocentesis). (See Chapter 23 and
final concentration of 6%weight/volume, and Method 5-3.)
2. Antibody identification. Some antibodies
the preparation may be frozen during stor-
that agglutinate virtuaily all reagent red cells
age. Eluates can also be prepared in antibody-
may give an indication of specificity by
free plasma, 6% albumin, or a similar protein
demonstrating reactivity of different
medium. When commercial elution kits are strengths with different red cell samples in
used, the manufacturer's instructions for titration studies. For example, potent undi-
preparation and storage should be followed. luted autoanti-I may be reactive with both
adult and umbilical cord blood red cells, but
Combined Adsorption-Elution titration studies may reveal reactivity with
Combined adsorption-elution tests can be used adult 1+ red cells at a higher dilution than
to separate a mixture of antibodies in a single- with cord blood I+wred cells. The reactivity
serum sample, detect weakly expressed antigens of most antibodies weakens progressively
on red cells, or help identify wealdy reactive an- with serial dilutions (ie, a 2+ reaction
tibodies. The process consists of first incubating becomes 1+ in the next dilution), and weak
serum with selected red cells and then eluting antibodies « 1+) may lose their reactivity
antibody from the adsorbing red cells. when diluted. Yet, some antibodies that have
Care must be taken when selecting the ad- weak reactions when they are undiluted con-
sorbing cells to separate a mixture of antibodies. tinue to react at dilutions as high as 1 in
The cells should express only one of the anti- 2048. Such antibodies include antl-Ch, -Rg,
gens corresponding to an antibody in the mix- -c-, _Yka,-Kna, -Mcca, and -JMH. Titration
ture so that the eluate from the cells will contain studies may be performed on a sample show-
only that antibody. Both the eluate and adsorbed ing unexplained weak IATreactions to deter-
serum can be used for further testing. Unmodi- mine whether the reactivity is consistent
fied red cells are generally used for adsorptions with the antibodies in this group; however,
when subsequent elutions are being prepared. not all examples of these antibodies demon-
strate such high-titer, low-avidity characteris-
Titration tics. ThUS,the serologic characteristics may
The titer of an antibody is usually determined by suggest certain spedflcities, but failure to do
testing serial twofold dilutions of the serum with so does not eliminate these possibilities. The
selected red cells. Results are expressed as the antibodies listed above are not expected to
reciprocal of the highest serum dilution that cause shortened red cell survival, although
shows macroscopic agglutination. Titration val- there are examples of other antibodies (eg,
ues can provide information about the relative anti-Lu", -Hy, and -Yt') that may mimic
amount of antibody present in a sample or the these serologic characteristics and cause
relative strength of antigen expression on red shortened red cell survival. Anti-CD38 may
cells. also show high-titer reactivity and is gener-
Titration studies are useful for the following ally nonreactive with Lu(a-b-) cells.48,6o If
purposes: administration of this therapy to the patient
is not disclosed to laboratory staff, the inves- components or to identify the need for further
tigation may conclude the sample contains monitoring for HDFN.
an antibody to a high-prevalence LU system
antigen. Details about titration are given in Significance of Identified Antibodies
Method 3-15. The phases in which an antibody is identified
3. Separating multiple antibodies. Titration and its specificity are the two primary means
results may suggest that one antibody is reac- used to predict an unexpected antibody's poten-
tive at higher dilutions than another anti- tial clinical significance.Antibodies that are re-
body. That information can ailow the serum active at37 C, in an IAT, or both are potentially
to be diluted before it is tested with a red cell clinicallysignificant.Antibodies that are reactive
at room temperature and below are usually not
panel, which effectively removes one anti-
clinically Significant;however, there are many
body and allowsthe other to be identified. For exceptions. Forexample, anti-Vel,-P,and _PP1pk
example, if a serum contains anti-lk' that is (-W) may be reactive only at cold temperatures,
reactive to a titer of 2 and anti-c that is reac- yet may cause red cell destruction in vivo. Anti-
tive to a titer of 16, it may be possible to Ch, antl-Rg, and many of the KN and COST
dilute the serum 1:8 (ie, one volume of serum antibodies have littie or no clinical significance
in a final volume of 8) to clearly detect and an
despite theirreactiyity in lAT. R_eportedexpe-
identify only the anti-c. This can be useful rience with examples of antibodies with the
same specificity can be used in assessing the
when the selection of reagent red cells is lim-
clinical Significance.
ited by resource availabilityor by the antibody Table 13-5 summarizes the expected reactivi-
specificitiespresent in the patient's serum. ty and clinical significance of commonly en-
countered alloantibodies. For some antibodies,
Other Methods little or no data exist, and the decision about
Methods other than traditional tube, gel, or clinical significance has to be based on the
solid-phasetechniques may be used for antibody premise that clinicallysignificant antibodies are
identification. Some methods are especiallyuse- those that are active at 37 C, in an IAT,or both.
Certain laboratory tests have been used to
ful for testing small volumes of samples or
predict the clinical significance of antibodies.
reagents. Such methods include testing in
The monocyte monolayer assay, which quanti-
capillary tubes, microplates, or enzyme-linked
fies phagocytosis, adherence of antibody-coated
immunosorbent assays. Other methods that are
red cells, or both, can be used to predict the in-
useful in laboratories with specialized equip- vivo clinical significanceof some antibodies.61,62
ment include immunofluorescence, flow cytom- The test for antibody-dependent cellular cyto-
etry, and immunoblotting. toxicity, whlch' measures -lysis of antibody-
coated red cells, and the chemiluminescence as-
say, which measures the respiratory release of
oxygen radicals after phagocytosis of antibody-
coated red cells, have been helpful in predicting
in-vivo antibody significance-particularly for
predicting the severity of HDFN. For cold-
Unexpected red cell antibodies are revealed in reactive antibodies, in-vitro thermal amplitude
antibody detection tests and characterized studies may be able to predict the likelihood of
through antibody identification testing. Informa- in-vivohemolysis/"
tion obtained from this process is then used to In-vivotests may also be used to evaluate the
help determine the potential clinicalsignificance Significanceof an antibody. The most common
of the unexpected antibody for the purpose of technique is a red cell survival study in which
providing an effective transfusion of red cell radiolabeled, antigen-positive red cells (usually
labeled with SICr) are infused into the patient. match procedure is required if the sample con-
After a specified period has elapsed, a sample of tains-or has previously contained-a clinically
blood from the patient is tested for radioactivity. significant antibody.9(PP38,77)
Exceptions to these
With this technique, it is possible to measure practices should be made only in extreme clini-
the survival of 1 mL of infused cells. Another in- cal emergencies under the direction of a physi-
vivo technique, flow cytometry, can also be used cian.
to measure the survival of infused red cells, but A potent example of the antibody should be
a larger aliquot of red cells (about 10 mL) is usu- used to identify antigen-negative RBCunits. Of-
ally required. Interpretation of in-vivo survival ten, the antibody is a commercial antiserum, but
test results is complicated by the fact that small to save expensive or rare reagents, units can be
aliquots of incompatible red cells may have a tested first (often referred to as screened) for
faster rate of destruction than an entire trans- compatibility with the patient's serum. Then,
fused unit of Red Blood Cells (RBCs).Compari- the absence of the antigen in compatible units
son with documented cases in the literature and can be confirmed with commercial reagents. If
consultation with an IRL should provide guid- the antibody is unusual and commercial antise-
ance about previous examples of similar specific- rum is not available, a stored sample from the
ities. sensitized patient may be used to select units for
transfusion at a later time, especially if the pa-
Subsequent Antibody Identification in tient's later samples lose reactivity. If any patient
Patients with a Known History of serum or plasma is used as a typing reagent, re-
Antibodies ferred to as a single-source antibody, the anti-
Once a clinically significant antibody has been body reactivity should be well characterized and
identified, the patient must receive red cells retain reactivity after storage. Appropriate nega-
negative for the corresponding antigen if at all tive and weak-positive controls (eg, from hetero-
possible, so it is rarely necessary to routinely re- zygous donors) should be used at the time of the
peat the identification of known antibodies in testing. The following criteria, established by
subsequent pretransfusion testing. AABB Stan- the FDA for licensing some reagents, should be
dards states that in patients with previously used as guidelines for human-source reagents
identified clinicallysignificant antibodies, testing used in lieu of commercial reagents?':
methods should be used that identify additional
clinically significant antibodies.9(p37)
Each labora- 1. Anti-K, ok, -Ik', -Py", and -C": dilution of 1:8
tory should define policies and methods for the must produce at least a 1+ reaction.
detection of additional antibodies in these pa- 2. Anti-S, -s, -PI, -M, -I, -c (saline), -e (saline),
tients. and -AI: dilution of 1:4 must produce at least
a 1+ reaction.
Selection of Donor Units for Patients
3. Most other spedftcities: undiluted reagent
Whose Serum Contains Antibodies
must produce at least a 2+ reaction.
Antigen-Negative Blood
RBC units selected for transfusion to a patient When selecting units for patients with clini-
with potentially clinically significant antibodies cally significant antibodies, some serologists rec-
should be negative for the corresponding anti- ommend typing the units with antibodies from
gents). Even if the antibodies are no longer de- two different sources, but others consider this
tectable, all subsequent RBC transfusions to step unnecessary-especially when potent corn-
that patient should lack the antigen to prevent a mercial reagents are available and an IAT cross-
secondary immune response. The transfusion match will be performed. Different lots of anti-
service must maintain records of all patients in body from the same manufacturer and even
whom clinically significant antibodies have different reagents from different manufacturers
been previously identified, and an IAT cross- may have been prepared from the same source
material because manufacturers often acquire cannot be demonstrated by routine testing.

these resources from the same entity. This is also true when an antibody has not
If a donor unit is tested for selected antigens been specifically demonstrated but decreased
and labeled by the blood center, the use of survival of transfused cellsis observed.
licensed (commerctal) reagents, if available, is For patients needing' chronic transfusion
required. If no licensed reagent is available, the therapy for sickle cell disease or thalassemia,
unit must be labeled with appropriate wording limited antigen-matching, specificallyfor Rh an-
(eg, "Tested and found negative for XXantigen tigens (generally C and E) and K is becoming
using unlicensed typing reagents'j." Except for common practice to prevent or mitigate allo-
results of ABOand D typing, there is no require- immunization. Transfusion. of phenotypica,l~y
ment that the hospital repeat testing of donor matched RBCunits, however, does not prevent
minor antigen typing if the results are on the la- formation of all new alloantibodies.
bel or on an attached tag.9(P36) Minor antigen
typing results on packing slips or not physically When Uncommon or Rare Blood Is
attached to the donor unit should be confirmed Needed
by the hospital, if possible, when the unit is in-
tended for transfusion to a patient with the cor- Rare blood includes units that are negative for
responding alloantibody. high-prevalence antigens « 1:1000 units) or
are negative for a combination of many. com-
Crossmatch for Compatibility mon antigens (<1:100). When a patient has
multiple antibodies, it is helpful to determine
For certain antibodies, typing the donor units the prevalence of compatible donors. To calcu-
may not be necessary, and the patient's serum late this prevalence, one must multiply the
can be used to select serologically compatible prevalence of donors who are negative for one
RBCunits. This is especially true for antibodies antigen by the prevalence of donors who are
that characteristically are reactive below 37 C negative for each of the other antigens. The
[eg, anti-M, -N, -PI, -Le",-Leb, and -AI) and that steps are outlined in Table 13-8, using the ex-
do not ordinarily produce a secondary immune ample of a group 0 patient's serum that con-
response following the transfusion of antigen- tains anti-c, anti-Py', and anti-So As shown,
positive RBCunits. only 1.3% of the general population would be a
compatible donor for a group 0 patient With
Phenotype-Matched Blood anti-c, anti-Py",and anti-So
Sometimes it may be best practice to provide If any of these antibodies is present alone,
phenotypically matched, antigen-negative RBC finding compatible blood is not very difficult,
units as a prophylactic measure-.For example, but the combination requires a large number of
when a patient of the RjRj phenotype produces units in order to find one compatible unit (ie,
antl-B, some serologists suggest that RBC units 1.3 compatible donors out of 100, or approxi-
should be negative for both the E and c anti- mately lout of 80). The calculation in the table
gens. This recommendation is based on the as- uses the prevalence in populations of European
sumption that the stimulus to produce antl-E ancestry, and prevalence may be different in
may also have stimulated anti-c or anti-cf that populations of non-European ancestry. In calcu-
remains undetected by routine tests." Similarly, lating the probability of compatible donors, one
for an RzRzpatient with demonstrable anti-C, should use the antigen prevalence that corre-
the use of e-negative donor blood may be con- sponds with the racial composition of the donor
sidered. population, if available.
It may be prudent to select RBC units that When units of rare or uncommon pheno-
are phenotypically matched with the patient for types are needed, the local IRL should be con-
clinicallysignificantantigens when a patient has tacted. LocalIRLsthat reside within or are asso-
a potent warm autoantibody or is receiving ciated with blood centers typically have an
monoclonal antibody therapy and compatibility inventory (fresh and/or frozen) of RBCunits of
TABLE 13-8. Calculating Prevalenceof Compatible Donors Requiredto Find Antigen-Negative Units
Identify the prevalenceof antigen-negativeindividualsfor each 18% C-, 34% Fy(a-), and 45% S-
of the applicableantigens
Calculatethe prevalenceof units negativefor all antigenscom- 0.18 x 0.34 x 0.45 = 0.028, or 2.8%
bined
Identifythe prevalenceof ABO-compatibledonors sothat it can 45%'

befactored into the calculation
Calculatethe prevalenceof ABO-compatibleantigen-negative 0.028 x 0.45 = 0.013, or 1.3%

units
'Prevalence of group 0 donors.
uncommon phenotypes and sometimes rare multiple antibodies or an antibody to a high-

phenotypes. When the local IRL does not have prevalence antigen, the mother (if she is ABO
RBCsof the necessary phenotype, they typically compatible)is often the logicaldonor.
have a mechanism for searching for the required
units. (SeeIRLsection below.)
If the clinical situation allows, autologous
RBC transfusions should be considered for pa-
tients with rare phenotypes who are expected to
need blood in the future. Additionally, family IRLstypicallyhave the skilled staff, procedures,
members are another potential source of rare and, most importantly, resources (such as frozen
blood donors. The absence of high-prevalence aliquots of fully phenotyped rare red cells lack-
antigens is usually associated with the inheri- ing high-prevalence antigens) to investigate and
tance of the same rare recessive blood group resolve many or most complex antibody prob-
gene from each heterozygous parent. Children lems. IRLs can also provide consultation and
from the same parents have one chance in four information to laboratories about unfamiliar or
of inheriting the same two rare genetic muta- infrequently encountered complex antibody
tions, making siblings much more likely than problems. Additionally, IRLsoften help facilities
the general population to have the rare blood procure units of specificphenotypes when such
type. In most cases, blood from the patient's par- units cannot be found in a routine transfusion
ents, children, and half of the patient's siblings service. Many IRLs also have access to the
express oniy one rare gene. If transfusion is es- American Rare Donor Program (ARDP),which
sential and there is no alternative to transfusing provides a network for finding RBCunits with
incompatible blood, these heterozygous (single- rare phenotypes throughout the United States
dose) donors may be preferable to random and also has connection to similar programs
donors. For infants with HDFN resulting from worldwide (SeeMethod 3-21).
KEY POINTS
1. A clinicallysignificantred cell antibody is an antibody that is frequently associated with HDFN,

hemolytic transfusion reactions, or a notable decrease in the survival of transfused red cells.
REFERENCES
1. Tremi A, King K. Red blood cell alloimmuniza- 4. Issitt PD, Anstee DJ.Applied blood group serolo-
tlon: Lessons from sickle cell disease. Transfu- gy. 4th ed. Durham, NC: Montgomery Scientific
sion 2013;53:692-5. Publications, 1998.
2. Chou ST, Jackson T, Vege S, et al. High preva- 5. Reid ME, Lomas-Francis C, Olsson M. The
lence of red blood cell alloimmunizatlon in sick-
blood group antigens factsbook. 3rd ed. London:
le cell disease despite transfusion from Rh-
matched minority donors. Blood 2013; 122: ElseyierAcademic Press, 2012.
1062-71. 6. Malyska H, KleemanJE, Masouredis SP,Victoria
3. Spanos T, KarageorgaM, LadisV, et al. Red cell EJ. Effectson blood group antigens from storage
alloantlbodies in patients with thalassemia. Vox at low ionic strength in the presence of neomy-
Sang 1990;58:50-5. cin. Vox Sang 1983;44:375-84.
426 A A B B TEe H N I CAL MAN UAL
7. Westhoff CM, Sipherd BD, Toalson LD. Red cell identification (abstract). Transfusion 2018;
antigen stabilityin K3EDTA.Immunohematolo- 58(S2):177A.
'1:! 1993;9: 109-11. 21. Howie HL, Delaney M, Wang X. Serological
8. Van Thof L, ed. Standards for Immunohematolo- blind spots for variants of human IgG3 and IgG4
'1:! reference laboratories. 11th ed. Bethesda, by a commonly used anti-immunoglobulin re-
MD: AABB,2019. agent. Transfusion 2016;56:2953-62.
9. Gammon R, ed. Standards for blood banks and 22. Oostendorp M, Lammerts van Bueren ]J, Doshi
transfusion services. 32nd ed. Bethesda, MD: P, et al. When blood transfusion medicine be-
AABB,2020. comes complicated due to interference by
10. Casina TS. In search of the holy grail: Compari- monoclonal antibody therapy. Transfusion20 IS;
son of antibody screening methods. Immunohe- 55: 1555-62.
matology 2006;22: 196-202. 23. Chapuy CL, Nicholson RT,Aguad MD, et al. Re-
11. Winters ]L, Richa EM, Bryant SC, et al. Polyeth- solving the daratumumab interference with
ylene glycol antiglobulin tube versus gel micro- blood compatibility testing. Transfusion 2015;
column: Influence on the incidence of delayed 55: 1545-54.
hemolytic transfusion reactions and delayed se- 24. Ananl WO, Duffer K, Kaufman RM, et al. How
rologic transfusion reactions. Transfusion 2010; do I work up pretransfusion samples containing
50:1444-52. anti-CD38? Transfusion2017;57:1337-42.
12. Bunker ML, Thomas CL, Geyer S]. Optimizing 25. Carreno-Tarrogona G, Cedena T, Montejano L,
pretransfusion antibody detection and identifica- et al. Papain-treated panels are a simple method
tion: A parallel, blinded comparison of tube for the identification of alloantibodies in multi-
PEG, solid-phase, and automated methods. ple myeloma patients treated with anti-CD38
Transfusion 2001 ;41:621-6. based therapies. Transfus Med 2019;29(3): 193-
13. Plsacka M, Kralova M, Sklenarova M. Solid- 6.
phase-membrane only antibodies-reactive only 26. Velliquette RW, Shakarian G, Lomas-FrancisC,
in Capture-R Readybut nonreactive by Capture- Westhoff CM. Testing samples from patients re-
R Select and in other techniques (abstract). ceiving anti-CD38 therapy with commercial pa-
Transfusion 2011 ;51(SuppI3):175A. pain treated reagent red cells (abstract). Transfu-
14. Lang N, Sulfridge DM, Hulina ], et al. Solid sion 20 18;58(S2):196A.
phase reactive only antibodies (abstract). Trans- 27. Velliquette RW, Kirkegaard ], Jones D, et al.
fusion 2011;51(SuppI3):I72A. Monoclonal anti-CD38 and anti-CD47 therapy
IS. Uu C, Grossman BJ. Antibody of undetermined interference with platelet antibody screen test
specificity: frequency, laboratory features, and methods (abstract). Transfusion 2018;58(S2):
natural history. Transfusion 2013;53:931-8. 51A.
16. Miller NM, Johnson ST, Carpenter E, et al. Pa- 28. Judd W], Steiner EA, Oberman HA, Nance S.
tient factors associated with unidentified reac- Can the reading for serologic reactivity follow-
tivity in solid-phase and polyethylene glycol an- ing 37 degrees C incubation be omitted? Trans-
tibody detection methods. Transfusion 2017;57: fusion 1992;32:304-8.
1288-93. 29. Reid ME, 0yen R, Storry], et al. Interpretation
17. Howard ]E, Winn LC, Gottlieb CE, et al. Clini- of RBC typing in multi-transfused patients can
cal Significanceof anti-complement component be unreliable (abstract). Transfusion 2000;40
of antiglobulin antisera. Transfusion 1982;22: (Suppl):123S.
269-72. 30. Lomas-FrancisC, DePalma H. 2007 Rock 0yen
18. Nedelcu E, Hall C, Stoner A, et al. Interference Symposium. DNA-basedassays for patient test-
of anti? CD47 therapy with blood bank testing ing: Their application, interpretation, and cor-
(abstract). Transfusion 2017;57(S3):148A. relation of results. Immunohematology 2008;
19. Velliquette RW, Degtyaryova D, Hong H, et al. 24:180-90.
Serological observations in patients receiving 31. Fisher RA. Statistical methods and scientific in-
Hu5F9? G4 monoclonal anti-CD47 therapy (ab- ference. 2nd ed. Edinburgh, Scotland: Oliver
stract). Transfusion 2017;57(S3): 159A. and Boyd, 1959.
20. Howard-Menk C, Crane J, Doshi L, Papari M. 32. Harris RE, Hochman HG. Revised p values in
HUSF9? G4 monoclonal anti-CD47 therapy: A testing blood group antibodies: Fisher's exact
first experience with interference in antibody test revisited. Transfusion 1986;26:494-9.
33. Kanter MH, Poole G, Garratty G. Misinterpreta- 47. Rolih S, Thomas R, Fisher F, Talbot J. Antibody
tion and misapplication of p values in antibody detection errors due to acidic or unbuffered sa-
identification: The lack of value of a p value. line. Immunohematology 1993;9: 15-18.
Transfusion 1997;37:816-22. 48. Velliquette RW, Shakarian G, Ihang J, et al. Da-
34. Moulds JM, Zimmerman PA, Doumbo OK, et al. ratumumab-derived anti-CD38 can be easily
Molecular identification of Knops blood group mistaken for clinically significant antibodies to
polymorphisms found in long homologous re- Lutheran antigens or to Knops antigens (ab-
gion D of complement receptor l. Blood 2001; stract). Transfusion 2015;55(3S):26A.
97:2879-85. 49. Issitt PD, Combs MR, Bredehoeft SJ, et al. Lack
35. Arndt PA, Leger RM, Garratty G. Serologicfind- of clinical significance of "enzyme-only" red cell
ings in autoimmune hemolytic anemia associat- alloantibodies. Transfusion 1993;33:284-93.
ed with immunoglobulin M warm autoantibod- 50. Advani H, Zamor J, Judd WJ, et al. Inactivation
ies. Transfusion2009;49:235-42. of Kell blood group antigens by 2-aminoethyliso-
36. Waligora SK, Edwards JM. Use of rabbit red thiouronium bromide. Br J Haematol 1982;
cells for adsorption of cold autoagglutinins. 51:107-15.
Transfusion 1983;23:328-30. 5l. Branch DR, Muensch HA, Sy Siok Hian AL,
37. Yuan S, FangA, Davis R, et al. Immunoglobulin Petz LD. Disulfide bonds are a requirement for
M red blood cell alloantibodies are frequently Kell and Cartwright (Yta) blood group antigen
adsorbed by rabbit erythrocyte stroma. Transfu- integrity. Br J Haematol 1983;54:573-8.
sion 2010;50: 1139-43. 52. Branch DR, Petz LD. A new reagent (ZZAP)
38. Arndt PA, Garratty G. The changing spectrum of having multiple applications in immunohema-
drug-induced immune hemolytic anemia. Semin tology, Am J Clin PathoI1982;78: 161-7.
HematoI2005;42: 137-44. 53. Liew YW, Uchikawa M. Loss of Er' antigen in
39. Judd WJ, Steiner EA, Cochran RK. Paraben- very low pH buffers. Transfusion 1987;27:442-
3.
associated autoanti-Ika antibodies: Three exam-
54. Swanson JL, Sastamoinen R. Chloroquine strip-
ples detected using commerciallyprepared low-
pingof HLAA,B antigens from red cells. Trans-
ionic strength saline containing parabens. Trans-
fusion 1985;25:439-40.
fusion 1982;22:31-5.
55. Morton JA, Pickles MM, Terry AM. The Sd'
40. Judd WJ, StorryJR, AnnesleyTD, et al. The first
blood group antigen in tissues and body fluids,
example of a paraben-dependent antibody to an
Vox Sang 1970;19:472-82.
Rh protein. Transfusion200 1;41:371-4.
56. O'Neill GJ, Yang SY,TegoliJ, et al. Chido and
4l. Dube VE, Zoes C, Adesman P. Caprylate- Rodgers blood groups are distinct antigenic
dependent auto-anti-e. Vox Sang 1977;33:359- components of human complement C4. Nature
63. 1978;273:668-70.
42. Rodberg K, Tsuneta R, Garratty G. Discrepant 57. Tilley CA, Romans DG, Crookston MC. Locali-
Rh phenotyping results when testing IgG- sation of Chido and Rodgersdeterminants to the
sensitized RBCswith monoclonal Rh reagents C4d fragment of human C4. Nature 1978;276:
(abstract).Transfusion 1995;35(Suppl):67S. 713-15.
43. Reisner R, Butler G, Bundy K, Moore SB. Com- 58. Judd WJ, Kraemer K,Moulds]]. The rapid iden-
parison of the polyethylene glycol antiglobulin tification of Chido and Rodgersantibodies using
test and the use of enzymes in antibody detec- C4d-coated red blood cells. Transfusion 1981;
tion and identification. Transfusion 1996;36: 21:189-92.
487-9. 59. Freedman J, Masters CA, Newlands M, Molli-
44. Issitt PD, Combs MR, Bumgarner DJ, et al. son PL. Optimal conditions for use of sulphydryl
Studies of antibodies in the sera of patients who compounds in dissociating red cell antibodies.
have made red cell autoantibodies. Transfusion Vox$ang 1976;30:231-9.
1996;36:481-6. 60. Aye T, Arndt PA, Leger RM, et al. Myelomapa-
45. Beattie KM, Zuelzer WW. The frequency and tientsreceiving daratumumab (anti-CD38) can
properties of pH-dependent anti-M. Transfusion appear to have an antibody with Lutheran-relat-
1965;5:322-6. ed specificity (abstract). Transfusion 2015;
46. Bruce M, Watt AH, Hare W, et al. A serious 55(3S):28A.
source of error in antiglobulin testing. Transfu- 61. Nance SJ, Arndt P, Garratty, G. Predicting the
sion 1986;26: 177-81. clinical Significanceof red cell alloantibodies us-
ing a monocyte monolayer assay. Transfusion 65. Food and Drug Administration. 7342.001: In-
1987;27:449-52. spection of licensed and unlicensed blood
62. Arndt PA, Garratty G. A retrospective analysis of banks, brokers, reference laboratories, and con-
the value of monocyte monolayer assay results tractors. Compliance Program guidance manual.
for predicting the clinical significance of blood Silver Spring, MD: CBEROffice of Compliance
group alloantibodies. Transfusion 2004;44: and Biologics Quality, 2010:50-3. [AVailableat
1273-81. https://www.fda.gov/media/84887/down
63. Petz LD, Garratty G. Immune hemolytic anemi- load.]
as. 2nd ed. Philadelphia: Churchill Livingstone, 66. Shirey RS,Edwards RE, Ness PM. The risk of al-
2004. loimmunization to c (Rh4) in RIRI patients
64. Code of federal regulations. Title 21, CFR Parts who present with anti-E. Transfusion 1994;34:
660.25 and 660.26. Washington, DC: US Gov- 756-8.
ernment Publishing Office, 2019 (revised annu-
ally).
SUGGESTED READINGS
Daniels G. Human blood groups. 3rd ed. Hoboken, NJ: Kanter MH. Statisticalanalysis.In: Busch MP, Brecher
Wiley-Blackwell,2013. ME, eds. Research design and analysis. Bethesda,
Daniels G, PooleJ, de SilvaM, et al. The clinicalsignif- MD: AABB,1998:63-104.
icance of blood group antibodies. Transfus Med Klein HG, Anstee DJ. Mollison's blood transfusion in
2002;12:287-95. clinicalmedicine. 12th ed. Oxford, UK:Wiley-Black-
Engelfriet CP, Overbeeke MA, Dooren MC, et al. Bio- well,2014.
assays to determine the clinical significance of red Menitove JE. The Hardy-Weinberger principle: Selec-
cell antibodies based on Fe receptor-induced destruction of compatible blood based on mathematic prin-
tion of red cells sensitized with IgG. Transfusion ciples. In: Fridey JL, Kasprisin CA, Chambers LA,
1994;34:617-26. Rudmann SV, eds. Numbers for blood bankers.
Garratty G. In vitro reactions with red blood cells that Bethesda, MD: AABB,1995:1-11.
are not due to blood group antibodies: A review. Gammon R, ed. Standards for blood banks and transfu-
sion services.32nd ed. Bethesda, MD: AABB,2020.
Immunohematology 1998;14: 1-11.
Reid ME, Lomas-Francis C, Olsson M. The blood
HamiltonJ,Johnson ST,Rudmann SV.Antibodyidenti-
group antigen factsbook. 3rd ed. London: Elsevier
fication: Art or science? A case study approach.
Academic Press, 2012.
Bethesda, MD: AABB,2013.
Rolih S. A review: Antibodies with high-titer, low-
Hamilton J, Johnson ST, Rudmann SV. Investigating
avidity characteristics. Immunohematology 1990;6:
positive DAT results: A case study approach.
59-67.
Bethesda, MD: AABB,2016. Rudmann SV,ed. Serologlcproblem-solving:A system-
Harmening DM. Modern blood banking and transfu- atic approach for improved practice. Bethesda, MD:
sion practices. 6th ed. Philadelphia:FADavis,2012. AABBPress, 2005.
Issitt PD, Anstee DJ.Appliedblood group serology.4th Weisbach V, Kohnhauser T, Zimmermann R, et al.
ed. Durham, NC: Montgomery Scientific Publica- Comparison of the performance of microtube col-
tions, 1998. umn systems and solid-phase systems and the tube
Judd WJ, Johnson S, StorryJ. Judd's methods in immu- low-ionic-strength solution additive indirect anti-
nohematology. 3rd ed. Bethesda, MD: AABBPress, globulin test in the detection of red cell alloantibod-
2008. ies. Transfus Med 2006; 16:276-84.
Van Thof L, ed. Standards for immunohematology ref- Westhoff C. 2007 Rock0yen Symposium.Potential of
erence laboratories. 11th ed. Bethesda, MD: AABB, blood group genotyping for transfusion medicine
2019. practice. Immunohematology 2008;24: 190-5.
The Positive Dlrect
Antig obulin Test and
Immune-lVIE!diateci Hemolysis
p. Dayand Borge Ir. MD, PhD, and Paul M. Mansfield, MT(ASCplMSBB
EM0 LYII CAN E M I A I S THE tic features of this rare type of hemolysis are
shortening of red cell survival. The nor- hemoglobinemia and, when the plasma hemo-
mal life span of red cells is approximately globin level exceeds the renal threshold, hemo-
110 to 120 days. In healthy individuals, 1% of globinuria. Conversely and more commonly,
red cells are.removed by the reticuloendothelial extravascular hemolysis results when macro-
system each day, but this is matched by red cell phages in the spleen and liver phagocytose red
production in the marrow. Normal marrow can cells completely or partially (producing sphero-
increase red cell production to compensate for cytes) or destroy red cells by cytotoxic events,
blood loss. Thus, in the absence of bleeding, an resulting in an increase in serum bilirubin. This
increased reticulocyte count is an indirect mea- distinction is a simplification, however, .because
sure of hemolysis. If the marrow is able to ade- hemoglobin can also be released into the plasma
quately compensate, a reduced red cell survival following extravascular destruction if hemolysis
may not result in anemia. is brisk.
Immune-mediated hemolysis, the subject of The serologic investigations carried out in
this chapter, is only one cause of hemolytic ane- the blood bank help determine whether the he-
mia, and many causes of hemolysis are unrelat- molysis has an immune basis and, if so, what
ed to immune reactions. Immune hemolytic type of immune hemolytic anemia is present.
anemia is the result of an immune response that This is important because the treatment for each
targets red cells. The diagnosis of hemolytic ane- type is different but generally involves some
mia rests on clinical findings and laboratory da- type of immunomodulatory therapy. Although
ta, such as hemoglobin or hematocrit values; re- there is no evidence-based algorithm for thera-
ticulocyte count; red cell morphology; and py, treatment options can include corticoste-
bilirubin, haptoglobin, and lactate dehydroge- roids, intravenous immunoglobulin (IYlG),sple-
nase (LDH)levels. nectomy,. rituximab' and other. more potent
In some cases, the destruction of red cells immunosuppressive medications.' As more pa-
takes place in the intravascular space with the tient outcome data with different treatment mo-
release of free hemoglobin into the plasma. The dalities become available, the determination of
red cells are ruptured following activation of the what is considered first-linetherapy will change.
classical complement cascade. The characteris- For example, although rituximab is considered
P. Dayand Borge Jr, MD, PhD, Chief Medical Officer, East Division; and Paul M. Mansfield, MT(ASCP)CMSBB,
Manager, IRL, Penn-Jersey Region and National Reference Laboratory for Blood Group Serology, American Red
Cross Blood Services, Philadelphia, Pennsylvania
429
to be first-line therapy for cold agglutinin syn- only 1.4% in a patient without hemolytic ane-
drome (CAS),eculizumab, an antibody that mit- mia.'
igates complement-mediated hemolysis, may be Small amounts of immunoglobulin G (IgG)
more effective in patients with acute, brisk he- and complement that are lower than the detec-
molysis.' tion limit of routine testing techniques appear to
The direct antiglobulin test (DAT)is a simple be present on all red cells. Using sensitive test-
test used to determine if red cells have been ing techniques, 5 to 90 IgG molecules/red cell"
coated in vivo with immunoglobulin, comple- and 5 to 40 C3d molecules/red celf have been
ment, or both. The DATis used primarily for the detected in healthy individuals. Depending on
investigation of hemolytic transfusion reactions the technique and reagents used, the DAT can
(HTRs),hemolytic disease of the fetus and new- detect 100 to 500 molecules of IgG/red cell and
born (HDFN), autoimmune hemolytic anemia 400 to 1100 molecules of C3d/red cell. Positive
(AlHA), and drug-induced immune hemolytic DAT results are reported in 1 in 1000 to I in
anemia (DIIHA).A positive DAT result mayor 14,000 blood donors and 1% to 15% of hospital
may not be associated with immune-mediated patients." These large differences in incidence
hemolysis. As shown in Table 14-1, there are are probably related to the different DAT tech-
many causes of a positive DATresult. niques used.
Most blood donors with a positive DATresult
appear to be healthy, and most patients with
H positive DAT results have no obvious signs of
hemolytic anemia. However, a careful evalua-
The DAT should be performed on every patient tion may show evidence of increased red cell de-
in whom the presence of hemolysis has been es- struction. Studies suggest that a positive DATre-
tablished to distinguish immune from nonim- sult in a healthy blood donor may be a marker of
mune hemolytic anemia. The DAT should also risk of future development of malignancy."
be performed when a positive autocontrol is A positive DATresult in a patient with hemo-
found in antibody identification studies (see lytic anemia indicates that the most likely diagno-
Chapter 13), but there is no benefit to perform- sis is one of the immune hemolytic anemias.
ing a DAT (or autocontrol) as part of routine pre- However, the DAT result can be positive, coinci-
transfusion testing. The DAT should not be per- dentally, in patients with hemolytic anemia that is
formed as a screening test for hemolytic anemia. not immune mediated. Conversely, some pa-
The predictive value of a positive DAT result is tients with immune hemolytic anemia have a
83% in a patient with hemolytic anemia, but negative DATresult. (See"DAT-NegativeAlHA.")
TABLE 14·1. Some Causes of a Positive DAT Result

~~---~.--~-,~~. ~------.~~-~---~~--------~
Autoantibodies to intrinsic red cell antigens
Hemolytic transfusion reactions
Hemolytic disease of the fetus and newborn
Drug-induced antibodies
Passivelyacquired alloantibodies (eg, from donor plasma, derivatives, or immunoglobulin)
Nonspecifically adsorbed proteins (eg, hypergammaglobulinemia, high-dose intravenous immune globulin,
or modification of red cell membrane by some drugs)
Complement activation due to bacterial infection, autoantibodies, or alloantibodies
Antibodies produced by passenger lymphocytes (eg, in transplanted organs or hematopoietic components)
OAT= direct antiglobulin test.
C HAP T E R 1 4 OAT/ImmuneHemolysis 431
The DAT can also be positive for IgG or ma globulins and complement; otherwise, the
complement without a clear correlation with antiglobulin reagent can be neutralized, leading
anemia in patients with sickle cell disease, beta- to a false-negative result. The saline used for
thalassemia, renal .disease, multiple myeloma, washing the red cells should be at room tem-
autoimmune disorders, AIDS, .or other. diseases perature; washing red cells with warm (eg,
associated with elevated serum globulin or 37 C) saline can result in the loss of red-cell-
blood urea nitrogen levels,"!' The interpretation bound, low-affinityIgG. Washing should be un-
of a positive DAT result should take into consid- interrupted, especially if performing manual
eration the patient's history, clinical data, and re- washing, and red cells should be tested immedi-
sults of other laboratory tests. ately after washing to prevent false-negativere-
Initial transfusion reaction investigations in- sults caused by the elution of IgG. Performing a
clude a DAT on a posttransfusion specimen. In DAT using a column agglutination test (eg, gel
the presence of immune-mediated hemolysis, test) does not require washing of the red cells
the DAT result may be positive if sensitized red before testing because plasma proteins do not
cells have not been destroyed, or negative if he- neutralize detection of red-cell-bound reactivity.
molysis and rapid clearance have occurred. This may represent a positive bias for detection
Preparation and testing of an eluate from DAT- of red-cell-boundlow-affinityIgG-
positive posttransfusion-reaction red cells is indi- Although any red cells may be tested, EDTA-
cated. Even if the DAT result is only weakly pos- anticoagulated blood samples are preferred. The
itive or negative, testing of an eluate may be in- EDTAprevents in-vitro fixation of complement
formative. If the DAT result is positive on the by chelating the calcium that Is needed for C1
postreaction specimen, a DAT should also be activation. If red cells from a clotted blood sam-
performed on the pretransfusion specimen for ple have a positive DAT result due to comple-
comparison and appropriate interpretation. ment, the results should be confirmed on red
cellsfrom freshly collected blood kept at 37 C or
The Principles of the DAT an EDTA-anticoagulatedspecimen if these re-
The DAT is based on the test developed by sults are to be used for diagnosticpurposes,
Coombs, Mourant, and Race" for the detection The DAT can be initially performed with a
of antibodies attached to red cells that do not polyspecificantihuman globulin (AHG)reagent
produce direct agglutination. This test, an indi- that is capable .of detecting both IgG and C3d.
rect antiglobulin test (JAT),was initiallyused to (See Method 3-14.) If the results are positive,
demonstrate antibody in serum, but it was later tests with monospecific reagents (anti-IgG and
applied to demonstrate the in-vivocoating of red anti-complement separately) need to be per-
cells with antibody or complement components formed to appropriately characterize the im-
(the DAT). mune process involved.and determine the diag-
Most of the antiglobulin reactivity is directed nosis. Because polyspecificreagents are usually
at the heavy chains (eg, Fc portion of the sensi- blended, and testing conditions for optimally de-
tizing antibody) or the complement compo- tecting IgG and C3d on red cells may differ,
nent, thus bridglng the gap between adjacent some laboratories perform the DATinitiallywith
red cells to produce visible agglutination. The anti-IgGand anti-C3d reagents separately. If the
strength of the observed agglutination is usually polyspecificreagent is polyclonal, proteins other
proportional to the amount of bound protein. than IgG or C3d (eg, IgM, 19A, or other comple-
The DAT is performed by testing freshly ment components) can occasionally be detect-
washed red cells directly with antiglobulin re- ed; however, specific reagents to distingulsh
agents containing anti-IgG and anti-C3d. In the these other proteins by serologic techniques are
United States, only polyspecific anti-IgG,-C3d not readily available. If umbilical cord blood
and monospecific anti-IgG, anti-C3d, and anti- samples are to be tested, it is appropriate to use
C3b,-C3d reagents are currently licensed. The anti-IgG only, because HDFN results from fetal
red cells need to be washed to remove free plas- red cell sensitization with maternally derived
IgG antibody, and complement activation rarely tosis; spherocytes observed on the peripheral
occurs." blood film; hemoglobinemia; hemoglobin-
It is important to follow the reagent manufac- uria; decreased serum haptoglobin; and ele-
turer's instructions and recognize any product vated levels of serum unconjugated (indirect)
limitations. False-negativeor weaker results can bilirubin or LDH, especially LDH1, may be
be obtained if the washed red cells are allowed associated with increased red cell destruc-
to sit before they are tested with anti-IgG or if
tion. These factors are indicative of hemo-
the reading of the results is delayed. Some anti-
lytic anemia but not specifically immune
complement reagents, in contrast, demonstrate
stronger reactivity if centrifugation is delayed for hemolytic anemia. If there is no evidence of
a short time after the reagent has been added. hemolytic anemia, no further studies are
When the DAT result is positive with both anti- necessary unless the patient requires a red
IgG and anti-Cs, the red cells should be tested cell transfusion and the serum contains
with an inert control reagent (eg, 6% albumin or incompletely identified antibodies to red cell
saline). Lack of agglutination of the red cells in antigens. Testing an eluate may be helpful for
the control reagent provides some assurance antibody identification. (See "Elution" sec-
that the test results are accurately interpreted. If tion below and Chapter 13.)
the control is reactive, the DAT result is invalid. 2. Recent trensfusion. When a patient has
[See sections below on warm AIHA (WAIHA) recently received transfusion, a positive DAT
and CAS.] Reactivity with this control reagent result may be the first indication of a devel-
can indicate spontaneous agglutination caused
oping immune response. Developing anti-
by heavy coating of IgG or rare warm-reactive
body sensitizes the transfused red cells that
IgM, or it can indicate IgM cold autoagglutinins
that were not dissociated during routine wash- have the corresponding antigen, and the
ing. DATresult becomes weakly positive (usually
<2+). The antibody may not be present in
Evaluation of a Positive DAY Result sufficient quantity to be detected in the
serum. Antibody may appear as early as 7 to
A positive DAT result alone is not diagnostic of
10 days after transfusion in a primary immu-
hemolytic anemia. Understanding the signifi-
nization or as early as 1 to 2 days in a sec-
cance of this positive result requires knowledge
of the patient's diagnosis; recent drug, pregnan- ondary response.v" These alloantibodies
cy, transfusion, and hematopoietic transplanta- could shorten the survival of red cells that
tion history; and the presence of acquired or un- have already been transfused or are adminis-
explained hemolytic anemia. Dialogue with the tered in subsequent transfusions. A mixed-
attending physician is important. Clinical con- field appearance in the posttransfusion DAT
siderations together with laboratory data should result (ie, agglutination of donor red cells and
dictate the extent to which a positive DATresult no agglutination of the patient's red cells)
is evaluated. mayor may not be observed.
3. Administration of drugs associated with
Patient History immune-mediated hemolysis. Many drugs
The following situations may warrant further in- have been reported to cause a positive DAT
vestigation of a positive DATresult. result and/or immune-mediated hemolysis,
but this occurrence is not common." (See
1. EVidence of in-vivo hemolysis (te, red cell "Drug-Induced Immune Hemolytic Anemia"
destruction). If a patient with anemia who section.)
has a positive DAT result shows evidence of 4. History of hematopoietic progenitor cell or
hemolysis, testing to evaluate a possible organ transplantation. Passenger lympho-
immune etiology is appropriate. Reticulocy- cytes of donor origin produce antibodies
directed against ABO or other blood group example, after transfusion. The eluate prepa-
antigens on the recipient's red cells, causing ration can concentrate small amounts of IgG
a positive DAT result." that may not be detectable in routine testing
5. Administration of !VIC or intravenous (IV) of the patient's plasma.
anti-D. WIG may contain ABO antibodies,
antl-D, or sometimes other antibodies." W Results of these tests combined with the pa-
anti-D used to treat immune thrombocytope- tient's history and clinical data should assist in
nia (previouslyknown as "immune thrombo- classificationof the problem involved.
cytopenic purpura") causes Rh-positive
patients to develop apositive DAT result." Elution
6. Administration of potentially interfering Elution can be informative in the following situ-
therapeutic agents that may react with tar- ations:
get antigen on red cells. For example, anti-
CD38 administered to treat myeloma (ie, Clinical signs and symptoms of immune he-
daratumumab) causes reactivity with all red molysisare present.
cells because of the presence of low amounts Serum test results are negative or inconclu-
of CD38 on red cells.17,18 The DAT mayor sive for a patient who has recently received
may not be positive. A complete history (eg, transfusion.
HDFN is suspected but no alloantibodies
diagnosis, medications) is important in these
were detected in the maternal plasma.
cases.
Performing an elution routinely on the red
Serologic Investigation cells of all patients who have a positive DATre-
Three investigative approaches are helpful in the sult is not recommended. The majority of pre-
evaluation ofa positive DATresult. transfusion patients with a positive DAT result
have a nonreactive eluate that is often associat-
1. Test the DAT-positivered cells with anti-IgG ed with an elevated serum globulin level,"!'
Elution frees antibody from sensitized red
and anti-C3d reagents to characterize the
cells and recovers antibody in a usable form.
type of protein(s) coating the red cells. This
Multiple elution methods have been described
will help to classify an immune-mediated and reviewed." Many laboratories use commer-
hemolytic anemia. cial acid elution kits, primarily for ease of use
2. Test the serum/plasma to detect and identify and decreased exposure to potentially harmful
clinically significant antibodies to red cell chemicals; these kits are suitable to recover anti-
antigens. Additional tests that are useful in body in most cases. False-positiveeluate results
classifying the immune hemolytic anemias associated with high-titer antibodies have been
and procedures for detecting alloantibodies reported when the low-ionicwash solution sup-
in the presence of autoantibodies are plied with the commercial acid eluates was
described later in this chapter. used." Because no single elution method is ide-
3. Test an eluate prepared from the DAT-positive al in all situations, an alternative elution method
red cells with reagent red cells to determine (eg, an organic solvent) may be used in some
high-complexity reference laboratories when a
whether the coating protein has red cell anti-
nonreactive acid eluate result is not in agree-
body specificity.When the only coating pro- ment with clinicaldata."
tein is complement, the eluate is likely to be Table 14-2 lists the uses of some common
nonreactive. However, an eluate from the elution methods. Typically, eluates are tested
patient's red cells coated only with comple- only at the antiglobulin phase. If an IgM anti-
ment should be tested if there is clinical evi- body is being investigated or suspected, howev-
dence of antibody-mediated hemolysis, for er, centrifugation and reading after the 37 C
TABLE 14-2. Antibody Elution Methods
Lui freeze-thaw ABOHDFN Quick;small volume of red cells needed;poor

recoveryof other antibodies
Heat(56 C) ABOHDFN,IgM agglutinat- Easy;poor recoveryof IgGallo- and autoantibodies
ing antibodies
Acid elution kits Warm auto- and alloantl- Easy;possiblefalse-positiveeluate resultswhen
(commercial) bodies high-titer antibody is present"
Chemical/organic Warm auto- and alloantl- Chemicalhazards-eg, flammability, toxicity, or
solvent bodies carcinogenicity
HDFN= hemolytic diseaseof the fetus and newborn; IgM = immunoglobulin M.
incubation should be performed. Technical con- rum and the patient has not recently received a
siderations for elution are discussed in Chapter transfusion,no further serologictesting ofan auto-
13. antibody detected onlyin the eluate is necessary.
In cases of HTR or HDFN, specific antibody The patient's complete history, including the
(or antibodies) is usually detected in the eluate presence of potential passive antibodies, needs
that mayor may not be detectable in the serum. to be reviewed when the serologic test results
For transfusion reactions, newly developed anti- are evaluated. If both the serum and eluate are
bodies that are initially detectable only in the el- nonreactive, there is evidence of immune he-
uate are usually detectable in the serum after molysis, and the patient has received a drug
about 14 to 21 days." If the eluate is nonreac- reported to have caused immune-mediated he-
tive and a non-group-O patient has received molysis, testing to demonstrate drug-related an-
plasma containing anti-A or anti-B (as a result of tibodies should be considered. Finally,if the elu-
the transfusion of group a platelets, for exam-
ate is disproportionately weaker than the
ple) and the recipient appears to have immune
strength of the positive DAT (eg, eluate is 2+
hemolysis, the eluate should be tested against Al
but DAT is 4+), along with clinical evidence of
and/or B cells. It may be appropriate to test the
eluate against red cells from recently transfused immune hemolysis, and the patient is receiving
donor units, which could have caused immuni- a drug therapy known to causeimmune-mediated
zation to a low prevalence antigen on the donor hemolysis, drug-induced immune hemolytic
red cell. For cases of HDFN when no maternal anemia is also a possibility. (See "Laboratory In-
antibody has been detected and paternal red vestigation of Drug-Induced Immune Hemoly-
cells are ABO incompatible with maternal plas- sis" below.)
ma, testing an eluate prepared from the infant's
red cellswith the paternal red cells may detect a
maternally derived antibody to a low-prevalence UN H
antigen.
When the eluate reacts with all cells tested,
autoantibody is the most likely explanation, es- Immune hemolytic anemias can be classifiedin
pecially if the patient has not had a recent trans- various ways. One classification system is
fusion. However, if the patient has recently re- shown in Table 14-3. The AIHAsare subdivided
ceived a transfusion, an antibody to a high- into the major types: WAIHA, CAS, mixed- or
prevalence antigen should be considered. When combined-type AIHA, and paroxysmal cold
no unexpected antibodies are present in the se- hemoglobinuria (PCH). Other classification
TABLE14-3. Classification of Immune Warm Autoimmune Hemolytic Anemia

Hemolytic Anemias
The majority ofAIHAcases are caused by warm-
Autoimmune hemolytic anemia (AIHA) reactive autoantibodies that are optimally reac-
tive with red cells at 37 C. The autoantibody is
Warm AIHA usually IgG, but it can be IgM or IgA.
• Cold agglutinin syndrome
• Mixed-type AIHA Serologic Characteristics
@ Paroxysmal cold hemoglobinuria The DAT result may be positive because of IgG
plus complement (67% of cases), IgG without
Alloimmune hemolytic anemia complement (20%),or complement without IgG
II Hemolytic transfusion reaction (13%).6 Performing an elution at initial diagnosis
and/or during pretransfusion testing is useful to
Hemolytic diseaseof the fetus and newborn
demonstrate that the IgG coating the patient's
Drug-induced immune hemolytic anemia red cells is auto-reactive.
Typically, in WAIHA, the eluate is reactive
with virtually all red cells tested, and reactivity
is enhanced in tests against enzyme-treated red
schemes consider PCH as one of the cold-reactive cells, with polyethylene glycol (PEG) enhance-
AIHAs.Not all cases fit neatly into these catego- ment, or in column agglutination and solid-
ries. Table 14-4 shows the typical serologic char- phase tests. The eluate usually has no serologic
acteristics of the AIHAs.Drugs (discussed in the activity if the only protein coating the red cells is
complement.
"Drug-Induced Immune Hemolytic Anemia"
If the autoantibody has been adsorbed by the
section) may also induce immune hemolysis; patient's red cells in vivo, the serum may not
the effects of drug-induced autoantibodies are contain detectable free antibody. The serum
serologicallyindistinguishable from WAIHA. contains free antibody when the amount of
TABLE14·4. Typical Serologic Findings in AIHA
OAT IgG C3 only IgG + C3 C3 only

(routine) IgG + C3 C3
C3
Ig type IgG IgM IgG,lgM IgG
Eluate IgG antibody Nonreactive IgG antibody Nonreactive
Serum By IAT, 35% agglu- IgM agglutinating IgG IAT-reactive Negative routine IAT
tinate untreated red antibody, titer antibody plus IgM result, IgG biphasic
cells at 20 C ::::1000(60%) at agglutinating anti- hemolysin in Donath-
4 C, reactive at 30 C body reactive at Landsteiner test
30 C
Specificity Broadly reactive, Usually antl-I Usually unclear Antl-P
multiple specifici-
ties reported
AIHA = autoimmune hemolytic anemia;WAIHA = warm AIHA; CAS = cold agglutinin syndrome; PCH = paroxysmal cold
hemoglobinuria; DAT = direct antiglobulin test; IgG = immunoglobulin G; IgM = immunoglobulin M; IAT = indirect
antiglobulin test.
autoantibody exceeds the available binding sites Serologic Problems

on the patient's red cells; thus, serum autoanti-
body reactivity is "left over," that is, what was Warm autoantibodies can cause technical diffi-
not adsorbed by the patient's red cells in vivo. culties during red cell testing. Spontaneous ag-
The DATresult in such cases is usually strongly glutination can occur if the red cells are heavily
positive. coated with IgG and the reagent contains a po-
Autoantibody in the serum typically is reac- tentiator, such as albumin. This has been ob-
tive against all cells by an lAT. Approximately served when high-protein Rh typing sera are
60% of patients with WAIHAhave serum anti- used. If the control reagent provided by the
bodies that react with untreated saline-suspended manufacturer for these antisera is reactive, the
red cells.When tested with PEG, enzyme-treated typing is invalid. 19Gcan less commonly cause
red cells, column agglutination, or solid-phase spontaneous agglutination in lower-protein re-
methods, >90% of these sera can be shown to agents (eg, monoclonal typing sera); this reactiv-
contain autoantibody. Agglutination at room ity is often weaker or more fraglle than true ag-
temperature is present in about one-third of pa- glutination and may not be detected by a 6%
tients with WAIHA, but these cold agglutinins albumin control." Spontaneous agglutination
have normal titers at 4 C and are nonreactive at caused by red cells heavily coated with IgG is
30 C and 37 C. Thus, these cold agglutinins are less frequently observed.
nonpathogenic and the patient does not have Warm-reactive IgM agglutinins can also
CASin addition to WAIHA.6 cause spontaneous agglutination, resulting in
An unusual subcategory of WAIHAis associ- ABO and Rh typing problems and/or reactivity
ated with 19Magglutinins in the plasma that are with the negative control reagent for the DAT.23
reactive at 37 C.6,23This type of WAIHAis char- In these cases, treatment with dithiothreitol
acterized by severe hemolysis, and the prognosis (DTT)or 2-mercaptoethanol (2-ME)(Method 2-
for these patients can be poor. The red cells are 18) to disrupt the IgM agglutinin is required to
typicallyspontaneously agglutinated in the DAT; accurately interpret typing and DAT results.
that is, the washed red cells are reactive with all When the spontaneous agglutination is disrupt-
reagents tested, including a control, such as 6% ed, the control reagent is nonreactive.
albumin or saline. (See "Serologlc Problems" When the DAT result is positive due to 19G,
section.) Complement is usually detected on the antiglobulin-reactive typing reagents cannot be
red cells; 19Gor 19Mmayor may not be detect- used unless the red-cell-bound 19G is first re-
ed. 19Magglutinins are often detected in an elu- moved. (See Methods 2-20 and 2-21.) An alter-
ate (eg, acid) when it is inspected for agglutina- native is to use low-protein antisera (eg, mono-
tion after the 37 C incubation and before the clonal reagents) that do not require an
antiglobulin test is conducted. Some serum 19M antiglobulintest. (Referto the manufacturer's in-
warm autoagglutinins may be difficult to detect; structions for the detection of spontaneous ag-
some are enhanced in the presence of albumin glutination.) It is helpful to know which of the
or at low pH. Optimal reactivity of the aggluti- common red cell antigens are lacking on the pa-
nin sometimes occurs between 20 C and 30 C tient's red cellsto predict which clinicallysignifi-
rather than at 37 C. These antibodies have low cant alloantibodies the patient may have pro-
titers at 4 C, usually <64, which easily duced or may produce in the future. Antigens
differentiates this IgM warm autoantibody from absent from autologous cells could well be the
those in CAS.To prevent misinterpretation of ti- target of present or future alloantibodies. The
tration results, titrations at different tempera- patient's phenotype for the common antigens
tures (eg, 37 C, 30 C, room temperature, and can be determined serologlcallyor predicted us-
4 C) need to be carried out with separate sets of ing DNA-basedmethods.
tubes to avoid carryover agglutination.v" Test- The presence of autoantibody in the serum
ing for the presence of a warm hemolysin can increases the complexity of the serologlc evalua-
sometimes define the AIHAas consistent with a tion and the time needed to complete pretrans-
warm IgMAIHA.23 fusion testing. If a patient who has warm-
reactive autoantibodies in the serum needs a rum contains high levels of autoantibody. Once
transfusion, it is important to determine wheth- autoantibody has been removed, the adsorbed
er alloantibodies are also present. Some alloantl- serum is tested for alloantibodyreactivity.
bodies may make their presence known by re- Autologous adsorption is not recommended
acting more strongly or at different phases than for patients who have received transfusion with-
the autoantibody, but quite. often, routine testin the last 3 months because a blood sample
ing maynot suggest the existence of masked al- may contain some transfused red cells that
loantibodies.25,26 might adsorb alloantibody. Red cells normally
Methods to detect alloantibodiesin the pres- survive for about 110 to 120 days. In patients
ence of warm-reactive autoantibodies are used with AIHA, autologous and transfused red cells
to attempt to remove, reduce, or circumvent the can be expected to have shortened survival.
autoantibody. Antibody detection methods that However, determining how long transfused red
use PEG, enzymes, column agglutination, or cells remain in circulation in patients who need
solid-phase red cell adherence usually enhance repeated transfusions is notfeasible. It has been
autoantibodies. Antibody detection tests using demonstrated that very small amounts (<10%)
low-ionic-strength saline (USS) or saline tube of antigen-positive red cells are capable of re-
methods may not detect autoantibodies but they moving alloantibody reactivity in in-vitro stud-
do detect most clinicallysignificant alloantlbodies.28 Therefore, it is recommended to wait for 3
ies. Other procedures involve adsorption; two months after transfusion before performing au-
widely used adsorption approaches are dis- tologous adsorptions.
cussed below.
Adsorption with Allogeneic Red Cells
Adsorption with Autologous Red Cells
The use of allogeneic red cells for adsorption (al-
In a patient who has not recently received trans- logeneic adsorption) may be helpful when the
fusion, adsorption with autologous red cells (au- patient has recently received transfusion or
tologous adsorption; see Method 4-8) is the best when insufficient autologous red cells are avail-
way to detect alloantibodies in the presence of able. The goal is to remove autoantibody and
warm-reactive autoantibodies. Only autoanti- leave the alloantibody in the adsorbed serum.
bodies are removed, and alloantibodies, if pres- The adsorbing red cells must not have the anti-
ent, remain in the serum. gens against which the alloantibodies are reac-
Autologous adsorption typically requires tive. Because the alloantibody specificity is un-
some initial preparation of the patient's red cells. known, red cells of different phenotypes are
At 37 C, in-vivoadsorption has occurred, and all usually used to adsorb several aliquots of the pa-
antigen sites on the patient's own red cells may tient's serum.
be blocked. A gentle heat elution at 56 C for 3 Given the number of potential alloantibodies,
to 5 minutes can dissociate some of the bound the task of selecting the red cells may appear for-
IgG. This can be followed by treatment of the midable. However, red cell selection is based
autologous red cellswith proteolytic enzymes to only on those few antigens for which alloanti-
increase their capacity to adsorb autoantibody. bodies of clinical Significance are likely to be
(Treatment with proteolytic enzyme alone does present. These include the common Rh antigens
not remove IgG coating the red cells.) Treat- (D, C, E, c, and e), K, Ft and Fyb,Jka and Jkb,
ment of the red cellswith ZZAP,a mixture of pa- and S and s. Red cell selection is made easier by
pain or ficin and DTT, accomplishes both of the fact that some of these antigens can be
these actions in one step. It is proposed that the destroyed by appropriate pretreatment (eg, with
sulfhydryl component makes the IgG molecules enzymes or ZZAP)before use in adsorption pro-
more susceptible to the protease and dissociates cedures. (See Chapter 13, Table 13-5.)Antibod-
the antibody molecules from the cell." Multi- ies to high-prevalence antigens cannot be ex-
ple sequential autologous adsorptions with new cluded by allogeneic adsorptions because the
aliquots of red cells may be necessary if the se- adsorbing red cells are expected to express the
antigen and adsorb the alloantibody along with only need to be E-, Jk(a-) because the 2-ME or
autoantibody. DTT in llAP denatures K, and ficin or papain
When the patient's phenotype is not known, denatures Fy". Caution should be used when
group 0 red cell samples of three different Rh matching red cellswith the patient's phenotype,
phenotypes (RjRj,~R2' and rr) should be select- as the patient may possess variant antigens; an
ed. (See Method 4-9.) One sample should lack antibody directed against variant antigen speci-
Ik", and another, Jkb. As shown in Table 14-5, ficities not expressed by the patient may be ad-
llAP or enzyme pretreatment of the adsorbing sorbed out and not detected.
red cells reduces the phenotype requirements. Adsorption using untreated red cells in the
Untreated red cells may be used, but the adsorb- presence of PEG (Method 4-10) or LISS29,30is a
ing red cells must include at least one sample modification that has been used to decrease the
that is negative for the S, s, pt, Fy", and K anti- incubation time for adsorptions and increase ef-
gens in addition to the Rh and JK (Kidd)pheno- ficiency. Adsorptions performed with the addi-
type requirements stated above. tion of PEG have been reported to result in un-
If the patient's phenotype is known or can be expected nondetection of alloantibodiesin some
determined, adsorption with a single sample of cases.31 ,32
red cells may be possible. Red cells can be se-
lected that match the patient's phenotype or at Testing of Adsorbed Serum
least match the Rh and JK phenotypes if llAP
treatment is used. For example, if a patient's In some cases, each aliquot of serum may need
phenotype is E-, K-, S-, Fy(a-), Jk(a-), untreat- to be adsorbed two or three times to remove the
ed adsorbing red cells need to lack all five anti- autoantibody. The fully adsorbed aliquots are
gens, but enzyme-treated red cells only need to then tested against reagent red cells known to
be E-, K-, Jk(a-) because enzyme treatment de- either lack or carry common antigens of the RH,
natures Fy"antigen, and ZZAP-treatedred cells MNS, KEL,FY, and JK blood group systems (eg,
TABLE 14-5. Selection of Red Cells for Allogeneic Adsorption

Step 1. Select red cells for each Rh phenotype.
Step 2. Onthe basis of the red cell treatment or lack of treatment (below), at least one of the Rh-phenotyped
cells should be negativefor the antigens listed below.
ZZAP-Treated Red Cells Enzyme-Treated Red Cells Untreated Red Cells
Jk(a-) Jk(a-) Jk(a-)

Jk(b-) Jk(b-) Jk(b-)
K- K-
Fy(a-)
Fy(b-)
S-
s-
antibody detection cells). If an adsorbed aliquot Specificity of Autoantibody

is reactive, the aliquot should be tested to identi-
In manycasesofWAIHA, no autoantibody spec:
fy the antibody. Adsorbing several aliquots with ificity is app~,epL The patient's serum reacts
different red cell samples provides a battery of with allred cell samples tested. If testing is per-
potentially informative specimens. For example, formed with cells of rare Rh phenotypes, s1.lchas
if the aliquot adsorbed with Jk(a-) red cells sub- D- - or RhnuIv, sorne ,autoanjjbodies are vvealdy
sequeptlyis reactive only with Jk(a+) red cells, reactive or ~e"nonreactive, and the autoanti-
the presence of alloanti-Ik' can be inferred confi- body appears to have broad specificityin the RH
dently. system. Apparent specificityfor simple Rh anti-
Sometimes, autoantibody is not completely gens (D, C, E, c, and e)is occasionally:seen, es-
removed by three sequential adsorptions. Addi- peciallyin salineor LISSIATs.A. "relative" speci-
tional adsorptions can be performed, but the fiCitybased on stronger reactivity with cells of
certain phenotypes may also be seen; relative
performance of multiple adsorptions has the po-
specificitymay alsobe apparent after adsorption.
tential to dilute the serum. If the adsorbing cells Autoantibody specificities can be clearer in the
do not appear to remove the antibody, the auto- serum than in the eluate.
antibody may have an unusual specificitythat is Apart from RHspeclfidty, warm autoantibod-
not reactive with the red cells used for adsorp- ies with many other specificitieshave been re-
tion. For example, autoantibodies with KEL, ported leg, Specificitiesin the LW, KEL,JK, FY,
LW, or EnaFSspecificity are not removed by andDI systems).34,35 Patients with autoantibod-
ZZAP-treated red cells. (See Table 14-5 for ies of KEL,RH, LW, GE, SC, LU,and LANspeci-
which antigens will be expressed on treated ad- ficities may have transiently depressed expres-
sorbing cells.) The possibility that the sample sion of the respective antigen, and the DAT
result may be negative or very weakly positive."
contains an auto- or alloantibody to a high-
In these cases, the autoantibody may inltially ap-
prevalence antigen should always be consid- pear to be alloantibody. Proof that it is truly au-
ered when adsorption fails to remove the reac- toantibody is demonstration that after the anti-
tivity. body and hemolytic anemia subside, the antigen
Autoantibodies sometimes have patterns of strength returns to normal, and stored serum
reactivity that suggest the presence of alloanti- containing antibody is reactive with the patient's
body. For example, the serum of a D- patient red cells.
may have apparent anti-C reactivity. The anti-C Tests against red cells of a rare phenotype
reactivity may reflect warm-reactive autoanti- and by specialtechniques to determine autoanti-
body even if the patient's red cells lack C. The body specificityhave limited clinical or practical
apparent alloanti-C would, in this case, be ad- application. If the autoantibody reacts with all
sorbed by C- red cells, both autologous and allo- red cells except those of a rare Rh phenotype
(eg, RhnuIl), compatible donor blood is unlikely
geneic. This is unlike the behavior ora true allo-
to be available. Such blood, if available, should
anti-C, which would be adsorbed only by C+ be reserved 'for alloimmunized patients of that
red cells. In one study, the serum adsorbed with uncommon phenotype.
autologous red cells often retained autoantibod-
ies that mimicked alloantibodies in addition to Selection of Blood for Transfusion
the true alloantibody(ies) present, whereas se-
The most important consideration is to exclude
rum adsorbed with allogeneicred cells most of-
the presence of potentially clinlcally significant
ten contained only alloantibodies." This re- alloantibodies before selecting Red Blood Cell
flects an inefficiency of autologous adsorption (RBC)units for transfusion. There are multiple
that is caused primarily by limited volumes of reports in the literature demonstrating that pa-
autologousred cells availablefor removing all of tients who have warm autoantibodies in their
the autoantibody reactivity from the serum. sera have a higher rate of alloimmunizatton (eg,
12% to 40%, with a mean of 32%).25,36-39 Al- gen is debatable. In the absence of hemolysis or
though these patients present a serologic chal- evidence of compromised survival of transfused
lenge, they deserve the same protection from cells, autoantibody specificity is not important.
HTRs as any other patient. Autoantibodies that However, donor units that are negative for the
react with all reagent red cells, even weakly, are antigen may be chosen because this is a simple
capable of masking alloantibody reactivity (ie, way to circumvent the autoantibody and detect
reactivity of red cells with both alloantibodyand potential alloantibodies.
autoantibody may not be any stronger than with It may be undesirable to expose the patient
autoantibody alone).25,26 to Rh antigens absent from autologous cells, es-
It is the exclusion of newly formed alloanti- pecially D and especiallyin females of childbear-
bodies that is of concern. Because of the pres- ing potential, merely to improve serologic com-
ence of autoantibodies, all crossmatches are in- patibility testing results with the autoantibody.
compatible. This is unlike the case of clinically (Forexample, when a D- patient has autoanti-e,
Significant alloantibodies without autoantibod- available e- units are likely to be D+; D-e-
ies, where a compatible crossmatchwith antigen- units are extremely rare.) Referral of the sample
negative red cells is possible. Monitoring for evi- for molecular investigation to determine the risk
dence of red cell destruction caused by slloenti- for allo- or autoantibody production will aid in
bodies is difficult in patients who already have decision-makingin these complex cases and po-
AIHA; these patients' own red cells and trans- tentially improve patient care.
fused red cellshave shortened survival. Some laboratories use the adsorbed serum to
If no alloantibodies are detected in adsorbed screen and select nonreactive units (units that
serum nor are previously identified, random are antigen-negative for clinically significant al-
units of the appropriate ABO group and Rh type loantibodies, if detected) for transfusion. Other
may be selected for transfusion unless there are laboratories do not perform a crossmatch with
contrary indications in the patient file. If clinical- the adsorbed serum because all units will be in-
ly significant alloantibodies are present, the compatiblein vivo due to the autoantibody. Issu-
transfused cells should lack the corresponding ing a unit that is serologically compatible with
antigen(s). For patients facinglong-term transfu- adsorbed serum may provide some assurance
sion support, it is prudent to obtain an extended that the correct unit has been selected and avoid
phenotype or predicted phenotype by genotyp- incompatibility because of additional antibodies
ing. Consideration can then be given to transfu- (eg, antl-Wr'), but this practice can also provide
sion with donor units that are antigen-matched a false sense of security about the safety of the
for clinically significant blood group antigens to transfusion for these patients.
avoid additional alloimmunization and poten- A transfusion management protocol using
tially decrease the number of adsorptions re- prophylactic antigen-matched units for patients
quired and the complexity of pretransfusion with warm autoantibodies, where feasible, in
workup. combination with streamlined adsorption proce-
If the autoantibody has clear-cut specificity dures has been described." The same antigens
for a single antigen (eg, anti-e) and active hemol- for the commonly occurring, clinically Signifi-
ysis is ongoing, blood lacking that antigen cant antibodies (D, C, E, c, e, K, Ft, Fyb, Jka,
should be selected according to the patient's his- Jkb, S, and s) are taken into account, as dis-
tory, availabilityof the unit, and medical obser- cussed in the previous section on adsorptions.
vation. A good practice is to verify the allele to The ability to implement such a protocol de-
avoid alloimmunization. There is evidence that pends on the ability of the transfusion service,
such transfused red cells survive longer than the and more often the blood supplier, to maintain
patient's own red cells." If the autoantibody an adequate inventory of phenotyped units to
shows broader reactivity-reacting with all cells meet the antigen-matching needs. In recent
but showing some relative specificity(eg, prefer- years, molecular technologies have been applied
entially reacting with e+ red cells), whether to to red cell genotyping for patients with warm
transfuse blood lacking the corresponding anti- autoantibodies to determine which common al-
C HAP T E R 1 4 OAT/Immune Hemolysis 441
loantibodies the patient can make. DNA tests The transfusion of patients with AIHA is a
are attractive for determining the predicted phe- clinicaldecision that should be based on the bal-
notype of patients with a positive DAT (IgG) re- ance between the risks and clinical need. Trans-
sult because IgG is not always successfully re- fusion should not be withheld solely because of
moved and some red cell antigens are sensitive serologicipcpmpatibility.The volume transfused
to IgG removal treatment.i'r" Recent transfu- should usually be the smallest amount required
sions do not interfere with molecular testing. It to maintain adequate oxygen delivery and not
must be remembered that genotyping may not necessarily the .amount required to reach an ar-
accurately predict the phenotype if uncommon bitrary hemoglobin level." The patient should be
or rare silencing mutations are present or the pa- carefullymonitored throughout the transfusion.
tient has received a stem cell transplant.
Some experts propose that an electronic OAT-NegativeAIHA
crossmatch can be safely used for patients with
autoantibodies when the presence of common, Clinical and hematologic evidence of WAIHAis
clinically significant alloantibodies has been ex- present in some patients whose DAT result is
cluded.43,44.This approach circumvents the need negative. The most common causes of AIHAas-
to issue units that are labeled "incompatible"; sociated with a negative DATresult are red-cell-
however, as discussed above, this practice can bound IgG below the detection threshold of the
also lead to a false sense of security. antiglobulin test, red-cell-bound IgM arid 19A
Although resolving serologic problems for that are not detectable by routine AHG re-
these patients is important, delaying transfusion agents, and low-affinityIgG that is washed off
in the hope of finding serologically compatible the red cells during the washing phase for the
blood may, in some cases, cause greater danger DAT.6,45
to the patient. Only clinical judgment can re- Nonroutine tests can be applied in these situ-
solve this dilemma; therefore, having a dialogue ations. Unfortunately, these assays require stan-
with the patient's physician is important. dardization and many have a low predictive val-
ue. One of the easier tests is for low-affinity
Transfusion of Patients with Warm-Reactive antibodies. Washing with ice-cold (eg, 4 C) sa-
Autoantibodies line or LISSmay help retaln antibody on the
Patients with warm-reactive autoantibodies may cells; a control (eg, 6% albumin) is necessary to
have no apparent hemolysis or may have life- confirm that cold autoagglutinins are not caus-
threatening anemia. Patients with little or no ev- ing the positive results.v" Methods that have
idence of significanthemolysis tolerate transfu- been used to detect lower levels of red-cell-
sion quite well. The risk of transfusion is some- bound IgG include the complement fixation an-
what increased in these patients because of the tibody consumption assay, the enzyme-linked
difficultieswith pretransfusion testing. The du- antiglobulin test, radiolabeled anti-IgG, flow cy-
ration of survival of the transfused red cells is tometry, solid-phasetesting, the direct PEG test,
about the same as that of the patient's own red the. direct Polybrenetest,. column agglutination,
cells. and concentrated eltlates.43 . .
In patients with active hemolysis, transfusion Anti~IgG,anti-C3d, and the combined anti-
may increase hemolysis, and the transfused red C3b,-C~d reagents are the only licensed prod-
cells may be destroyed more rapidly than the pa- ucts available in the United States for use with
tient's own red cells. This is related to the in- human red cells. AHG reagents that react with
creased red cell mass avallablefrom the transfu- IgA or IgM are avallable commercially but prob-
sion and the kinetics of red cell destruction." ably have not been standardized for Use with
Destruction of transfused cells may increase he- red cells in agglutination tests. They must be
moglobinemia and hemoglobinuria. Disseminat- used cautiously, and their hemagglutination re-
ed intravascular coagulation can develop in pa- activity must be carefully standardized by. the
tients with severe posttransfusion hemolysis. user." Outside the United States, AHG reagents
for the detection of IgM and IgAin tube tests or can sometimes be demonstrated at 20 C to 25
column agglutination tests may be available. C. Except in rare cases with anti-Pr speciftctty,
enzyme-treated red cells are hemolyzed in the
Cold Agglutinin Syndrome presence of adequate complement.
To determine the true thermal amplitude or
CAS,which is less common than WAIHA,is the
titer of the cold autoagglutinin, the specimen is
hemolytic anemia that is most commonly associ-
collected and maintained strictly at 37 C until
ated with autoantibodies that react preferential·
the serum and red cells are separated to avoid
ly in the cold. CASoccurs as an acute or chronic
in-vitro autoadsorption. Alternatively, plasma
condition. The acute form is often secondary to
can be used from an EDTA-anticoagulatedspeci-
Mycoplasma pneumoniae infection. The chron-
ic form is often seen in elderly patients and is men that has been warmed for 10 to 15 minutes
sometimes associated with lymphoma, chronic at 37 C (with repeated mixing) and then sepa-
lymphocytic leukemia, or Waldenstrom macro- rated from the cells, ideally at 37 C. This pro-
globulinemia. Acrocyanosis and hemoglobinuria cess should release autoadsorbed antibody back
may occur in cold weather; thus, patients into the plasma.
should be advised to avoid cold. CAS is often In chronic CAS, the IgM autoagglutinin is
characterized by agglutination, at room tern- usually a monoclonal protein with kappa light
perature, of red cells in an EDTA specimen, chains. In the acute form induced by Mycoplas-
sometimes to the degree that the red cells ap- ma or viral infections, the antibody is polyclonal
pear to be clotted. IgM with normal kappa and lambda light-chain
distribution. Rare examples of IgAand IgG cold-
Serologic Characteristics reactive autoagglutinins have also been de-
scribed.?
Complement is the only protein detected on red
cells in almost all cases of CAS. If the red cells Serologic Problems
have been collected properly and washed at
37 C, there will be no immunoglobulin on the Problems with ABO and Rh typing and other
cells and no reactivity in the eluate. If other pro- tests are not uncommon. Often, it is only neces-
teins are detected, a negative control for the sary to maintain the blood sample at 37 C im-
DAT (eg, 6% albumin or saline) should be tested mediately after collection and to wash the red
to ensure that the cold autoagglutinin is not cells with warm (37 C) saline before testing. Al-
causing a Ialse-positlveresult. The cold-reactive ternatively, an EDTAsample can be warmed to
autoagglutinin is usually IgM, which binds to 37 C for about 10 minutes, after which the red
red cells in the lower temperature of the periph- cells are washed with warm saline. It is helpful
eral circulation of an individual and causes com- to perform a parallel control test with 6%bovine
plement components to attach to the red cells. albumin to determine whether autoagglutina-
As the red cells circulate to warmer areas, the tion persists. If the control test result is nonreac-
IgM dissociatesbut the complement remains. tive, the results obtained with anti-A and antl-B
IgM cold-reactive autoagglutinins associated are usually valid. If autoagglutination still oc-
with immune hemolysis usually react at 30 C, curs, it may be necessary to treat the red cells
and 60% have a titer of :2:1000 when tested at with sulfhydryl reagents. Because cold-reactive
4 C.6 If 22% to 30% bovine albumin is included autoagglutinins are almost always IgM and sulf-
in the test system, pathologic cold agglutinins hydryl reagents denature IgM molecules, re-
will react at 30 C or 37 C.6 On occasion, patho- agents (such as 2-ME or DTT) can be used to
logic cold agglutinins have a lower titer (ie, abolish autoagglutination. (See Method 2-18.)
<1000), but they have a high thermal amplitude The red cells can also be treated with ZZAPre-
(ie, reactive at 30 C with or without the addi- agent, as in the preparation for adsorptions. (See
tion of albumin). The thermal amplitude of the Method 4-8.)
antibody has greater significance than the titer. When the serum agglutinates group 0 re-
Hemolytic activity against untreated red cells agent red cells, ABOserum tests are invalid. Re-
peating the tests using prewarmed serum and ly associated with infectious mononucleosis. On
group AI, B, and 0 red cells and allowing the rare occasions, other specificities are seen.
red cells to "settle" after incubation at 37 C for Autoantibody specificity is not diagnostic for
1 hour (instead of centrifuging the sample) often CAS. Autoanti-I may be seen in healthy individ-
resolves any discrepancy. (See Method 2"11") By uals as well as in patients with CAS. The non"
eliminating the centrifugation step, interference pathologlc forms. of autoanti-I, however, rarely
by cold-reactive autoantibodies might be avoid" react at titers above 64 at 4 C and are usually
ed. Weak anti-A end/or-B in some patients' sera nonreactive with h(cord i and adult i) red cells
may not react at 37 C. Alternatively, adsorbed at room temperature. In contrast, the autoanti-I
serum (either autoadsorbed or adsorbed with al- of CASmay react quite strongly with 1- red cells
logeneic group 0 red cells) can be used. Serum in tests at room temperature, and equal or even
adsorbed with rabbit erythrocyte stroma should stronger reactions occur with 1+ red cells. AtM"
not be used for ABO serum tests because anti"B anti-I reacts in the opposite manner, demonstrat-
and anti-Al' may be removed.46,47 ing stronger reactions with 1- red cells than
with red cells that are 1+. Anti-I', originally
Detection of Alloantibodies in the Presence of thoughtto recognize a transition state of i to I
Cold-Reactive Autoantibodies (explaining the designation "IT"), reacts strong"
ly with cord red cells, weakly with normal adult
Cold-reactive autoagglutinins rarely mask clint- I red cells; and most weakly with the rare adult i
cally significant alloantibodies if serum tests are red cells. In rare cases, the cold agglutinin speci-
conducted at 37 C and IgG"specificreagents are ficity may be anti-Pr, which reacts equally well
used for the antiglobulin phase. The use of po- with untreated red cells of I or i phenotypes but
tentiators (eg, albumin or PEG) is not recom- does not react with enzyme-treated red cells.
mended because they may increase the reactivi- Procedures to determine the titer and specificity
ty of the autoantibodies. In rare instances, it of cold-reactive autoantibodies are given in
may be necessary to perform autologous adsorp- Methods 4"6 and 4" 7. Typical reactivity patterns
tion at 4 C. (See Method 4"5.) Achieving the of cold autoantibodies are shown in the table in
completeremoval of potent cold-reactive auto" Method 4"6.
agglutinins is very time-consuming and usually
unnecessary. Sufficient removal of cold autoag- Mixed-TypeAIHA
glutinins may be facilitated by treating the Although about one-third of patients with
patient's cells with enzymes or ZZAP before WAlHA have nonpathologic IgM antibodies that
adsorption. One or two cold autologous adsorp- agglutinate at room temperature, another group
tions should remove enough autoantibody to of patients with WAlHA have cold agglutinins
make it possible to.detect alloantibodies at 37 C that react at or above 30 C. This latter group is
that were otherwise masked by the cold-reactive referred to as having "mixed" or "combined
autoantibody. As an alternative, the allogeneic warm and cold" AIHA and can be subdivided
adsorption process used for WAIHA can be per" into patients with hlgh-ttter, high-thermal-
formed at 4 C. Rabbit erythrocyte stroma, amplitude IgM cold antibodies (the rare WAIHA
which removes autoantl-I and "IH from sera, plus classic CAS) and patients with normal-titer
should be used with caution because this meth- (<64 at 4 C), high-thermal-amplitude cold anti"
od can remove clinically significant alloantibod- bodies.5o·52 Patients with mixed-type AlBA often
ies-notably antl-D, "E, and "Vel, and IgM anti" present with hemolysis and complex serum re-
bodies regardless of blood group specificity.48,49 activity in all.phases of testing.
Specificity of Autoantibody Serologic Characteristics

The autoantibody specificity in CAS is most of. In mixed-type AIHA, both IgG and C3 are usual"
ten anti-I but is usually of academic interest on" ly detectable on patients' red cells; however, C3,
ly. Anti-i is found less commonly, and it is usual" IgG, or IgA alone may be detectable on the red
cells." An eluate contains a warm-reactive IgG for the DAT are usually coated only with com-
autoantibody. Both warm-reactive IgG autoanti- plement, but IgGmay be detectable on cells that
bodies and cold-reactive, agglutinating IgM have been washed with cold saline and tested
autoantibodies are present in the serum. These with cold anti-IgG reagent." Keeping the test
autoantibodies usually result in reactivity at all system close to its optimal binding temperature
phases of testing and with virtually all cells test- allows the cold-reactive IgG autoantibody to re-
ed. The IgM agglutinating autoantibody reacts main attached to its antigen. Because comple-
at 30 C or above. If adsorptions are performed ment components are usually the only globulins
to detect alloantibodies, it may be necessary to present on circulating red cells, eluates prepared
perform them at both 37 C and 4 C. from the red cells of patients with PCH are al-
most always nonreactive.
Specificity of Autoantibodies The IgG autoantibody in PCH is classically
described as a biphasic hemolysin because bind-
The unusual cold-reactiveIgM agglutinating au- ing to red cells occurs at low temperatures, but
toantibody can have specificitiesthat are typical hemolysis does not occur until the complement-
of CAS (ie, antt-I or -i)but often has no apparent coated red cells are warmed to 37 C. This is the
specificity.50,5!The warm-reactive IgG autoanti- basis of the diagnostic test for the disease, the
body often appears to be serologicallyindistin- Donath-Landsteiner test. (See Method 4-11.)
guishable from autoantibodies encountered in The autoantibody may agglutinate normal red
typicalWAIHA. cells at 4 C but rarely to titers >64. Because the
antibody rarely reacts above 4 C, pretransfusion
Transfusion for Patients with Mixed-Type AIHA antibody detection tests are usually nonreactive,
and the serum is usually compatible with ran-
If blood transfusions are necessary, the consider-
dom donor cells by routine crossmatch proce-
ations for the exclusion of alloantibodiesand the dures.
selection of blood for transfusion are identical to
those described for patients with acute hemoly-
Specificity of Autoantibody
sis caused by WAIHAand CAS.(See above.)
The autoantibody of PCH has most frequently
Paroxysmal Cold Hemoglobinuria been shown to have P specificity.The autoanti-
body reacts with all red cells by the Donath-
PCH is the rarest form of DAT-positiveAIHA. Landsteiner test (includingthe patient's own red
Historically, PCH was associated with syphilis, cells), except those of the very rare p or pk phe-
but this association is now unusual" More notypes.
commonly, PCH presents as an acute transient
condition that is secondary to a viral infection, Transfusion for Patients with PCH
particularly in young children. In such cases, the
biphasic hemolysin may be only transiently de- Transfusion is rarely necessary for adult patients
tectable. PCH can also occur as an idiopathic with PCH, unless their hemolysis is severe. In
chronic disease in older people. young children, the thermal amplitude of the
antibody tends to be much wider than in adults
Serologic Characteristics and hemolysis is often more brisk, so transfusion
may be required as a lifesaving measure. Al-
PCHis caused by a cold-reactiveIgGcomplement- though there is some evidence that p red cells
binding antibody. As with IgM cold-reactive au- survive longer than P (P1+ or P 1-) red cells, the
toagglutinins, reactivity occurs with red cells in prevalence of p blood is approximately 1 in
colder areas of the body (usually the extremi- 200,000, and the urgent need for transfusion
ties) and causes C3 to bind irreversibly to red usually precludes attempts to obtain this rare
cells. The antibody then dissociates from the red blood. Transfusion of donor blood should not be
cells as the blood circulates to warmer parts of withheld from patients with PCH whose need is
the body. Red cells washed in a routine manner urgent. Transfusion of red cells that are negative
for the P antigen should be considered only for Theoretical Mechanisms of Drug-
those patients who do not respond adequately Induced Antibodies
to randomly selected units of donor blood."
Numerous theories have been suggested to ex-
plain how drugs induce immune responses and
what relation such responses may have to the
positive DAT result and immune-mediated cell
destruction observed in some patients." For
Drugs rarely cause immune hemolytic anemia; many years, drug-associated positive DAT re-
the estimated incidence is I in 1 million peo- sults were classified by four mechanisms: drug
ple.54Many drugs have been implicated in he- adsorption (penicillin-type), immune complex
molytic anemia over the years, as can be seen in formation, autoantibody production, and NIPA.
the list provided in Appendix 14-1 and reviewed This classification has been useful serologically,
elsewhere." but many aspects lack definitive proof. In addi-
Drugs sometimes induce the formation of an- tion, some drugs demonstrate serologic reactivi-
tibodies against the drug, red cell membrane ty that appears to involve more than one mecha-
components, or an antigen formed by the drug nism. A more comprehensive approach, termed
and the red cell membrane. These antibodies a "unifying hypothesis," is shown in Fig 14-1.
may cause a positive DAT result, immune red One or more populations of antibodies may be
cell destruction, or both.14,54In some instances, present. In addition, NIPA, which is indepen-
a positive DAT result can be caused by nonim- dent of antibody prcduction, appears to playa
munologic protein adsorption (NIPA) onto the role in drug-induced immune hemolytic ane-
red cell, which is caused by the drug." mia."
ANTIBODY
TODR~;$
ANTIBODY TO
(MAINLY) MEMBRANE DRUG
COMPONENTS
.~
~~
REDCEll MEMBRANE
FIGURE14-1. Proposed unifying theory of drug-induced antibody reactions (based on a cartoon by
Habibi as cited by Garratty"), The thicker lines represent antigen-binding sites for the Fab region of
the drug-induced antibody. Drugs (haptens) bind loosely or firmly to cell membranes, and antibodies
may be made to 1) the drug [producing in-vitro reactions typical of a drug adsorption (penicillin-
type) reaction); 2) membrane components or mainly membrane components (producing in-vitro
reactions typical of autoantibody); or 3) part-drug, part-membrane components (producing an in-
vitro reaction typical of antibody reactive in the presence of a drug).34(P55)
Serologic Classification action unless the patient also has alloantibodies

to red cell antigens present on these cells).
Drug-induced antibodies can be classified into
Penicillin and the cephalosporins are beta-
two groups: drug dependent (those that require
lactam antibiotics. It was thought for some time
the presence of the drug in the test system to be
detected) and drug independent (those that do that antibodies to any drug in the penicillin and
not require the in-vitro addition of the drug for cephalosporin familiescould be detected by test-
detection]." Drug-dependent antibodies are sub- ing red cells with drug-treated cells using meth-
divided into those that react with drug-treated ods previously described for penicillin and ceph-
red cells (eg, antibodies to penicillin and some alothin. It is now known that this is not the
cephalosporins) and those that react with un- case. Synthetic penicillins and newer cephalo-
treated red cells in the presence of a solution of sporins cannot be assumed to have the same
the drug (eg, antibodies to quinine and ceftriax- red-cell-bindingcharacteristics as penicillin and
one). Drug-independent antibodies (eg, autoan- cephalothin (a first-generation cephalosporin).
tibodies induced by methyldopa and fludarabine) Cefotetan (a second-generation cephalosporin)
have serologic reactivity that is independent of binds very well to red cells, and antibodies
the drug despite the fact that the drug originally caused by cefotetan typicallyreact to very high
induced the immune response. Because the titers with cefotetan-treated red cells. However,
drug does not need to be added to the test sys- ceftriaxone (a third-generation cephalosporin)
tem, drug-independent antibodies behave like does not bind well to red cells; therefore, anti-
autoantibodies that are serologically indistin- bodies to ceftriaxone cannot be tested by this
guishable from idiopathic warm autoantibodies. method." Piperacillin, a semisynthetic penicil-
If a patient is suspected of having DIIHA,the lin, binds to red cells at high pH. However, a
suspected drug should be stopped. Laboratory large percentage of plasma from healthy blood
testing to detect drug-dependent antibodies can donors and patients reacts with piperacillin-
be performed, but DIIHA caused by drug- treated red cells, so this method is not recom-
independent antibodies or NIPA can be suggest- mended for testing for piperacillin antibodies."
ed only by showing a temporal association of the For drug-dependent antibodies detected using
drug administration and hemolysis. drug-treated red cells, the following are expect-
The historical details of penicillin- and ed:
methyldopa-induced antibodies are not de-
scribed in this chapter but have been extensive- The DAT result is usually positive for IgG,
ly reviewed elsewhere.s" DIIHA caused by but complement may also be present.
high-dose intravenous penicillin therapy is no Serum contains an antibody that reacts with
longer seen, and methyldopa, the prototype for drug-treated red cells but not untreated red
drug-independent antibodies, is not used as fre- cells.
quently as in the past. Currently, the drugs most Antibody eluted from the patient's red cells
commonly associated with DIIHA are piperadl- reacts with drug-treated red cells but not un-
lin, ceftriaxone, and cefotetan; there has also treated red cells.
been a small increase in DIIHA caused by drugs
in the platinum family.14 Hemolysis develops gradually but may be
life-threatening if the etiology is unrecognized
Drug-Dependent Antibodies Reactive with and drug administration is continued. The pa-
Drug-Treated Red Cells tient mayor may not have been previously ex-
Some drugs (eg, penicillin, ampicillin, and many posed to the drug, and in the case of cefotetan,
cephalosporins) covalently bind to red cells, even only a single dose given prophylactically
thus making it possible to coat red cells in the can result in severe hemolysis. Normal plasma
laboratory with the drug. Antibodies directed to has been shown to react with some drug-treated
these drugs will react with the drug-treated red red cells (eg, red cells treated with cefotetan,
cells but not with untreated red cells (ie, no re- piperacillin, or oxallplatinj." suggesting prior
exposure to these drugs through environmental pared from .the patient's .red. •cells, and
routes. autoantibody in the serum persists. Consequent-
ly, DIIHAcan be misdiagnosed as WAIHA,espe-
Drug-Dependent Antibodies Reactive with cially if the eluate reacts. Differentiation of
Untreated RedCellsin the Presenceof Drug warm-reactive autoantibody from DIIHA is im-
Antibodies to many drugs that have been report· portant for clinical management.'"
ed to cause immune hemolytic anemia are de-
Drug-Independent Antibodies: Autoantibody
tected by testing untreated red cells in the pres-
ence of the drug. Piperacillin and some of the Production
second- and third-generation cephalosporins re- Some drugs induce autoantibodies that appear
act by this method; anti-ceftriaxone has been de- serologically indistinguishable from those of
tected only by testing red cells in the presence of WAIHA. Red cells are coated with IgG, and the
drug.55The following observations are charac- eluate as well as the serum react with virtually
teristic: all cells tested in the absence of the drug. The
antibody has no direct or indirect in-vitro inter-
IgG and complement are detected but com- action with the drug. As mentioned earlier, the
plement may be the only protein present. prototype drug for such cases is methyldopa,
Serum antibody can be IgM, IgG, or IgM which.is now used much less frequently than in
with IgA. the past. Currently, fludarabine, used to treat
A drug (or metabolite) must be present in vi- chronic lymphocytic leukemia, is the most
tro for the antibody in the patient's serum to commonly used drug :that produces drug-
be detected. Antibodies may cause hemoly- independent antibodies and AIHA.54
sis, agglutination, and/or sensitization of red
cells in the presence of the drug. Nonimmunologic Protein Adsorption
The patient need only take a small amount of
the drug (eg, a single dose). The positive DAT result associated with some
Acute intravascular hemolysis with hemoglo- drugs is caused by modification of the red cell
binemia and hemoglobinuria is the usual pre- membrane by the drug and is independent of
sentation. Renal failure is quite common. antibody production. Hemolytic anemia associ-
Once antibody has been formed, severe heated with this mechanism is rare.
molytic episodes may recur after exposure to Cephalosporins (primarily cephalothin) are
very small quantities of the drug. the drugs with which positive DAT results and
NIPA were originally associated. In vitro, red
On occasion, it appears that a patient's serum cells coated with cephalothin in pH 9.8 buffer
contains an "autoantibody" in addition to a drug and incubated with normal plasma adsorb albu-
antibody reacting in the presence of the drug. min, IgA, IgG, IgM, C3, and other proteins in a
Rather than a true autoantibody, it is believed nonimmunologic manner. For this reason, the
that this reactivity results from the presence of IAT result with virtually all plasma will be posi-
circulating drug or drug-pIus-antibodycomplex- tive. Other drugs that cause NIPA and a positive
es." In these cases, an eluate is usually nonreac- DAT result include diglycoaldehyde, cisplatin,
tive when the drug is not present in the system. oxaliplatin, and beta-lactamase inhibitors (clavu-
However, in some cases, especiallythose involv- lanic add, sulbactam, and tazobactam)."
ing piperacillin, the eluate reacts while the pa- NIPA should be suspected when a patient's
tient is still taktng the drug. A sample collected plasma/serum and most normal plasma/sera
several days after the drug has been discontin- are reactive in an IATwith drug-treated red cells
ued will be nonreactive. A true warm autoanti- but the eluate from the patient's red cells is non-
body is expected to be reactive in an eluate pre- reactive with the drug-treated cells.
Laboratory Investigation of Drug- drug.55 The patient's serum and an eluate from
Induced Immune Hemolysis the patient's red cells can be tested against the
drug-treated red cells. (See Method 4-12.) This
The drug-related problems that are most com- is the method of choice when cephalosporins
monly encountered in the blood bank are those (except for ceftriaxone) are thought to be impli-
associated with a positive DATresult and a non- cated. Results that are definitive for a drug-
reactive eluate. Recent red cell transfusions induced positiveDATresult are reactivity of the
and/or dramatic hemolysismay result in a weak eluate with drug-treated red cellsand absence of
DAT result by the time hemolysis is suspected. reactivitywith untreated red cells.
When other, more common causes of immune- Drug-treated red cells should always be test-
mediated hemolysis have been excluded and a ed with saline and normal serum (or plasma) as
temporal relationship exists between the admin- negative controls. This approach ensures that
istration of a drug and the hemolytic anemia, a the observed reactivity with the patient's se-
drug antibody investigation should be pursued. rum/plasma is appropriately interpreted. Anti-
The patient's serum should be tested for un- bodies reactive with red cells treated with some
expected antibodies by routine procedures. If drugs (eg, beta-Iactams and platinums) have
the serum does not react with untreated red been detected in the plasma from blood donors
cells, the tests should be repeated with the and patients without hemolytic anemia and are
drug(s) suspected of causing the problern" thought to be caused by environmentai expo-
Some drug formulations contain inert ingredi- sure. Therefore, misinterpretation of reactivity
ents (eg, pill or capsule forms), and other drugs in a patient's serum is possible.55
are combinations of two drugs (eg, piperacillin Whenever possible, a positive control should
plus tazobactam). Although it would seem logi- be tested with drug-treated red cells. Negative
cal to test the patient's serum with the actual results of a patient's serum and eluate without a
drug that the patient received, inert ingredients positivecontrol can be interpreted only as show-
or drug combinations can make preparation of ing that antibodies to that drug were not detect-
drug-treated red cells difficult or make the re- ed. The drug mayor may not be bound to the
sults confusing. It is preferable to test serum us- test red cells.
ing pure drug formulations as well as separate If the drug in question is known to cause NI-
components of combination drugs. PA, the patient's serum and the controls (nega-
If the drug has already been reported to tive and positive)should also be tested at a dilu-
cause hemolytic anemia, testing methods may tion of 1 in 20. Normal sera at this dilution do
be described in the case reports. Far more drug- not usually contain enough protein for NIPA to
dependent antibodies are detected by testing se- be detected.
rum in the presence of drug; therefore, when a When a patient is receiving more than one
previous report of antibodies to a drug is not drug that has a temporal relationship to hemoly-
available, an initial screening test can be per- sis, the coadministered drugs should be tested.
formed with a solution of the drug at a concen- Antibodies to more than one drug have been re-
tration of approximately 1 mg/mL in phosphate- ported in cases where chemotherapeutics have
buffered saline/" (See Method 4-13.) Serum, been administered multiple times." In addition,
rather than plasma, is the preferred specimen an immune response may be caused by a metab-
for testing for hemolysis to be observed; this also olite of a drug rather than the drug itself. If the
allows for the addition of fresh normal serum (as clinical picture is consistent with immune-
a source of complement) to the test system. The mediated hemolysisand the above tests are non-
addition of the fresh complement increases the informative,it may be helpful to test metabolites
sensitivityof the test for the detection of in-vitro of the parent drug that are present in the serum
hemolysis resulting from complement activa- or urine of an individual who is taking that
tion. drug.57 Antibodies to some nonsteroidal anti-
If these tests are not informative, attempts inflammatory drugs have required testing in the
can be made to coat normal red cells with the presence of metabolite.f The metabolism and
half-lifeof the drug determines when the drug serum or urine, as well as previous reports for
metabolite should be collected. Pharmacology the drug under investigation should be consult-
information for the metabolite(s) detectable in ed.
REFERENCES
1. Zanella A, Barcellini W. Treatment of autoim- 2. Shapiro R, Chin-Yee I, Lam S. Eculizumab as a

mune hemolytic anemias. Haematologica 2014; bridge to immunosuppressive therapy in severe
99:1547-54. cold agglutinin disease of anti-Pr specificity.Clin
Case Rep 2015;3:942-4.
3. Kaplan HS, Garratty G. Predictive value of di- 17. Chapuy CI, Nicholson RT,Aguad MD, et al. Re-
rect antiglobulin test results. Diagnostic Med solving the daratumumab interference with
1985;8:29-32. blood compatibility testing. Transfusion 2015;
4. Garratty G. The significance of IgG on the red 55:1545-54.
cell surface. Transfus Med Rev 1987;1:47-57. 18. Oostendorp M, Lammerts van Bueren JJ, Doshi
5. Freedman J. The significanceof complement on P, et al. When blood transfusion medicine be-
the red cell surface. Transfus Med Rev 1987;1: comes complicated due to interference by
58-70. monoclonal antibody therapy. Transfusion
6. Petz LD, Garratty G. Immune hemolytic anemi- 2015;55:1555-62.
as. 2nd ed. Philadelphia: Churchill-Livingstone, 19. Judd WJ. Elution-dissociation of antibody from
2004. red blood cells:Theoretical and practical consid-
7. Rottenberg Y, YahalomV, Shinar E, et al. Blood erations. TransfusMed Rev 1999;13:297-310.
donors with positive direct antiglobulintests are 20. Leger RM, Arndt PA, CiesielskiOJ, Garratty G.
at increased risk for cancer. Transfusion 2009; False-positive eluate reactivity due to the low-
49:838-42. ionic wash solution used with commercial acid-
8. Hannon JL. Management of blood donors and elution kits. Transfusion 1998;38:565-72.
blood donations from individuals found to have 21. Judd WJ, Johnson ST, StorryJR.Judd's methods
a positive direct antiglobulin test. Transfus Med in immunohematology. 3rd ed. Bethesda, MD:
Rev 2012;26:142-52. AABBPress, 2008.
9. Toy PT, Chin CA, ReidME, Burns MA. Factors 22. Judd WI, Barnes BA,Steiner EA, et al. The eval-
associated with positive direct antiglobulin tests uation of a positivedirect antiglobulintest (auto-
in pretransfusion patients: A case control study. control) in pretransfusion testing revisited.
Vox Sang 1985;49:215-20. Transfusion 1986;26:220-4.
10. Heddle NM, Kelton JG, Turchyn KL, Ali MAM. 23. Arndt PA, LegerRM, Garratty G. Serologlcfind-
Hypergammaglobulinemia can be associated ings in autoimmune hemolytic anemia associat-
with a positive direct antiglobulin test, a nonre- ed with immunoglobulin M warm autoantibod-
active eluate, and no evidence of hemolysis. ies. Transfusion2009;49:235-42.
24. Rodberg K, Tsuneta R, Garratty G. Discrepant
Transfusion 1988;28:29-33.
Rh phenotyping results when testing IgG-
II. Clark JA, Tanley PC, Wallas CH. Evaluation of
sensitized RBCswith monoclonal Rh reagents
patients with positive direct antiglobulin tests
(abstract).Transfusion I 995;35(Suppl):67S.
and nonreactive eluates discovered during pre-
25. Leger RM, Garratty G. Evaluation of methods
transfusion testing. Immunohematology 1992;
for detecting alloantibodies underlying warm
8:9-12. autoantibodies. Transfusion 1999;39: 11-16.
12. Coombs RRA,Mourant AE, RaceRR.A new test
26. Church AT, Nance SJ, KavitskyOM. Predicting
for the' detection of weak and "incomplete" Rh the presence of a new alloantibodyunderlying a
agglutinins.BrJ Exp Pathol 1945;26:255-66. warm autoantibody (abstract). Transfusion
13. Heddle NM, Soutar RL,O'Hoski PL,et al. A pro- 2000;40(Suppl):121S.
spective study to determine the frequency and 27. Branch DR, Petz LD. A new reagent (1IAP)
clinical significance of alloimmunization post- having multiple applications in immunohema-
transfusion.BrJ HaematoI1995;91: I 000-5. tology.AmJ Clin PathoI1982;78:161-7.
14. Garratty G, Arndt PA. Drugs that have been 28. Lalne EP, Leger RM, Arndt PA, et al. In vitro
shown to cause drug-induced immune hemolyt- studies of the impact of transfusion on the de-
ic anemia or positive direct antiglobulin tests: tection of alloantibodies after autoadsorption.
Some interesting findings since 2007. Immuno- Transfusion2000;40: 1384-7.
hematology 2014;30:66-79. 29. Chiaroni J, TouinssiM, Mazet M, et al. Adsorp-
15. Desborough MJ, Miller J, Thorpe SJ,et al. Intra- tion of autoantibodies in the presence of LISSto
venous immunoglobulin-induced haemolysis: A detect alloantibodiesunderlying warm autoanti-
case report and review ofthe literature. Transfus bodies. Transfusion2003;43:651-5.
Med 2014;24:219-26. 30. Magtoto-JocomJ, Hodam], Leger RM, Garratty
16. Rushin J, Rumsey DH, Ewing CA, Sandler SG. G. Adsorption to remove autoantibodies using
Detection of multiple passivelyacquired alloan- allogeneic red cells In the presence of low ionic
tibodies following infusions of N Rh immune strength saline for detection of alloantibodies
globulin. Transfusion 2000;40:551-4. (abstract). Transfusion2011 ;51(SuppJ):174A.
31. Judd WJ, Dake L. PEG adsorption of autoanti- warm autoantibodies (letter). Vox Sang 2007;
bodies causes loss of concomitant alloantibody. 93:92.
Immunohematology 2001;17:82-5. 45. Leger RM, Co A, Hunt P, Garratty G. Attempts
32. Combs MR, Eveland D,Jewet-Keefe. B, Telen to support an immune etiology in 800 patients
MJ. The use of polyethylene glycol in adsorp- with direct antiglobulin test-negative hemolytic
tions: More evidence that antibodies may be anemia. Immunohematology 2010;26: 156-60.
missed (abstract). Transfusion 2001 ;41 (Sup- 46. Waligora SK, Edwards JM. Use of rabbit red
pl):30S. cells for adsorption of cold autoagglutinins.
33. Issitt PD, Combs MR, Bumgarner DJ, et al. Transfusion 1983;23:328-30.
Studies of antibodies in the sera of patients who 47. Dzik WH, Yang R, Blank J. Rabbit erythrocyte
have made red cell autoantibodies. Transfusion stroma treatment of serum interferes with rec-
1996;36:481-6. ognition of delayed hemolytic transfusion reac-
34. Garratty G. Target antigens for red-cell-bound tion (letter). Transfusion 1986;26:303-4.
autoantibodies. In: Nance SJ, ed. Clinical and 48. Mechanic SA,Maurer JL, IgoeMJ, et al. Anti-Vel
basic science aspects of immunohematology. Ar- reactivity diminished by adsorption with rabbit
lington, VA:MBB, 1991:33-72. RBCstroma. Transfusion2002;42: 1180-3.
35. Garratty G. Specificity of autoantibodies react- 49. Storry JR, Olsson ML, Moulds JJ. Rabbit red
ing optimally at 37° C. Immunohematology blood cell stroma bind immunoglobulin M anti-
1999;15:24-40.
bodies regardless of blood group specificity(let-
36. Branch DR, Petz LD. Detecting alloantibodies in ter). Transfusion2006;46: 1260-1.
patients with autoantibodies (editorial). Transfu-
50. Sokol RJ, Hewitt S, Stamps BK. Autoimmune
haemolysis: An 18-year study of 865 cases re-
sion 1999;39:6-10.
37. Young PP, Uzieblo A, Trulock E, et al. Autoanti-
ferred to a regional transfusion centre. Br Med J
1981;282:2023-7.
body formation after alloimmunization: Are
51. Shulman lA, Branch DR, NelsonJM, et al. Auto-
blood transfusions a risk factor for autoimmune
immune hemolytic anemia with both cold and
hemolytic anemia? Transfusion 2004;44:67-72.
warm autoantibodies.JAMA1985;253: 1746-8.
38. Maley M, Bruce DG, Babb RG, et al. The inci-
52. Garratty G, Arndt PA, Leger RM. Serological
dence of red cell alloantibodies underlying pan-
findings in autoimmune hemolytic anemia
reactive warm autoantibodies. Immunohematol-
(AIHA)associated with both warm and cold au-
ogy 2005;21: 122-5.
toantibodies (abstract). Blood 2003;102(Suppl):
39. Ahrens N, Pruss A, Kahne A, et al. Coexistence
563a.
of autoantibodies and alloantibodies to red 53. Eder AF. Review:Acute Donath-Landsteinerhe-
blood cells due to blood transfusion. Transfusion molytic anemia. Immunohematology 2005;21:
2007;47:813-16. 56-62.
40. ShireyRS, BoydJS, ParwaniAV, et al. Prophylac- 54. GarrattyG. Immune hemolyticanemia associated
tic antigen-matched donor blood for patients with drug therapy. BloodRev2010;24:143-50.
with warm autoantibodies: An algorithm for 55. LegerRM,Arndt PA,GarrattyG. How we inves-
transfusion management. Transfusion 2002;42: tigate drug-induced immune hemolytic anemia.
1435-41. Immunohematology 2014;30:85-94.
41. Hillyer CD, Shaz BH, Winkler AM, Reid M. In- 56. Leger RM, Jain S, Nester TA, Kaplan H. Drug-
tegrating molecular technologies for red blood induced immune hemolytic anemia associated
cell typing and compatibility testing into blood with anti-carbop1atin and the first example of
centers and transfusion services. Transfus Med anti-paclitaxel.Transfusion2015;55:2949-54.
Rev 2008;22:117-32. 57. Salama A, Mueller-Eckhardt C, Kissel K, et al.
42. Denomme GA. Prospects for the provision of Ex vivo antigen preparation for the serological
genotyped blood for transfusion. Br J Haematol detection of drug-dependent antibodies in im-
2013; 163:3-9. mune haemolytic anaemias. Br J Haematol
43. Lee E, Redman M, Burgess G, Win N. Do patients 1984;58:525-31.
with autoantibodies or clinically insignificant allo- 58. Johnson ST, Fueger ]I, Gottschall JL. One cen-
antibodies require an indirect antiglobulin test ter's experience: The serology and drugs associ-
crossmatch? Transfusion 2007;47:1290-5. ated with drug-induced immune hemolytic ane-
44. Richa EM, Stowers RE, Tauscher CD, et al. The mia-a new paradigm. Transfusion 2007;47:
safety of electronic crossmatch in patients with 697-702.
APPENDIX 14-1
Drugs Associated with Immune Hemolytic Anemia
Aceclofenac +Drug
Acetaminophen +Drug
Acyclovir DT
Alemtuzumab AA
Aminopyrine DT
Amoxicillin DT
Amphotericin B +Drug
Ampicillin DT +Drug
Antazoline +Drug
Azapropazone AA DT
Bendamustine AA
Butizide +Drug
Carbimazole AA DT -Druq
Carboplatin AA DT +Drug NIPA
Carbromal DT
Cefamandole DT
Cefazolin DT
Cefixime DT +Drug
Cefotaxime DT +Drug
Cefotetan AA DT +Drug NIPA
Cefoxitin AA DT -Druq
Cefpirome + Drug
Ceftazidime AA DT +Drug
Ceftizoxime DT +Drug
Ceftriaxone +Drug
Cefuroxime DT
Cephalexin DT
Cephalothin DT +Drug NIPA
Chloramphenicol AA DT
Chlorinated hydrocarbons AA DT +Drug
Chlorpromazine AA +Drug
Chlorpropamide +Drug
APPENDIX 14-1
Drugs Associated with Immune Hemolytic Anemia (Continued)
Cimetidine DT +Drug
Ciprofloxacin +Drug
Cisplatin DT +Drug NIPA
Cladribine AA
Clavulanate NIPA
Cyanidanol AA DT +Drug
Cyclofenil AA + Drug
Cyclosporine DT
Diclofenac AA DT +Drug
Diethylstilbestrol +Drug
Diglycoaldehyde NIPA
Dipyrone DT +Drug
Erythromycin DT
Etodolac -Druq
Fenoprofen AA +Drug
Fluconazole DT + Drug
Fludarabine AA
Fluorescein DT +Drug
Fluorouracil + Drug
Furosemide + Drug
Hydralizine DT
Hydroch10rothiazide DT +Drug
Hydrocortisone DT + Drug
9-Hydroxy-methyl-ellipticinium -Druq
Ibuprofen +Drug
Imatinib mesylate DT
Insulin DT
Isoniazid DT +Drug
Levodopa AA
Levofloxacin DT -Druq
(Continued)
APPENDIX 14-1
Mefenamic acid AA
Mefloquine DT +Drug
Melphalan + Drug
6-Mercaptopurine DT
Methadone DT
Methotrexate AA DT +Drug
Methyldopa AA
Nabumetone +Drug
Nafcillin DT
Naproxen +Drug
Oxaliplatin DT +Drug NIPA
Paclitaxel DT + Drug
p-Aminosalicylic acid +Drug
Pemetrexed + Drug
Penicillin G DT
Phenacetin +Drug
Phenytoin DT
Piperacillin DT -Druq
Probenicid +Drug
Procainamide AA
Propyphenazone +Drug
Pyrazinamide DT + Drug
Pyrimethamine DT
Quinidine DT +Drug
Quinine +Drug
Ranitidine DT +Drug
Rifabutin + Drug
Rifampicin DT +Drug
Sodium pentothal/thiopental +Drug
Stibophen +Drug
Streptokinase DT
APPENDIX 14-1
Streptomycin AA DT -Druq
Sulbactam NIPA
Sulfamethoxazole +Drug
Sulfasalazine +Drug
Sulfisoxazole +Drug
Sulindac AA DT -Druq
Suprofen AA +Drug
Tazobactam NIPA
Teicoplanin AA +Drug
Teniposide AA +Drug
Tetracycline DT
Ticarcillin AA DT
Tolbutamide DT
Tolmetin AA -Druq
Triamterene DT -Druq
Trimethoprim +Drug
Vancomycin + Drug
Vincristine DT + Drug
Zomepirac AA -Druq
AA = drug-independent autoantibody; DT = testing with drug-treated red cells; +Drug = testing in the presence of drug; NIPA
= nonimmunologic protein adsorption.
Platelet and .Granulocyte Anti.gens an.d.
Antibodies
David F. Stroncek, MD, and Ralph R. Vassallo, MD, FACP
HIS CHAPTER DISCUSSESANTIGENS nucleotide polymorphisms (SNPs) in the genes

expressed on platelets and granulocytes that encode them. The amino acid changes re-
and the antibodies formed by sensitized sulting from these SNPs in turn result in glyco-
individuals. These antigens and the immune re- protein structural and antigenic changes capable
sponses to them are of importance in alloim- of eliciting alloantibody responses after pregnan-
mune, autoimmune, and drug-induced immune cy or transfusion. Currently, 35 different HPAs
syndromes involving platelets and granulocytes. expressed on six different platelet membrane
glycoproteins (GPs)-GPIIb, GPIIIa, GPIba,
GPIb~, GPIa, and CD109-have been formally
N recognized (Table 15-1).4 These antigens are of-
ten referred to as "platelet specific" and, al-
though some are found on cells other than plate-
Platelets express a variety of antigenic markers lets (especially leukocytes and endothelial cells),
on their surface. Some of these antigens are their. chief clinical importance appears to be
shared with other cells, such as ABH and HLA linked to their presence on platelets.
determinants, whereas others are essentially Twelve antigens are clustered into six import-
platelet specific,such as human platelet alloanti- antbiallelic groups (HPA-1, HPA-2, HPA-3, HPA-
gens (HPAs). 4, HPA-5, and HPA-15). The nomenclature for
HPAs consists of numbering the antigens in their
HPA order of discovery,with the higher-frequencyan-
Platelets play roles in inflammation, immune re- tigens designated "a" and the lower-frequency
sponses, cardiovascular disease, and even can- antigens designated "b."s HPAs for which anti-
cer.'? However, their primary function is hemo- bodies against only one of the two antithetical
stasis. Platelets perform all these functions (non-wild-type)antigens have been detected are
through multiple ligand-receptor interactions in- labeled with a "w" for "workshop," such as
volving glycoproteins expressed on their cell HPA-6bw. New low-frequency HPAs continue
surface membranes. to be discovered but have not yet been con-
Platelet membrane glycoproteins are ex- firmed in the Immuno Polymorphism Database
pressed in different forms as a result of single (https.Zzwww.ebl.ac.uk/ipd/hpa/j."
David F. Stroncek, MD, Director, Center for Cellular Engineering, Department of Transfusion Medicine, Clinical
Center, National Institutes of Health, Bethesda, Maryland; and Ralph R.Vassallo,MD, FACP,Executive Vice Pres-
ident/Chief Medical and Scientific Officer, Vitalant, Scottsdale, Arizona
D. Stroncek has disclosed no conflicts of interest. R. Vassallo has disclosed a financial relationship with Cerus
Corp and Terumo BCT.
457
TABLE 15·1. Human Platelet Alloantigens
Cqrrent Le!J~cy Phenoty~ic AlllinoAcid Gellel

Nomenclature Nomenclature Frequency· Glycoprotein Change Nucleotide Change
HPA-1a z«, PI A1 72% ala
ITGB3
HPA-1b Zwb, PIA2 26% alb GPllla Leu33Pro
176T>C
2% bib
HPA-2a KOb 85% ala
GPIBA
HPA-2b Koa, Siba
14% alb GPlba Thr145Met
482C>T
1% bib
HPA-3a Baka, Leka 37% ala
ITGA2B
HPA-3b
48% alb GPllb IIe843Ser
Bakb 2621T>G
15% bib
HPA-4a Yukb, Pen' >99.9% ala
ITGB3
HPA-4b Yu~, Penb
<0.1% alb GPllla Arg143Gln
506G>A
<0.1% bib
HPA-5a Brb, Zavb 88% ala
ITGA2
HPA-5b Bra, Zava, Hca 20% alb GPla Glu505Lys
16ooG>A
1% bib
<1% ITGB3
HPA-6bw Caa,Tua GPllla Arg489Gln
alb or bib 1544G>A
<1% ITGB3
HPA-7bw Moa GPllla Pro407Ala
alb or bib 1297C>G
<1% ITGB3
HPA-8bw sr alb or bib
GPllla Arg636Cys
1984C>T
HPA-9bw <1%
ITGA2B
(assoc. with Maxa alb or bib GPllb Val837Met
2602G>A
HPA-3b)
<1% ITGB3
HPA-10bw Laa GPllia Arg62Gln
alb or bib 263G>A
<1% ITGB3
HPA-11bw Groa GPllla Arg633His
alb or bib 1976G>A
<1% GPIBB
HPA-12bw Iya GPlbl3 Gly15Glu
alb or bib 119G>A
<1% ITGA2
HPA-13bw Sita GPla Met799Thr
alb or bib 2483C>T
HPA-14bw <1% bib
ITGB3
(assoc. with Dea GPllla Lys611 del
1909-1911de1AAG
HPA-1b)
HPA-15a Gov" 35% ala
CD 109
HPA-15b
42% alb CD109 Ser682Tyr
Gova 2108C>A
23% bib
C HAP T E R 1 5 Plateletand GranulocyteAntigens and Antibodies 459
TABLE15·1. Human Platelet Alloantigens (Continued)
HPA-16bw Duvaor Duv" GPllla Thr1401le

497C>T
<1% ITG83
HPA-17bw Vaa GPllla Thr195Met
alb or bib 662C>T
<1% ITGA2
HPA-18bw Caba GPla Gln716His
alb or bib 2235G>T
<1% ITG83
HPA-19bw Sta GPllla Lys137Gln
alb or bib 487A>C
<1% ITGA28
HPA-20bw Kno GPlib Thr619Met
alb or bib 1949C>T
<1% ITG83
HPA-21bw Nos GPllla Glu628Lys
alb or bib 1960G>A
<1% ITGA28
HPA-22bw Sey GPlib Lys164Thr
alb or bib 584A>C
<1% ITG83
HPA-23bw Hug GPllla Arg622Trp
alb or bib 1942C>T
<1% ITGA28
HPA-24bw Cab2at GPlib Ser472Asn
alb or bib 1508G>A
<1% ITGA2
HPA-25bw Swia GPla Thr1087Met
alb or bib 3347C>T
<1% ITG83
HPA-26bw Seca GPllla Lys580Asn
alb or bib 1818G>T
<1% ITGA28
HPA-27bw Cab3at GPlib Leu841 Met
alb or bib 2614C>A
<1% ITGA28
HPA-28bw War GPlib Val740Leu
alb or bib 2311G>T
<1% ITG83
HPA-29bw Khab GPllla Thr7Met
alb or bib 98C>T
'Phenotypic frequencies are for people of Europeanancestry who live in North America. Human platelet alloantigen (HPA)
frequencies in other racesand ethnic groups can be found in the Immuno Polymorphism Databaseat http://www.ebLac.uk/
ipd/hpalfreqs_j .htm!.
Platelet Alloantigens on GPllblllla Integrins are a broadly distributed family of

HPA-l a is the platelet alloantigen that was dis- adhesion molecules consisting of an a and a p
covered first and is most familiar." Originally chain held together by divalent cations in a het-
named "Zw"" and more commonly referred to as erodimeric complex." Integrins are essential for
"PIAl," it is expressed on GPIIIa, the p-subunit platelet adhesion and aggregation because they
of the integrin GPIIb/IIIa (aIIb/b3) complex. serve as receptors for ligands, such as fibrinogen,
460 AA B B TE C H N I CAL MAN UA L
collagen, fibronectin, von Willebrand factor antigens were all discovered in cases of FNAIT
(vWF),and other extracellular matrix proteins. by specific antibodies in maternal sera reactive
Postactivation binding of fibrinogen by solelywith paternal GPIIb/IIIa. The vast majori-
GPIIb/IIIa results in platelet aggregation,which ty of these antigens are private antigens restrict-
forms the "platelet plug" to stop bleeding. ed to the single families in which they were
GPIIb/IIIa's hemostatic importance is demon- discovered. HPA-6bwand HPA-21bware excep-
strated by the serious bleeding in patients with tions, having antigen frequencies of 1% and 2%,
Glanzmann thrombasthenia, a rare disorder that respectively,in people of Japanese ancestry, and
is caused by congenital absence or dysfunction HPA-9bwhas been implicatedin several cases of
of the GPIIb/IIIa genes fTGA2B and/or FNAIT.15-18
fTGB3.11 Patients with Glanzmann thrombas-
thenia who are exposed to normal platelets by Platelet Alloantigens on GPlbNI/X
transfusion or pregnancy can make isoantibodies
against GPIIb/IIIa. The GPlblVIlK complex forms the vWF recep-
GPIIb/IIIa is the most abundantly expressed tor on platelets, and platelets express approxi-
(SO,000-80,000 molecules/platelet) glycopro- mately 12,SOO copies of this seven-unit com-
plex." Followingvascular injury, binding of the
tein complex on the platelet membrane, making
it highly immunogenic.F Antibodies against GPlblVIlK complex to vWF facilitates platelet
adhesion to vascular subendothelium and initi-
HPA-Ia account for the vast majority (>80%) of
ates signaling events within adherent platelets
the HPA-specificplatelet antibodies detected in
that lead to platelet activation, aggregation, and
the sera of alloimmunized people of European
hemostasis. GPlb is composed of two a (GPlba)
ancestry. HPA-1a antibodies are produced by the
2% of individuals with the platelet type HPA- and two ~ (GPlb~) subunits (2S,000 surface
copies) that form noncovalent associationswith
1b/1b.
two GPIK and one GPV.GPlba carries HPA-2al
Twenty-five of the 3S recognized HPAs are
2b, and GPlb~ carries HPA-12bw. Antibodies
carried by GPIIb (8) and GPIIIa (17). Like HPA-
against HPA-2a, -2b, and -12bw have all been
1a/1b, the HPA-4a/4b antigens are also ex-
implicatedin FNAIT.
pressed on GPIIIa and have been implicated in
Deficiencyof the entire GPlblVIlK complex
fetal and neonatal alloimmune thrombocytope-
can occur from mutations in the encoding genes
nia (FNAIT),posttransfusion purpura (PTP),and
GPfBA, GPfBB,or GP9, and is the cause of Ber-
platelet transfusion refractoriness. The low-
nard Soulier syndrome (BSS).BSSis a disorder
frequency HPA-4b antigen is more common in
characterized by prolonged bleeding time,
populations ofJapanese and Chinese ancestry.12
thrombocytopenia, and the presence of "giant
The HPA-3a/3b antigens are expressed on
platelets," and affectsapproximately one person
GPIIbbut, despite the relativelyhigh rate of ho-
per million." BSS patients whose platelets are
mozygosity of the low-frequency antigen in the
devoid of GPlblVIlK can produce isoantibodies
general population, detection of HPA-3antibod-
when they are exposed to the protein complex
ies is uncommon. Some HPA-3 antibodies are
on normal platelets through transfusions or
difficultto detect in monoclonal antigen capture pregnancy.19
assays (ACAs), such as the modified antigen
capture enzyme-linked immunosorbent assay
Platelet Alloantigens on GPlallla
(MACE) and monoclonal antibody-specificim-
mobilization of platelet antigens (MArPA), in The integrin GPla/IIa, also known as integrin
which GPIIb is extracted from platelets with a2b1, is a major collagen receptor on platelets.
detergents that can denature the antigenic The GPla protein carries the HPA-Sa/Sb anti-
epitopes recognized by various HPA-3 antibod- gens. Antibodies against HPA-Santigens are the
ies.13,14 second most frequently detected, after anti-HPA-
In addition to HPA-1b,-3b, and -4b, 19 other 1a, in patients with FNAIT and are also fre-
low-frequency platelet antigens are expressed quently detected in patients with PTPand trans-
on either GPIIb or GPIIIa (Table IS-I). These fusion refractoriness.About 3000 to SOOOmole-
C HAP T E R 1 5 Plateletand GranulocyteAntigensand Antibodies 461
cules of the GPla/IIa heterodimeric complex Other Antigens on Platelets

are expressed on platelets." HPA-13bw, -18bw,
ABO and Other Blood Groups
and -25bw are low-frequency antigens that are
also expressed on GPla and have all been impli- Most of the ABH antigen on platelets is carried
cated in FNAIT. Interestingly, the HPA-13bw on saccharides attached to the major platelet
polymorphism has been reported to cause func- membrane glycoproteins (Table 15-2). GPIIb
tional defects that reduce platelet responses to and platelet endothelial cell adhesion molecule
collagen-induced aggregation and spreading on 1 (PECAM-l/CD31) carry the largest amounts
collagen-coated surfaces." of A and B antigens." Platelet A and B antigen
levels are quite variable from individual to indi-
Platelet Alloantigens on CD709 vidual, with 5% to 10% of non-group-O individ-
uals expressing extremely high levels of A or B
CD109 is a glycosylphosphatidylinositol (GPI)- on their platelets.z5,z6 These "high expressers"
linked protein and a member of the az- have highly active glycosyltransferases, which
macroglobulin/complement superfamily. Its are much more efficient at attaching A or B anti-
function is still not completely understood, but gens."
CD109 has been reported to bind to and nega- Interestingly, although individuals with the
tively regulate the signaling of transforming subgroup Az red cell phenotype express lower
growth factor beta. CD109 is also expressed on levels of A on their red cells than AI individuals,
activated T lymphocytes, CD34+ hematopoietic they do not express detectable A antigens on
cells, and endothelial cells, and it carries the their platelets. As a result, Az platelets may be
HPA-15 antigens. successfully transfused to group 0 patients with
Platelets express an average of 2000 mole- high-titer immunoglobulin G (IgG) anti-A or
cules of CD109, although interindividual copy -A,Bwho are refractory to non-group-O plate-
numbers vary signiftcantly." Studies show the lets."
presence of HPA-15 antibodies in 0.22% to 4% Although platelets are often transfused with-
of maternal sera in patients with suspected out regard to ABO compatibility, the use of
FNAIT,and several reports suggest that HPA-15 major-mismatched platelets (eg,A or B platelets
antibodies are more frequently detected in sera into group 0 recipients) frequently results in
from patients with immune platelet refractori- lower posttransfusion recovery rates, while mi-
ness.":" nor mismatches (eg, 0 platelets into A or B
TABLE 15-2. Other Platelf3tAntigens
ABO Same as for red cells Multiple ABO Multiple
HLA-A,-B,
Same as for leukocytes Class I HLA Multiple MHC Multiple
and -C
90%-97% (African ancestry) 1264T>G*
Tyr325Thr*
GPIV 90%-97% (Asian ancestry) CD36 CD36 478C>T*
Pro90Ser*
99.9% (European ancestry) Exons 1-3 del
GPVI N/A GPVI N/A GP6 N/A
saccharides GPsduring their glycosylation.
t Only the most common changesare shown.
t Only the most common mutations are shown.
recipients) do not.28,29Clinical trials comparing expressed as integral membrane proteins,

ABO-identical to -unmatched platelets in pa- whereas smaller amounts may be adsorbed from
tients with cancer who require multiple platelet surrounding plasma. HLA-Aand -B locus anti-
transfusions have suggested that rates of refrac- gens are significantlyrepresented, but there ap-
toriness are significantly higher when un- pears to be only minimal platelet expression of
matched components are used." Although oth- HLA-C.41With rare exceptions, Class II HLAis
er red cell antigens (eg, Lea,Le", I, i, P, P\ and not present on the platelet membrane.
Cromer) are alsopresent on platelets, there is no Transfusion-associated HLA alloimmuniza-
evidence that these antigens Significantlyreduce tion appears to be influenced by the underlying
platelet survivalin vivo.31,32 disease, immunosuppressive effectsof treatment
regimens, and whether the blood components
GPIV/CD36 contain a significant amount of leukocytes.
Widespread use of leukocyte-reduced blood
Platelets, monocytes/macrophages, and nucle- components has considerablyreduced HLAallo-
ated erythrocytes are the only blood cells that immunization from transfusion.f Despite white
express GPN/CD36 (Table 15-2). GPNbelongs cell inactivation by pathogen inactivation tech-
to the Class B scavenger receptor family and nologies, pathogen-reduced platelet transfu-
binds a number of different ligands, including sions are associated with more HLAalloimmuni-
low-density lipoprotein cholesterol, thrombos- zation and immune transfusionrefractoriness.S"
pondin, types I and N collagen, and malaria- HLAantibodies also commonly develop follow-
infected red cells. A number of mutations in ing pregnancy and are present in the sera of
CD36 have been described that result in a com- >32% of women who have had four or more
plete lack of protein expression on both platelets pregnancies." HLA antibodies have also been
and monocytes in populations of Asian and Afri- identified in 1.4% to 3.3% of women who have
can ancestry.30-32 CD36-deficient individuals ex- never been pregnant or received transfusion and
posed to normal platelets can produce antibod- men with no previous transfusions, and not all
ies to CD36 that have been reported to cause such antibodies recognize only denatured bead-
FNAIT, PTP, and platelet transfusion refractori- coating antigens." Sensitization to HLA anti-
ness.31,33-37 gens becomes important in platelet transfusion
recipients when HLAantibodies cause destruc-
GPVI tion of allogeneicplatelets, contributing to plate-
GPVI is a major collagen receptor on platelets let transfusion refractoriness.
and a member of the immunoglobulin superfam-
ily. GPVIinteractions with collagen exposed on Alloimmune Platelet Disorders
the extracellular matrix result in platelet activa- Platelet Transfusion Refractoriness
tion and aggregation. To date, no HPAs have
been identified on GPVI, but platelet autoanti- A less-than-expected increase in platelet count
bodies formed against GPVIhave been reported occurs in about 25% to 70% of patients with
to cause a mild form of autoimmune thrombocy- thrombocytopenia who have received multiple
topenia.38,39Interestingly, GPVI autoantibodies transfusions." Patients treated for malignant he-
induce shedding of GPVI from platelets, result- matopoietic disorders are particularly likely to
ing in reduced collagen binding and clinically become refractory to platelet transfusions. Re-
significantbleeding. sponses to platelet transfusions are often deter-
mined 10 to 60 minutes after transfusion by cal-
HLA
culating either a corrected platelet count
increment (CCI) or a percentage platelet recov-
HLAis present on all nucleated cells of the body. ery (PPR),both of which normalize transfusion
(See Chapter 16.) HLAassociated with platelets responses for patient blood volume and platelet
is the main source of Class I HLA in whole dose. (See Chapter 19 for more on CCls.) Most
blood." Most Class I HLA on platelets is experts would agree that a l-hour posttransfu-
sion CCI of <5000 to 7500 or a PPR of platelets alloimmune refractoriness. When HLA antibod-
transfused or a PPR <30% after two consecutive ies are present, a widely used approach is to sup-
transfusions adequately defines the refractory ply apheresis platelets from donors whose HLA-
state. A and -B antigens match those of the patient. A
HLA sensitization is the most common im- pool of 1000 to 3000 or more HLA-typed apher-
mune cause of refractoriness and can be diag- esis donors is generally necessary to find HLA-
nosed by demonstration of significant levels of identical donors for most patients."
antibodies to Class I HLA in the refractory pa- When exact matches are unavailable, it is
tient's serum. (See Chapter 16 for more informa- useful to directly determine the specificityof the
tion on detecting HLA antibodies.) Other im- patient's HLA antibodies and select donors
mune causes to be considered include whose platelets lack the corresponding anti-
antibodies to HPA, ABO incompatibility, and gens.48,49,SI This antibody specificity prediction
drug-induced antibodies. (ASP)method is at least as effective as selection
Poor platelet recovery [l-hour CCI) is usually by HLAmatching or platelet crossmatching and
caused by antibody-mediated destruction, se- is superior to random selection of platelets. Fur-
vere splenic sequestration, massive bleeding, or thermore, many more potential HLA-typeddo-
a combination of nonimmune factors affecting nors are identified by the ASP method than are
platelet survival (18- to 24-hour. CCI). Some of available using traditional HLA-matching crite-
the most commonly cited nonimIIlWle factors ria."
associated with refractoriness are listed in Table A popular, although outmoded, system grad-
15-3. Even when possible immune causes of re- ed components by their degree of mismatch,
fractoriness are identified, nonimmune factors that is, within or outside so-called cross-reactive
are often simultaneously present48,49 groups (CREGs). An "A grade" was a perfect
four-out-of-fourmatch for the HLA-Aand HLA-B
Selection of Platelets for Transfusion in
alleles. A BU/B2U grade contained a single mis-
match with one or two antigens, respectively. A
Patients with Alloimmune Refractoriness
BXgrade had a single mismatch, with one anti-
Several strategies may be considered when se- gen cross-reactive with a known antibody with-
lecting platelets for transfusion to patients with in a CREG. C and D grade matches had three
TABLE 15·3. Causesof Platelet Retractorlness"
HLA antibodies
Medications (eg, amphotericin, vancomycin) ABO incompatibility
Hepatic or splenic sequestration Human platelet antigen (HPA) antibodies
Sepsis Druq-dependent.autoantibodles
Disseminated intravascular coagulation Autoantibodies secondary to Iymphoproliferative
disease
Hemorrhage Immune thrombocytopenia
Graftcvs-host disease
Prolonged platelet storage
Thrombotic microangiopathy (TIP; HUS,
drug~induced)
TTP = thrombotic thrombocytopenia; HUS = hemolytic uremic syndrome.
and four mismatched alleles, respectively. Since Hl.Aselected and -crossmatched platelets
the advent of the single-bead assay, the impor- should be irradiated to prevent transfusion-
tance of CREGs has diminished, as antigen- associated graft-vs-hostdisease (TAGVHD).55(P41)
specific reactivity is known (see Chapter 16 for Because these platelets are selected to mini-
more detail). mize incompatible antigens, they are more like-
A best-mismatch estimation can also be per- ly to cause TAGVHDbecause the recipient's im-
formed in a more automated fashion by compar- mune system may fall to recognize donor T
ing donor and recipient immunogenic epitopes, lymphocytes as foreign. Irradiation prevents the
stereochemical amino acid "eplets," on ex- lymphocyteproliferationrequired for TAGVHD.
posed quaternary HLA protein structures. An
Excel spreadsheet is availablethat computes the Fetal and Neonatal Alfoimmune
total number of mismatches between donor and Thrombocytopenia
recipient HLA antigens, with <11 mismatches
FNAIT (also known as neonatal alloimmune
resulting in better posttransfusion CCls.52
thrombocytopenia and abbreviated as NATP or
Pretransfusion crossmatching of the patient's
NAIT)is a syndrome involvingimmune destruc-
serum against platelets from potential donors is
tion of fetal platelets by maternal antibody anal-
an additional approach to provide effective
ogous to red cell destruction in hemolytic
platelet transfusions to patients with alloim-
disease of the fetus and newborn. During preg-
mune retractortness." Each potential platelet nancy, a mother may become sensitized to an
unit is tested in the crossmatch assaywith a cur- incompatible paternal antigen on fetal platelets.
rent sample of the patient's serum. The solid- IgG specific for the platelet antigen crosses the
phase red cell adherence (SPRCA)test is the placenta, causing immune platelet destruction
most widely used method." Although less suc- and thrombocytopenia.
cessful than published HLA-identicalor antigen- FNAITis the most common cause of severe
negative success rates, crossmatching is more fetal/neonatal thrombocytopenia, and affected
widely available and Significantly faster than infants are at risk of major bleeding complica-
waiting for HLA test results." It avoids exclu- tions, especially intracranial hemorrhage. The
sion of HLA-mismatchedbut compatible donors most commonly implicated platelet antigen in-
and has the added advantage of facilitating the compatibility in FNAITis HPA-1a,but all HPAs
selection of platelets when platelet-specificanti- identified to date have been implicated.56 A se-
bodies are present. rologic diagnosis of FNAITmay be made by: 1)
Platelet crossmatching, however, will not al- testing maternal serum for platelet antibodies
ways be successful, particularly when patients using assays that can differentiate platelet-
are highly alloimmunized or have interfering specificfrom non-platelet-specificreactivity, and
ABO antibodies, which can make finding suffi- 2) performing platelet genotyping on parental
cient amounts of compatible platelets problem- DNA.57 Demonstration of both a platelet-specific
atic. Although the incidence of platelet-specific (HPA) antibody in the maternal serum and the
antibodies causing patients to be transfusion- corresponding incompatibility for the antigen in
refractory is very small, this possibilityshould be the parental platelet types confirms the diagno-
investigated when most of the crossmatches are sis.
unexpectedly incompatible or when HLA- Treatment of acutely thrombocytopenic new-
matched transfusions fail. If platelet-specific an- borns includes the administration of intravenous
tibodies are present, donors of known platelet immune globulin(MG) with or without antigen-
antigen phenotype or familymembers, who may compatible platelet transfusions, occasionally
be more likely to share the patient's phenotype, provided by the mother as washed components
should be tested. Platelet crossmatching or HPA when higher-frequency antigen-negative donors
genotyping should be considered for patients are unavailable." When antigen-negative plate-
who do not respond to ABO-and HLA-compatible lets are not available, random units may also be
platelets. effective." Once the diagnosis of FNAIT has
been made in a family, subsequent fetuses are at Drug-Induc~d Thrombocytop~nia

risk. Antenatal' treatment with WIG with or
Thrombocytopenia caused by drug-induced
without steroids has proven to be an effective platelet antibodies is a recognized complication
means of moderating' fetal thrombocytopenia of drugtherapy..Drugs commonly implicated in-
and preventing intracranial hemorrhage." (Fora clude quinine, sulfa drugs, vancomycin, pipera-
more in-depth discussion of FNAIT,see Chapter cillin, GPIIb/IIIa antagonists, and heparin.65,66
23.) Both drtig'dependent antibodies and non-drug-
dependent antibodies may be produced. Non'
Posttransfusion Purpura drug-dependent antibodies, although stimu-
PTP is a rare syndrome characterized by the de- lated by 'drugs, do not require the continued
velopment of dramatic, sudden, and self-limiting presence of the drug to be reactive with plate-
thrombocytopenia 5 to 10 days after a blood lets and are serologicallyindistinguishable from
transfusion in patients with a previous history of other platelet autoantibodies.
Although several mechanisms for drug-
HPAsensitization by pregnancy or transfusion."
induced antibodyformationhave been described,
Coincident with the thrombocytopenia is the re-
most clinicallyrelevant drug-dependent platelet
crudescence of a potent platelet-specificalloanti- antibodies are thought to result when a drug in-
body, usually anti-HPA-la, in the patient's plas- teracts with platelet membrane glycoproteins,
ma. Other specificities have been implicated; inducing conformational changes recognized by
these are almost always associatedwith antigens the humoral immune system and development
on GPIIb/IIIa. The patie~t'~ own. antigen- of drug-dependent antibodies.67,68 These anti-
negative platelets as well as transfused platelets bodies can cause the sudden, rapid onset of
are destroyed. The pathogenesis of autologous thrombocytopenia that usually resolves within 3
platelet destruction is not fully understood. to 4 days after drug discontinuation.
Mounting evidence suggeststhe development of Immune responses triggered by exposure to
transient platelet autoantibodies along with the heparin are particularly important because of
alloantibodtes.f These panreactive autoantibod- both the widespread use of this anticoagulant
ies often target the same glycoprotein that ex- and the devastating thrombotic complications
presses the allo-targetedHPA. associated with the heparin-induced thrombocy-
Platelet antibody assaysusually reveal serum topenia (HIT)syndrome." The incidence rate of
antibody specificity,usually anti-HPA-la. Geno- HITis unknown, but it may develop in up to 5%
typing documents the absence of HPAl a or oth- of patients treated with unfractionated heparin.
er platelet-specific antigens. IVIG is first-line Low-molecular-weightheparin is less likely to be
therapy, successfullyelevating platelet counts in associated with HIT than unfractionated hepa-
days. Plasma exchange, which is less effective rin.
A reduction in baseline platelet count by 30%
than the primary therapy, is employed in the
to 50% generally occurs within 5 to 14 days af-
10%to 15% ofIVIG failures." Antigen-negative
ter primary exposure to heparin, or sooner if the
platelets transfused following the institution of patient has been exposed to heparin within the
treatment appear to survive' somewhat better last 3 months. The platelet count is often
than unselected units/" <lOO,OOO/]1Lbut usually recovers within 5 to
After recovery, future platelet transfusions 7 days upon discontinuation of heparin. More
should be from antigen-negative donors. Inter- than 50% of patients with HIT develop throm-
estingly, the frequency of PTP has significantly bosis, which can occur in the arterial or venous
decreased with the introduction of leukocyte- system, or both." Patients may develop some-
reduced blood components." Although no data times-fatalstroke, myocardial infarction, limb or
have been reported to explain this trend, use of other organ ischemia, or deep venous thrombo-
leukocyte-reduced components may also be of sis. It is thus of critical importance to discontin-
benefit in reducing the risk of PTP. ue heparin therapy when a diagnosis of HIT is
suspected. Moreover, strong considerationshould my, if used, is reserved for children whose dis-
be given to using an alternative (nonheparin) an- ease is severe and lasts >6 months; this condi-
ticoagulant (eg, a direct thrombin inhibitor) to tion is similar to chronic ITP in adults.
prevent thrombosis." Rituximab and various thrombopoietin receptor
The mechanism of HIT involves formation of agonistshave been used as second-line therapies
a complex between heparin and platelet factor 4 for acute ITP.72,73
(PF4), a tetrameric protein released from plate- Studies of both sera and washed platelets
let exgranules. IgG antibodies to the complex at- from patients with ITPhave identified IgG, IgM,
tach secondarily to platelet FcyRIIareceptors via and 19A autoantibodies that are reactive with a
their Fc region, resulting in platelet activation, number of platelet surface-membrane struc-
thrombin generation, and thrombosis. tures that most often include GP complexes lIb/
IlIa, Ia/lIa, and Ib/IX but can alsoinclude GPIV,
Autoimmune or Immune GPV,and GPVI.75In the majorityof cases,platelet-
Thrombocytopenia associated autoantibodies are reactive with two
or more platelet glycoproteins." There is no
Immune thrombocytopenia (ITP)is an immune
compelling evidence to date suggesting that a
platelet disorder in which autoantibodies are di-
patient's profile of autoantibody specificitiescor-
rected against platelet antigens, resulting in
relates with the severity of the disease or pre-
platelet destructton.l'-? Chronic ITP, which is
dicts that patient's response to therapy.
most common in adults, is characterized by an
insidious onset and moderate thrombocytopenia
Testing for Platelet Antigens and
that may exist for months to years before diag-
Antibodies
nosis. Females are twice as likely to be affected
as males. Laboratory detection of platelet antibodies pro-
Spontaneous remissions are rare, and treat- vides important results to aid in clinical diagno-
ment is usually required to raise the platelet sis of an immune platelet disorder. A compre-
count. First-line therapy consists of steroids or hensive workup for platelet antibodies requires
MG followed by more potent immunosuppres- the use of multiple test methods, including a
sive agents and, less frequently, splenectomy in glycoprotein-specific assay, a test employing
nonresponders." Many other therapies have intact/whole platelets, and HPAgenotyping.F:"
been used in patients who do not respond to Glycoprotein-specificassays are the most sensi-
splenectomy, with variable results. tive and specificfor identifyingthe HPAspecific-
Chronic ITP may be idiopathic or associated ity of serum antibodies (Fig15-1).
with other conditions, such as human immuno- The inclusion of assays that use intact plate-
deficiency virus infection, malignancy, or other let targets is critical for detection of antibodies
autoimmune diseases. Acute ITP is mainly a that can be missed by glycoprotein-specifictests
childhood disease characterized by the abrupt because the process of platelet lysis with deter-
onset of severe thrombocytopenia and bleeding gent and capture of glycoprotein with a specific
symptoms, often after a viral infection. The ma- monoclonal antibody can disrupt HPA epitopes
jority of cases resolve spontaneously over a 2- to recognized by some antibodies. HPAgenotyping
o-month period. If treatment is required, MG by DNA methods is helpful to confirm the HPA
or anti-D immunoglobulin infusions given to specificityof the alloantibodies and for prenatal
D-positive patients are usually effective in rais- typing of a fetus in suspected cases of FNAIT.
ing platelet counts. Ostensibly blockading retic- The test methods that follow are examples of
uloendothelial system Fc receptors, these mo- the current state-of-the-artmethods used by ref-
dalities may have additional mechanisms of erence laboratories. For in-depth descriptions of
action." Steroids are used less often because of platelet antibody and antigen testing, readers
their serious side effects in children. Splenecto- should consult recent reviews.57,78,79
Murine GP-
specific MoAb
+ r/ Incubate, wash,
/~Iyse
Incubate,
wash, lyse
\
Enzyme-labeled
anti-human IgG
Patient GP-
specific IgG Ab
PlateletGP
Murine GP- ntl-rnouse IgG
specific MoAb
ACE MACE MAIPA
FIGURE 15-1. Enzyme-linkedimmunosorbent assay(ELISA)testing. Antigen capture ELISA(ACE)employs

microtiter well plates precoatedwith adherent glycoproteins (GPs) to screen patient serum for GP-specific
antibody (Ab). It can be modified (modified-ACE, or MACE)by preincubation of patient serum with target
platelets, followed by washing and lytic solubilization of platelet GPs. The lysate is added to microtiter
plate wells for capture of platelet GP by a specific mouse immunoglobulin G (lgG) monoclonal antibody
(MoAb). Platelet GP-specific IgGAb in the patient's serum bound to the GP is detectedby the addition of
an enzyme-labeledgoat antihuman IgGand chromogenic substrate. The monoclonal Ab immobilization of
platelet antigens (MAIPA) assay is very similar to the MACE, but the patient's serum and MoAb are
incubated with platelets before washing and platelet GPsolubilization, and the patient GP-specificAb/GPI
MoAb complex is captured by goat antimouse IgG adherent to the well bottom. There is a progressive
increase in test sensitivity from ACEto MAIPA.
Assays Using Intact Platelets with low surface copy numbers may also be dif-
ficult to detect.
An assay that is widely used for the detection of
Flow cytometry is commonly used for immu-
platelet-speclflcantibodies and for platelet cross- nofluorescent detection of platelet antibodies us-
matching is the SPRCAassay." Intact platelets ing intact platelets.57 Following incubation of
are immobilized in round-bottomed wells of a platelets with the patient's serum, platelet-
microtiter plate and are then incubated with the bound antibodies are detected with a fluores-
patient's serum. After washing, detector red cently labeled antiglobulin reagent specific for
cells coated with antihuman IgG are added, and human IgG or IgM. The results can be ex-
the mixture is centrifuged and examined visual- pressed as a ratio of mean or median channel
ly. The method's main limitations are its subjec- fluorescence of platelets sensitized with patient
tive endpoint and failure to distinguish platelet- serum to that of platelets incubated with nega-
specific (ie, platelet glycoprotein-directed/HPA) tive control serum. Platelet autoantibodies coat-
from non-platelet-specificantibodies (le,ABO or ing patient platelets can also be detected in a di-
HLA antibodies). Antibodies to glycoproteins rect flow cytometry assay."
468 AA B B TE C H N I CAL MAN UAL
Flow cytometry has proven to be a very These techniques are reliable, but they are also
sensitive method for detection of antibodies to laborious and time consuming. Higher-through-
platelets. Alloantibodies specific for labile epi- put methods have been developed, such as real-
topes and unreliably detected by ACM can be time PCR, melting curve analysis, allele-specific
detected with intact platelets using flow cytome- fluorescent beadchip probes, and next-
try." Flow cytometry does not differentiate be- generation sequencing.78,84
tween platelet-specific and non-platelet-specific
antibodies. This is a drawback when investigat- Testing for Platelet Autoantibodies
ing FNAITor PTPbecause more relevant plate1et-
Numerous assayshave been developed to detect
specific antibodies can be obscured by non-
platelet autoantibodies in patients with ITP. Al-
platelet-specificreactivity.
though many tests are quite sensitive, particular-
ly in detecting cell-surface, platelet-associated
Antigen Capture and Other Assays
immunoglobulins, none has been sufficientiy
Platelet glycoprotein ACM are used to deter- specificto be particularly useful in either the di-
mine the HPA that is recognized by platelet agnosis or the management of ITP. The Ameri-
antibodies in a patient's serum. Commonly can Society of Hematology practice guidelines
employed assays include enzyme-linked immu- for ITP state that serologic testing is unneces-
nosorbent assays (ACE, MACE, and MAIPA) sary, assuming that the clinicalfindings are com-
(Fig 15_1).57,81 The assays require the use of patible with the diagnosis." However, platelet
monoclonal antibodies that recognize the target antibody tests may be helpful in the evaluation
antigens of interest but do not compete with the of patients suspected of having ITP when non-
patient's antibody. These assays capture specific immune causes may be present. The goal of se-
platelet glycoproteins on plastic wells of a micro- rologic testing in ITP is to detect autoantibody
plate after sensitization with the patient's se- bound to the patient's own platelets with or
rum. The patient's bound antibody is detected without demonstration of similar reactivity in
with an enzyme-labeled antihuman globulin. patient plasma.
Because only the glycoproteins of interest are Newer assays are designed to detect immu-
immobilized, interference by reactions from noglobulin binding to platelet-specific epitopes
non-platelet-specific antibodies, especially anti- found on platelet GPIIb/IIIa, GPla/IIa, andlor
HLA,is eliminated. A different solid-phaseassay GPlblVIIX complexes. These solid-phase, GP-
affixes HPA antigens to microbeads that are ex- specificassaysappear to have improved specific-
posed first to patient serum and then a fluores- ity in distinguishing ITP from nonimmune
cently labeled antihuman globulin. Specific thrombocytopenia, but this benefit is often bal-
binding can be detected with the use of a Lu- anced by a decrease in sensitivity.76,85One com-
minex mini-flowp1atform.82,83 Care must be tak- mercially available test uses eluates prepared
en when using commercial ACM that antibod- from the patient's washed platelets." The elu-
ies to the HPAsmost commonly associated with ates are tested against a panel of monoclonal-
specific disease states are detectable by the asantibody-immobilizedplatelet glycoprotein com-
say. plexes, and platelet antibodies are detected
using an enzyme-linked antihuman globulin. In
Platelet Genotyping the indirect phase of the assay,patient plasma is
Genotyping for the SNPs in the genes encoding tested against the same glycoprotein panel. Al-
HPA can be performed by any of the myriad though autoantibodies are most often detected
molecular methods available. Allele-specific in the eluates, they are infrequently detected (in
polymerase chain reaction (PCR)and restriction approximately 17% of cases) in the plasma. Pa-
fragment length polymorphism analysis are two tients with ITP may have antibodies that are re-
methods that have been used successfully.77 active with one or several GP targets."
Testing for Drug-Dependent Platelet phosphate release, phosphatidylserine exposure,

Antibodies etc). In asymptomatic patients receiving hepa-
rin or an.ticipatingits use, neither PF4 ELI~J\nor
Any plqtel~t serologytest.used to detect platelet· functional tests are sufficiently predictive of HIT
bound immunoglobulin can be modified to de- to warrant use as screening tests."
tect drug-dependent antibodies. Each patient se-
rum sample should be tested against normal
platelets in the presence and absence of the RANUlOC N IGENS
drug. Moreover, at least one normal serum sam- AND ANTIBODiES
ple should be tested with and without the drug
to control for nonspecific antibody binding that Antibodies against granulocyte (neutrophil) anti-
may occur in the drug's presence. A positive gens are implicated in the following clinical
control sample known to be reactive with the syndromes: neonatal aIloimmune neutropenia
drug being assayed should be tested with and (NAN), transfusion-related acute lung injury
without the drug to complete the evaluation. A (TRALI),febrile transfusion reactions, primary
positive result demonstrates greater reactivity or secondary autoimmune neutropenia (AIN),
against normal platelets in the presence of the refractoriness to granulocyte transfusion,
drug than without and demonstrates that the transfusion-related alloimune neutropenia, and
drug did not nonspecificallycause a positive re- immune neutropenia ..after hematopoietic .pro-
sult with normal serum controls. Flow cytorne- genitor cell (HPC) transplantation. To date, 10
try is the most sensitive and most commonly neutrophil antigens carried on five different gly-
used method to detect both IgG and IgM drug- coproteins have been characterized and given
dependent antibodies.57,86 Limitations to detec- human neutrophil alloantigen (HNA).deslgI"la-
tion of drug-dependent platelet antibodies in- tions by the Granulocyte Antigen Working Party
clude the following: 1) for many drugs, the opti- of the International Societyof BloodTransfusion
mal concentration for antibody detection has (Table 15-4).90This nomenclature system fol-
not been determined.. and hydrophobic drugs lows a convention similar to that used for HPA
are difficultto solubilize;2) the presence of non- nomenclature. Several of the antigens on granu-
drug antibodies can mask drug-related antibod- locytes are shared with other cells and are not
ies; and 3) a patient may be sensitized to a drug granulocyte specific. ..
metabolite and not the native drug.
Assays for heparin-dependent antibodies in- HNA
clude ELISAusing microtiter wells coated with
complexes of PF4 and heparin or heparin-like Antigens on FcyRlllb
molecules (eg, polyvinyl sulfonate)_87 Optical The first granulocyte-specific antigen detected
density values above cutoff that can be inhibited was NA1, later named HNA 1a. Five alleles of
by added high-dose heparin confirms the pres- HNA-1 have now been identified, encoding
ence of heparin-dependent antibodies. Al- three antigens: HNA1a, HNAlb, and HNA1c,
though IgG antibodies are the most clinically which are located on the protein FcyRIIIb
relevant antibodies, afew patients with HITap- (CD16b).90Recently,another epitope onHl-l-Ib,
pear to have only.IgM or 19Aantibodies detect- termed HNA ld, was described." FcyRlllbis a
able in variants of this assay. GPHinlcedprotein receptor for the Fc region of
These assays are sensitive but not specificfor IgG and is present oniy on the surfaces of
the subset of antibodies that result in clinically neutrophils. Three alleles encode HNA lb: one
significant platelet activation and thrombosis. encoding HNA 1b alone, one along with HNA
The HC-serotonin release assay (SRA)is a func- 1c, and another with HNA 1d.91Neutrophils ex-
tional assay for detection of this type of anti- press 100,000 to 200,000 molecules of
body." Other functional tests include heparin- FcyRIIIb,but there are rare individuals (approxt-
induced platelet aggregation and various other mately 0.1 %) whose neutrophils express no
measures of platelet activation (adenosine tri- FcyRIIIb(CD16 null) and who can produce anti-
TABLE 15·4. Human Neutrophil Antigens

___ ~,~~~_ ~__ , ~~"'_~~'~' "'_~,~" =-"~--===-==<r-=,_
Phenotypic Amino Acid Encoding Nucleotide

Antigen Frequency* Glycoprotein Change Gene Change
HNA-1a 12% ala CD16b Multiplet FCGR38 Multiplet
HNA-1b 54% alb
46% bib
HNA-1c 5%
HNA-2 97% CD177+ CD177 N/A CD177 N/A
HNA-2 null 3% CD177-
HNA-3a 56%-59% ala CTL2 Arg152Gln SLC44A2 455G>A
HNA-3b 34%-40% alb
3%-6% bib
HNA-4a 78.6% ala CD11b Arg61 His ITGAM 230G>A
HNA-4b 19.3% alb
2.1% bib
HNA-5a 54.3% ala CD11a Arg766Thr ITGAL 2466G>C
HNA-5b 38.6% alb
7.1% bib
Nucleotide Changes Amino Acid Changes

141 147 227 266 277 349 36 38 65 78 82 106
HNA-1a G C A C G G Arg Leu Asn Ala Asp Val
HNA-1b C T G C A A Ser Leu Ser Ala Asn lie
HNA-1c C T G A A A Ser Leu Ser Asp Asn lie
'Phenotypic frequencies are for people of Europeanancestry who live in North America.
tHNA-1 amino acid and nucleotide changesare shown in a separatesection of the table.
HNA = human neutrophil antigen; N/A = not applicable.
bodies that are reactive with FcyRIIIb when CD177 protein, which is missing from the neu-
they are exposed to it through transfusion or trophils of immunized individuals. Neutrophils
pregnancy.92,93Antibodies to HNA-la and -lb from approximately 1% to 11% of people lack
have been implicated in TRALI,NAN, and AIN, expression of CD177.94-96The genetic basis of
and antibodies to HNA-lc and -1d have resulted CD177 deficiency is SNPs leading to truncated
in NAN.91,94,95
mRNAswith premature stop cedens." Interest·
tngly, CD177 is expressed only on a neutrophil
Antigens on CD 177 subpopulation in CD177-positive individuals."
HNA-2 (previouslyknown as NB1) is not an al- The proportion of the CD177-positive neutro-
loantigen because HNA-2 antibodies are isoanti- phil population ranges from 0% to 100%, with a
bodies that recognize common epitopes on median of 60%.99,100 Among individuals homo-
zygous for CD177, neutrophil subpopulations but several have been found during screening of
are due to the epigenetic silencing of either the the serum of multiparous blood donors and re-
maternal or paternal allele, with both alleles in- cently reported to result in NAN.108
activated in some neutrophils.l'" Both CD177
alleles are expressed in hematopoietic stem Antigens on CD7 la andCtntb
cells, but one of the two alleles is silenced
during neutrophil differentiation.100This ex- HNA-4a/4b and HNA-5a antigens are present
plains the increased CD177 subpopulation size on monocytes and lymphocytes as well as gran-
in healthy people given granulocyte colony- ulocytes. HNA-4a is carried on the CDllb/18
stimulating factor (G-CSF)and in patients with (Mad, CR3, umb2) glycoprotein." CD11b/18
polycythemia vera. plays a role in neutrophil adhesion to endotheli-
In neutrophils, CD177 is physically associat- al cells and phagocytosis of C3bi opsonized mi-
ed with the neutrophil serine proteinase 3 (PR3) crobes. There is some evidence showing that
and the b2 integrin Mad (CDllb/CD18). pathogenic alloantibodies against HNA-4a inter-
CD177 recognizes platelet and endothelial cell fere with CDllb/18-dependent neutrophil ad-
adhesion molecule 1 (PECAM-1), an immuno- hesion and enhance neutrophil respiratory
globulin family molecule expressed on endothe- burst.'?" Antibodies against HNA-4a and -4b
lial cells, platelets, and some leukocytes. The have been implicated in NAN, and autoantibod-
migration, through human umbilical vein endo- ies against CD11b/18 have also been de-
thelial cells, of neutrophils that express CD177 scribed. 110,111
is more efficient than that.of CD177-negative HNA-5a is carried on CDlla/18 (LFA-l,
neutrophils due to a .. PECAM-l-dependent uLb2) glycoproteln.!" CDlla/18, like CD11b/
mechanism. Recent studies have shown that 18, plays a role in neutrophil adhesion to endo-
CD177 signals in a b2-integrin-dependent man- thelial cells. Antibodies that are reactive with
ner, which leads to impaired neutrophil migra- HNA-5a have been found in a chronic transfu-
tion.'?' sion recipient with aplastic anemia and have
Antibodies against HNA-2 have been impli- also been reported to be associated with
cated in NAN, TRALI,AIN, and neutropenia in NAN.lll,113
hematopoietic stem cell transplant recipients.102-104
Interestingly, autoantibodies directed against the
Other Neutrophil Antigens
CD177-associated protein PR3 are found in pa-
tients with antineutrophil cytoplasmic antibody Neutrophils do not express ABH or other red
(ANCA)-associated vasculitis. Patients with cell group antigens, but they do express modest
ANCAvasculitis have been found to have a larg- amounts of HLAClass I, as well as HLA Class II
er mean CD177 subpopulation size than upon activation.
healthy people. 100
Immune Neutrophil Disorders
Antigens on CTL2
Neonatal Alloimmune Neutropenia
HNA-3a and HNA-3b are carried on the choline
transporter-like protein 2 (CTL2),and an SNP in NAN is caused by maternal antibodies against
the gene (SLC44A2) accounts for the polymor- the antigens on fetal neutrophils; the: most fre-
phism (Table 15_4).105,106 CTL2 is also expressed quent specificities are those against HNA-1a,
on both T and Blymphocytes, platelets, and vas- HNA-1b, and HNA-2 antigens, although
cular endothelium. HNA-3a antibodies are usu- HNA-1c, -Id, -3a, -3b, ,Aa, -4b, and -Sa have
ally agglutinins. They occasionally develop in caused this syndrome. NAN may also occur in
women after pregnancy, and HNA-3a antibodies the children of women who lack the FcyRIIIb
are the most frequent cause of fatal TRALI, in protein. Neutropenia in NAN can occasionally
addition to causing febrile reactions and be life-threatening because of increased suscepti-
NAN.107HNA-3b antibodies are rarely detected, bility to infection. 114Management with antlbiot-
ics, MG, G-CSF,and/or plasma exchange may able. Class I HLAantibodies that are often pres-
be helpful. ent in patient sera complicate detection and
identification of granulocyte antibodies. For
TRALI these reasons, it is critical that granulocyte anti-
body and antigen testing be performed by an ex-
TRALIis an acute, often life-threatening reac-
tion characterized by respiratory distress, hypo- perienced laboratory using appropriate controls.
tension or hypertension, and noncardiogenic
Granulocyte Agglutination Test
pulmonary edema that occurs within 6 hours of
a blood component transfusion.115TRALIrepre- This was one of the first tests developed for the
sented 30% of transfusion-associated fatalities detection of granulocyte antibodies. It is typical-
reported to the US Food and Drug Administra- ly performed by overnight incubation of small
tion over the last 5 reported fiscalyears, second volumes of isolated fresh neutrophils with the
only to circulatory overload fatalitles.!" In se- patient's serum in a microplate. The wells are
vere TRALI,causative antibodies are most often viewed under an inverted phase microscope for
found in the plasma of blood donors. When neutrophil agglutination or aggregation.
these antibodies are transfused, they cause acti-
vation of primed neutrophils that are seques- Granulocyte Immunofluorescence Test
tered in the lungs of certain patients. The acti-
vated neutrophils undergo oxidative burst, This test also requires fresh target cells that are
releasing toxic substances that damage pulmo- incubated, usually at room temperature for 30
nary endothelium and resulting in capillaryleak minutes, and washed in EDTA and phosphate-
and pulmonary edema. HNA and Class II HLA buffered saline. Neutrophil-bound antibodies are
antibodies are thought to be more pathogenic then detected with fluorescein-isothiocyanate-
than Class I HLA antibodies.!" (For a more in- labeled antihuman IgG or IgM using either a flu-
depth discussion ofTRALI,see Chapter 22.) orescence microscope or a flow cytometer. 120A
combination of agglutination and immunofluo-
Autoimmune Neutropenia rescence tests is beneficial.106Other methods in-
clude chemiluminescence, SPRCA, and the
AIN may occur in adults or in infants. When monoclonal antibody-specificimmobilization of
present in adults, it is generally persistent and granulocyte antigens (MAIGA)assay, which is
may be idiopathic or secondary to autoimmune similar to the MAIPAassaybut uses monoclonal
diseases, malignancies, or infections.!" In AIN antibodies to capture the various glycoproteins
of infancy, the autoantibody has neutrophil anti- that express HNA. The MAIGAassay is used to
gen specificity (usually HNA-la, but occasional- differentiate between HLA- and HNA-specific
ly -Ib or -4a) in about 60% of patients.!" This antibodies.
condition is generally self-limitingand relatively
benign, with recovery usually occurring in 7 to
luminex-Based Antibody Identification
24 months. I19
Microbeads coated with HNA antigens have
Testing for Granulocyte Antibodies and been developed that can detect anti-HNA-la,
Antigens -Ib, -Ic, -2, -3a, -3b, -4a, -Sa, and -5b antibod-
ies.121Although rates of false positives remain
Granulocyte antibody testing is technically com- high (5.5%), for anti-HNA-la, -Ib, -l c, -2, and
plex and labor-intensive. The inability to main- -3a, sensitivityhas been reported to be ::::90%.121
tain the integrity of granulocytes at room tem-
perature, in refrigerated conditions, or by
HNATyping
cryopreservation requires that cells be isolated
from fresh blood on each day of testing. This de- As with HPA, typing for HNA is performed
mands that readily availableblood donors typed largely using molecular methods to detect the
for the various granulocyte antigens be avail- allelic variants that determine the antigens. Any
methods used in HPA typing can be applied to HNA-2/CD 177 has traditionally required sero-
HNA typing with simple modifications to the logic testing on freshly isolated neutrophils us-
primer and probe sequences. Readers are re- ing specificmonoclonal antibodies. Testing may
ferred to several publications on this sub- evolve as the SNPsresponsible for some individ-
ject.122,123 Because the molecular defect that re- uals' loss of expression are confirmed."
sults in CD177 deficiency is variable, typing for
REFERENCES
1. Deppermann C, Kubes P. Start a fire, kill the 3. PasaIic L, Wang SS, Chen VM. Platelets as bio-
bug: The role of platelets in inflammation and markers of coronary artery disease. Semin
infection. Innate Immun 2018;24(6):335-48. Thromb Hemost 2016;42:223-33.
2. Franco AT, Corken A, Ware J. Platelets at the in- 4. Immuno polymorphism database. All HPA ge-
terface of thrombosis, inflammation and cancer. netic information. Hinxton, UK: European Bio-
Blood 2015;126:582-8. informatics Institute, 2019. [AVailableat http://
www.ebLac.uklipd/hpa/table2.html (accessed 17. Peterson JA, Balthazor SM, Curtis BR, et al.
June 7, 20l9).J Maternal alloimmunization against the rare
5. Metcalfe P, Watkins NA, Ouwehand WH, et al. platelet-specificantigen HPA-9b[Max"]is an im-
Nomenclature of human platelet antigens. Vox portant cause of neonatal alloimmune thrombo-
Sang 2003;85:240·5. cytopenia. Transfusion2005;45: 1487-95.
6. Iallu V, Beranger T, Bianchi F, et al. Cab4b, the 18. Kaplan C, Porcelijn L, Vanlieferinghen P, et al.
first human platelet antigen carried by glycopro- Anti-HPA-9bw(Mar) fetomaternal alloimmuni-
tein IX discovered in a context of severe neona- zation, a clinicallysevere neonatal thrombocyto-
tal thrombocytopenia. J Thromb Hemost 2017; penia: Difficulties in diagnosis and therapy and
15:1646-54. report on eight families. Transfusion 2005;45:
7. Sullivan MJ, Kuhlmann R, Peterson JA, Curtis 1799-803.
BR. Severe neonatal alloimmune thrombocyto- 19. Andrews RK, Berndt MC. Bernard-Souliersyn-
penia caused by maternal sensitization against a drome: An update. Semin Thromb Hemost
new low-frequency alloantigen [Dom'') located 2013;39:656-62.
on platelet glycoprotein IIIa. Transfusion 2017; 20. Corral J, RiveraJ, Gonzalez-ConejeroR, Vicente
57:1847-8. V. The number of platelet glycoproteinIa mole-
8. Wihadmadyatami H, Heidinger K, Roder L, et cules is associated with the genetically linked
al. Alioantibodyagainst new platelet alloantigen 807 CIT and HPA-5polymorphisms. Transfu-
(Lap') on glycoprotein IIb is responsible for a sion 1999;39:372-8.
case of fetal and neonatal alloimmune thrombo- 21. Ertel K, Al-Tawil M, Santoso S, Kroll H. Rele-
cytopenia. Transfusion2015;55:2920-9. vance of the HPA15 (Gov) polymorphism on
9. Aster RH, Newman Pl. HPAla!b(PIA1/ CD109 in alloimmune thrombocytopenic syn-
AZ,Zwa/b): The odyssey of an alloantigen sys- dromes. Transfusion2005;45:366-73.
tem. Immunohematology 2007;23:2-8. 22. Mandelbaum M, Koren D, EichelbergerB, et al.
10. Bennett JS, Berger BW, BillingsPC. The struc- Frequencies of maternal platelet alloantibodies
ture and function of platelet integrins.J Thromb and autoantibodies in suspected fetal/neonatal
Haemost 2009;7(Suppl 1):200-5. alloimmune thrombocytopenia, with emphasis
11. Nurden AT, Pillois X, Wilcox DA. Glanzmann on human platelet antigen-5 alloimmunization.
thrombasthenia: State of the art and future di- Vox Sang2005;89:39-43.
rections. Semin Thromb Hemost 2013;39:642- 23. Berry JE, Murphy CM, Smith GA, et al. Detec-
55. tion of Gov system antibodies by MAIPA reveals
12. Lucas GF, Metcalfe P. Platelet and granulocyte an immunogenicity similarto the HPA5 alloan-
glycoprotein po1ymorphisms. Transfus Med tigens. BrJ Haemato12000;110:735-42.
2000;10:157-74. 24. Vassallo RR. Recognition and management of
13. Harrison CR, Curtis BR,McFarlandJG, et al. Se- antibodies to human platelet antigens in platelet
vere neonatal alloimmune thrombocytopenia transfusion-refractory patients. Immunohema-
caused by antibodies to human platelet antigen tology 2009;25: 119-24.
3a (Bak")detectable only in whole platelet as- 25. Curtis BR,Edwards IT, Hessner MJ, et al. Blood
says.Transfusion2003;43: 1398-402. group A and B antigens are strongly expressed
14. Socher I, Zwingel C, Santoso S, KrollH. Hetero- on platelets of some individuals.Blood2000;96:
geneity of HPA-3 alloantibodies: Consequences 1574-81.
for the diagnosisof alloimmune thrombocytope- 26. Ogasawara K, Ueki J, Takenaka M, Furihata K.
nic syndromes.Transfusion2008;48:463-72. Study on the expression of ABH antigens on
15. Koh Y, Ishii H, Amakishi E, et al. The first two platelets. Blood 1993;82:993-9.
cases of neonatal alloimmune thrombocytopenia 27. SkogenB, RossebeHansen B, HusebekkA, et al.
associated with the low-frequencyplatelet anti- Minimal expression ofblood groupA antigen on
gen HPA-21bw (Nos) in Japan. Transfusion thrombocytes from AZ individuals. Transfusion
2012;52:1468-75. 1988;28:456-9.
16. Peterson JA, Pechauer SM, Gitter ML, et al. The 28. Slichter SJ, Davis K, Enright H, et al. Factors af-
human platelet antigen-21bw is relatively com- fecting posttransfusion platelet increments,
mon among Asians and is a potential trigger for platelet refractoriness, and platelet transfusion
neonatal alloimmune thrombocytopenia. Trans- intervals in thrombocytopenic patients. Blood
fusion 2012;52:915-16. 2005; 105:4106-14.
29. Triulzi DJ, Assmann SF, Strauss RG, et al. The 41. Saito S, Ota S, Seshimo H, et al. Platelet transfu-
impact of platelet transfusion characteristics on sion refractoriness caused by a mismatch in
post-transfusion platelet increments and clinical HLA-C antigens. Transfusion 2002;42:302-8.
bleeding in patients with hypo-proliferative 42. Seftel MD, Growe GH, Petraszko T, et al. Uni-
thrombocytopenia. Blood 2012; 119:5553-62. versal prestorage 1eukoreduction in Canada de-
30. Heal JM, Rowe JM, Blumberg N. ABO and creases platelet al1oimmunization and refractori-
platelet transfusion revisited. Ann Hematol ness. Blood 2004; 103:333-9.
1993;66:309-14. 43. Estcourt LJ, Malouf R, Hopewell S, et al.
31. Dunstan RA, Simpson MB. Heterogeneous dis- Pathogen-reduced platelets for the prevention of
tribution of antigens on human platelets demon- bleeding. Cochrane Database Syst Rev 2017;7:
strated by fluorescence flow cytometry. Br J CD009072.
Haematol 1985;61 :603-9. 44. Saris A, Kerkhoffs JL, Norris PJ, et al. The role of
32. Spring FA, Judson PA, Daniels GL, et al. A hu- pathogen-reduced platelet transfusions on HLA
man cell-surface glycoprotein that carries alloimmunization in hemato-oncological pa-
Cromer-related blood group antigens on erythro- tients. Transfusion 2019;59:470-81.
cytes and is also expressed on leucocytes and 45. Triulzi DJ, Kleinman S, Kakaiya RM, et al. The
platelets. Immunology 1987;62:307-13. effect of previous pregnancy and transfusion on
33. Ghosh A, Murugusan G, Chen K, et al. Platelet HLA alloimmunization in blood donors: Implica-
CD36 surface expression levels affect functional tions for a transfusion-related acute lung injury
responses to oxidized LSL and are associated risk reduction strategy. Transfusion 2009;49:
1825-35.
with inheritance of specific genetic polymer-
46. Vassallo RR, Hsu S, Einarson M, et al. A compar-
phisms. Blood 2011; 117:6355-66.
ison of two robotic platforms to screen platelet-
34. Curtis BR, Ali S, Glazier AM, et al. Isoimmuni-
pheresis donors for HLA antibodies as part of a
zation against CD36 (glycoprotein IV): Descrip-
transfusion-related acute lung injury mitigation
tion of four cases of neonatal isoimmune throm-
strategy. Transfusion 2010;50: 1766- 77.
bocytopenia and brief review of the literature.
47. Kerkhoffs JL, Eikenboom JC, Van De Watering
Transfusion 2002;42: 1173-9. LM, et al. The clinical impact of platelet refrac-
35. Rac ME, Safranow K, Poncyljusz W. Molecular
toriness: Correlation with bleeding and survival.
basis of human CD36 gene mutations. Mol Med
Transfusion 2008;48: 1959-65.
2007;13:288-96. 48. Hod E, Schwartz J. Platelet transfusion refracto-
36. Bierling P, Godeau B, Fromont P, et al. Post-
riness. Br J Haemato12008;142:348-60.
transfusion purpura-like syndrome associated 49. Vassallo RR Jr. New paradigms in the manage-
with CD36 [Nak"] isoimmunization. Transfu- ment of alloimmune refractoriness to platelet
sion 1995;35:777-82. transfusions. Curr Opin Hematol 2007; 14:655-
37. Ikeda H, Mitani T, Ohnuma M, et al. A new 63.
platelet-specific antigen, Nak", involved in the 50. Bolgiano DC, Larson EB, Slichter SJ, A model to
refractoriness of Hl.Arnatched platelet transfu- determine required pool size for Hl.A'typed
sion. Vox Sang 1989;57:213- 17. community donor apheresis programs. Transfu-
38. Boylan B, Chen H, Rathore V, et al. Anti-GPVI- sion 1989;29:306- 10.
associated ITP: An acquired platelet disorder 51. Petz LD, Garratty G, Calhoun L, et al. Selecting
caused by autoantibody-mediated clearance of donors of platelets for refractory patients on the
the GPVI/FcRy-chain complex from the human basis of HLA antibody specificity. Transfusion
platelet surface. Blood 2004; 104: 1350-5. 2000;40: 1446-56.
39. Akiyama M, Kashiwagi H, Todo K, et al. Pres- 52. Brooks EG, MacPherson BR, Fung MK. Valida-
ence of platelet-associated anti-glycoprotein tion of HLAMatchmaker algorithm in identify-
(GP)VI autoantibodies and restoration of GPVI ing acceptable HLA mismatches for thrombo-
expression in patients with GPVI deficiency. cytopenic patients refractory to platelet
J Thromb Haemost 2009;7: 1373-83. transfusions. Transfusion 2008;48:2159-66.
40. BialekJW, Bodmer W, Bodmer J, Payne R. Dis- 53. Vassallo RR, Fung M, Rebulla P, et al. Utility of
tribution and quantity of leukocyte antigens in cross-matched platelet transfusions in patients
the formed elements of the blood. Transfusion with hypoproliferative thrombocytopenia: A sys-
1966;6: 193-204. tematic review. Transfusion 2014;54: 1180-91.
476 AA B B TEe H N I CAL MAN U A L
54. Rachel JM, Summers TC, Sinor LT, Plapp FV. 66. Reese JA, Li X, Hauben M, et al. Identifying
Use of a solid phase red blood cell adherence drugs that cause acute thrombocytopenia: An
method for pretransfusion platelet compatibility analysis using 3 distinct methods. Blood 2010;
testing. Am J Clin PathoI1988;90:63-8. 116:2127-33.
55. Gammon R, ed. Standards for blood banks and 67. Aster RH, Bougie DW. Drug-induced immune
transfusion services. 32nd ed. Bethesda, MD: thrombocytopenia. N Engl J Med 2007;357:
AABB,2020. 580-7.
56. Curtis BR.Recent progressin understanding the 68. BougieDW, Wilker PR,Aster RH. Patients with
pathogenesis of fetal and neonatal alloimmune quinine-induced immune thrombocytopenia
thrombocytopenia. Br J Haematol 2015; 171: have both "drug-dependent"and "drug-specific"
671-82. antibodies. Blood2006; 108:922-7.
57. Curtis B, McFarland J. Detection and identifica- 69. Greinacher A, Selleng K, Warkentin TE. Auto-
tion of platelet antibodies and antigens in the immune heparin-induced thrombocytopenia.
clinical laboratory. ImmunohematoI2009;25: J Thromb Haemost 2017;15:2099-114.
125-35. 70. Linkins LA, Dans AL, Moores LK, et al. Treat-
58. Peterson JA, McFarland JG, Curtis BR, Aster ment and prevention of heparin-induced throm-
RH. Neonatal alloimmune thrombocytopenia: bocytopenia: Antithrombotic therapy and pre-
Pathogenesis, diagnosis and management. Br J vention of thrombosis. 9th ed. American
Haemato12013;161:3-14.
Collegeof Chest Physiciansevidence-basedclin-
59. Kiefel V, Bassler D, Paes B, et al. Antigen-
ical practice guidelines. Chest 2012; 141(Suppl
positive platelet transfusion in neonatal alloim-
2):e495S-530S.
mune thrombocytopenia (NAIT).Blood 2006;
71. Neunert C, Lim W, Crowther M, et al. The
107:3761-3.
American Society of Hematology 2011
60. Pacheco LD, Berkowitz RL, Moise KJ Ir, et al.
Fetal and neonatal alloimmune thrombocytope- evidence-based practice guideline for immune
nia: A management algorithm based on risk thrombocytopenia.Blood2011; 117:4190-207.
72. Lambert MP, Gernsheimer TB. Clinicalupdates
stratification. Obstet Gynecol 2011; 118:1157-
63. in adult immune thrombocytopenia. Blood
61. McFarlandJG. Posttransfusionpurpura. In: Pop- 2017; 128:2829-35.
ovsky MA, ed. Transfusion reactions. 4th ed. 73. Chaturvedi S,Arnold DM, McCrae KR.Splenec-
Bethesda, MD: AABBPress, 2012:263-87. tomy for immune thrombocytopenia:Down but
62. Taaning E, Tonnesen F. Pan-reactiveplatelet an- not out. Blood2018; 131:1172-82.
tibodies in post-transfusion purpura. Vox Sang 74. Lazarus AH, Crow AR. Mechanism of action of
1999;76:120-3. IVIG and antt-D in ITP. Transfus Apher Sci
63. BrecherME, Moore SB,Letendre L. Posttransfu- 2003;28:249-55.
sion purpura: The therapeutic value of PIAI- 75. McMillan R. Antiplatelet antibodies in chronic
negative platelets. Transfusion1990;30:433-5. immune thrombocytopenia and their role in
64. Bolten-MaggsPHB, ed, Poles D, et al on behalf platelet destruction and defective platelet pro-
of the Serious Hazards of Transfusion (SHOT) duction. Hematol Oncol Clin North Am 2009;
Steering Group. The 2015 Annual SHOT Re- 23:1163-75.
port. Manchester, UK: SHOT Office, 2016. 76. Davoren A, BusselJ, Curtis BR, et al. Prospec-
[Availableat https://www.shotuk.orglwp-con tive evaluation of a new platelet glycoprotein
tent/uploads/myimages/SHOT-20l5-Annual- (GP)-specificassay (PakAuto)in the diagnosisof
ReportWeb-Edition-Final-bookmarked-l.pdf(ac- autoimmune thrombocytopenia (AITP). Am J
cessed October 16, 2019).] HematoI2005;78:193-7.
65. Drug-induced immune thrombocytopenia: Re- 77. Wu GG, Kaplan C, Curtis BR,Pearson HA. Re-
sults of the testing for drug-dependent platelet- port on the 14th International Society of Blood
reactive antibodies by the BloodCenter of Wis- Transfusion Platelet Immunology Workshop.
consin, 1995-2015. Linked from: George IN. Vox Sang2010;99:375-81.
Platelets on the web: Drug-inducedthrombocy- 78. Veldhuisen B, Porcelijn L, van der Schoot CE,
topenia. Oklahoma City, OK: OUHSC, 2015. de HaasM. Molecular typing of human platelet
[Availableat http://www.ouhsc.edu/platelets/ and neutrophil antigens (HPAand HNA).Trans-
ditp.html (accessedJune 7, 2019).] fus Apher Sci 2014;50: 189-99.
79. ReilA, Bux J. Geno- and phenotyping of human and clinical consequences: A study of 21 indio
neutrophil antigens. Methods Mol BioI 2015; viduals from 14 families. Blood 1995;86:2403-
1310:193-203. 13.
80. Christopoulos CG, Kelsey HC, Machin SJ. A 93. Stroncek DF, Skubitz KM, Plachta LB, et al. AI-
flow-cytometric approach to quantitative estima- loimmune neonatal neutropenia due to an anti-
tion of platelet surface immunoglobulin G. Vox body to the neutrophil Fc-yreceptor IIIwith ma-
Sang 1993;64:106-15. ternal deficiency of CD16 antigen. Blood 1991;
81. Kiefel V, Santoso S, Weisheit M, Mueller- 77:1572-80.
Eckhardt C. Monoclonal antibody-specific im- 94. Muschter S, Bertold T. Greinacher A. Develop-
mobilization of platelet antigens (MArPA): A ments in the definition and clinical impact of
new tool for the identification of platelet- human rieutrophil antigens. Curr Opin Hematol
reactive antibodies. Blood 1987;70: 1722-6. 2011; 18:452-60.
82. Porcelijn L, Huiskes E, Comijs-van Osselen I, et 95. Moritz E, Norcia AMMI, Cardone JDB, et al.
al. A new bead-based human platelet antigen Human neutrophil alloantigens systems. An
antibodies detection assay versus the monoclo- Acad Bras Cienc 2009;81 :559-69.
nal antigen immobilization of platelet antigens 96. Sachs UJ, Andrei-Selmer CL, Manlar A, et al.
assay. Transfusion 2014;54: 1486-92. The neutrophil-specific antigen CD 177 is a
83. Cooper N, Bein G, Heidinger K, et al. A bead- counter-receptor for platelet endothelial cell ad-
based assay in the work-up of suspected platelet hesion molecule-I (CD31). J Bioi Chern 2007;
alloimmunization. Transfusion 2016;56: 115- 282:23603-12.
18. 97. Li Y, Mair DC, .Schuller RM, et al. Genetic
84. Davey S, Navarrete C, Brown C. Simultaneous mechanism of human neutrophil antigen 2 defi-
human platelet antigen genotyping and detec- ciency and expression variations. PLoS Genet
tion of novel single nucleotide polymorphisms 2015;11 :el005255.
by targeted next-generation sequencing. Trans- 98. Moritz E, Chiba AK, KimuraEY,et al. Molecular
fusion 2017;57:1497-504. studies reveal that A134T, G156A and G133A
85. McMillan R, Tani P, Millard F, et al. Platelet-
SNPs in the CD177 gene are associated with
associated and plasma anti-glycoprotein autoan-
atypical expression of human neutrophil
tibodies in chronic ITP. Blood 1987;70:1040-5.
antigen-2. Vox Sang 2010;98: 160-6.
86. Curtis BR, McFarland lG, Wu GG, et al. Anti-
99. Matsuo K, Lin A, Procter JL, et al. Variations in
bodies in sulfonamide-induced immune throm-
the expression of granulocyte antigen NB1.
bocytopenia recognize calcium-dependent epi-
topes on the glycoprotein lIb/IIIa complex.
100. Eulenberg-Gustavus C, Bahring S, Maass PG, et
Blood 1994;84: 176-83.
al. Gene silencing and a novel monoallelic ex-
87. McFarland J, Lochowicz A, Aster R, et al. Im-
proving the specificity of the PF4 ELISAin diag- pression pattern in distinct CD 177 neutrophil
nosing heparin-induced thrombocytopenia. Am subset. J Exp Med 2017 ;214:2089-10 I.
J HematoI2012;87:776-81. 101. Bal M, Grieshaber-Bouyer R, Wang J, et al.
88. Sheridan D, Carter C, Kelton JG. A diagnostic CD177 modulates human neutrophil migration
test for heparin-induced thrombocytopenia. through activation-mediated integrin and che-
Blood 1986;67:27-30. moreceptor regulation. Blood 2017;130:2092-
89. Favaloro EJ, McCaughan G, Pasalic L. Clinical 100.
and laboratory diagnosis of heparin induced 102. Lalezari P, Murphy GB,AIlenFH Jr. NB1, a new
thrombocytopenia: An update. Pathology 2017; neutrophil-specific antigen involved in the
49:346-55. pathogenesis of neonatal neutropenia. J Clin In-
90. Flesch BK, Curtis BR, de Haas M, et al. Update vest 1971;50:1108-15.
on the nomenclature of human neutrophil anti- 103. Bux J, Becker F, Seeger W, et al. Transfusion-
gens and alleles. Transfusion 2016;56: 1477-9. related acute lung injury due to HLA-A2-specific
91. ReilA, Sach UJ, Siabanidou T, et al. HNA-ld: A antibodies in recipient and NBl-specific antibod-
new human neutrophil antigen located on Fcy ies in donor blood. BrJ Haematol 1996;93:707-
receptor IIb associated with neonatal immune 13.
neutropenia. Transfusion 2013;53:2145-51. 104. Stroncek DF, Shapiro RS, Filipovich AH, et al.
92. de Haas M, Kleijer M, van Zwieten R, et al. Prolonged neutropenia resulting from antibodies
Neutrophil Fc gamma RIIIb deficiency, nature, to neutrophil-specific antigen NB1 following
marrow transplantation. Transfusion 1993;33: globulin G antibodies against human neutrophil

158-63. antigen-Sa.Transfusion2011 ;51:574-7.
105. Curtis BR, Cox NJ, SullivanMJ, et al. The neu- 114. Farruggia P. Immune neutropenias of Infancy
trophil alloantigen HNA-3a (5b) is located on and childhood. WorldJ Pediatr 2016;12: 142-8.
choline transporter-likeprotein 2 and appears to 115. Kleinman S, CaulfieldT, Chan P, et al. Toward
be encoded by an R>0154 amino acid substitu- an understanding of transfusion-related acute
tion. Blood2010;115:2073-6. lung injury: Statement of a consensus panel.
106. Greinacher A, Wesche J, Hammer E, et al. Char- Transfusion2004;44:1774-89.
acterization of the human neutrophil alloanti- 116. Food and Drug Administration. Fatalitiesreport-
gen-3a. NatMed 2010;16:45-8. ed to FDAfollowingblood collection and trans-
107. Reil A, Keller-StanislawskiB, Gunay S, Bux J. fusion: Annual summary for fiscal year 2017.
Specificities of leucocyte alloantibodies in SilverSpring,MD: CBEROfficeof Communica-
transfusion-related acute lung injury and results tion, Outreach, and Development, 2017. [Avail-
ofleucocyte antibody screening ofblood donors. able at https://www.fda.gov/media/124796/
Vox Sang2008;95:313-17. download (accessedOctober 16,2019).]
108. Lopes LB,Abbas SA,Moritz E, et al. Antibodies 117. Toy P, Gajic 0, Bacchetti P, et al. Transfusion-
to human neutrophil antigen HNA-3bimplicat- related acute lung injury: Incidence and risk fac-
ed in cases of neonatal alloimmune neutropenia. tors. Blood2012;119:1757-67.
Transfusion2018;58:1264-70. 118. AfzalW, Owlia MB, Hasni S, Newman KA. Au-
109. Sachs UJ, Chavakis T, Fung L, et al. Human toimmune neutropenia updates: Etiology, pa-
a11oantibodyanti-Mart Interfereswith Mac-I-de- thology, and treatment. South Med J 2017;110:
pendent leukocyte adhesion. Blood 2004; 104: 300-7.
727-34. 119. Audrain M, Martin J, Fromont P, et al. Autoim-
110. Fung YL, Pitcher LA, Willett JE, et al. Alloim- mune neutropenia in children: Analysis of 116
mune neonatal neutropenia linked to anti-HNA- cases. Pediatr AllergyImmuno12011;22:494-6.
4a. Transfus Med 2003;13:49-52. 120. Clay ME, Schuller RM, Bachowski GJ. Granulo-
111. Hartman KR,Wright DG. Identification of auto- cyte serology:Current concepts and clinical sig-
antibodies specific for the neutrophil adhesion nificance. Immunohematology 2010;26: 11-21.
glycoproteins CD11b/ CD18 in patients with 121. Schulz U, ReilA, KiefelV, et al, Evaluation of a
autoimmune neutropenia. Blood 1991 ;78: new microbeads assayfor granulocyte antibody
1096-104. detection. Transfusion2017;57:70-81.
112. Simsek S, van der Schoot CE, Daams M, et al. 122. Stroncek DF, Fadeyi E, Adams S. Leukocyte an-
Molecular characterization of antigenic poly- tigen and antibody detection assays:Toolsfor as-
morphisms (Ond(a) and Mart(a)) of the beta 2 sessing and preventing pulmonary transfusion
family recognized by human leukocyte alloanti- reactions. TransfusMed Rev 2007;21 :273-86.
sera. Blood 1996;88:1350-8. 123. Bux J. Molecular genetics of granulocyte poly-
113. Porcelijn L, Abblnk F, Terraneo L, et al. Neona- morphisms. Vox Sang 2000;78(Supp12):125-
tal alloimmune neutropenia due to immuno- 30.
The HLA System
Jeremy Ryan Peiia, MD, PhD; Arthur B. Eisenbrey III, MD, PhD; and
Patricia M. Kopko, MD
HE HLA SYSTEM IS COMPOSED OF molecular methods. (Historically significant

a complex array of genes located within methods of identifying HLA polymorphisms us-
the human major histocompatibility coming antibody binding to surface antigens and
plex (MHC) on the short arm of chromosome 6. mixed lymphocyte culture reactions have been
Thelr protein products, the HLA antigens, con- discussed in previous editions of the Technical
tribute to the recognition of self and nonself, the Manual, and the reader is encouraged to review
immune responses to antigenic stimuli, and the those methods for perspective.)
coordination of cellular and humoral immunity. Understanding the relationships .between the
HLAmolecules playa key role in antigen pre- polymorphisms of the HLA genes and antigen
sentation and the initiation of the immune reo presentation by the HLA molecules has permit-
sponse. The HLA system is generally viewed as ted the analysis of peptide-binding restriction
second in importance only to the ABO antigens parameters needed for effective vaccine devel-
in influencing the survival of transplanted solid opment. The MHC and HLA genetic polymer-
organs. In hematopoietic stem cell transplanta- phisms have been used by anthropologists and
tion (HSCT), the HLAsystem is paramount with population geneticists as accurate tools for popu-
regard to graft rejection and graft-vs-hostdisease lation studies. Because of the complexity of the
(GVHD). HLA antigens and antibodies are also MHC and the extent of polymorphism in the
important in complications of transfusion thera- HLAgenes, a complex nomenclature was devel-
py, such as platelet refractoriness, febrile nonhe- oped (and continues to evolve) to define unique
molytic transfusion reactions (FNHTRs),transfu- allele sequences based on the relationship of
sion-related acute lung injury (TRAtI), and each allele's protein sequence to the serologic
transfusion-associated GVHD (TA·GVHD). specificity of the corresponding antigen. 1,2
The biologic rolepofthe genes in the MHC
continue to be identified (neither transfusion
nor transplantation are natural events), and the
tremendous polymorphism of the HLA genes is
Rue
used outside of transplantation. Studies correlat- Characteristics of Class I and
ing HLA genes with disease susceptibility and
Class IIAntigens
disease resistance began soon .after serologic
techniques for HLA Class I typing were .devel- Class I antigens (HLA·A,·B, and .C) have a mo-
oped and have resurged with the adoptlon of lecular weight of approximately 57,000 Daltons
Jeremy Ryan Pefia, MD, PhD, Assistant Professor of Pathology, Harvard Medical School, and Director, HLALabo-
ratory, Beth Israel Deaconess Medical Center, Boston, Massachusetts; Arthur B. Eisenbrey III, MD, PhD, Associ-
ate Professor of Pathology, University of Toledo College of Medicine, Toledo, Ohio, and Associate Professor of
Pathology, Wayne State University School of Medicine, Detroit, Michigan; and Patricia M. Kopko, MD, Professor
of Pathology, University of California San Diego Medical Center, San Diego, California
The authors have disclosed no conflicts of Interest.
479
and consist of two protein chains: a glycoprotein and HLA-A28(A68 or A69), respectively. Plate-
heavy chain (45,000 Daltons) encoded on the lets express primarily HLA-Aand HLA-Banti-
short arm of chromosome 6 and, as a light gens. HLA-Cantigens are present at very low
chain, the ~2-microglobulinmolecule (12,000 levels, and Class II antigens are generally not
Daltons) encoded by a gene on chromosome 15. present on platelets.
The heavy chain penetrates the ce1lmembrane, Class II antigens (HLA-DR,-DQ, and -DP)
whereas ~2-microglobulindoes not. Rather, ~2- have a molecular weight of approximately
microglobulin associates (noncovalently) with 63,000 Daltons and consist of two structurally
the heavy chain through the latter's nonvariable similar glycoprotein chains (a and ~), both of
(a3) domain. (See Fig 16-1.) The external por- which traverse the membrane. (See Fig 16-1.)
tion of the heavy chain consists of three amino The extramembranous portion of each chain has
acid domains [o l , a2, and (3), of which the two amino acid domains, of which the outer-
outermost domains, a 1 and a2, contain the ma- most contains the variable regions of the ClassII
jority of polymorphic regions conferring serolog- alleles. The expression of Class II antigens is
ic HLAantigen specificity. more restricted than that of Class I antigens.
The "classical" HLAClass I molecules (HLA- Class II antigens are expressed constitutively on
A, -B, and -C) are present on platelets and most B lymphocytes, monocytes, macrophages, den-
nucleated ce1lsin the body, with some excep- dritic ce1ls, intestinal epithelium, and early
tions that include neurons, corneal epithelial hematopoietic ce1ls. There is also constitutive
ce1ls,trophoblasts, and germinal ce1ls.Only ves- expression of ClassII antigens on some endothe-
tigial amounts remain on mature red ce1ls,with lial ce1ls,especially those lining the microvascu-
certain a1lotypesbetter expressed than others. lature. However, in general, endothelium, par-
These Class I types were independently recog- ticularly that of larger blood vessels, is negative
nized as red ce1lantigens by serologists and des- for Class II antigen expression, although ClassII
ignated as "Bennett-Goodspeed" (Bg) antigens. antigen expression can be readily induced (for
The specificities called "Bga," "Bgb,"and "Bt" instance, by interferon-gamma during immune
are actually HLA-B7, HLA-B17 (B57 or B58), activation). Resting T lymphocytes are normally
FIGURE16·1. Stylized diagram of Class I and Class II major histocompatibility complex molecules
showing a and ~ polypeptidechains,their structural domains, and attachedcarbohydrateunits.
CHAPTER 16 TheHLASystem 481
negative for Class II antigen expression and be- ent sequences. The peptide-binding groove is
come positive when activated. critical for the functional aspects of HLA mole-
Soluble HLA Class I and Class II antigens cules. (See "BiologicFunction" section below.)
shed from cells are present in blood and body
fluids and may play a role in modulating im- Nomenclature for HLAAntigens
mune reactivlty.' Levels of soluble HLA in-
An international committee sponsored by the
crease with infection [including with human im-
World Health Organization (WHO) establishes
munodeficiency virus (HIV)], inflammatory
the nomenclature of the HLA system. This no-
disease, and transplant rejection, but HLA levels
menclature is updated regularly to incorporate
decline with progression of some malignancies.
new HLA alleles.' HLA antigens are designated
Levels of soluble HLA in blood components are
by a number following the letter that denotes
proportional to the number of residual donor
the HLA series (eg, HLA-A1or HLA-B8).Previ-
leukocytes and the duration of storage. Soluble
ously, antigenic specificities that were not fully
HLA in blood components may be involved in
confirmed carried the prefix "w" (eg, HLA-
the immunomodulatory effect of blood transfu-
Aw33) for "workshop." When the antigen's
sion."
identification became definitive, the WHO No-
menclature Committee dropped the "w" from
Configuration
the designation. (The committee meets regular-
A representative three-dimensional structure of ly to update nomenclature by recognizing new
Class I and Class II molecules can be obtalned specificities or genetic loci.) The "w" prefix is no
by x-ray crystallographic analysis of purified longer applied In this manner and is now used
HLA antigens. (See Fig 16-2.) The outer do- only for the following: 1) Bw4 and Bw6, to dis-
mains, which contain the regions of greatest tinguish such "public" antigens (see "'Public'
amino acid variability and the antigenic epitopes Antigens" section below) from other B-locus
of the molecules, form a structure known as the alleles; 2) all serologically defined C-locus speci-
"peptide-binding groove." Alleles that are de- ficities, to avoid confusion with components of
fined by polymorphisms in the HLA gene se- the complement system; apd 3) Dw specificities
quences encode unique amino acid sequences that were defined by mixed leukocyte reactions
and therefore form unique binding grooves, but are now known to be caused by RLA-DR,
each of which is able to bind peptides of differ- RLA-DO, and RLA-DPpolymorphisms. The nu-
Class I Class II
FIGURE16·2. Ribbon diagram of HLA Class I and Class II molecules. Note the peptide in the groove of
each molecule.
meric designations for the HLA-Aand HLA-B identified, many of which share common sero-
specificitiesare assigned according to the order logic phenotypes. The minimum requirement
of their officialrecognition. for designation of a new alleleis the sequence of
exons 2 and 3 for HLA Class I and exon 2 for
Splits and Cross-Reactive Groups HLA Class II. These exons encode the variable
amino acids that confer HLAantigen specificity
Refinement of serologic methods permitted anti- and much of the biologic function of the HLA
gens that were previously believed to represent molecule.
a single specificityto be "split" into specificities A uniform nomenclature has been adopted
that were recognized as serologically(and, later, that takes into account the locus, major serolog-
genetically) distinct. The designation for an indi- ic specificity, and allele group determined by
vidual antigen that was split from an earlier rec- molecular typing techniques. For example, al-
ognized antigen often includes the number of though many alleles have been sequenced only
the parent antigen in parentheses leg, HLA-B44 for exons 2 and 3, nucleotide sequencing has
(12)]. identified at least 300 unique amino acid se-
In addition to "splits," certain apparently dis- quence variants (alleles) of HLA-DR4as of late
tinct HLA antigens may have other epitopes in 2019 (see http://hla.alleles.org/alleles/class2.
common. Antibodies that are reactive with htmll.' The first HLA-DR4variant is designated
these shared determinants often cause cross- DRB1*04:01, indicating the locus (DR),protein
reactions in serologic testing. The collective (~l chain), major serologic specificity (04 for
term for a group of HLA antigens that exhibit HLA-DR4),and sequence 2 variation allelenum-
such cross-reactivity is "cross-reactive epitope ber (variant 01). The asterisk indicates that an
group" (CREG). allele name follows (and that the typing was de-
termined by molecular techniques). There may
"Public" Antigens be up to four sets of digits after the asterisk;
In addition to splits and CREGs,HLAantibodies each set of digits is separated by a colon and is
may recognize epitopes present across many dif- called a field. The digits in the first field corre-
ferent HLA specificities. Called "public" anti- spond to the antigen's serologic specificity in
gens, these common amino acid sequences ap- most cases. The digits in the second field make
pear to represent less variable portions of the up the code for a unique amino acid sequence in
HLA molecule. Two well-characterized public exon 2 (ClassII) or exons 2 and 3 (ClassI), with
antigens, HLA-Bw4and HLA-Bw6,are present numbers being assigned in the order in which
in almost all HLA-Bmolecules." The HLA-A lo- the DNA sequences were determined. There-
cus molecules A23, A24, A25, and A32 also fore, B*27:04 represents the HLkB locus, has a
have a Bw4-likeepitope. serologic specificityof B27, and was the fourth
Public antigens are clinically important be- unique exon 2 and 3 sequence allele described
cause patients exposed to them through preg- in this family. (See Table 16-1.) A third field in
nancy, transfusion, or transplantation can make the allele name is added for alleles that differ
antibodies to these antigens. A single antibody, only by synonymous ("silent") nucleotide substi-
when directed against a public antigen, can re- tutions in exons 2 and 3 for Class I or in exon 2
semble multiple discrete alloantibodies, and this for Class II. For example, A *01:01:02 differs
has significant consequences for identifying from A *01:01:01 only in that the codon for iso-
compatible donors for transplantation and plate- leucine in position 142 is ATTinstead ofATC.A
let transfusion. fourth field in the allele name can be added for
alleles that differ only in sequences within in-
trons or in 3' or 5' untranslated regions. Finally,
Nomenclature for HLA Alleles
the nomenclature accommodates alleles with
Nucleotide sequencing has replaced serologic null or low expression or other characteristics
methods for investigating the HLAsystem, and by the addition of an UN"or "L," respectively, or
increasing numbers of HLA alleles are being another letter as appropriate to the end of the al-
C HAP TE R 1 6 TheHLASystem 483
TABLE16·1. Current HLA Nomenclature
HLA DRB1 * 04
Examples:
DR4 . Serology
DRB1*04:xx . Serologic Equivalent
>
DRB1*04:02 - Allele
DRB1*04:01 :01;··DRB1*04:01:02 •- Silent Mutations
A*02:15N; DRB4*01:03:01:02N - Null Alleles (exon, intron)
A*24:02:01 :02L
B*44:02:01:02S - ExpressionModifiers
B*32:11 Q
lele name. The other officialexpression modifi- Role ofC/ass I Molecules

ers are as follows: S (secreted, not on cell sur- Class I molecules are synthesized, and peptide
face), Q (expression level questionable), A antigens are inserted into the peptide-binding
(unlmown but aberrant expression, perhaps groove, in the endoplasmic reticulum. Peptide
null), and C (cytoplasmic expression only). The antigens that fit into the Class I peptide-binding
last two have not been used to date. groove are typicallyeight or nine amino acids in
length and are derived from proteins made by
Biologic Function the cell (endogenous proteins], Such endoge-
The essentialfunction of the HLAsystem is self/ nous proteins-which may be normal self-
nonself discrimination, which is accomplished proteins; altered self-proteins, such as those in
by the interaction of T lymphocytes with pep- cancer cells; or viral proteins, such as those in
tide antigens presented by HLAproteins. T lym- virus-infectedcells-are degraded in the cytosol
phocytes interact with peptide antigens only by a large multifunctional protease (LMP) and
when the T-cell receptor (TeR) for antigen en- are transported to the endoplasmicreticulum by
a transporter associated with antigen processing
gages both an HLAmolecule and the antigenic
(TAP).The LMP and TAP genes are both local-
peptide contained within the TCR's peptide-
ized to the MHC.
binding groove. This limitation is referred to as Class I molecules are transported to the cell
"MHC restriction."? surface, where the molecules are availableto in-
In the thymus, T lymphocytes with TCRs
teract with CDS-positiveT lymphocytes. If the
that bind to self HLA molecules are selected TCR of a CDS cell can bind the antigenic pep-
(positiveselection), with the exception of those tide in the context of the specific Class I mole-
with TCRs that also bind to a peptide derived cule displaying it, then TCR binding activates
from a self-antigen,in which case the T lympho- the cytotoxic properties of the T cell, which at-
cytes are deleted (negative selection).Some self- tacks the cell, characteristically eliciting an
reactive T cells escapenegative selection and, if inflammatory response. The presentation of
not functionally inactivated (forinstance, by the antigen by ClassI molecules is especiallyimport-
mechanism of anergy), may become involved in ant in host defense against viral pathogens and
an autoimmune process. malignant transformation. Tumor cells that do
not express Class I antigens escape this form of sponding Class II antigens. Located between the
immune surveillance. Class I and Class II genes is a group of non-Hl.A
genes that code for molecules that include the
Role ofC/ass /I Molecules complement proteins C2, Bf, C4A, and C4B; a
steroid enzyme (21-hydroxylase); and a cyto-
Like Class I molecules, Class II molecules are
kine (tumor necrosis factor) and other genes in-
synthesized in the endoplasmic reticulum, but
volved in immune responses. This non-Hl.A re-
peptide antigens are not inserted into the peptide-
gion is referred to as "MHC Class III" even
binding groove there. Instead, an invariant
though it does not contain any HLAgenes.
chain (Ii) is inserted as a placeholder. The Class
Il-lnvariant chain complex is transported to an
Organization of HLA Genetic Regions
endosome, where the invariant chain is re-
moved by a specialized Class II molecule called The HLA Class I region contains (in addition to
"DM." The DM locus is also localized to the the classical genes HLA-A, HLA-B, and HLkC)
MHC. A Class II antigenic peptide is then insert- other gene loci designated HLA-E,HLA-F,HLA-G,
ed into the peptide-binding groove. HLA-H, HFE, HLAJ, HLkK, HLA-L,MICA, and
Peptide antigens that fit into the Class II MICB. The latter genes encode nonclassical, or
peptide-binding groove are typically 12 to 25 Class Ib, HLAproteins, which have limited poly-
amino acids in length and are derived from pro- morphism, low levels of expression, and limited
teins that are taken up by the cell through endo- distribution of tissue expression." Some Class Ib
cytosis (of exogenous proteins). Exogenous pro- genes express nonfunctional proteins or no pro-
teins, which may be normal self-proteins or teins whatsoever (termed "pseudogenes" and
proteins derived from pathogens, such as bacte- are presumed evolutionary dead ends). Other,
ria, are degraded to peptides by enzymes in the expressed, nonclassical HLAproteins have been
endosomal pathway. Class II molecules are then associated with a variety of functions. For exam-
transported to the cell surface, where the mole- ple, HLA-Eis associated with the surveillance
cules are available to interact with CD4-positive system of one subset of natural killer cells. HLA-G
T lymphocytes, which secrete lmmunostimula- is expressed by the trophoblast and may be in-
tory cytokines in response. That mechanism is volved in the development of maternal immune
especially important for the production of anti- tolerance of the fetus.
bodies. The genomic organization of the MHC Class
II (HLA-D) region is more complex. An MHC
Class II molecule consists of a noncovalent
GEN l'H He complex of two structurally similar chains: the
a-chain and the ~-chain. Both of these chains
Class I and II HLAmolecules are cell-surfacegly- are encoded within the MHC. The polymor-
coproteins and are products of closely linked phism of HLA Class II molecules results from
genes mapped to the p21.3 band on the short differences in both the a-chain and the ~-chain;
arm of chromosome 6 (Fig 16-3). That genomic this polymorphism depends on the Class II iso-
region, the MHC, is usually inherited en bloc as form. For example, with HLA-DR,the a-chain is
a haplotype. Each of the many loci has multiple essentially monomorphic, but the ~-chain is
alleles with codominant expression of the prod- very polymorphic. Multiple loci code for the a-
ucts from each chromosome. The HLAsystem is and ~-chains of the Class II MHC proteins.
the most polymorphic genetic system described Different haplotypes have different numbers
in humans. 7 of Class II genes and pseudogenes. The proteins
The genes HLA-A, HLA-B, and HLkC encoded by DRAl and DRBl result in HLA-DR1
code the corresponding Class I A, B, and C through HLA-DR18 antigens. The products of
antigens. The genes HLA-DRAl, -DRBl, DRAl and DRB3 (if present) express HLA-
-DRB3, -DRB4, -DRB5; HLA-DOAl, -DOEl; DR52; those of DRAI and DRB4 (ifpresent) ex-
and HLA-DPAI and -DPBl encode the corre- press HLA-DR53;and those of DRAI and DRB5
CHAPTER 16 The HLA System 485
E ~
0
:I:j :.s:
Q.)
0 c::
..0 ~
--:>
Q.)
..c::
+-' E
0 0
...... I-
'+-
Q.)
I- c::
Q.) 0
E 'C/i
en
~
~-
0
a:; 'E
+-' .._
Q.) Q.)
..c:: e..
+-'
Q.)'-- ..c::
......
~ :.s:
8 0"0
"'" .-
-en
Q.)::::J
Q.)
~ _,-...'_
"f-'___':O
oEo
0 ¢:::_N
(;t Q.) ..c:::: ai
"'J -xu
e..Q.)c::
.s ~.~
Q.)
"(i) CI)
..c::
..... Q.) "0
R o a:; c::
.......='ct$
~ en~1::
.- en ~
Q.)I-C!l
t:!) ._ 0
Q.) • ..s.:
I-
E ~ 0
0 :-§ >-
0
.....
I-
<ti'Q.):.s:
8 c::
Q.)
u
Q.)
:mz
..c:: ~ "0
Q.)
l- e..
:I:j ..c::
<:I <C ..c:: +-'
L!')
ij. Q.)
0
rr'.} en cD
E <ti
~=
Cl
0
en
0
C/)
Q.)
Q.)
~
Q.)
(/)
E CI) .-
_"0
0
"0
§ l-
..c:: t:::
~ c::
'" -u
E
:.s:
0..c::
0 ..c:: .....
C/) <ti
~ .~
Q.)
I- ..c::
(/)
g t= c::
0 0
c::
(1j ..c:: 's, E

(/)
Q.)
Q.)
Q.)
l-
.....
(/)
»
0
-- ~
..c::
c::
0
-
"0
c::
(/)
Q.)
c::
:=s
g "0
t'> Q.) " E
1<1
u - .§
Q.)
= ..2 (/)
(/) ..c::
U·_ .~ ..£S! I-
o X c..:> <ti
~ l!:'!Q.) .....
-
t'>
= e.. ..c:: Q.)
+-'
"..
E 2"
0 0
u
o c:: t=
1:l
u §f ::5
:c ~
0 0
e..
<ti
i !!!"
..c:: 'i::
Q.)
~
N
5:
l- E' 0.."
cv:i 0 (/)
I-
....wr.b ..c::
Q.) Q.)
>
a::: I---~
....1-
::::l..c:: -
C!:I 0>«
u::: .;:: c..:>
(if present) express HLA-DR51.The HLA-D01 Mendelian genetics. Every person has two dif-
through D09 antigens are expressed on the gly- ferent copies of chromosome 6 and possesses
coproteins coded by DOA! and DOE! in the two HLAhaplotypes, one from each parent. The
DO cluster. Many of the other genes of the DO combination of HLAalleles inherited from both
cluster are likely pseudogenes. A similar organi- parents constitutes the genotype. The expressed
zation is found in the HLA-DPgene cluster. HLAgenes then result in the phenotype, both of
Although not generally considered part of the which can be determined by typing for HLAan-
HLAsystem, the MHC Class III region contains
tigens or alleles. Because HLAgenes are autoso-
four complement genes with alleles that are typ-
mal and codominant, the phenotype represents
ically inherited together as a unit, termed a
"complotype." More than 10 different complo- the combined expression of both haplotypes.
types are inherited in humans. Two of the Class However, to define haplotypes, parents (and
III genes, C4A and C4B, encode for variants of possibly other family members) also need to be
the C4 molecule and antigens of the Chido/ typed to determine which alleles are inherited
Rodgers blood group system. These variants together. Figure 16-4 illustrates inheritance of
have distinct protein structures and functions; haplotypes and demonstrates that four possible
the C4A molecule (ifpresent) carries the Rg an- combinations of haplotypes are possible in the
tigen, and the C4B molecule (ifpresent) carries offspring, assuming no recombination occurs.
the Ch antigen. Both of these antigens are ad- The chance that two siblingswill be phenotypi-
sorbed onto the red cells of individualswho pos- cally HLAidentical is 25%. The chance that any
sess the genets). one patient with "n" siblings will have at least
one HLA-identicalsibling is 1 - (3/4)n. Having
Patterns of Inheritance two siblingsprovides a 44% chance of finding an
Although MHC organization is complicated, its HLA-identicalSibling,and having three siblings
inheritance follows the established principles of provides a 58% chance.
Chromosome 6
en bloc Inheritance of the
HLAcomplex
a c
d
a a ..JL
c d c d
FIGURE 16·4. The deslqnattons alb and c/d represent paternal and maternal HLA haplotypes,
respectively. Except for crossovers, the HLA complex is transmitted en bloc from parent to offspring.
Absence of Antigens tionship evaluations, the possibility of recombi-

nation should always be considered.
Before the advent of molecular-based HLA typ-
ing; the absence of an antigen in serologic phe-
Linkage Disequllibrtum
notyping results was attributed to homozygosity
at a locus (eg, inheritance of AI·from both par- The MHC system is so polymorphic that the
ents, which in reality represented only an appar- number of possible unique HLA phenotypes is
ent absence of the antigen as a result of limita- theoretically greater than that of the global hu-
tions of phenotyping methods) or to a null man population. Moreover, new HLAalleles are
(nonexpressed) allele. With DNA sequencing constantly being discovered and characterized.
and other molecular HLA typing methods, ho- As of August 2018, 4340 HLA-Aalleles, 5212
mozygosity can now be presumed with a higher HLA-Balleles, 2593 DRB1 alleles, and 1257
degree of confidence. However, proving homo- DOB1 alleles had been identified.2,lo Usually,
zygosityrequires family studies or methods per- HLAgenes are inherited together as haplotypes.
mitting hemizygous typing (ie, typing of an indi- Many HLA alleles are overrepresented com-
vidual haplotype). A null allele is characterized pared with what would be expected if the distri-
by one or more DNA sequence changes, within bution of HLA genes were random. The phe-
or outside the gene's coding region, that prevent nomenon of linkage disequilibrium describes
expression of a functional protein at the cell sur- the discrepancy between expected and observed
face. Such inactivation of a gene may be caused HLAallele frequencies. .
by nucleotide substitutions, deletions, or inser- Expected frequencies for HLA alleles are de-
tions that lead to a premature cessation in the rived by multiplication of the frequencies of
protein's synthesis. In the absence of a family each allele. For example, in individuals of Euro-
study, a phenotyping study revealing a single al- pean ancestry, the overall frequency of HLA-A1
lele at any locus offers only presumptive evi- is 0.15 and that for HLA-B8is 0.10; therefore,
dence for homozygosity. In this situation, the al- 3.0% (0.15 x 0.10 x 2) of all HLAhaplotypes in
lele should be listed only once because it is people of European ancestry would be expected
unknown whether that alleleis present twice (a to contain both HLA-Al and HLA-B8if the al-
true homozygote) or there is another allele not leles were randomly distributed. The actual hap-
detected by the availablemethod. lotype frequency of the A1 and B8 combination,
however, is 7%to 8%in that population.
Crossovers Certain allelic combinations occur with in-
The genes of the HLA region occasionally creased frequency in different racial groups and
demonstrate chromosome crossover, in which constitute common haplotypes in those popula-
segments containing linked genetic material are tions. These common haplotypes are called "an-
exchanged between the ..two chromosomes cestral haplotypes" because they appear to be in-
during meiosis or gametogenesis. The .recombi- herited from a single common ancestor or to be
nant chromosomes are then transmitted as new conserved within the population, because of ei-
haplotypesto the offspring.Crossover frequency ther survival advantage of carriers or resistance
is related partly to the physical distance between to recombination. The most common ancestral
the genes and partly to the resistance or suscep- haplotype in people of Northern European an-
tibility of specific A, B, and DR antigens to re- cestry-AI, B8, DRl7 (DRBI*03:0I), D02-
combination. (See below.) The HLA-A, HLA-B, includes both Class I and ClassII regions.
and HLA-DR loci are close together, with 0.8% Some alleles in apparent linkage disequilibri-
crossover between the A and B loci and 0.5% um may represent relatively young haplotypes
between the B and DR loci. Crossoversbetween that have not had sufficient time to undergo re-
the HLkB and HLA-Cloci or between the HLA- combination, whereas some old haplotypes are
DR and HLA -DO loci are extremely rare, where- resistant to recombination because of selection
as crossovers between the DO and DP loci are or physlcal.limltattons. For example, th,eAI, B8,
relatively common," In family studies and rela- DRB1*03:01 haplotype appears to be resistant
to recombination because of deletion of the HLAtyping has several advantages over past se-
complement C4A gene, which results in de- rologic assays: 1) high sensitivity and specifictty,
creased distance between HLA-B and HLA- 2) use of small sample volumes, and 3) no need
DRB1 in those individuals. Linkage disequilibri- for cell-surface-antigenexpression or cell viabili-
um affects the likelihood of finding suitable ty. Although serologic methods can identify a
unrelated donors for HLA-matched platelet limited number of HLAspecificities, high-
transfusions and HSCT. resolution DNA-basedmethods have the poten-
tial capability to identify all known alleles as
well as new alleles.
l TVPIN Polymerase chain reaction (PCR)technology
allows amplification of large quantities of a par-
Clinical identification of HLA antigens and al- ticular target segment of genomic DNA. Low-to
leles is now performed almost exclusively by intermediate-resolution typing detects the HLA
molecular methods. Historically, serologic serologic equivalents with great accuracy (eg, it
(antibody-based)and cell-based assayswere also distinguishes DR15 from DR16), whereas high-
used. The reader is referred to previous editions resolution typing distinguishes individual alleles
of the AABB Technical Manual and Bontadini11 (eg, DRBI*OI:OI:OI from DRBI*Ol:02:0l).
for detailed descriptions of the use of the lym- Several molecular methods are PCR-based;the
phocytotoxicity method. most common molecular methods for HLAtyp-
Detailed procedures for commonly used as- ing are described below.
says are available from reagent and kit manufac-
turers and have been summarized in reviews of Oligonucleotide Probes
availablemethodology. 11 Depending on the clin- Sequence-specific oligonucleotide probes (SSO
ical situation, a particular HLA antigen/allele or SSOP) use arrays of labeled oligonucleotide
detection or typing method may be preferable to probes to detect HLA nucleotide sequences
determine the HLAgenotype or phenotype. (See present in immobilized DNA.ll Reverse SSO
Table 16-2.) Current molecular (DNA-based) (rSSO)has been widely adopted and uses probes
TABLE 16-2. HLA Typing Methods and Appropriate Applications

Method Clinical Application Resolution
SSP (PCR) Solid-organ and HSCT Serologic to allele level, higher
resolution with large number of
primers
Forward SSOPhybridization Solid-organ and HSCT(can Serologic to allele level
accommodate high-volume testing)
ReverseSSOPhybridization Solid-organ and HSCT Serologic, higher resolution with
larger number of probes
DNAsequencing Unrelated HSCT,resolution of typing Allele level
problems with other methods,
characterization of new alleles
Lymphocytotoxicity Supplementaltesting for DNA-based Serologic specificity
HLAtypings without defined HLA
antigen assignments, and research
support for HLAallele and antigen
designation
SSP = sequence-specific primer; PCR = polymerase chain reaction; HSCT = hematopoietic stem cell transplantation;
SSOP= sequence-specificoligonucleotide probe.
C HAP T E R 1 6 TheHLASystem 489
individually attached to a solid-phase matrix (for tions are assigned to different alleles. The
example, each probe may be attached to a differ- haplotype to assign individual base pairs in
ent microbead). DNA from a target locus is then polymorphic combinations may be determined
amplified and labeled in PCR reactions, and the with additional results from selected SSPor SSO
binding to the different probes is evaluated. reactions.
Commercially available microbead array assays
use rSSO methods for HLA Class I and Class II Next-Generation Sequencing
low-to-high-resolutiontissue typing and comput- Massively parallel sequencing ("next-generation
er proprietary algorithms to match binding pat- sequencing," or NGS) has allowed for sequenc-
terns to allele databases. 12 ing of whole genes and reduced the frequeI1cy
of the ambiguities that occur with Sanger-
Sequence-Specific Primers chemistry SBTof HLAbecause NGS sequences
A second major technique uses sequence-specific single strands of DNA. HLA typing kits are
primer (SSP)pairs that target and amplify a par- commercially available for both clinical and re-
ticular DNA sequence." This sequence-specific search instruments." NGS methods obtain se-
method requires the performance of multiple quences, from libraries, formed from fractured
PCR assays in which each reaction is selected pre- or post-PCR cellular DNA. The very large
for a particular allele or group of alleles. The am- number of sequences obtained allow for identi-
plification products are directly visualized after fication of overlapping sequences and arrange-
agarose gel electrophoresis. Because SSPs have ment of the resulting sequences through
specific targets, the amplified material indicates computer analysis (requiring very powerful
the presence of the allele or alleles that have processors, large sequence databases, and com-
that sequence. The pattern of positive and nega- plex programming).
tive PCR amplifications is examined to deter- Two families of NGS are sequencing by syn-
mine the HLAallele(s) present. Primer pair sets thesis ~d sequencing by hybridization ~d liga-
are commercially available that can determine tion. Sequencing by synthesis has three method-
HLI\-A,-B, -C, -DR, -DQAl, -DOBI, and -DPBI ologies: pyrosequencing; ion semiconductor
phenotypes.and may be combined to determine sequencing; and fluorescently labeled, reversible
common alleles. nucleotide terminator chemistry. Nucleotide se-
quence detection isby photon release from dide-
Sequence-Based Typing
oxynucleotide incorporation, detection of hy-
drogen ion release, or laser interrogation of dye
("Sequencing")
terminator incorporation. A much less common
High-resolution typing is necessary for assign- method is matrix-assisted laser desorption/
ment of HLA alleles." Sanger-chemistry se- ionization time-of-flight mass spectrometry
quence-based typing (SBT)can be used to identi- (MALDHOF MS). NGS is rapidly evolving, and
fy known alleles and characterize new alleles.II selection of methods and instrumentation
Although SBTis considered the"gold standard" should be determined by individual laboratory
for HLAtyping,ambiguities occurwhen two dif- needs.
ferent base pairs are found at the same position
and can result in two different possible combina-
tions of alleles. These ambiguities occur because HER N
SBT evaluates both maternal and paternal HLA HIS CO y
genes (haplotypes) simultaneously. Pairs of sin- ER IN
gle nucleotide substitutions can be encountered
in which cis (same parental haplotype) or trans Although HLA and ABO remain the most im-
(nucleotide assignment of the second polymor- portant histocompatibility systems in transplan-
phic site is on the other haplotype) can give am- tation, the following is a brief review of some
biguous results when those nucleotide combina- non-Hl.A factors with emerging relevance to
transplantation. Clinical testing for these mark- to as "lymphocyte crossmatching." Crossmatch-

ers is most commonly performed in the HLA ing consists of incubating serum from a potential
laboratory. recipient with lymphocytes (unfractionated or
separated into T and B lymphocytes) from pro-
Non-HLA in Solid-Organ Transplantation spective donors. Variations of the lymphocyto-
Even in identical twins or HLA-matcheddonor- toxicity test include extended incubations, in-
recipient pairs, allograft rejection has been ob- clusion of wash steps, and use of an antiglobulin
served, suggesting the role of non-Hl.Adetermi- reagent. Flow cytometry has largelyreplaced the
nants." In solid-organ transplantation these cytotoxicity crossmatch method with greater
other non-HLA antigens can elicit an antibody sensitivity than the antiglobulin-enhanced cross-
response from the recipient. Some of the better- match.
studied targets include MICA (major histocom-
Solid-Phase Assays
patibility class I chain related gene A), AIl R
(angiotensin II type I receptor), ETAR(endothe- The current approach to identify HLAantibodies
lin type A receptor), vimentin, and perlecan relies on the use of beads or microparticles (ie,
(LG3).16With the exception of MICA, these an- solid-phase methodology) coated either with
tigens are often targets of autoantibody respons- clusters of HLAClass I or Class II antigens from
es. It is estimated that <3% of antibody-mediated cultured lymphocytes (ie, an HLAphenotype) or
rejection is due to non-HLA antibodies.I? The with individually purified or recombinant HLA
two best-studied antibodies for which there are antigens (single-antigen beads)." Antibody-
commercial clinical assays currently in use are binding is detected by staining with fluorescent-
MICA and AIlR. ly labeled antihuman globulin (AHG). The
presence of antibody is detected with flow cy-
Non-HLA Hematopoietic Stem Cell tometry, flow microarrays, or enzyme-linked im-
Transplantation munosorbent assay (ELISA). Only an ELISA
method is FDA-approved for donor screening
In HSCT, killer immunoglobulin-like receptor
(August 2018). Flow cytometry and flow mi-
(KIR)typing of donors is becoming increasingly
croarray methods are more sensitive than lym-
more common. Natural killer (NK)cells use KIR phocytotoxicity and focus on the detection of
to recognize self vs nonself through interactions IgG antibodies. The use of single-antigen bead
mediated by KIRligands on target cells." Byse- assays are of particular importance for highly
lecting donors with more activating KIRalleles sensitized patients where multiple HLA anti-
(called group B KIR)and/or by givingthe recipi- body specificities cannot be reliably distin-
ents who lack the ligands that signal, there may guished and identified with either cell-based
be increased graft-vs-leukemia effect and im- cytotoxic assays or solid-phaseassaysusing clus-
proved HSCToutcomes. ters of HLAmolecules.
Although HLA antibody has been demon-
strated in transplantation population studies to
ING AND
be detrimental to transplanted organ and patient
HlA survival, the clinical significance of low-level
ANTI antibodies detectable only by solid-phase assays
cannot be predicted for individual patients.
Cell-Based Assays
Newer adaptations of the solid-phasetechnology
Compatibility testing similar to the red cell can determine whether the antibodies do or do
crossmatch has been performed by lymphocyto- not fix complement and may improve the pre-
toxicity for over 50 years" and is best referred dictive value of testing for individual patients.
N Identifying Compatible Donors

Selected donor platelets may be either HLA-
matched (identical or zero-mismatch), antigen-
HLA system antigens and antibodies play im- negative (antibody avoidance), or crossmatch-
portant roles in a number of transfusion-related compatible units. A perfect antigen-matched
events, including platelet refractoriness,FNHTRs, unit (HLA identical) might yield no significant
TRALI,and TA-GVHD.HLAantigens are highly posttransfusion count increment because the
immunogenic. In response to pregnancy, trans- majority of allogeneic units have been typed
fusion, or transplantation, immunologically only at low resoluticn, and some patients dis-
competent individuals are more likely to form play allele-specific antibodies. The authors
strongly believe that in the age of molecular
antibodies to HLAantigens than to any other an-
HLA typing and single-antigen testing for HLA
tigens. antibodies, the antigen-match grade system (A,
BU,B2U, BX,C, and D matches) is obsolete and
Platelet Refractoriness should no longer be used. In this older, HLA-
The incidence of HLA alloimmunization and match grade system, a grade A match represents
platelet refractoriness has been substantially re- a donor who has identical HLA-Aand -B anti-
duced by the implementation of nearly universal gens with the patient. The assumption is that
the patient should not have any antibodies to an
transfusion of leukocyte-reduced cellular blood
A match. With the advent of molecular typing, a
components in Canada and the United States." BU or B2U match should also be equally com-
The refractory state exists when a transfusion of patible (as an A match) with the recipient be-
suitably preserved platelets fails to increase the cause, unlike serologic methods, molecular typ-
recipient's platelet count. Platelet refractoriness ing methods should not "miss" an antigen.
may be caused by clinical factors, such as sepsis, However, a BXmatch could be incompatible if a
high fever, disseminated intravascular coagulop- patient makes antibodies to the cross-reactive
athy, bleeding, medications, hypersplenism, antigen present on the donor platelets. Con-
complement-mediated destruction, or a combi- versely, a grade C or D match may be compati-
nation of these factors; alternatively,it may have ble if the patient does not have antibody against
any of the donor antigens."
an immune basis." Single-antigen testing for Class I HLA anti-
bodies allows for the selection of antigen-
Antibody Development negative units. This practice has been shown to
Antibodies against HLA Class I antigens are a be as effective as using HLA-matched units for
common cause of immune-mediated platelet re- refractory patients, and in many cases, it can be
fractoriness, but antibodies to platelet-specificor easier to find antigen-negative units than a per-
ABH antigens may also be involved. Although fect four-locus match. However, when patients
are highly alloimmunized with multiple specific-
platelets express HLAClass I antigens, the most ities, lt may be necessary to honor only the
likely cause of HLA sensitization from transfu- strongest [highest mean-fluorescent-intensity
sion is the leukocytes in the blood component. (MFI)jantibodies. Finally,the provision of cross-
This is demonstrated by the development of an- match-compatible units is a useful alternative
tibodies to HLAClass II antigens, which are not when the HLAtype and antibody profile of a pa-
expressed on platelets. Leukocyte reduction to tient is not yet known.
<5 x 106 per component was shown to reduce
alloimmunization from 19% to 7% and alloim- Febrile Nonhemolytic Transfusion
mune platelet refractoriness from 14% to 4% in Readions
patients undergoing chemotherapy for acute HLA, granulocyte, and platelet-specificantibod-
leukemia or stem cell transplantation." ies have been implicated in the pathogenesis of
FNHTRs. The recipient's antibodies, reacting same haplotype from each parent, and child 1 is
with transfused antigens, elicit the release of cy- homozygous for the shared parental HLAhaplo-
tokines (eg, interleukin-I] that are capable of type. Transfusion of blood from child 1 to an
causing fever. Serologic investigation, if under- unrelated recipient with different haplotypes
taken, may require multiple techniques and tar- would have no untoward consequences. If,
get cells from a number of different donors. Cy- however, child 1 were a directed donor for a rel-
tokines in stored cellular blood components, ative who was heterozygous for that haplotype
particularly non-leukocyte-reduced components, (eg, one of the parents or child 3), the recipient's
are also a cause of FNHTRs.23(See Chapter 22.) body would fail to recognize the antigens on the
transfused lymphocytes as foreign and would
TRALI not eliminate them. The donor cells would rec-
ognize the recipient's other haplotype as foreign
In TRALI,a potentially fatal transfusion reaction
and would become activated, proliferate, and at-
that may occur with transfusion of plasma-
tack the host.
containing blood components, acute noncardlo-
To avoid this situation, it is recommended
genic pulmonary edema develops in response to
that all cellular components from blood relatives
transfusion. The pathogenesis of TRALIappears
be irradiated before transfusion. Irradiation
to reflect the presence of HLAor neutrophil an-
causes damage to nucleic acids of residual Iym-
tibodies in donor blood that react with the tar-
phocytes in blood components, which prevents
get antigens in the recipient. Studies have
them from dividing, thereby preventing them
shown that 2% (male donors) to 17% (female
from attacking the host. (See Chapter 17.) Oth-
donors) of blood components can contain de-
er specially chosen donor units, including HLA-
tectable amounts of HLA antibodies." If pres-
matched platelets, may also present an in-
ent, such antibodies can be reactive with and fix
creased risk of TA-GVHDand should be irradiat-
complement to the recipient's granulocytes,
ed. Rarely, TA-GVHD has occurred after the
leading to severe capillary leakage and pulmo-
transfusion of blood from an unrelated donor,
nary edema. Rarely,the recipient's HLAantibod-
usually within populations in which shared HLA
ies are reactive with transfused leukocytes from
haplotypes are common.
the donor (reverse TRALI).(See Chapter 22 for
Chimerism is proposed to be responsible for
more on TRALI.)
the maintenance of tolerance in some organ
transplant recipients as well as for the mainte-
Chimerism and TA-GVHD
nance of HLAsensitization." It has been postu-
"Chimerism" refers to the presence of two cell lated that scleroderma is a form of GVHDresult-
populations, such as transfused or transplanted ing from chimeric cells derived from fetal cells
donor cells and recipient cells, in an individual. transferred across the placenta during pregnan-
Persistent chimerism after blood transfusion cy.26Furthermore, the persistence of donor lym-
may lead to the development of TA-GVHDin phocytes originally present in and transplanted
the recipient. The development of TA-GVHDde- with a solid-organ allograft has been document-
pends on the following factors: 1) the degree to ed to cause fatal GVHDin recipients of these or-
which the recipient is immunocompromised, 2) gans." Although donor lymphocytes are poten-
the number and viability of lymphocytes in the tially detectable by molecular typing for HLAin
transfused component, and 3) the number of HLA all but HLA-identical transplants, current stan-
alleles shared by the donor and recipient. The dards require chimerism testing for marrow en-
development of TA-GVHDwith the use of fresh graftment monitoring using a different method.
blood components from blood relatives has high- The test for chimerism after transplantation in-
lighted the pathogenic role ofthe HLAsystem. volves identifying the genetic profiles of the re-
Figure 16-5 illustrates the conditions for in- cipient and donor and then evaluating the
creased risk of TA-GVHD.The parents have one extent of mixture in the recipient after trans-
HLA haplotype in common. Each child, there- plantation. The technique commonly employed
fore, has one chance in four of inheriting the uses DNA analysis of short tandem repeat (STR)
Father Mother
Child 1 Child 2 ChildS
Ai Ai
B8 B8
DR17 DR17
FIGURE16-5.HLA haplotypes in a family at risk for transfusion-associated gratt-vs-host disease

(GVHD).In contrast to the family shown in Fig 16-4, each parent shares a common HLA haplotype,
HLA-A 1,BB,DR17. Child 1 is homozygousfor the haplotypeshared by the parentsand by child 3. The
lymphocytes of child 1 are capable of producing posttransfusion GVHD if they are transfused to
either parent or to child 3.
sequences that are amplified, separated by capil- Hematopoietic Stem CellTransplants

lary electrophoresis, and evaluated by the DNA
It has long been recognized that disparitywithin
fragment slzes."
HLAincompatibilityhas rarely been implicat- the HLA system is a significant barrier to suc-
ed in shortened red cell survivalin patients with cessful HSCT.29 HLAsimilarityand compatibility
antibodies to HLA antigens, such as Bg' (B7), between the donor and the recipient are re-
Bt (B17-B57 or B58), and Bg<(A28-A68 or quired for engraftment and to reduce the risk of
A69). These antigens are expressed, although GVHD. However, some degree of rejection or
weakly, on red cells. Such incompatibility may GVHDis a common problem for recipients of al-
not be detected by conventional pretransfusion logeneic stem cells, despite immunosuppressive
testing. conditioning.
The goal of HLAtyping is to match the alleles
of the prospective donor and recipient at the
HLA-A, -B, -C, -DRBI, and -DOBlloci.30 Some
transplant programs also attempt to match do-
HLAtesting is an integral part of solid-organand nors and recipients for HLA-DPalleles. When
HSCT. The extent of testing differs depending matched stem cell donors are unavailable, hap-
on the type of transplantation. (SeeChapter 26.) loidentical stem cell transplants are considered
for patients with high-risk hematologic malig- eluding inflammatory conditions. Serum used
nancies." for crossmatching is often obtained within 48
Although HLA-identical sibling donors re- hours of surgery for sensitized potential recipi-
main the best choice for HSCT, there is increas- ents and may be retained in the frozen state for
ing use of unrelated donors identified by search- any subsequent testing. An incompatible cross-
ing the files of more than 20 million donors match with unfractionated or T lymphocytes is
listed in the National Marrow Donor Program's typically a contraindication to kidney transplan-
registry of volunteer donors, cord blood regis- tation. A positive B-cellcrossmatch is significant
tries, and international registries." when caused by donor-specific HLA Class I or
ClassII antibodies.
Kidney Transplants Sera from patients awaiting deceased-donor
kidney transplant surgery are tested at regular
ABO and lymphocyte crossmatch compatibility intervals to screen for HLAantibodies and to de-
remain the most important factors in determin- termine the specificitiesof the detected antibod-
ing the immediate outcomes of kidney trans- ies. If an antibody with a defined HLAspecificity
plantations. ABO- or crossmatch-incompatible is identified in a recipient, a common practice is
transplantations may be facilitated using proto- to avoid donors who express the corresponding
cols that include various combinations of anti- HLAantigen(s). Such antigens are deemed "un-
body suppression, splenectomy, plasmapheresis, acceptable." Using a standardized algorithm in-
infusion of intravenous immune globulin (MG), volving the HLAfrequencies from 12,000 HLA-
and other treatments to remove preexisting anti- typed donors, a calculated "panel-reactive anti-
bodies and promote accommodation of the body" (cPRA)is obtained and is a more specific
transplanted organ. measure of the sensitization of the patient and
Unlike in HSCT, renal transplants are not the percentage of probability of encountering
routinely HLA matched. Analogous to red cell incompatible or "unacceptable" random do-
transfusions, in renal transplantation the donor nors." Frozen serum samples used for periodic
and recipient need to be compatible, meaning antibody testing are often stored so that "histori-
that the donor does not express preformed anti- cal" samples with the greatest reactivity can be
bodies (ABHor HLA) directed against antigens used in addition to a preoperative sample for
on the donor kidney. Recipients and donors are pretransplantation crossmatching.
routinely typed for ABO and HLA. Recipients Prospective crossmatching is often not per-
are typed for HLA-A,-B,and -DRat a minimum. formed for recipients for whom HLAantibodies
Donors are typed for HLA-A, -B, -C, -DRB, are not detected (ie, cPRA = 0%) after repeat
-DOA, -DQB, and -DPB antigens by molecular testing. Prompt transplantation with reduced
methods. Before transplantation, a crossmatch cold-ischemia time for the renal allograft may
between recipient serum and donor lympho- provide greater benefit to the patient than pro-
cytes is required. Clinical Laboratory Improve- spective crossmatching, provided that 1) a very
ment Amendments (CLIA)regulations [Code of sensitive method for antibody detection, such as
Federal Regulations (CFR) Title 42, Part flow cytometry or microarrays, has been used,
493.1278( e)1 and US federal Organ Procure- and 2) it is certain that the patient has had no
ment and Transplant Network regulations re- additional sensitizing event (ie, immunizations
quire a sensitive crossmatch method." Flow cy- or transfusions in the 2 weeks before or at any
tometry is the most sensitive method and has time after the serum was screened]." "Virtual
been credited with predicting early acute rejec- crossmatching" for renal transplantation re-
tion and delayed graft function, both of which quires careful review of the patient history and
are strong predictors of chronic rejection (if re- HLA antibody test results and is similar to the
sults are positive) and long-term allograftsurviv- concept of an "electronic crossmatch" for red
al (ifresults are negative).33 cell transfusion. In other words, a virtual cross-
HLA antibody levels are dynamic and match is an inferred crossmatch that involves a
change with new immunologic challenges, in- determination of the presence or absence of do-
C HAP TE R 1 6 The HLA System 495
nor HLA-specific antibodies in a patient by compatibility between the donor and recipient is re-
paring the patient's HLA antibody specificity quired. Young pediatric heart or liver transplant
profile to the HLA type of the proposed donor recipients, who have low levels of ABO isoag-
without carrying out a physical crossmatch such glutinins, have had successful outcomes with
as a complement-dependent cytotoxicity or ABO-incompatiblehearts or livers.39,40 HLAanti-
flow-cytometric crossmatch. body screening and typing of potential redpients
Long-term allograft survival is longer with liv- of nonrenal organs is recommended to improve
ing donors than deceased organ donors. One- deceased-donor organ transplantation outcomes.
year graft survival rates from living and de- Likewise, a crossmatch should be available be-
ceased renal donors were 97% and 91.3%, re- fore transplantation when the recipient has
spectively, and the half-lives of living-donor and demonstrated HLA antibody, except for emer-
deceased-donor renal allograft recipients were gency situations. Although the degree of HLA
14.2 years and 9.9 years, respecttvely." compatibility correlates with graft survival after
The significantly better graft survival rate for heart, lung, small-intestine, and liver transplan-
recipients of living- vs deceased-donor renal al- tations, prospective HLA matching is generally
lografts, even when donors and recipients are not performed for these procedures because of
completely unrelated, coupled with inadequate the relative scarcity of donors. Pancreas trans-
numbers of deceased-organ donors has led to plantation generally follows the same guidelines
kidney paired donations (KPD), which permit as kidney transplantation.
patients with an ABO- or Hl.Aincompatible po-
tentialliving donor to exchange their donor for
the donor of other patients in the same situa- H (LiN; AL
tion." KPDs have been facilitated through local I NIFIC NT F
and national registries. As a simple example, a H A
blood group A transplant candidate with an in-
compatible blood group B potential living kid- For some conditions, especially those believed
ney donor could exchange that donor for the in- to have an autoimmune etiology, an association
compatible blood group A livingkidney donor of exists between HLA phenotype and the occur-
a blood group B transplant candidate. Patients rence of, or resistance to, clinical dtsease.'!"
with HLA-incompatible potential donors have (SeeTable 16-3.) HLA-associateddisease suscep-
similar possibilities for donor exchange, and
multiple "pairs" can be involved in one continu-
ous exchange (chain) process. TABLE 16-3. HLA-Associated Diseases
The introduction of altruistic donors (ie, indi-
viduals Who choose to donate a kidney without
having a specific intended recipient) can stgnlfl-
cantlyexpand KPDoptions. Briefly,an altruistic
Ankylosing spondylitis 827 >150
donor donates a kidney to a patient with an in-
compatible potential living donor, who then do- Narcolepsy. OQ6 >38
nates to a different recipient With an incompati- Subacutethyroiditis 835 14 ....
ble donor, starting a chain with the possibilityof
a large number of livtng-donor transplants. A Type'1 diabetes OQ8
chain of 10 transplants has been reported." Multiple SClerosis OR15,OQ6 12
Rheumatoid arthritis OR4 9
Other Solid-Organ Transplants
Juvenile rheumatoid OR8 8
For liver, heart, lung, and heart/lung trans- arthritis
plants, ABO compatibility remains the primary
immunologic concern for donor selection, and Grave disease OR17 4
pretransplantation determination of ABO com' RR = relative risk.
tibilitiesare known or suspected to be inherited, into the peptide-binding grooves of) two specif-
and the diseases display a clinical course with ic HLAmolecules."
acute exacerbations and remissions and usually Peptide-binding specificity is important to
have characteristics of autoimmune disorders. consider in the development of vaccines. For ex-
Although linkage to disease-susceptibility ample, a vaccine to enhance immune responses
genes was favored as an explanation for the to melanoma using a melanoma-specificpeptide
HLA associations with individual diseases, evi- that binds only to the cells of individuals with
dence has been accumulating that implicates the HLAtype HLA-A *02:01 was selected for de-
the HLA molecules themselves. The most fre- velopment because A *02:01 is the most com-
quentiy stated hypothesis is abnormal presenta- mon allele in virtually all populations.50
tion of peptides by particular HLA molecules Some HLAalleles have been noted to be as-
that result in autoreactivity from presentation of sociated with increased risk for hypersensitivity
cross-reactive nonself peptides or improper pre- reactions, such as toxic epidermal necrolysis
sentation of self peptides. The ancestral haplo- (TEN),with certain drugs. Among the growing
type HLA-A1, B8, DR17 (DRB1*03:01), D02, list of associationsare HLA-B*57:01 with abaca-
discussed in the "Linkage Disequilibrium" sec- vir, HLA-B*15:02 with carbamazepine, and
tion above, is associated with susceptibility to HLA-B*58:01 with allopurinol." The use of mo-
type I diabetes, systemic lupus erythematosus, lecular genetics in pharmacology is called phar-
celiac disease, common variable immunodefi- macogenetics.
ciency, immunoglobulin A (IgA)deficlency,and The degree of association between a given
myasthenia gravis." This haplotype is also asso- HLA type and a disease is often described in
ciated with an accelerated course of HN infec- terms of relative risk (RR),which is a measure of
how much more frequently a disease occurs in
tion. A problem with the abnormal peptide pre-
individuals with a specific HLAtype than in in-
sentation hypothesis is the lack of commonality .
dividuals not having that HLAtype. Calculation
in the associated diseaseprocesses."
of RRis usually based on the cross-productratio
One of the first disease associationsidentified
of a 2 x 2 contingency table. However, because
was between HLA-B27and ankylosingspondyli-
the HLA system is so polymorphic, there is an
tis. Although >90% of patients of European increased possibilityof finding an associationbe-
ancestry with ankylosing spondylitis express
tween an HLAantigen and a disease by chance
HLA-B27 (most commonly HLA-B*27:02 or alone. Therefore, calculating RRs for HLA dis-
-B*27:05j, the test's specificityis low; only 20% ease associationsis more complex and is typical-
of individuals with B27 develop ankylosing ly accomplished by use of Haldane's modifica-
spondylitis.One of the strongest associationsbe- tion of Woolf's formula.52,53 The RRvalues for
tween an HLAallele and a medical condition is some diseases associated with HLA types are
narcolepsy and the HLA-D06 HLA allele shown in Table 16-3.
DOB1 *06:02.45As with HLA-B27and ankylos- HLAtesting was previously used to establish
ing spondylitis, >90% of individualswith narco- relationships or for forensic testing purposes.
lepsy are positive for HLA-DOB1 *06:02, but However, HLAtypinghas been replaced by anal-
only a minority of the individualswith the allele ysis of STR polymorphisms at numerous well-
develop the disease. For some autoimmune dis- defined polymorphic loci for relationship and
eases, the specificpeptide that might trigger the other forensic testing. STRanalysisis sometimes
autoimmune response has been at least tenta- used to confirm that an apparent twin who is
tively identified: a gluten peptide, gliadin,for ce- HLA-identicalto the recipient is a true monozy-
liac disease; cyclic citrullinated peptides for gotic twin through an analysis of these loci on
rheumatoid arthritis; and a peptide from glutam- other chromosomes. These reliable methods
ic acid decarboxylasefor type I diabetes.46·48 Re- and established population databases allow ex-
sistance to cerebral malaria seems to result from clusion or identification of individuals using ex-
a strong cytotoxic f-cell response to particular tremely small samples of DNA-containingbio-
malarial peptides that are restricted by (ie, fit logic material such as body fluids and tissues.
C HAP T E R 1 6 TheHLASystem 497
STR analysis may also be used to identify unla- outcomes. There is still debate as to what level
beled or mislabeled surgical pathology samples. of HLAantibody is clinicallysignificant,especial-
ly in the era of solid-phaseunmunoassays," but
it is not uncommon to transplant organs despite
(LIN L N U NI the low-level (low concentration) presence of
HLA DSAs.This practice can result in risk for an an-
amnestic response in the recipient. This risk is
Clinical consultation in HLAis essential yet un- often minimized by increasing or altering the pa-
derutilized. Not unlike transfusion medicine tient's immunosuppressive regimen. The deci-
and blood banking, the field of clinical histo- sion to proceed with transplantation despite the
compatibility requires expertise in laboratory presence of DSAs is predicated upon assigning
methods, interpretation, and clinical application clinical relevance to the DSAs, and the proce-
of test resuits, combined with an understanding dure can be performed only in the context of a
of overarching regulatory processes, to resolve particular recipient and donor pair.
issues. According to CLlA(42 CFR493.1455), a
"clinical consultant must be qualified to consult Consultation in HSCT
with and render opinions to the laboratory's cli-
ents concerning the diagnosis, treatment and The primary role of the HLAexpert in HSCT is
management of patient care," and be a licensed in allogeneic donor selection. Donor selection is
physician or meet the CLlArequirements to be straightforward when fully matched related do-
a laboratory director. This tralning is not a part nors are available. Queries from the transplanta-
of most routine postgraduate medical education tion team may be made when multiple HLA-
(or clinical laboratory technologist education). matched donors are available and the question
Marques et al54have provided an excellent re- becomes, which is the better/best one? Con-
view of the role and importance of a laboratory- versely, the consult question of which is the
medicine-trained expert. least problematic arises when only mismatched
It is imperative that the HLAlaboratory direc- donors are available. In both cases, the HLAex-
tor, technical supervisor, and clinicalconsultant pert might consider the role of low-expressed
(HLA expert) develop relationships with trans- HLA antigens,56permissive mismatches.F-" or
plantation surgeons, physicians, and other clini- other factors such as age, gender, and CMV sta-
cal care staff (eg, transplantationcoordinators], tuS.59For mismatched allogeneicdonors (includ-
Implicit in consultation is that the referring phy- ing haploidentical and cord donors), identifying
sician is seeking the advice of the HLAexpert in DSAsagainst any donors is critical for successful
either developing a treatment plan or managing HSCT.60
care, with assistance in selectingor interpreting
HLAtests. Consultathmin Solid-Organ
One major distinction between the practice Transplantation
of transfusion medicine and HLApractice is the
assignment of clinicalrelevance of antibodies.In The most commonly transplanted solid organ is
general, blood banks will avoid transfusion of the kidney; therefore, renal transplantation con-
antigen-positive red cell units to a patient who sultation is used as a model here. In pretrans-
has (ever) had an antibody against that antigen, plantation consultation, the general approach to
in order to mitigate hemolytic transfusion reac- kidney transplantation is to determine if an in-
tions. For all the major red cell antibodies, the tended donor is compatible-that is, are there
mere concern for presence is considered clini- HLA immunologic barriers that make the
cally significant. In contrast, HLAantibodies are transplant undesirable? For patients awaiting
not treated the same way. HLA antibodies tar- deceased-donor kidneys, there is a requirement
geting donor allograft/tissue are referred to as to test patients for HLAantibodies and to deter-
donor-specific antibodies (DSAs).The presence mine whether any preexisting antibodies are an
of DSAscan adversely impact long-term allograft impediment to future transplantations. The HLA
director (or clinical consultant) plays a critical the Accreditation of CellularTherapy (FACT)ac-
role in assigning clinically relevant antibodies credits HSCTprograms.
because this determines which future donors
will be viable for a patient. In the United States,
the computer program that assigns donor kid- u u IR N
neys to potential recipients excludes "matches"
when "unacceptable antigens" are present. The The role of HLA in medicine has historically
unacceptable antigens are assigned by the pa- been focused on transplantation and limited to
tient's transplantation program and HLAlabora- studies of the traditional structural HLA anti-
tory. If assignments are ill-assigned (ie, listing gens (ie, HLA-A,-B, -C, and -DR, and more re-
too many unacceptable antigens), donor kidneys cently, -DO and -DP). As biomedical knowledge
may be excluded that may have provided func- increases as a result of the acceleration in tech-
tioning allografts. Conversely, failure to list nologies like NGS and computational methods,
clinicallyrelevant antibodies may result in unex- understanding of the role of HLAin health and
pectedly incompatible crossmatches, delayed disease will continue to grow, and the role of
transplantations, and potential for accelerated the HLAexpert and HLAlaboratory will contin-
rejection. After transplantation, there may be uallyevolve.
programmatic or clinicallydirected screening or Individual immune responses to HLA anti-
monitoring for DSA formation to help guide gens and alleles remain difficult to predict. Ob-
management" or as an adjunct for the diagnosis servations that patients exposed to only a few al-
of rejection." loantigens can become highly sensitized, even
against antigens to which they have not been
exposed, are contrasted with those of other or-
gan recipients whose donors are complete HLA
mismatches, yet no detectable alloresponsesare
made. Some of these phenomena may be due to
epitopes and/or eplets (immunogenic determi-
The field of clinical histocompatibility (also nants) shared by seemingly disparate HLAanti-
known as HLA)is a highly regulated field, and gens (even from different loci). In the first case,
has similaritiesto blood banking and transfusion exposure to an allogeneic but common epitope
medicine. The CFR has requirements (42 CFR results in alloreactivityto all antigens that carry
493.1278) that apply to laboratories performing that epitope. In the latter case, it may be a result
histocompatibility testing/" and the Centers for of shared epitopes between mismatched anti-
Medicare and Medicaid Services (CMS)has giv- gens that prevent an alloresponse.Although not
en select US organizations deemed status for ac- widely used, there is some evidence to suggest
crediting laboratories that perform histocompati- that eplet/epitope-based matching (vs antigen-
bility testing, such as the College of American matching) may result in better outcomes." A
Pathologists(CAP)and the American Societyfor limitation of eplet/ epitope-based matching is
Histocompatibilityand Immunogenetics (ASHI). that eplet/epitope identification requires high-
In addition to the rules governingcllnicalIabora- resolution typing. In the United States, HLAtyp-
tories, clinical transplantation programs also ad- ing for transplantation must be by molecular
here to guidelines set by organizations as a re- methods, but only low-resolution (antigen-
quirement for membership, participation, and equivalent) typing is required for solid-organ
reimbursement by CMS. For solid-organ trans- transplantation. As NGS becomes more widely
plantation, the United Network for Organ Shar- availableand demand for eplet/epitope-matching
ing (UNOS) holds the federal contract for the increases, it is possible that high-resolutionHLA
Organ Procurement Transplant Network typing for solid-organ transplantation will be-
(OPTN).The National Marrow Donor Program come the standard of care same as it currently is
(NMDP)oversees HSCT,and the Foundation for for HSCT.
u additional unrecognized polymorphisms within

the HLAcomplex (ie, single nucleotide polymor-
In conclusion, the HLAsystem is a complex and phisms, or SNPs). In the future, the translation
highly polymorphic set of genes that are collec- of this basic information will undoubtedly lead to
tively involved in all aspects of the immune re- new clinical applicationsin transplantation, auto-
sponse. The recent development of molecular immune diseases, vaccine development, phar-
tools to explore this genetic oasisis providing ad- macogenetics, and infectious diseases.
ditional information, such as the elucidation of
REFERENCES
1. Marsh SG, Albert ED, Bodmer WF, et al, No- 6. Zinkernagel RM, Doherty PC. The discovery of
menclature for factors of the HLAsystem, 2010. MHC restriction. ImmunolToday 1997; 18:14-
Tissue Antigens 2010;75:291-455. 17.
2. RobinsonJ, HalliwellJA, HayhurstJD, etal. The 7. Mungall AI, Palmer SA, Sims SK, et al. The
IPD and IMGT/HLA database:Allelevariant da- DNA sequence and analysis of human chromo-
tabases. Nucleic Acids Res 2015;43:D423-31. some 6. Nature 2003;425:805-11.
3. Tabayoyong WB, Zavazava N. Soluble HLA re- 8. Horton R, Wilming L, Rand V, et al. Gene map
of the extended human MHC. Nat Rev Genet
visited. LeukRes2007;3l:l21-5.
2004;5:889-99.
4. Ghio M, Contini P, Mazzei C, et al. SolubleHLA
9. Buchler T, Gallardo D, Rodriguez-LuacesM, et
class I, HLA class II, and Fas ligand in blood
al. Frequency of HLA-DPB1 disparities detected
components: A possible key to explaln the im- by reference strand-mediated conformation
munomodulatory effects of allogeneic blood analysis in HLA-A,-B, and -DRB1matched sib-
transfusions. Blood 1999;93: 1770-7. lings. Hum Immunol2002;63: 139-42.
5. Voorter CE, van der VliesS, Kik M, et al. Unex- 10. IPD-IMGT/HLA: Statistics. Hinxton, UK: Euro-
pected Bw4 and Bw6 reactivity patterns in new pean Molecular Biology LaboratoryIEuropean
alleles. Tissue Antigens 2000;56:363-70. Bioinformatics Institute, 2019. [Available at
http://www.ebLac.uk/ipd/imgt/hla/stats. 26. Artlett CM, SmithJB,Jimenez SA.Identification

html (accessedOctober 18, 2019).1 of fetal DNA and cellsin skin lesions from wom-
11. Bontadini A. HLAtechniques: Typing and anti- en with systemic sclerosis. N Eng!J Med 1998;
body detection in the laboratory of immunoge- 338:1186-91.
netics. Methods 2012;56:471-6. 27. Pollack MS, Speeg KY, Callander NS, et al. Se-
12. Erlich H. HLA DNA typing: Past, present, and vere, late-onset graft-versus-hostdisease in a liv-
future. TissueAntigens 2012;80: 1-11. er transplant recipient documented by chime-
13. Nunes E, Heslop H, Fernandez-Vina M, et al. rism analysis.Hum Immuno12005;66:28-31.
Definitions of histocompatibility typing terms. 28. ClarkJR, Scott SD,Jack AL, et al. Monitoring of
Blood2011;118:e180-3. chimerism following allogeneic haematopoietic
14. De Santis D, Dinauer D, Duke J, et al. 16(th) stem cell transplantation (HSCT):Technicalrec-
IHIW: Review of HLAtyping by NGS. Int J Im- ommendations for the use of short tandem re-
munogenet 2013;40:72-6. peat (STR)based techniques, on behalf of the
15. Opelz G; Collaborative Transplant Study. Non- United Kingdom National External Quality As-
HLA transplantation immunity revealed by sessment Service for Leucocyte Immunopheno-
1ymphocytotoxic antibodies. Lancet 2005;365: typing Chimerism Working Group. BrJ Haema-
1570-6. to12015;168:26-37.
16. DelvilleM, Charreau B, Rabant M, et al. Patho- 29. Thomas ED. Bone marrow transplantation:A re-
genesis of non-HLA antibodies in solid organ view. Semin Hematol1999;36:95-1 03.
transplantation: Where do we stand? Hum Im- 30. Petersdorf EW. Optimal HLAmatching in hema-
munol2016;77: 1055-62.
topoietic cell transplantation. Curr Opin Immu-
17. Amico P, Honger G, Bielmann D, et al. Inci-
no12008;20:588-93.
dence and prediction of early antibody-mediated
31. Ricci MJ, Medin JA, Foley RS. Advances in
rejection due to non-human leukocyte antigen-
haplo-identical stem cell transplantation in
antibodies. Transplantation 2008;85: 1557-63.
adults with high-risk hematological malignan-
18. RajalingamR. The basics of KIR-HLAcomplexi-
ty. ASHIQuarterly 2015;39:22-7. cies. WorldJ Stem Cells2014;6:380-90.
19. Terasaki PI. A personal perspective: 100-year 32. Organ Procurement and Transplant Network.
history of the humoral theory of transplantation. Policy 4: Histocompatibility. Rockville, MD:
Transplantation 2012;93:751-6. Health Resources and Services Administration,
20. Klein HG, Anstee DJ. Immunology of leuco- 2017.
cytes, platelets and plasma components. In: 33. Bryan CF, Baier KA, Nelson PW, et al. Long-
Mollison's blood transfusion in clinical medi- term graft survival is improved in cadaveric re-
cine. 12th ed. Oxford, UK: Wiley-Blackwell, nal retransplantation by flow cytometric cross-
2014:549-610. matching. Transplantation 1998;66: 1827-32.
21. Hod E, Schwartz J. Platelet transfusion refracto- 34. Cecka JM. Calculated PRA (CPRA):The new
riness. BrJ Haemato12008; 142:348-60. measure of sensitization for transplant candi-
22. Kopko PM, Warner P, Kresie L, et al. Methods dates. AmJ Transplant 2010; 10:26-9.
forthe selectionofplateletproductsfora1loimmune- 35. Gebel HM, Bray RA.Sensitization and sensitivi-
refractory patients. Transfusion 2015;55:235- ty: Defining the unsensitized patient. Transplan-
44. tation 2000;69:1370-4.
23. Davenport RD, Kunkel SL.Cytokine roles in he- 36. Scientific Registry of Transplant Recipients;Or-
molytic and nonhemolytic transfusion reactions. gan Procurement and Transplantation Network.
TransfusMed Rev 1994;8:157-68. SRTR/OPTN 2012 annual data report. [Avail-
24. Triulzi DJ, Kleinman S, KakaiyaRM, et al. The able at https:llsrtr.transp1ant.hrsa.gov/.1
effect of previous pregnancy and transfusion on 37. Terasaki PI, CeckaJM, Gjertson DW, et al. High
HLAalloimmunizationin blood donors: Implica- survivalrates of kidney transplants from spousal
tions for a transfusion-related acute lung injury and living unrelated donors. N Engl J Med
risk reduction strategy. Transfusion 2009;49: 1995;333:333-6.
1825-35. 38. Rees MA, KopkeJE, Pelletier RP, et al. A nonsi-
25. SivaSaiKS,Jendrisak M, DuffyBF,et al. Chime- multaneous, extended, altruistic-donor chain.
rism in peripheral blood of sensitized patients N Eng!J Med 2009;360: 1096-101.
waiting for renal transplantation: Clinicalimpli- 39. Daebritz SH, Schmoeckel M, Mair H, et al.
cations. Transplantation 2000;69:538-44. Blood type incompatible cardiac transplantation
C HAP TE R 1 6 The HLA System 501
in young infants. Eur J Cardiothorac Surg 2007; 53. Woolf B. On estimating the relation between
31 :339-43; discussion, 43. blood group and disease. Ann Hum Genet
40. Heffron T, Welch D, Pillen T, et al. Successful 1955;19:251-3.
ABO-incompatible pediatric liver transplantation 54. Marques MB, AnastasiJ, Ashwood E, et al. The
utilizing standard immunosuppression with se- clinical pathologist as consultant. Am J Clin
lective postoperative plasmapheresis. Liver Pathol20 11;135:11-12.
Transpl2006; 12:972-8. 55. Bettinotti MP, ZacharyAA, LeffellMS. Clinically
41. Thorsby E. Invited anniversary review: HLA as- relevant interpretation of solid phase assays for
sociated diseases. Hum Immunol 1997;53: 1-11. HLA antibody. Curr Opin Organ Transplant
42. Pile KD. Broadsheet number 51: HLA and dis- 2016;21 :453-8.
ease associations. Pathology 1999;31 :202-12. 56. Femandez-Vina MA, KleinJP, Haagenson M, et
43. Price P, Witt C, Allcock R, et al. The genetic ba- al. Multiple mismatches at the low expression
sis for the association of the 8.1 ancestral haplo- HLA loci DP, DO, and DRB3/4/5 associate
type (AI, B8, DR3) with multiple immuno- with adverse outcomes in hematopoietic stem
pathological diseases. Immunol Rev 1999;167: cell transplantation. Blood2013; 121:4603-1O.
257-74. 57. Fleischhauer K, Shaw BE, GooleyT, et al. Effect
44. Holoshitz J. The quest for better understanding of T-cell-epitope matching at HLA-DPB1 in re-
of HLA-diseaseassociation: Scenes from a road cipients of unrelated-donor haemopoietic-cell
less travelled by. DiscovMed 2013; 16:93-101. transplantation: A retrospective study. Lancet
45. Pelin Z, Guilleminault C, Risch N, et al. HLA- Oncol20 12;13:366-74.
DOB1* 0602 homozygosity increases relative 58. Fernandez-Vina MA, Wang T, Lee SJ,et al. Iden-
risk for narcolepsy but not disease severity in tification of a permissible HLAmismatch in he-
matopoietic stem cell transplantation. Blood
two ethnic groups. US Modafinil in Narcolepsy
2014; 123:1270-8.
Multicenter Study Group. Tissue Antigens
59. Spellman SR, Eapen M, Logan BR, et al. A per-
1998;51:96-100.
spective on the selection of unrelated donors
46. Cinova I, Palova-JelinkovaL, SmythiesLE,et al.
and cord blood units for transplantation. Blood
Gliadin peptides activate blood monocytes from
2012;120:259-65.
patients with celiac disease. J Clin Immunol 60. Yoshihara S, Taniguchi K, Ogawa H, et al. The
2007;27:201-9. role of HLAantibodies in allogeneic SCT: Is the
47. van Gaalen FA,van Aken J, HuizingaTW, et al. 'type-and-screen' strategy necessary not only for
Associationbetween HLAclass II genes and au- blood type but also for HLA? Bone Marrow
toantibodies to cyclic citrullinated peptides Transplant 2012;47: 1499-506.
(CCPs)influences the severity of rheumatoid ar- 61. Tait BD, Susal C, Gebel HM, et al. Consensus
thritis. Arthritis Rheum 2004;50:2113-21. guidelines on the testing and clinical manage-
48. Mayr A, SchlosserM, Grober N, et al. GAD au- ment issues associated with HLAand non-HLA
toantibody affinityand epitope specificityidenti- antibodies in transplantation. Transplantation
fy distinct immunization profiles in children at 2013;95:19-47.
risk for type 1 diabetes. Diabetes2007;56: 1527- 62. Haas M, Loupy A, Lefaucheur C, et al. The
33. Banff 2017 KidneyMeeting report: Revised di-
49. Hill AV. The immunogenetics of resistance to agnostic criteriafor chronicactiveT cell-mediated
malaria. Proc Assoc Am Physicians 1999; 111: rejection, antibody-mediated rejection, and
272-7. prospects for integrative endpoints for next-
50. SlingluffCL Ir, YamshchikovG, Neese P, et al. generation clinicaltrials. Am J Transplant 2018;
Phase I trial of a melanoma vaccine with 18:293·307.
gp100(280-288) peptide and tetanus helper 63. Code of federal regulations. Standard: Histo-
peptide in adjuvant: Immunologic and clinical compatibility. Title 42, CFR Part 493.1278.
outcomes. Clin Cancer Res2001 ;7:3012-24. Washington, DC: US Government Publishing
51. Pavlos R, Mallal S, PhillipsE. HLAand pharma- Office,2019 (revisedannually).
cogenetics of drug hypersensitivity. Pharmacog- 64. Wiebe C, Pochinco D, Blydt-HansenTD, et al.
enomics 2012; 13:1285-306. Class II HLA epitope matching-A strategy to
52. Haldane JB. The estimation and significanceof minimize de novo donor-specificantibody devel-
the logarithm of a ratio of frequencies. Ann opment and improve outcomes. Am J Trans-
Hum Genet 1956;20:309-11. plant 2013; 13:3114-22.
Transfusion-Service-Related .:Activities:
Pretran.sfuSion Testing and Storage,
Monitoring, Processing, .Distribution,
and Inventory Management of Blood
Components
Caroline R.Alquist, MD, PhD, and Sarah K. Harm, MD
OSPITAL-BASEDTRANSFUSIONSER- any special needs (eg,irradiation) or special pro-

vice activities ensure transfusion of cessing required (eg,volume reduction) and the
pure, potent, safe, and efficaciousblood name of the ordering physician or authorized
components to recipients needing this life- health professional. Additional useful informa-
sustaining therapy. Safe transfusion practice be- tion to guide testing and/or product/component
gins at the bedside with proper recipient identi- selection may include the recipient's gender,
fication, sample collection, and labeling; contin- age, weight, diagnosis, and transfusion and/or
ues with pretransfusion testing; and culminates pregnancy history.
in the selection and distribution of compatible
blood and blood components that have been ap- Recipient Identification and Sample
propriately processed, tested, monitored, and Labeling
stored. Collection of a properly labeled pretransfusion
blood sample from the intended recipient is crit-
ical to safe blood transfusion. At the time of
A A U T sample collection, the phlebotomist must 1) ac-
Transfusion Requests curately identify the potential transfusion recipi-
ent before collecting the pretransfusion sample
All requests for blood and blood components and 2) ensure that the correct recipient informa-
must be legible, accurate, and complete, with tion is placed on the sample label before the
two independent identifiersfor correct recipient sample leaves the side of the recipient.1(P38) A
identification.1(pp38-39) Requests should include mechanism must be in place to identify the
the type and amount of component requested, phlebotomist, as well as the date and time of
Caroline R. Alquist, MD, PhD, Section Head of Clinical Pathology, Department of Pathology and Laboratory Med-
Icine, Ochsner Health System, and Assistant Professor, Ochsner Clinical School, University of Queensland School
of Medicine, New Orleans, LOUisiana;and Sarah K. Harm, MD, Medical Director of Blood Bank, University of
Vermont Medical Center, and Associate Professor, Department of Pathology, Robert Larner MD College of Medi-
cine, Burlington, Vermont
503
sample collection.llp38)ABO-incompatible Red RETRANlSFU I TING

Blood Cell (RBC)transfusions should never oc- RECIPIENT B
cur, but when they do, they most commonly re-
sult from misidentification of recipients or pre- Serologic Testing Overview
transfusion sample labeling errors.' These errors The recipient's ABO group and RhD type must
could result in wrong blood in tube (WBIT),a be determined before transfusion, with the ex-
situation where the blood in the tube is not that ception of emergent situations (discussed later
of the recipient identified on the sample label.' in this chapter). In addition, testing for unex-
The risk of WBIT when using manual methods pected antibodies to red cell antigens (eg, anti-
of patient identification is reported to be at least body detection test or screen) is required before
transfusion of Whole Blood, RBCs,and Granulo-
1 in 3000 samples."
cytes.llp39)Results of testing on the current sam-
Because of the risk for WBIT, for allogeneic
ple must be compared with previous transfusion
transfusions, two determinations of a transfu- service records to identify any discrepancy be-
sion recipient's ABO group are required. Elec- tween previous and current determinations of
tronic identification systems that use machine- ABO group and RhD type.llp41)Current testing
readable information (eg, bar codes or embed- results must also be compared to previous re-
ded radio-frequency-emitting chips) are avail- cords to identify any history of clinicallysignifi-
able to perform and integrate many functions, cant antibodies, adverse events to transfusion,
including recipient identification, sample label- or special transfusion requirernents.t=" Cross-
ing, blood unit identification, and linkage of match testing is required on any blood compo-
nent containing ;:::2 mL of red cells. Unless the
blood components to their intended recipi-
need for blood is urgent or emergent, a cross-
ent(s).5-7The use of an electronic patient identi- match must be performed before a Whole Blood
fication system at the time of pretransfusion or RBC transfusion. Crossmatching is also re-
sample collection Significantlydecreases the risk quired before granulocyte transfusion and plate-
ofWBIT to about 1 in 15,000 samples." let transfusions unless the component is
prepared by a method known to result in a com-
Confirming Sample Linkage ponent containing <2 mL of red cells.1(p42) When
clinically significant antibodies are not detected
When a pretransfusion sample is received in the
using current antibody detection tests and there
laboratory, laboratory personnel must confirm is no record of previous detection of such anti-
that the information on the sample label and the bodies, then a method must be used to detect
information on the pretransfusion testing re- ABO incompatibility.llp42)The advantages and
quest are in agreement. If there is any doubt limitations of various pretransfusion testing
about the identity of the recipient or the labeling schemes are shown in Table 17-1.
of the sample, a new sample must be ob-
tained.1(p38)
One study found that samples failing Serologic Testing Principles
to meet acceptability criteria were 40 times The basis of pretransfusion testing using test
more likely to have a blood grouping discrepan- tube methods, solid phase, or column agglutina-
cy, and a larger multicenter study found a WBIT tion is the detection of in-vitro red cell antigen!
rate of approximately I in 30 mislabeled (reject- antibody reactions by observation of agglutina-
tion or hemolysis. Agglutination is a reversible
ed) samples.v' ThUS,adherence to a strict policy
chemical reaction that occurs in two stages: 1)
of rejecting incorrectly labeled samples for test- sensitization, in which the antibody attaches to
ing should help avoid blood grouping errors and the red cell antigen, and 2) agglutination, in
decrease the risk for transfusion of ABO-incom- which the sensitized red cells are bridged to
patible blood components. form a macroscopically detectable lattice. The
C HAP T E R 1 7 Transfusion-Service-RelatedActivities 505
TABLE17-1. Pretransfusion Testing Schemes
Hold A sample has been ABO, RhO,and antibody

collected. detection testing are not
performed.
Type and hold ABOand RhO A sample has beencol- Antibody detection test-
lected;the recipient's ABO ing is not performed.
group and RhOtype are
known.
Type and screen ABO, RhO,and antibody Most of the pretranstu- Does not include cross-
detection test/identification sion testing has beenper- match.
formed; compatible blood
can be provided in most
situations.
Type and screen ABO, RhO,antibody detec- Routine pretransfusion Units are removed from
with crossmatch * tion test/identification, Red testing has been per- general inventory and
Blood Cell unit selection or formed; compatible blood may not be availablefor
phenotyping, and cross- can be provided in most timely use by other
match situations. recipients.
*"Prepare" (or other term) may be used instead of the word "crossmatch" by some hospital electronic ordering systems.
antihuman globulin (AHG)phase, or indirect an- tors affect the incubation time necessary to
tiglobulin test (IAT,sometimes referred to as the achieve the detectable endpoint. The strength of
indirect Coombs test), detects bound red cell an- agglutination or degree of hemolysis, or both,
tibodies that do not produce direct agglutina- observed during serologic testing should be re-
tion. AHG reagent consists of immunoglobulin corded immediately after reading. Refer to
M (IgM) antibodies directed against the Fc por- Method 1-9 for details on the grading and scor-
tion of IgG molecules." In the past, AHG sera ing of serologictest reactions.
were produced primarily by injecting animals
with human globulins to stimulate antibody pro- Test Methods
duction against the foreign human protein. Now Pretransfusion testing may be performed using
AHG is most often manufactured from hybrid- the traditional tube method or with other auto-
oma cell lines as a monoclonal antibody. This mated and semiautomated testing platforms us-
antiserum attaches to and causes agglutination ing column-agglutination(sometimescalled "gel"
of red cells sensitized with human globulins, as or "beads"), microplatesolid-phase,or hemagglu-
illustrated in Fig 17-1. Causes of false-positive tination-microplate technologies. Fluid-phase
and false-negative results in antiglobulin tests and solid-phase assays are discussed in greater
using test tube methods are listed in Appendices detail in Chapter 8. Automated testing systems
17·1 and 17-2. require validation before implementation and af-
Various factors may enhance or decrease sen- ter any alteration in software functionality. Mo-
sitization and agglutination, including tempera- lecular testing is not routinely performed for
ture, immunoglobulin class, and interactions be- pretransfusion testing. However, testing by mo-
tween antigen configuration and the antigen- lecular methods may be employed to resolve
binding fragment site of the antibody. These fac- ABO group and/or RhD typing discrepancies
FIGURE17-1. The antihuman globulin (AHG) reaction. AHG (lgM anti-lgG) molecules are shown reacting with
the Fc portion of human IgG coating adjacentred cells. (Image courtesy of J. O'Connor.)
and to determine red cell antigen genotyping (See Method 1-3.)At times, it may be necessary
when serologic typing cannot be performed. to obtain a pretransfusion blood sample from the
Molecular testing is discussed in greater detail in same extremity in which there is an intravenous
Chapter 8. Additionally, platelet crossmatch infusion. If this is the case, steps should be taken
compatibility testing may also occur in the pre- to avoid dilution of the sample, which might re-
transfusion setting, using commercial solid- sult in a failure to detect unexpected red cell an-
phase adherence assays when patients have antibodies. Plasma is also preferred for automated
tibodies to HLA or human platelet antigen methods to avoid fibrin interference and clots.
(HPA) identified in their serum. Further infor-
mation on selection of HLA-matched,Hl.A-anti- Sample Age
gen-negative, or crossmatch-compatible platelet
When pretransfusion testing is performed for a
units for the treatment of platelet refractoriness
recipient who has been pregnant or received
is covered in Chapters 16 and 19.
transfusion within the previous 3 months, or
when pregnancy and/or transfusion history are
Sample Requirements
uncertain, the pretransfusion sample used for
Pretransfusion testing uses recipient red cells testing must be no more than 3 days old at the
and either serum or plasma. Because the end- time of the intended transfusion because a re-
point of pretransfusion testing is the visualiza- cent transfusion or pregnancy may stimulate
tion of agglutination, the use of hemolyzed or production of unexpected antibodies. The day of
lipemic samples may create difficultiesin evalu- collection is counted as day 0; therefore, a sam-
ating test results. Plasma is often the preferred ple collected on a Monday can be used for a
sample, as incompletely clotted serum samples transfusion until 11:59 pm on Thursday of the
may contain small fibrin clots that trap red cells same week. 1(p40) Although the specification of 3
into aggregates, which may cause false-positive days is arbitrary, the 3-dayrequirement was cre-
results. In addition, clotting may be incomplete ated as a practical approach to ensure that the
in anticoagulated serum samples, such as those sample used for testing reflects the recipient's
from recipients who have been treated with current immunologic status." If the histories of
heparin; addition of thrombin or protamine sul- transfusion and pregnancy are known with cer-
fate to the sample may correct this problem. tainty and if no transfusion or pregnancy has oc-
curred in the previous 3 months, pretransfusion Comparison with previous records.

testing may be completed in advance of a sched- Retesting the same sample if patient identifi-
uled surgicalprocedure. Pretransfusion testing is cation was verified using a validated elec-
performed at some centers up to 45 days before tronic identification system.
a scheduled surgical procedure to reduce delays
and cancellations associated with unexpected Routine testing for ABO and resolution of
antibodies identified when samples are drawn ABO discrepancies are described in greater de-
on the day of surgery." tail in Chapter 10. Table 17-2 lists blood compo-
nent selection criteria for ABO groups when
ABOGroup and RhDTyping ABO-identicalcomponents are not available.
To determine the recipient's RhD type, the
To determine the recipient's ABO group, the re- recipient's red cells must be tested for the
cipient's red cells must be tested with anti-Aand RhD antigen using antl-D reagent (Method 2-
antl-Breagent(forward or front typing). In addi- 13). 1(pp39-40) Weak RhD testing is not required for
tion, the recipient's serum or plasma must be recipient samples except when assessingthe red
tested against Al and B red cells (reverse or back cells of an infant born to an RhD-negativemoth-
typing). (SeeMethod 2-2.) Donor red cells must er (Method 2_15).I{P51)If problems in RhD typing
be ABO compatible with the recipient's plasma. arise, especially if the recipient is a female of
Any discrepant ABO typing results must be re- childbearing potential, it is prudent to limit
solved before type-specificblood can be given transfusion to blood components containing red
(Method 2-4). If urgent or emergent transfusion cells that are RhD negative, at least until the
is necessary before ABO typing can be complet- problem is resolved. Testing for the RhD antigen
ed, the recipient should receive group 0 red is described in greater detail in Chapter 11.
cells.I{p39) Group 0 RBCunit selection should be
continued, as needed, until the patient's ABO
Detection of Unexpected Antibodies
group can be confirmed by two determina-
tions.I{pp40-41) Antibody detection tests, also known as. anti-
A second determination can be accomplished body screens, are designed to detect clinically
as follows: significant (predominantly IgG) antibodies, in-
cluding those associated with hemolytic disease
Testing a second sample collected at a time of the fetus and newborn (HDFN), hemolytic
different from the first sample, including a transfusion reactions, or notably decreased sur-
new verification of patient identification. vival of transfused red cells. The procedure uses
TABLE17·2. Blood Component ABO Requirements

Whole Blood Must identical to that of the recipient. Low-titer group a Whole Blood may
be used for emergent transfusions.
Red Blood Cells Must be compatible with the recipient's plasma.
Granulocytes Must be compatible with the recipient's plasma.
Plasma Selectedto.be compatible with the recipient's red cells. Some centers are using
thawed group A plasma for emergent transfusions.
Platelets All ABO groups are acceptable. Although ABO-identical platelets are preferred,
components that are compatible with the recipient's red cells are recom-
mended.
Cryoprecipitated All ABO groups are acceptable.
Antihemophilic Factor
the IAT method to demonstrate in-vitro reac- with a negative antibody screen include reduc-
tions between red cells with known antigen tions in turnaround time, workload, and reagent
expression and any antibodies that might be costs.
present in the patient. The recipient serum or In the IS saline technique, recipient serum or
plasma is incubated with unpooled reagent plasma is mixed with saline-suspended donor
antibody-screeningcells. Panels of two, three, or red cells at room temperature. The tube is cen-
four screening cells may be used. Following in- trifuged immediately and observed for the pres-
cubation at 37 C, the cells are then washed to ence of agglutination. Failure to properly per-
remove unbound globulins. The presence of ag- form the IS crossmatch test can lead to false-
glutination with the addition of AHG reagent in- negative results and the failure to detect ABO-
dicates antibody binding to a specificred cell an- incompatible RBCunits. IS This technique is de-
tigen. scribed in greater detail in Method 3-1.
When reactivity is identified in the antibody
detection test (screen), additional testing is per- Computer/Electronic Crossmatch
formed to identify the target red cell antigen(s).
Additional testing strategies used for antibody ABO compatibility may be verified alternatively
identification may include use of enhancement by using a computerized/electronic crossmatch,
media [albumin additive, low-ionic-strength sa- provided that the following conditions have
been met1(pp42-43):
line (LISS),polyethylene glycol (PEG)] and/or
enzyme or chemical treatment of the panel
The computer system has been validated on-
screening cells. Antibody identification is dis-
site to ensure that only ABO-compatible
cussed in greater detail in Chapter 13.
Recipients with clinically significant red cell Whole Blood or RBCsare selected for trans-
antibodies or a history of such antibodies must fusion.
The computer system contains the donation
receive crossmatch-compatible Whole Blood,
RBCs,or Granulocytes that lack the correspond- identification number (DIN), component
ing antigen. Clinicallysignificantred cell alloan- name, ABO group, and RhD type of the com-
tibodies may become undetectable in a recipi- ponent; the ABO group confirmation of the
ent's plasma over time. Between 30% and 35% component; the two unique recipient identi-
of antibodies are undetectable within 1 year, fiers; the ABO group and RhD type of the re-
and nearly 50% are undetectable after 10 or cipient and the results of the antibody screen;
more years.12Failure to detect a weakly reactive and the interpretation of compatibility.
A method exists to verify correct entry of
red cell alloantibody can be followed by a rapid
anamnestic production of antibody and/or a de- data before the release of blood components.
layed hemolytic transfusion reaction. 13 The system contains logic to alert the user 1)
to discrepancies between the donor ABO
group/RhD type on the unit label and the
Immediate-Spin Crossmatch
ABO group/RhD type determined by blood
The immediate-spin (IS) crossmatch method is group confirmatory tests and 2) to ABO in-
the serologic method used to detect ABO incompatibility between the recipient and do-
compatibilitybetween donor red cells and recip- nor unit.
ient serum. This method can be used as the sole
crossmatch method only if the recipient has no This method can be used as the sole cross-
present or previously detected, clinicallysignifi- match method only if the recipient has no pres-
cant antibodies.1(p42)When IS crossmatch is used ent or previously detected, clinically significant
for recipients with a negative antibody detection antibodies.1(pp42-43)
Potential advantages of a
test result, the risk of an overt hemolytic reac- computer crossmatch include decreased work-
tion from an undetected alloantibody is lOW.14 load, reduced sample volume required for test-
The potential benefits of using an IS crossmatch ing, reduced exposure of personnel to blood
instead of a full AHG crossmatch in recipients samples, and better use of blood inventory."
Antiglobulin Crossmatch blood and blood components. Transport require-

ments must be adhered to anytime blood com-
Bloodlacklng relevant antigens should be select-
ponents are transferred between locations, in-
ed for transfusion in a recipient with a clinically
cluding 1) from the collection site to the
significant antibody identified currently or his-
processing facility, 2) from the supplier to the
torically, even if the antibody is presently nome-
blood bank, 3) from the blood bank to the recip-
active.1(p41) This crossmatch includes incubation
ient, and 4) back to the blood bank if not trans-
at 37 C and the IAT (AHGtest). The AHG cross-
fused. Storage requirements and expiration
match may be performed using tube, column-
dates vary by component type and are based on
agglutination ("gel" or "beads"), or solid-phase
factors such as in-vitro red cell metabolism for
systems. For a routine tube AHG crossmatch,
RBCcomponents in various storage solutions or
the donor red cells obtained from a segment of
coagulation protein stabilization solutions for
tubing that was originallyattached to the donor
plasma components (Table 17-3). Failure to ad-
unit to be transfused are washed and resuspend-
here to these storage and expiration require-
ed to between 2% and 5% concentration in sa-
ments can result in decreased component poten-
line, Recipient serum or plasma is then mixed
cy and/or safety.
with the washed donor red cells. Following in- Temperature requirements during transport
cubation at 37 C, the cells are again washed to of blood components differ from those during
remove unbound immunoglobulins. The pres-
storage." (See Table 17-3.) Shipping from the
ence of agglutination with the addition of AHG supplier to the hospital blood bank is considered
reagent indicates incompatibility(Method 3-2). transport, and applicable temperature require-
ments must be met. When blood components
Interpretation of Antibody Detection are issued from the blood bank to the recipient-
Testing (Screening) and Crossmatch care area, maintenance of appropriate tempera-
Results ture requirements allows for the possibilityof re-
Most samples tested have a negative antibody turning the components to inventory if they are
detection test and are crossmatch compatible not transfused.
with the donor RBCunits selected. However, a Refrigerators, freezers, and platelet incuba-
negative antibody screening result does not tors for blood and blood component storage
guarantee that the serum or plasma does not can be equipped with continuous-temperature-
contain clinically significantred cell antibodies; monitoring devices to allow detection of tem-
it shows only that the sample contains no de- perature deviations before components are af-
tectable antibodies that are reactive with the fected. Automated electronic monitoring devic-
screening cells or techniques used. Further- es that are available include 1) weekly pen-and-
more, a compatible crossmatch does not guaran- chart recorders, 2) sets of hard-wired or radio-
tee normal red cell survival. See Appendix 17-3 frequency temperature-recording devices, iand
for a summary of positive pretransfusion test re- 3) centralized temperature-monitoring systems.
sults and their possible causes. Requirements Thermometers or thermocouples should be
for compatibility testing for neonates/infants strategicallyplaced in the equipment for optimal
<4 months of age are discussedin Chapter 24. temperature monitoring. If an automated tem-
perature-recording device is not used, tempera-
tures of the blood storage environment must be
recorded manually every at least. every 4
IE hourS.1(p16) This requirement includes ambient
room temperature monitoring of platelets 'that
are not stored in a platelet chamber or incuba-
General Considerations
tor.
Recorded temperatures should be checked
Transport and storage requirements must be fol- daily to ensure proper operation of the equip-
lowed to optimize the safety and efficacy of ment and recorder. Deviations from acceptable
,.... "'0
N CD
C:::i
N e
Cl.
Cl.
~~
dO
c, c,
""
<t
o
(.) (.) u....
(.) (.) (.) (.) (.)

o
,.... o
,.... o
,.... o
,.... o
,....
, ,...., , , ,....,
"'0
~
(.) (.) (.) ec.
,....,
(.0
,
(.0
,
(.0
c.
""
(J)
(.)
co
cc
"'0 "'0
Ji
c
"'0
CD
o o .!::l
o
ii5
o
zs
II)
C
Q
CI..
eCD
CD CD E §;, en
o
.c:
o
.c:
Q 0, (.)
(.) CD co
$: $:
(jj
o cc
(.)
"Q
Q
Q
iii
"Q
N II) (.0
a:
CHAPTER 17 Transfusion-Service-Related Activities 511
:0-
w
=>
<= c:
0 ~0
~<= 2-
CD
>
::::>
.~
,_
CD
q::
'"
CD
N
CD
~
,.-
I
CI)
«
« I
Cl
u;
.,_ Ol
<= C/l
e
'5.
>. "0 '5 -,_
C/l
t>,
.0
"0
CD
Cl..CD OJ <=
CD e::::> ~
~ '" e
"0
e::::> CD
> e 0.0
Ci5 0
E
:;:::;
Cl..
CD t>
til
<=
"0 ::::> +-'
..- C'J 0
.<::
0
si:
2
Cl..
'"
CD
>.
6)
"0 ~ :c C/l
"0
en E $
::::> C/l :~
N
C/l """<= """ -e- Cl.. a CD
'"...... >. %; <= <=
0
0
C::i
C'J
a...
'>,
'"
0
"0 '';:::
C'J
E
C'J
,.- '",_
C/l
,.-
.0
,_ til ::::> 0
<=
......
C/l
'E
"0
'"
"0 >.
C/l '"
E t5
,_
E
_g
(.) LO "0
::::>
« I
'" «I
s:
C/l
>.
LO
.s
<= 0 ::::>
......
'<,
Cl
a...
cry
>.
C/l ......
CD
C/l Cl
a...
0
e::::> Cl >.
a...
CD
'"
C/l <=
>. CD
N "0
CD
.0
e::::> ~
0
'';:::
(.)
Ol
<=
"Ii)
.=
CD
Ol
<= CD
_g :::l ~
,.- CD
(.) cry .<:2 tti
~ > C/l
<:( :;:::; <=
(.)
0 "0
CD
<= 0 0 "0 E <=
CD '"
.<:: "0
ci .;;; 0
Cl Cl '5 CD
(.) a...
Cl
a,
.<::
a...
I Cl..
e ......
.<:: sz: (.)
(.)
~
(.)
<=
« ::s ~ e
"0 Cl.. CI) <= (.) 0 0
« (.) «a (.) """
N (.)
"""
N """
C'J 0 « 5: 2: :;:::;
...... ......
CD
......
0 :::l
0
0
...... ::::>
0
tti
CD .s:
...... C/l CD .<::
tl
zs .§: ,_ zs ,_
C/l
<=
'Cii
C/l
::::>
0
'zs
C/l
~ ::::>
0
0 CD 0 CD
CD .<:: .<::
N Cl.. E a Cl.. E a
0,_
....,.. :;:::; cry :;:::; cry
'" ••
C/l
'" ••
C/l
<= CD (.)
E <= CD (.)
E <=
:::l :::l
0
~<= E ~0
(.) (.) C/l C/l
(.) (.)
0 E .~ (.) 0
a
,.-
a
,.-
a
,.- .;;; a
..-I """
C3 t;J 'x +-'
a C3 N """
I 'x ......
I I
,.-
I
2: «C/l a
N '" '0,
2:
'7"
«
C/l a '"
2:
'0,
'" N
'"
I I
:::l ::::>
<= ""<= <= ""<=
:;:::; 0 :;:::;0
,_ <= .~ <=
~
CD
0
(.) ...... <=
0
0
(.) :t::
"0 '0, Ol
.<::
"0
(.)
,_ ~ '"
CD
:;:::;
E
"0_
:5
.§: '"
CD
:;:::;
0
(.)
(.) <=
CD ~ til
<=
(.) <=
CD
(.) (.) (.)
LO
(.)
"""
N
Ol (.) 0
:;:::; """
N
Ol
~
,.-
~ ~ (0 ~ 6 C/l
::::>
(0
I Cl..
I
a
C/l
::::>
T- I N 0 T- o N 0
"0 "0
CD ""
C/l
*
(.) (.)
::::> <= co
"0
CD
QJ
> cr::.
cr::. C/l
::::>
'(i;'
"0
CD
C/l (.)
co cr::. tti
~ cr::. "0 <=
CD C/l
(.)
"0
CD > (.) :t=
~0:::l .!:::! ::::> co "0
CD
tti e '(i;' cr::. ~ "0
cr::.
-
""en
CD 0
<= 0 C/l
...... C/l
*
CD "0
-l CD
>
(.) <= CD
c:
~ "'ci) CI) "'ci)
(/)
(.) ::::> -§;(3 CD
N
.<::
(/) CI.)
Ci5 Ci5 "0 Ci5 '5
co co tti tti "0 tti ~
cr::.
'(i;'
cr::.
CD
Cl cr::. 2
u.. '"
$:
c:
C)
CI..
0:: 0:: (.) 0::
E
--
C)
c.:I
CI.)
as
co en a N
~ cry
c::: """
LO
co
.;::
II)
...
:=
c:..)
~
.S!
:=
"CI
~
c:
o
0>
c:
:a
c:
Q.)
Cl..
~
:::I
Q.)
"0
o
.c:
.-c: -e-
o E
Q.)
~
'i5.. ~
><
L.I.J
c:
Q.)
Cl..
"0
C
a
e<:S
C
o
o
+-'
Q.)
+-'
o
:::I a
Q.)
+-'
o
:::I
:5 :5
~
o
:0
'en
en
'~ f3
Q.) 0
:0
'en
en
'~ f3
Q.) 0
c. o E.c: o E.c:
Cl.. Cl..
C/)
c en ~g en +='~
e
I-
ell
Q.)
enO
•
• E
:::I
ell
Q.)
enO
•
• E
:::I
CD-
0",," E 0",," E
C> 13SJ '~ TIC;J '~
eo
-
e,,_n
~~ ::a: ~~ ::a:
-
B
c
o
CD
E Q.)
.~
::::l o
:;:::;
c:
c- """
NO> Q.)
CD
a: 6 ~
JOC! N 0
..-
io
1:
e<:S
en
-
"0
C 1i)
e<:S
en
enE
1i) Q.)
Q5
1<l
CD
(,)
c
e
===
II)
CI.
~~
c:: c:
't:l Q.)
Cl..
c::
en
'en
e
'*
Q.) Q.)
E (5 0
a: =
c:..)
o
c,
.c:
Cl..
-c
c:i
z
E
~ co o
N
CHAPTER 17 Transfusion-Service-RelatedActivities 513
e
I I
"0. ~'-'
X Ci
(J.) '-'
Ci'j Ci'j
c:: c::
"0, "0,
";::
o
";::
o e
:::l
E o
E s:
~ ~ co
(J.) (/) .~
C> :5 c::
~~
x: "C o
c:: :::l
(J.)
'-' c:: E c:: c;,

o 0 N 0 c::
2':.j::i :;::; U5
o
.....,
(J.)
:E
"U5
(/)
+-'
:::l
o
-'=
.-t=
:s: ~
:::l
-
o
(J.)
:E
"U5
(/)
o
+-'
(J.)
:E
"U5
(/)
o
+-'
(J.)
:E
"U5
(/)
.s
(J.)
:E
"U5
. .
(J.) 0 (/)
o
c. E-'= o
c.
o
c.
o o
c. c.
'oP ~
~ E ~ (/)
ro
(/)
ro
(/)
ro
(/)0
(J.) • :::l (/)0
(J.) • 510 510 510
E .g~
0"<:1"
C3Ci" "~
(/)0
0"<:1"
TIN I
0"<:1"
TIN .g~
(/) 0 (/)0 (/)0 (/)0
«N 2.: «N «N «N «N
,_ ,_
(J.) (J.)
"C "C
Ci Ci
'-'
,_ '-'
o
"<:I"
o
o o
o
N
o
N
00
'I
00
'I
N C'Q t.n co 00
N N N N N
E
2
......
(/)
.c c:: c::
1:: o o
o :g ~
E .92 ..!!2
N o
'-'
o
'-'
E E
~ 2
......
(/) (/)
:5 s:
c:: 1::
o o
E E
'" '"
o
+-'
<l.)
:0
"00
(/)
o
c.
(/)
C<S
~(.)
(.)
0"","
TISJ o
T""
(/)0 I
<:(N
._
o
._ ._ ._ ._
<l.) <l.) <l.) <l)
"0 "0 "0 "0
00 o o
'-'
._ '-'
'-' ._ '-'
._ ._
o 0 o o
(.) (.) (.) (.) (.)
co L!) co co co
11 I
I
I
$l
<h>
e::-
LL
!:::.-
C<S
E
(/)
C<S
0::
c::
fl:l
2
LL
.c
(/)
e
LL
Q
Z
e
-s. en o
'"
'" M M
CHAPTER 17 Transfusion-Service-RelatedActivities 515
, , =0-
<=
2
C.
e
:::J
:t::: 2
c. ,
Q.)
:::J
<=
ro "0 0 .c: Q,) :;:::;
0>
<=
<=
ro
.c: ,S:<
.c:
'<l" $:
"0
<=
ro
e
0>
<=
0
'1ii
:::J
"0 E Q,)
N "0 E
Q,)
ro 2.-
Q,)
ts 1il <= E Q,)
ts 1il
>.
>.
.c
.... ~0 e-,
(J)
:E
:t::: _g .!E>. (J)
Ci
c.
0
Q,)
$: D- o :::J
(.)
(.)
':;
(.)
<=
"0
Q,)
(J) <=
Q,)
u;
u; (.)
(.)
<=
"0
Q,)
(J)
:c (J)
1:::
r--
Q,)
"0 Q) 0 N Q,) r-, Q,) 0 M 0
c:;
x: zs
.c:
M
"0
Q,)
.c 2 s:
..... M Q,) LO
c:) (J)
6 ero ro '§
"0
6 ro
M
.....
ro Q,)
c:;
Q,)
>
ro
.c: "0
,= Q,)
>
.c '§: Q,)
0>
<= ';::
Q,)
M
1;;
Q,)
>
ro
.c: "0
,= ..-
c.o
a:
ro
(J)
e 0
$:
de
Q,) ro "0 $: Q,) u; '3 .....
0
Cii .c: .....

(J) (J) (.)
ro Cii (J) (J)
ro Cii (J) 0- <=
.c: Cl s: Q,) .c: .,_ x: .s: Q,) Q,) Q,)
I- u; en (.) en 0 ~ I- en (.) N a: E
, ,
~ ,.!..
..... ,
e- _gE
0 0> Q,)
:::J 'c_ ';:: <= (.)
0 (J)
'<3 (.) 0 Q,)
x: ro <=-
$: 0 ec. ro (J) ....0
0
,!':Q
.s:
5:
E
Q,)
.!E
0
'<l"
N
......
(.)
~
<=
0 0
E >.
ro
"0 "0
.... '+- e (.)
-<t :::J
"0 'a 'n.!E i:::.' ~
(.)
(J)
ro
0:: LO
Q,)
~
Q,)
>
Q,)
0
<=
0>
ro Cii
<=
N
I-
0
.... X
Q,) .....
0 ro .c:
:::J
0 "0 i:i .....
.c:
.c:
(.)
0 >.
Ci
'0,
';::
a: c. -"- <=
Q,)
(.)
E Q,) Q,)
:E ~ c.
Q,) ro (.) 0
'<l" '<l" $: (J)
N <= 0
1;; '0, E (J)
ro :::J ro $: 'c_ :::J
E
u; 0 ro N "0 $: <=- x
(J)
D- "0 ';:: (J)

,!':Q
....
0
'+-
0:: "0
.c:
I-
Q,)
a: t:>
Q,)
.... 1::: 0
.:::
(J)
ro E 0
..... (J)
(J) Q,) (J) 0
Q,)
0
.c:
"0 _g 0 Q,)
> -5
ro
"0 "0
:::J
(.)
ro
"0
.....
Q,) :::J
.l:l
"0 ~
.... 'E (J)
(J)
x:
......
c:
'c_ 2 'c_
Q,) "0 Q,) <= Q,) Q,) Q,) ......
0
:::J
(J)
(J)
>.
ro ~
.c:
(.)
0
E
:::J
(J) a: :::J
(J) '<3
c. X
Q,)Q,)
Q,)
<=
(J)-o
ii;'o
....
Q,)
0
E
,!':Q ,!':Q Q,) <=
::!::
"0 ro
s: :E
...... $: N
::!:: 13
,!':Q
....
Q,)
c. "0
1;; ro
0
0
"0 0
co ~
a: N 0
LO ::!:: <= (J) LO '.p
Q,) >. Q,)

Ci 1;;
~ c.
:::J 1il
<=
Q,)
(J)
<=
1::: CD
N N
0
2
'+-
.s: ...... ....
0
(J) <= '+-
<= Q,) <=
(.) (.)
19 (.) (.) (.) .....
0
..... E Q,) ~<=
0 0 <= 0 0 0
..-, ~ e
,....,
..-
, ,
,....,
,- ,- 'c;:l ,- 'c;:l
Q,) 0>
:2: a: ro :2:
>.
Ci
c.
....
Q,)
:::J
(J) ....
Q,)
"0 1::: "0
0 0 0
(.) x: ...... (.)
0
(J)
0
<=
Q,) ....
...... E 0
(.) (.) (.) (.) (.) (.) ..... Q,) (.)
9
,.... 9 00 c.o, ,
c.o c.o, ~Q,)
e
0>
00
'I ..- ..- ..- a: ro 'I
"0 ,
::c .....
Q30 >. 'c_ W>
"0
'<l" c. E '<3 Q,)
(.)
N >':=l ......
0
Q,) S .....
Q)
<= E
Q,) :::J
Q,) 0 1;; 1;; <= C.
o .....c
...... '§: 0
"0
Q,)
:E 0 :::J Q,) :t::: :t:::
i:::.' a:
c.
.... ::c:
:t::: .c 1;; c. ro (.) ro
5: $l '<3 '<3 .c: E <=
<=
Q,)
::c: c. Q,) D-
.....
Q,) S
ro
E ec. .........
Q,)-
c. Q,)
ro
E ro
(J)
ro
c:
Q,)
Q,)
0>
Q,) D-
E
..... <= (J) 0 (J) "0 E 0:: N 0
.c:.
N
2 I- ~
Q,)
Q,)
'§: ro i:::.'
04=
i:::.'~ ro Q,)
(.) (J)
ro "0
....
0
1;;
0:: 0:: '+-
.....
u..
ro ~ E ~
:::J .c:
.....
ro
"0
(.)W>
ro
-0
Q,)
(.)
ro
"0
Q,)
:::J
"0 "0 0:: e
CD
0
D-
ro
0 0 ....
Q,) Q,)
E ~ 0 ::c $: E (.)
:::J E (.)
:::J
Q)
a: "0
'3
>
0
"0
'3 E
(J)
ro :::J
0
a: Q,)
ro
.c:
(J)
ro "0
Q,)
(J)
ro "0
Q,)
~
.c:
Q,)
0-
(.)
Q,)
(J)
ro
0:: ::c 1;; '<l"
N ~ I- 0:: a: 0:: a: I-
1;;
...... :.:J a: ~ 0::
LO c.o 00
M M M
..
c
o
~
.0..
><
L.U
"0
C
~
C
o
~
o
c..
(/)
c
e
I-
m
Q)
eo
......
U)
._
-
B
c
o
0.:>
E
e
::s
c-
o.:>
0:
(/)
Q)
.:::
Q)
::l
~
(/)
(/) 'E
Q)
F Cl
temperature ranges should be documented and the disposition of blood components when using
explained (including any actions taken and an temperature-monitoring indicators.
assessment of blood component acceptability for
transfusion), dated, and initialed by the person Specific Considerations
noting the deviation.
Holding blood components that have been dis-
Most .blood component-storage devices are
equipped With audible alarms to alert personnel pensed from the blood bank to other hospital ar-
that temperature ranges are approaching unac- eas before transfusion is considered "storage." If
ceptable levels. Central alarm monitoring allows the blood components are not kept in a moni-
facilitiesthat do not have personnel in the vicin- tored device, they must be stored in containers
ity of the equipment to alert designated staff at (eg, boxes or coolers) validated to maintain the
another location when an alarm is activated. Be- correct temperature during storage (Table 17-3).
cause platelets must be gently agitated during If the temperature of the blood or blood compo-
storage, typically using horizontal flatbed or el- nent exceeds the temperature range, it must be
liptical rotators, alarm systems should also.emit discarded.
alerts when the platelet agitator has malfunc-
tioned. Red Blood Cells
Transfusion services may place blood storage
RBC components are stored in plastic bags of
refrigerators in other areas of the hospital to al-
low immediate access to blood in emergency sit- different types with various added anticoagu-
uations. Such a practice requires that the same lants and additive solutions that modify the cel-
blood component storage monitoring standards lular and protein environment. The storage tem-
be met in these other areas. perature of RBC units must be maintained at
If an equipment failure occurs and prevents 1 to 6 C throughout the duration of storage (Ta-
acceptable temperature ranges from being main- ble 17-3).
tained, the facility should have polides, process- During storage of RBC units, biochemical
es, and procedures in place to relocate blood and morphologic changes to the red cells occur
and blood components. The secondary storage that collectively have been referred to as the
location may be another on- or off-siterefrigera- "storage lesion." These changes include cell
tor or freezer, qualified storage boxes, or coolers membrane shape change and microvesiculation;
used with a validated process that has been decreased pH, adenosine triphosphate, and 2,3-
shown to maintain required temperatures diphosphogiycerate; and increased lysophospho-
during storage. Because the safety, purity, poten- lipids, potassium, and free hemoglobin." In ad-
cy, and quality of the blood components could dition to decreased posttransfusion in-vivo red
be affected by delays in relocation to a second- cell recovery, the red cell storage lesion directly
ary storage location, it is recommended that the affects how long RBCunits may be stored. Prod-
relocation occur before upper or lower accept- uct approval by the Food and Drug Administra-
able storage temperatures are exceeded. This
tion (FDA) requires in-vivo labeling studies
can be accomplished by setting the alarm points
demonstrating that at least 75% of the trans-
of the storage devices so that an alarm sounds
before the unacceptable temperature limit is fused red cells are present in circulation
reached. 24 hours after transfusion with less than 1% he-
In some facilities, temperature-monitoring in- molysis. Although the in-vitro observations of
dicators may be used for each blood component the storage lesion are well described, there is
container. Such indicators monitor the liquid emerging evidence that red cell storage duration
temperature of the immediate inner bag, not the does not correlate with worse clinical out-
liquid core temperature in the unit, which may comes. 19·21 One situation where evidence sup-
be cooler. Policies, processes, and procedures ports the use of fresher RBC units is large-
should specify how the facility will determine volume transfusions (>25 mLlkg) in neonates.f
Whole Blood Plasma and Cryoprecipitate

Whole Blood,which contains red cells, plasma, Plasma and cryoprecipitate storage conditions
and platelets, is collected and stored in plastic and shelf life affect coagulation factor activi-
collection bags that contain various anticoagu- ty.24,25Fresh Frozen Plasma (FFP),Plasma Fro-
lant and storage solutions approved for Whole zen Within 24 Hours After Phlebotomy (PF24),
Blood donations. The storage temperature must Plasma Frozen Within 24 Hours After Phleboto-
my HeldAt RoomTemperature Up To 24 Hours
be maintained at 1 to 6 C throughout the dura-
After Phlebotomy (PF24RT24), and Plasma
tion of storage (Table 17-3). Whole Blood units
Cryoprecipitate Reduced must be stored at
are subject to the same storage lesion changes as -18 C or colder (Table 17-3). Frozen plasma
described above for RBC components, and the and cryoprecipitate must be thawed before
expiration date of the unit depends on the stor- transfusion as described in the "Pretransfusion
age solution used during collection. Processing"section below.
Platelets Granulocytes
Platelet metabolic, morphologic, and functional Granulocytes should be transfused as soon as
changes associated with storage define the shelf possibleafter receipt from the supplier. Granulo-
life and storage conditions of platelet compo- cytes are stored at 20 to 24 C (Table 17-3),
nents. Metabolic changes include the glycolytic should not be agitated, and must never be leu-
production of lactic acid and the oxidative me- kocyte reduced. In addition, granulocyte com-
tabolism of free fatty acids, resulting in carbon ponents should be irradiated because recipients
dioxide production. Platelet pH is maintained who require granulocytes are severely immuno-
compromised.
above 6.2 via buffering of lactic acid by bicar-
bonate and promotion of oxidative metabolism
facilitatedby the diffusionof oxygen and carbon N
dioxide across a gas-permeable storage bag un-
dergoing gentle agitation." The maximum total
time for platelets to be without agitation is 30 Thawing Plasma and Cryoprecipitate
hours (Table 17-3). Platelets are stored at 20 to Frozen plasma (FFP, PF24, PF24RT24, Plasma
24 C. Investigation of cold-stored and cryopre- Cryoprecipitate Reduced) must be thawed at
served platelets is ongoing. 30 to 37 C using a waterbath or other FDA-
Platelet shelf life is also limited by increased approved device. Thawing in a waterbath re-
risk of bacteria growth due to room-temperature quires the frozen component to be in a plastic
storage. Blood banks and transfusion services overwrap before submersion into the water to
are required to have methods to detect or inacti- prevent contamination of the container entry
vate bacteria in all platelet components.v'" Re- ports. Once thawed, plasma is stored at 1 to 6 C
cent changes have allowed platelet shelf life to and expires 24 hours after thawing (Table 17-3).
be extended from 5 days to 7 days if the plate- Thawed FFP, PF24, and PF24RT24 must be
lets are stored in bags approved for 7-day stor- relabeled as "Thawed Plasma" if stored for lon-
ger than 24 hours. Although not licensed by the
age, are stored in 100% plasma, and have
FDA, Thawed Plasma is included in the MBB
screened negative for bacterial contamination
Standards for Blood Banks and Transfusion
with an FDA-approved"safetymeasure" test. At Services 1(p30) and the Circular 0/ In/ormation/or
the time of writing, the shelf life of platelets the Use 0/ Human Blood and Blood Compo-
stored in platelet additive solution or treated nents.25 Thawed Plasma is stored at 1 to 6 C
with pathogen inactivation systems in the Unit- and expires 5 days after it was originally
ed States cannot exceed 5 days. thawed. Facilitiesmay label such components as
C HAP T E R 1 7 Transfusion-Service-Related
Activities 519
"Thawed Plasma" at the initial time of thawing. ately so that a segment may be detached and
By maintaining a Thawed Plasma inventory, available for ABO/RhD confirmation and cross-
transfusion services may decrease wastage of match testing.
thawed plasma components and have plasma The shelf life of Deglycerolized RBCs de-
immediately available for emergent need such pends on the type of system used. Closed-
as for trauma recipients," system devices allow storage for up to 14 days,
Levels of labile coagulation factors (Factor V and components prepared using open systems
and Factor VIII) and stable factors are well expire within 24 hours of the start of the deglyc-
above 50% of immediate post-thaw levels in erolization process.
Thawed Plasma that has been stored for up to 5
days." Thawed Plasma does, however, contain Platelet Gel Production
reduced concentrations of Factor V, Factor VII, Platelet gel is produced when thrombin and cal-
and Factor VIII as compared to thawed FFP, cium are added to platelet-rich plasma to pro-
PF24, and PF24RT24. For this reason, Thawed duce a glue-like substance for surgical applica-
Plasma is not suitable for single-factor replace- tlon." This product is typicallyprepared at the
ment when antihemophilic factor derivatives bedside immediately before use. Facilities in-
are unavailable. volved in the production pf. this component
Cryoprecipitate is thawed at 30 to 37 C and should refer to the current edition of the AABB
gently resuspended; it can be pooled for ease of Standards for Pettoperstive Autologous Blood
transfusion using small quantities of 0.9% sodi- Collection and Administration for guidance and
um chloride injection (USP) to rinse the con- quality oversight of this manufacturing process.
tents of the bag into the final container by the
transfusion service (Method 6-11) or pooled be- Irradiation
fore storage by the blood supplier. Thawed cryo-
precipitate is stored at 20 to 24 C and expires Irradiation of cellular components is intended to
within 4 hours of pooling if it is pooled in. an prevent transfusion-associated graft-vs-host dis-
open system, or within 6 hours for single units, ease (TA-GVHD),which is caused by prolifera-
prestorage-pooled units, or units pooled using tion of donor T lymphocytes. People at in-
an FDA-clearedsterile connection device." creased risk of TA-GVHDinclude profoundly
immunocompromised recipients; recipients of
Thawing and DeglyceroBizing RBts intrauterine transfusion; recipients undergoing
marrow, umbilical cord blood, or peripheral
RBCunits may be frozen and stored for up to 10 blood stem cell transplantation; and recipients
years following the addition of glycerol as a of cellular components from blood relatives or
cryopreservation agent (Methods 6-6 and 6- donors selected for HLAcompatibility or platelet
7).28,29Frozen RBCunits can be thawed using a crossmatch compatibility.
37 C dry heater or 37 C waterbath. After units Sources of ionizing radiation include gamma
are thawed, the glycerol must be removed be- rays (cesium-I37 or cobalt-60 radioisotopes)and
fore the component is transfused. Commercial x-rays. The required irradiation dose for either
instruments for batch or continuous-flow wash- source needed to prevent proliferation of donor
ing are available for deglycerolization.The man- T lymphocytes in the recipient is a minimum of
ufacturer's instructions should be followed to 25 Gy (2500 cGy/rad) to the central point of
ensure maximal red cell recovery and minimal the blood container and 15 Gy(1500 cyy/rad)
hemolysis. Measurement of free hemoglobin in to any other part of the container. 1(p27) Confirma-
the final wash can be used to confirm adequate tion that the blood container has received an ad-
free hemoglobin removal and as a surrogate equate radiation dose can be achieved with the
marker for adequate deglycerolization (Method use of commercially available radiographic film
6-8). indicators.
Integrally attached tubing must be filledwith Irradiation is associated with damage to the
the deglycerolizedred cells and sealed appropri- red cell membrane, which may result in increas-
es in extracellular free hemoglobin and potassi- duce exposure to plasma proteins or additives,
um during component storage. For this reason, or achieve a target hematocrit level.
the expiration date of irradiated RBCsis 28 days Volume reduction of platelets is described in
after irradiation or the originai expiration date, Method 6-13. The speed of centrifugation may
whichever is earlier." affect the degree of platelet loss. Higher g forces
Hospitai transfusion services may purchase are associated with better platelet retention but
irradiated blood components from their supplier raise the theoretical concern of platelet damage
or perform irradiation within the blood bank us- and activation as platelets are forced against the
ing approved and monitored radiation devices. container wall. When platelet component vol-
Hospitaisthat perform their own irradiation may umes are reduced, platelets should rest at room
irradiate their inventory on demand or in batch- temperature for 20 to 60 minutes followingcen-
es. Maintenance of a duai inventory (irradiated trifugation and before resuspension in remaining
and nonirradiated) requires policies and proce- plasma or added saline. The manufacturer's in-
dures to ensure that transfusion recipients re- structions must be followed regarding the mini-
ceive the appropriate component for their clini- mum volume necessary to maintain proper air
cai situation. exchange across the gas-permeable platelet-
storage bag. The shelf life of volume-reduced
Leukocyte Reduction platelets is 4 hours. In addition to centrifuga-
Prestorage leukocyte reduction is the preferred tion, RBCunits can be volume reduced through
method for leukocyte reduction because it pre- settling by gravity overnight with the ports of
vents accumulation of cytokines during compo- the unit facing upward and simple removai of
nent storage. Quality control of bedside filtra- the overlying plasma and preservative solution.
tion is challenging, and the process has been The shelf life of volume-reduced RBCunits is 24
associated with hypotensive transfusion reac- hours when stored at 1 to 6 C.
nons."
Poststorage leukocyte reduction can be per- Washing
formed by the blood bank before issuing a com- Cellular components may be washed to remove
ponent using a leukocyte reduction filter at- plasma proteins. Washing is aiso performed to
tached by a sterile connection. It can also be remove glycerol from frozen RBC units after
performed at the bedside during transfusion us- thawing. Indications for washing RBC or plate-
ing a blood-administrationfilter designed for this let components include a recipient history of se-
purpose. Leukocyte reduction filters are de- vere allergicreactions to components containing
signed to remove >99.9% of white cells (3-10g plasma, the presence of antibodies against im-
reduction) and meet the AABBstandard of <5 x munoglobulin A (IgA)in an IgA-deficientrecipi-
106 leukocytes in 95% of sampled units for ent when IgA-deficientcellular components are
RBCs and Apheresis Platelets, and <8.3 x 105 not available, the presence of maternal antibod-
leukocytes in :::::95%of sampled units for whole-
ies to human platelet antigens (eg, HPA-1a,
blood-derived p1ate1ets.!(p27)
The manufacturer's
when using maternal blood for a neonatal trans-
instructions must be followed for the filtration
fusion), and the need for complement removal
device used to achieve acceptable leukocyte re-
for recipients experiencing posttransfusion pur-
duction.
pura. RBC units for intrauterine transfusions
may be washed to remove some of the preserva-
Volume Reduction
tive solutions and excess potassium that accu-
Volume reduction results when plasma and ad- mulates during storage.
ditive solutions are partially removed from RBC Washing is accomplished with the use of 1 to
or platelet components, typically following cen- 2 L of sterile normai saiine (preferablyusing au-
trifugation. This process may be used to aggres- tomated equipment]. As with volume reduction,
sively manage volume in recipients at risk of washed platelets should rest at room tempera-
transfusion-associated Circulatory overload, re- ture for 20 to 60 minutes without agitation be-
tween centrifugation and resuspension with Reconstituted Whole Blood consists of ABO/
normal saline. Up to 20% of the red cell yield or RhD-compatible RBCs combined with ABO-
33% of the platelet yield may be lost during compatible plasma. This component can be used
washing. Becausewashing creates an "open sys- for neonatal exchange transfusion. The conven-
tem" and removes anticoagulant-preservative tional approach is to combine group 0 RBCs
solutions, washed RBCunits expire 24 hours af- (RhD compatible with the neonate) and group
ter the start of washing, and washed platelet ABplasma to achieve a 50% ± 5% hematocrit of
units expire 4 hours after the start of washing. It the final product. The volumes of the two com-
is recommended that hospitals performing ponents before pooling can be adjusted to
washing complywith the manufacturer's recom- achieve a desired hematocrit level. Followingre-
mendations regarding minimum volumes need- combination, the component can be stored at
ed for component storage bags to maintain opti- 1 to 6 C for up to 24 hours. US sites performing
mal storage conditions unless the component is this manufacturing step are required to register
used shortly thereafter. with the FDA.
Current FDA uniform guidelines should be
Pooling followed when pooled components are la-
beled." A unique pool number should be af-
Certain blood components (whole-blood- fixed to the final container, and all units in the
derived platelets, cryoprecipitate, or reconstitut- pool must be documented in electronic or man-
ed Whole Blood)may need to be pooled to pro- ual records.
vide clinically effective transfusion therapy
without the need to transfuse multiple single Aliquoting
components.
Pooled whole-blood-derived platelets may Recipients requiring low-volume transfusions
contain a significant number of red cells, and may receive aliquots of smaller volumes derived
from the original unit via an FDA-clearedsterile
therefore ABOcompatibilityand risk for RhD al-
connection device or integrated transfer bags.
loimmunization in the recipient must be consid-
Available products designed for use with sterile
ered. If whole-blood-derivedplatelets are pooled
connection devices include transfer packs,
using an open system, the expiration time is 4
smaller-volumebags, and tubing with integrally
hours from the start of pooling. A commercially attached syringes.
available, FDA-cleared,prestorage, whole-blood- The expiration date of the aliquot and mini-
derived platelet pooling system allows storage mum residual volumes that must be maintained
for up to 5 days and the ability to perform depend on the storage container used. Hospital
culture-based bacteria testing." When this sys- transfusion services must develop policies and
tem is used, the pool maintains the expiration procedures for aliquot preparation and storage
date of the earliest collected component in the that comply with manufacturer specifications.
pool. The use of aliquots for neonatal transfusion has
Single cryoprecipitate units are pooled after been shown to result in a decreased number of
thawing in a manner similar to that used for donor exposures." The process of preparing .a.li-
platelets (Method 6-11). The expiration time of quots for small-volume transfusion in neonates
cryoprecipitate pools depends on the method and children is discussed in greater detail in
used for pooling. Cryoprecipitate pooled in an Chapter 24. Lower-volume.components (split
open system expires within 4 hours of the start units) may also be prepared for adult recipients
of pooling. Thawed single concentrates and who require slow rates of transfusion because of
pooled concentrates using sterile connection de- concerns about fluid overload. Split units are
vices expire 6 hours after thawing. Thawed recommended when the component volume
cryoprecipitate is stored at 20 to 24 C. As an al- cannot be transfused at a rate that ensures com-
ternative, the blood center may pool single con- pletion of the transfusion within 4 hours or for
centrates before freezing. small volume transfusions (ie, for pediatric and
neonatal patients.) Split units, especiallyaphere- confirm appropriate transport conditions, com-
sis platelet split units, may also be considered ponent appearance, and expiration date. Any
during times of inventory shortages, depending deviation from routine shipping or component
on the clinical situation of the patient and under conditions should be reported to the shippingfa-
the guidance of the blood bank medical director. cility and documented according to each loca-
Split units are discussed in greater detail in the tion's policies,processes, and procedures.
"Inventory Management" section.
Whole Blood, RBCs,and Thawed Plasma
Components
I TRi U
Whole Blood, RBCs,and thawed plasma compo-
Inspection nents must be transported at a temperature of 1
to 10 C. A variety of options exist for maintain-
Visualinspection of the blood component unit is
ing transport temperature, including bagged wet
a critical control point in blood component man-
ice, commercial cooling packs, and speciallyde-
ufacturing and must occur before labeling, be-
fore shipping, upon receipt, and before issue for signed containers. All transport coolers must be
transfusion. Proper documentation of this pro- qualified to maintain the transport temperature
cess includes the 1) date of inspection, 2) DIN, when packed using a validated process.
3) description of any visible abnormalities, 4) ac- Blood components transported at 1 to 10 C
tion(s) taken, and 5) identity of the staffmember and stored at 1 to 6 C may need to be temporar-
performing the inspection. Visible abnormalities ily removed from those temperatures for entry
may include discoloration of the segments, com- into inventory, irradiation, or other processing.
ponent, or supernatant fluid or the presence of The maximum number of units that can be ma-
visible clots, particulate matter, or other foreign nipulated before the component reaches an un-
bodies. Detection of any such abnormalities acceptable temperature should be determined
should result in component quarantine for fur- and not exceeded. Validation of this process
ther investigation that may include returning may be accomplished using manual temperature
the component to the supplier. monitoring indicators affixed to the blood com-
If a component is determined to be bacterial- ponents or electronic devices that can measure
ly contaminated, the component manufacturer the temperature of the blood components.
must be notified so that an immediate investiga-
tion can take place. Other components prepared Platelets, Thawed Cryoprecipitate, and
from that collection should be quarantined until Granulocytes
the investigation is complete. If the component
(or co-component) has been transfused, the re- Platelets, thawed Cryoprecipitate, and Granulo-
cipient's attending physician should be notified, cytes must be transported at a temperature as
and consultation with the medical director is close to 20 to 24 C as possible (Table 17-3). All
recommended. transport coolers must be qualified to maintain
this transport temperature when packed using a
Shipping validated process. For platelets, the total maxi-
mum time without agitation is 30 hours.
Blood components may be transported between
blood centers, between hospitals, and between Frozen Components
blood centers and hospitals. All containers used
to transport blood components must be quali- Frozen components should be packaged to mini-
fied before use to ensure that the proper compo- mize breakage and maintain a frozen state. Dry
nent transport temperature is maintained. The ice in a suitable container has historically been
shipping transit time, mode of transport, and cli- used for shipping these components. Any dry ice
mate conditions must also be validated. All com- alternative should be qualified in the properly
ponents should be inspected upon receipt to packed shipping container. All transport coolers
C HAP T E R 1 7 Transfusion-Service-Related
Activities 523
must be qualified to maintain the transport tem- grally attached to the donor unit to be trans-
perature when packed using a validated process. fused.
The recipient's sample and a segment from
Receiving any red cell-containing component must be
stored at refrigerated temperatures for at least 7
The.receiving facility should notify the shipping days after each transfusion.1(p39) Retaining both
facilityand provide documentation of any devia- the recipient's sample and donor unit segment
tion from usual shipping container packing or containing red cells allows for repeat and/or ad-
the usual appearance of the shipped blood com- ditional testing if the recipient has a transfusion
ponents. Any blood component not in compli- reaction. Testing of stored samples should be
ance with the facility's policies, processes, and based on the sample storage limitations in the
procedures should be. quarantined. Only after reagent manufacturer's packageinsert.
investigation of the deviation and determination Lack of appropriate storage space may limit
that the component meets acceptance criteria the length of time that samples are stored. Insti-
may the component be removed from quaran- tutions that limit the use of pretransfusion recip-
tine and released into the general inventory. ient samples to those that have been stored for
Blood components should be fully traceable no more than 3 days frequently store these sam-
from collection to final disposition. Electronic or ples for 10 days (ie, 3 days + 7 days). Institu-
manual records indicating compliance with poli- tions that permit the testing of pretransfusion
cies, processes, and procedures should be gener- samples that have been stored for >3 days need
ated and maintained for the applicable record- to ensure that each pretransfusion sample is re-
retention time. Any deviation must be recorded, tained for at least 7 additional days after the
and blood components not meeting require- transfusion for which the sample was used for
ments should be quarantined. Deviations must donor selection.
be investigated to determine appropriate compo- Donor red cells may be obtained from the re-
nent disposition and possible corrective action. mainder of the segment used in the crossmatch-
The results of any corrective action should be ing or from a segment removed before the blood
reported to the blood supplier as needed. Inven- was issued. If the opened crossmatching seg-
tory management should consist of routine de- ment is saved, it should be placed in a tube la-
termination that all blood components are ac- beled with the unit number and then sealed or
counted for and transfused .or appropriately stoppered.
discarded.
Component Testing NG ON N
Before transfusion, the ABO group of all units
Donor RBCUnit Seledion
and RhD type of any units labeled "Rh negative"
must be confirmed by serologic testing for all The results of compatibility testing, as well as a
red-cell-containing components. (RBCs, Whole visual inspection, should guide donor unit selec-
Blood, and Granulocytes).Any typing discrepan- tion (see "Transfusion Documentation and Re-
cies identified must be reported immediately to cipient Identification" section below). Compati-
the supplier and resolved before the component bility considerations vary according to ABO,
is issued for transfuston.P'" RhD, and other blood group antigens of the re-
cipient.
Retention and Storage of Donor
Samples ABO Group Compatibility
IS crossmatch and AHG/IAT crossmatch are Whenever possible, recipients should receive
both performed using the recipient's serum or ABO-identicalblood components; however, oc-
plasma and donor red cells, which must be ob- casionally it may be necessary to select alterna-
tained from a segment of tubing that was inte- tive components. If the component to be trans-
fused contains 22 mL of red cells, the donor's limited group 0, RhD-negative RBCinventory,
red cells must be ABO compatible with the re- AABBand the Choosing Wisely campaign have
cipient's plasma.1{P42) Because plasma-containing
components can also affect the recipient's red
°
emphasized selective use of group Rhli-nega-
tive RBCsin known group 0, RhD-negativepa-
cells, anti-A and/or anti-B antibodies in plasma tients and for emergency transfusions for fe-
for routine transfusions should be compatible males of childbearing potential. Group 0, RhD-
with the recipient's red cells when feasible." positive RBCs should be used for emergency
However, it is not uncommon for blood banks transfusions to males and females who are not
and transfusion services to transfuse incompati- of childbearing potential.f The patient should
ble plasma-containing blood components during be switched to type-specificblood as soon as the
emergent transfusion situations or platelet inpatient's ABO/RhD type is known and con-
ventory shortages. Some centers use thawed firmed (see "Emergent Transfusion" below).
group A plasma for emergent transfusions be-
cause of the relative scarcity of group AB plas-
°
rna." Low-titer group Whole Blood is now
part of trauma resuscitation protocols due to the
Other Blood Groups
Antigens other than ABO and RhD are not rou-
tinely considered in the selection of units of
evidence supporting a balanced ratio approach
to massive hemorrhage." As a result of frequent blood for transfusion to nonalloimmunized re-
platelet shortages, ABO-incompatible platelet cipients. However, for recipients with certain
components are often transfused. Requirements medical conditions, such as sickle cell disease,
some institutions may elect to transfuse RBC
for components and acceptable alternative
units that are phenotypically matched to various
choices are summarized in Table 17-2.
degrees, to prevent alloimmunization in a popu-
lation of frequent transfusion recipients." One
RhO Type
study showed that North American hospital
RhD-positiveblood components should be rou- transfusion service laboratories most commonly
tinely selected for RhD-positiverecipients. RhD- match for the C, E, and Kantigens when pheno-
negative units are compatible with RhD-positive type-matched RBCsare transfused to nonalloim-
recipients but should be reserved for RhD- munized recipients with sickle cell disease."
negative recipients. RhD-negative recipients Matching for more than three antigens may be
(especially females of childbearing potential) of incremental benefit, although sustaining an
should receive red cell-containing components inventory of such components may be extreme-
that are RhD-negative to avoid alloimmuniza- ly challenging.
tion to the RhD antigen and prevent the poten- If the reciplent has clinically significant and
tial for possible HDFN. When ABO-compatible, unexpected antibody(ies), blood lacking the cor-
RhD-negative components are not available for responding antigen(s) should be selected for
an RhD-negativerecipient, the blood bank phy- crossmatching. If there is an adequate quantity
sician and the recipient's physician should of the recipient's serum or if another recipient's
weigh alternative courses of action. The risk of serum with the same antibody specificity is
alloimmunization to the RhD antigen in hospi- available, and if that antibody reacts well with
talized RhD-negative recipients of RhD-positive antigen-positive red cells, that serum may be
RBCunits is approximately 22%, while the inci- used to screen for antigen-negative donor RBC
dence of alloimmunization after apheresis plate- units. When red cells are found to be antigen
let transfusion is 0 to 1.4%.38-41 Depending on negative, this result must be confirmed with a li-
the clinical situation (especiallychildbearing po- censed reagent when such a reagent is available.
tential) of the recipient and the volume of red When licensed reagents are not available (eg,
cells transfused, it may be desirable to adminis- anti-Lan or anti-Yt'), expired reagents or stored
ter Rh Immune Globulin to an RhD-negativere- serum samples from recipients or donors may be
cipient who is given RhD-positiveblood compo- used, provided that the results of controls tested
nents." In order to help hospitals manage the on the day of use are acceptable." When
crossmatch-compatible units cannot be found, Special transfusion requirements leg, cyto-

the transfusion service's attending physician megalovirus .(CMV)-reduced-risk,irradiated,
should be involved in the decision about how to or antigen-negative components], if applica-
manage the recipient. Antigen-negative donor ble.
RBCunits are not usually provided for recipients Expiration date and, if applicable,time.
whose antibodies are not clinicallysignificant. Date and time of issue.
Visualinspection of the product.
Transfusion Documentation and
Recipient Identification The transfusion. service must confirm that
the recipient identifying information, transfu-
The recipient's medical record must include ac- sion request, testing records, and blood compo-
curate and complete documentation of all trans- nent labeling and compatibility are accurate and
fusions. For each transfusion, this documenta- in agreement. Any discrepancies identified must
tion must contain the transfusion order, consent be resolved before issue. Additional records that
for transfusion, component name, DIN, date may be useful include those that identify the
and time of transfusion, pre- and posttransfusion person issuing the blood, the. person to whom
vital signs, volume transfused, identification the blood was issued, and the unit's destina-
of the transfusionist, and, if applicable, any tion. After the transfusion, a record of the trans-
transfusion-related adverse events. fusion becomes part of the. recipient's perma-
Ensuring that the correct blood component is nent electronic or paper medical record.
transfused to the correct recipient is paramount Records must contain the identity of the per-
for transfusion safety. All requests for blood son(s) performing the crossmatch and, if blood is
components must contain at least two.indepen- issued before the resolution of compatibility
dent identifiers so that the. intended recipient problems, the finalserologicfindings.
can be uniquely identified. Recipientcompatibil- Final identification of the transfusion recipi-
ity testing records must also be reviewed. ent and blood component rests with the transfu-
Current testing results must be compared with sionist(s). The Joint Commission requires hospi-
historical records, if available, and any discrep- tals to use a two-person verification process
ancies must be resolved before component se- before initiating a blood or blood component
lection. transfusion. Alternatively, an. automated ID
Personnel must visually inspect and docu- technology (eg, bar coding) may be used in
ment that the selected component is acceptable place of one of the individuals." The individu-
for use. This inspection must include confirma- al(s) must identify the recipient and donor unit
tion that the component does not have an ab- and certify that the identifying information on
normal color or appearance and that the conforms, tags, and labels is in agreement.
tainer is intact. Once selected for transfusion,
the blood component must have an attached la- Spedal.Clinical Situations
bel or tie tag that contains the intended reclpi-
Emergent Transfusion
ent's two independent identifiers, the DIN, the
donor ABO/RhD type, and the compatibility When blood is urgently or emergently needed,
test result interpretation, if performed. At the the recipient's physician must weigh the risk of
time of issue, there must be a finalcheck of each transfusing uncrossmatched or partially compat-
unit that includes the following1(PP46-47): ible blood against the risk of delaying transfu-
sion until compatibility testing is complete or
Intended recipient's two independent identi- fully compatible blood components are identi-
fiers, ABOgroup, and RhD type. fied. A transfusion service physician should be
DIN, donor ABO group, and, if required, availablefor consultation as needed.
RhD type. If transfusion is deemed medically necessary
Interpretation of the crossmatch test results, and blood is released before pretransfusion test-
if performed. ing is complete, the records must contain a
signed statement from the requesting physician Exchange transfusion of a neonate/infant is also
indicating that the clinical situation was suffi- considered a massive transfusion.
ciently urgent to require release of blood compo- Many hospitals have developed massive
nents before completion of compatibility test- transfusion protocols to standardize the re-
ing.1(p48) Such a statement does not need to be sponse to hemorrnage.w" Massive transfusion
obtained before a lifesaving transfusion takes protocols are designed to rapidly provide blood
place, and it does not absolve blood bank per- components in a balanced ratio of plasma and
sonnel from their responsibilityto issue properly platelets to RBCs, particularly when laboratory
labeled donor blood that is ABO compatible testing is not rapid enough to guide transfusion
with the recipient. support. Typicalratios of plasma and platelets to
When emergency release is requested, blood RBCsrange from 1:2 to 1:1, with evidence that
bank personnel should take the following ac- there is no statistically significant difference in
tions: recipient survival in the trauma setting if a 1:2
or a 1:1 ratio is used." Additional studies are
Issue uncrossmatched group 0 RBCs,or low- needed to clarifywhether the use of these proto-
titer group 0 Whole Blood (see note below), cols is associated with improved recipient out-
if the recipient's ABOgroup is unknown. It is comes.
preferable to give RhD-negativeRBCsif the To ensure the ability to accurately interpret
recipient is a female of childbearing poten- ABO group testing results, the recipient sample
tial, while other recipients may receive group should be obtained for testing as early as possi-
0, RhD-positiveRBCs. ble during massive transfusion. If the recipient
Issue blood that is ABO and Rh compatible if ABO group cannot be determined, continued
there has been time to test a current sample. support with group 0 RBCs is required, and
Indicate in a conspicuous fashion on the tag consideration can be given to using group A
or label attached to the unit that compatibili- plasma rather than AB plasma. Unexpected and
ty testing was not completed at the time the significantusage of group 0 RBCsin the setting
unit was issued. of massive transfusion should be considered
Begin compatibilitytests and complete them when determining component inventory levels.
promptly (for massive transfusion, see be- In massive transfusion situations where large
low). If incompatibilityis detected, the recip- amounts of blood may be required, polldes may
ient's physician and the transfusion service be developed to provide RhD-positiveRBCsto
physician should be notified as soon as possi- select RhD-negativerecipients, such as all adult
ble. males and females without childbearing poten-
tial.
Note: The use of low-titer group 0 Whole In the massive transfusion setting, an abbre-
Blood for emergent transfusion of recipients viated crossmatch, such as IS crossmatch,
whose ABO group is not known requires the should be performed, if possible, to confirm the
blood bank/transfusion service to define low- ABO compatibility of the units administered. If
titer group 0 Whole Blood and have policies, the recipient has no transfusion testing history,
processes, and procedures for its use, the maxi- collection of a second sample is recommended
mum volume allowed, and patient monitoring to verify the ABO group and RhD type, which
for adverse effects.1(p48) may allow for the use of electronic/computer
crossmatching. If the recipient qualifiesfor elec-
Massive Transfusion
tronic/computer crossmatching (two separate
ABO group and RhD typings, current negative
There are several different definitions of massive antibody screen, no history of clinically signifi-
transfusion. In this chapter, it is defined as the cant antibodies), electronic crossmatching can
administration of greater than 8 RBCunits in an save Significantamounts of time when issuing
adult recipient in <24 hours or acute adminis- blood components in an emergency. If a more
tration of more than 4 RBCunits within 1 hour. limited pretransfusion testing protocol is used,
the protocol should be described in a written IN N N

standard operating procedure. The transfusion
service should have a policy that addresses com- General Considerations
patibility testing in recipientsof massive transfu- A sufficientnumber of units of varying ABOand
sion;:such. as abbreviation or omission of cross- RhD specificities should be available to meet
matching. 1Ip45) routine hospital needs, allow for unanticipated
increases in utilization from emergency situa-
Blood Administration after Non-Group- tions, and minimize component expiration and
Specific Transfusion wastage. Factors that influence determination of
blood bank component inventory levels include
Once a sample is received and the recipient's historic usage patterns, expiration/wastage
ABO group and RhD type are determined, the rates, and distance from suppliers. Inventory
recipient can begin receiving transfusions of levels should be periodically evaluated in re-
group-specificcomponents. Group 0 RBCunits sponse to institutional changes that may affect
stored in additive solution contain minimal re- component usage, including expansion of inpa-
tient beds or operating rooms; implementation
sidual plasma, which minimizes concerns re-
of new surgicalprocedures; or changes in hospi-
garding passive transfusion of A and B antibod- tal guidelines or medical practice that may influ-
ies. Therefore, switching to ABO-identicaiRBC ence transfusion behavior.
components can be done safely,although an oc- The blood bank should also maintain a re-
casional recipient may exhibit a transient posi- serve of universally compatible RBCs for emer-
tive direct antiglobulin test (DAT) result. In gency use and have a reliable emergency deliv-
some cases, such as when large volumes of ery system to ensure adequate availability of
RBCsare transfused or small children or infants blood components in unexpected situationswhen
demand exceeds supply. Although the practice
receive transfusions, passively acquired anti-A
is not universally adopted, many transfusion ser-
and/or anti-Bmay be detected in the recipient's vices choose to use oniy leukocyte-reduced RBC
serum or plasma." In these cases, a transfusion inventories to decrease transfusion-associated al-
medicine physician should be consulted before loimmunization and subsequent platelet refrac-
switching to type-specific blood components. toriness, as well as to reduce the incidence of fe-
Transfusion of RBCsthat lack the corresponding brile nonhemolytic transfusion reactions.52,53 In
A and/or B antigen(s) may be indicated. times of platelet inventory shortage, medical di-
RhD-negativeRBCunits should be selected if rectors may approve the splitting of apheresis
the recipient is a female of childbearing poten- platelet units (split units) or pooling of3 whole-
blood-derived platelet units (vs the usual adult
tial and RhD type is unknown, or if the recipi-
dose of 4-6 pooled units) for prophylactic trans-
ent, regardless of gender, has anti-D or a history fusion orders to conserve inventory. This prac-
of anti-D. In massive transfusion situations tice is rooted in the findings of the platelet-dose
where large amounts of blood may be required, trial called PLADO, which identified that pro-
policies may be developed to provide D-positive phylactic doses ranging from 1.1 to 4.4 x lOll
RBCsto select recipients. If a recipient receives platelets per square meter had no difference in
blood of an RhD type that is different from that effect on subsequent patient bleeding.54 Stan-
of his or her own blood, it may become difficult dard platelet doses traditionally range from 3 to
6 X 1011 platelets, but have not been shown to
to determine the recipient's true RhD type. If
be superior to lower dosing for prophylactic he-
there is any question about the recipient's true mostasis.55,56 Using this logic, many pediatric
RhD type, it may be prudent to administer RhD- centers also routinely utilize platelet splitting to
negative blood, especiallyif the recipient is a fe- better ration platelet supplies. Inventory conser-
male of childbearingpotential. vation strategies, particularly for use in disaster
plans, should be developed and tested periodi- be available, but also for which surgical proce-
cally. dures require a type and screen or not. The ad-
Routine inventory levels should be moni- vent of IS and electronic crossmatching have de-
tored daily to facilitate timely ordering from creased the utility of the MSBOS, and this
blood suppliers and maintain adequate invento- practice is useful now primarily in hospital trans-
ry levels. This can be particularly challenging for fusion services that lack the ability to perform
platelets because of their limited shelf life. In- electronic crossmatching. A discussion with the
ventory management plans should also take into clinician and the blood bank staff should occur
consideration desirable inventory levels of spe- to determine how many units to crossmatch for
cial products, such as leukocyte-reduced (ie, recipients undergoing elective surgery who are
CMV-reduced-risk)and irradiated components. known to have clinically significant alloantibod-
Antigen-negative RBC units as well as HLA- ies and requlre crossmatch-compatible, antigen-
matched, HLA-selected,or crossmatched plate- negative blood. If an MSBOS has been estab-
lets may be ordered on an as-needed basis from lished, the transfusion service routinely cross-
suppliers, but clear communication between the matches the predicted number of units for each
clinical team and the blood bank is required to recipient undergoing the designated procedures.
anticipate patient needs given procurement de- Routine orders may need to be modified for re-
lays inherent in the process and component ex- cipients with anemia, bleeding disorders, or oth-
piration. er conditions in which increased blood use is
anticipated. As with other circumstances that
Surgical Blood Ordering Practices requlre rapid availability of blood components,
Component outdate rates are influenced by sur- the transfusion service staff should be prepared
gical ordering practices. For example, when to provide additional blood components if the
RBC units are crossmatched for surgical recipi- need arises.
ents, they are unavailable to other recipients Unfortunately, it is not uncommon for an ini-
and have a greater likelihood of expiring if the tial pretransfusion sample to be received by the
component is not promptiy returned to the un- transfusion service laboratory on the morning of
crossmatched inventory. The crossmatch-to- a same-day-admissionsurgical procedure, which
transfusion (C:T) ratio is the number of RBC gives the laboratory limited time to complete
units crossmatched divided by the number of pretransfusion red cell compatibility testing.58
RBCunits actually transfused and is most useful Up to 9% of type and screen samples may not be
in determining individual physician or specialty- tested completely until after the recipient's sur-
specific ordering practices. When C:T ratios are gery has begun. Such testing may offer the first
monitored, a C:T ratio of >2.0 may indicate ex- and only opportunity for a laboratory to deter-
cessive ordering of crossmatched blood and may mine a recipient's ABO group and RhD type. In
identify instances when a preoperative type and addition, approximately 3% of samples received
screen order is more appropriate. have a serologic finding that requires further in-
One approach to reducing excessive C:T ra- vestlgation." The discovery of a serologic find-
tios is to identify procedures that do not typical- ing in the immediate preoperative or periopera-
ly requlre blood, and use this information to de- tive period requiring further investigation may
velop guidelines for the use of type and screen cause dangerous delays in blood availabilityfor a
orders or hold-sample orders (samples that are recipient. Thus, each recipient's blood sample
received in the transfusion service but do not should be received in the laboratory in advance
have any testing orders) instead of crossmatch of the scheduled procedure, and sufficient time
orders. Maximum surgical blood order sched- must be available to complete all preoperative
ules (MSBOSs)for common elective procedures pretransfusion testing before surgery begins.
can also be developed based on local transfusion Collection of a type and screen sample days or
utilization patterns." The MSBOS serves as a even weeks in advance of surgery, with collec-
guideline not only for how many units should tion of a second sample on the morning of a
scheduled surgery, is one approach to mitigate Individual-unit temperature indicators or

the problem. temperature-reading devices can be used. to de-
termine the acceptability of components for re-
Return of Blood Components and turn to inventory. Blood and blood components
Reissue may also be transported or. stored in qualified
containers using a validated process that has
The transfusion service may receive back into been shown to. maintain acceptable tempera-
inventory units that meet acceptance specifica- tures for a defined.interval. If time frames are
tions. These conditions include the follow- used to determine the acceptability of a compo-
ing1(P47): nent's.return to inventory, the time frame must
be validated by the individual facility. The vali-
The container closure has not been dis- dation should demonstrate that for the defined
period, the appropriate temperature of the com-
turbed.
ponent has been maintained.
The component has been maintained at the
Components meeting the acceptance criteria
appropriate temperature. may be returned to the general blood inventory
At least one sealed segment remains integral- and reissued. Components not meeting the ac-
lyattached to the container, if RBCs. ceptance criteria must be quarantined for fur-
Documentation indicates that the compo- ther investigation or discarded in a biohazard
nent has been inspected and is acceptable for container to prevent inadvertent return to in-
reissue. ventory.
and quality of the blood components may be affected by improper storage, alarm settings
should be configured to notify necessary personnel be/ore the upper or lower acceptable stor-
age temperatures are exceeded.
8. Temperature requirements during transport of blood components differ from those during stor-
age. Blood components held outside the blood bank before transfusion are considered to be in
storage. Validated processes must ensure that acceptable storage temperatures are maintained.
9. Acceptable time frames for returning blood components to inventory after issue should be vali-
dated by individual facilities. Individual-unit temperature indicators or temperature-reading de-
vices may be used to determine component acceptability for return to inventory.
10. Thawed FFP, PF24, and PF24RT24 expire within 24 hours of thawing. These components may
be labeled as "Thawed Plasma" to allow for a 5-day shelf life (from the date the component
thawed) if they were originally collected in a closed system.
REFERENCES
1. Gammon R, ed. Standards for blood banks and 10. Shulman IA. When should antibody screening
transfusion services. 32nd ed. Bethesda, MD: tests be done for recently transfused recipients?
AABB,2020. Transfusion 1990;30:39-41.
2. LindenN, Wagner K,VoytovichAE, Sheehan J. 11. Boisen ML, Collins RA, Yazer MH, Waters JH.
Transfusion errors in New York state: An analy- Pretransfusion testing and transfusion of un-
sis of 10 years' experience. Transfusion 2000; crossmatched erythrocytes. Anesthesiology
40:1207-13. 2015;122: 191-5.
3. Bolton-Maggs PH, Wood EM, Wiersum- 12. Ramsey G, Smietana SJ. Long-term follow-up
OsseltonJC. Wrong blood in tube-potential for testing of red cell alloantibodies. Transfusion
serious outcomes: Can it be prevented? Br J 1994;34:122-4.
Haemato12015;168:3·13. 13. Hendrickson JE, HillyerCD. Noninfectious seri-
4. Kaufman RM, Dinh A, Cohn CS, et al for the ous hazards of transfusion. Anesth Analg 2009;
BESTCollaborative. Electronic patient identifi- 108:759-69.
cation for sample labeling reduces wrong blood 14. Shulman IA, Odono V. The risk of overt acute
in tube errors. Transfusion2019;59(3):972-80. hemolytic transfusion reaction followingthe use
5. Knels R, Ashford P, Bidet F, et al for the Task of an immediate-spin crossmatch. Transfusion
Force on RFIDofthe Working Party on Informa-
1994;34:87-8.
tion Technology; International Society of Blood
15. Shulman IA, Calderon C. Effect of delayed cen-
Transfusion. Guidelines for the use of RFID
trifugation or reading on the detection of ABO
technology in transfusion medicine. Vox Sang
incompatibility by the immediate-spin cross-
2010;98(Supp12):1-24.
6. Askeland RW, McGrane S, Levitt JS, et al. Im- match. Transfusion 1991;31:197-200.
proving transfusion safety: Implementation of a 16. Mazepa MA, RavalJS, Park YA;Education Com-
comprehensive computerized bar code-based mittee of the Academy of Clinical Laboratory
tracking system for detecting and preventing er- Physicians and Scientists. Pathology consulta-
rors. Transfusion2008;48: 1308-17. tion on electronic crossmatch. Am J Clin Pathol
7. Murphy MF, Fraser E, Miles D, et al. How do 2014;141 :618-24.
we monitor hospital transfusion practice using 17. Nunes E. Transport versus storage: What is the
an end-to-end electronic transfusion manage- difference?AABBNews 2013;15(2):4-5.
ment system?Transfusion2012;52:2502·12. 18. Klein HG, Spahn DR, Carson JL. Red blood cell
8. Lumadue JA, BoydJS, Ness PM. Adherence to a transfusion in clinical practice. Lancet 2007;
strict specimen-labelingpolicy decreases the in- 370:415-26.
cidence of erroneous blood grouping of blood 19. Fergusson DA, Hebert P, Hogan DL,et al. Effect
bank specimens. Transfusion 1997;37: 1169-72. of fresh red blood cell transfusions on clinical
9. Coombs RRA,Mourant AE, RaceRR.A new test outcomes in premature, very low-birth-weight
for the detection of weak and "incomplete" Rh infants: The ARIPI randomized trial. JAMA
agglutinins.BrJ Exp PatholI945;26:255-66. 2012;308: 1443-51.
20. LacroixJ, Hebert P, Fergusson DA, et al. Age of 33. Food and Drug Administration. Guidance: Unit-
transfused blood in critically ill adults. N EnglJ ed States industry consensus standard for the
Med 2015;372:1410-18. uniform labeling of blood and blood compo-
21. Steiner ME, Ness PM, Assmann SF,et al. Effects nents using ISBT 128. (June 2014) Silver
of red-cell storage duration on recipients under- Spring, MD: CBER Office of Communication,
going cardiac surgery. N Engl J Med 2015; Outreach, and Development, 2014.
372:1419-29. 34. Liu EA, Mannino FL, Lane TA. Prospective, ran-
22. Strauss RG. Data-drivenblood banking practices domized trial of the safety and efficacy of a limit-
for neonatal RBC transfusions. Transfusion ed donor exposure transfusion program for pre-
2000;40: 1528-40. mature neonates. J Pediatr 1994; 125:92-6.
23. Shrivastava M. The platelet storage lesion. 35. Fung M, Downes KA, Shulman IA. Transfusion
TransfusApher Sci 2009;41: 105-13. of platelets containing ABO-incompatible plas-
24. Scott E, Puca K, HeralyJC, et al. Evaluation and ma: A survey of 3,156 North American labora-
comparison of coagulation factor activity in tories. Arch Pathol Lab Med 2007; 131 :909-16.
fresh-frozenplasma and 24-hour plasma at thaw 36. Dunbar NM, Yazer MH; Biomedical Excellence
and after 120 hours of 1 to 6 C storage. Transfu- for Safer Transfusion Collaborative. A possible
sion 2009;49: 1584-91. new paradigm? A survey-based assessment of
25. AABB,American Red Cross, America's Blood the use of thawed group A plasma for trauma re-
Centers, Armed Services Blood Program. Circu- suscitation in the United States. Transfusion
lar of information for the use of human blood 2016;56:125-9.
and blood components. Bethesda, MD: MBB, 37. Sehault IN, Bahr M, Anto V, et al. Safety profile
2017. of uncrossmatched, cold-stored, low-titer, group
26. Werhli G, Taylor NE, Haines, AL, et al. Institut- 0+ whole blood in civilian trauma patients.
ing a thawed plasma procedure: It just makes Transfusion 2018;58(10):2280-8.
sense and saves cents. Transfusion 2009;49: 38. Yazer MH, Triulzi DJ. Detection of anti-D in D-
2625-30. recipients transfused with D+ red blood cells.
27. Tholpady A, Monson J, Radovancevic R, et al. Transfusion 2007;47:2197-201.
Analysis of prolonged storage on coagulation 39. Cid J, Lozano M, Ziman A, et al. Low frequency
Factor (F)V,FYII,and FYIIIin thawed plasma: Is of antl-D allotmmuntzation following D+ plate-
it time to extend the expiration date beyond 5 let transfusion: The Anti-D Alloimmunization af-
days? Transfusion 2013;53:645-50. ter D-incompatible Platelet Transfusions
28. Meryman HT, Hornblower M. A method for (ADAPT) study. Br J Haemato12015;168:598-
freezing and washing RBCsusing a high glycerol 603.
concentration. Transfusion 1972; 12:145-56. 40. O'Brien KL, Haspel RL, Uhl L. Anti-D alloimmu-
29. Valeri CR, Ragno G, Pivacek LE, et al. A multi- nlzation after D-incompatible platelet transfu-
center study of in vitro and in vivo values in hu- sions: A 14-year smgle-instiution retrospective
man RBCs frozen with 40-percent (wt/vol) review. Transfusion 2014;54:650-4.
glyceroland stored after deglycerolizationfor 15 41. Weinstein R, Simard A, Ferschke J, et al. Pro-
days at 4 degrees C in AS-3:Assessment of RBC spective surveillance of D- recipients of D+
processing in the ACP 215. Transfusion 2001; apheresis platelets: Alloimmunization against D
41:933-9. is not detected. Transfusion 2015;55: 1327-30.
30. Borzini P, Mazzucco L. Platelet gels and releas- 42. PollackW, Ascari WO, Crispen JF, et al. Studies
ates. Curr Opin HematoI2005;12:473-9. on Rh prophylaxis II: Rh immune prophylaxis af-
31. CyrM, Hume H, SweeneyJD, et al. Anomaly of ter transfusion with Rh-positiveblood. Transfu-
the des-Arg9-bradykininmetabolism associated sion 1971; 11:340-4.
with severe hypotensive reactions during blood 43. Callum JL,Waters JH, Shaz BH, et al. The AABB
transfusions: A preliminary report. Transfusion recommendations for the Choosing Wisely cam-
1999;39: 1084-8. paign of the American Board of Internal Medi-
32. Benjamin RJ, Kline L, Dy BA, et al. Bacterial cine. Transfusion 2014;54:2344-52.
contamination of whole-blood-derived platelets: 44. Afenyi-Annan A, Brecher ME. Pre-transfusion
The introduction of sample diversion and phenotype matching for sickle cell disease recip-
prestorage pooling with culture testing in the ients. Transfusion 2004;44:619-20.
American Red Cross. Transfusion 2008;48: 45. Osby M, Shulman IA. Phenotype matching of
2348-55. donor red blood cell units for nonalloimmu-
nized sickle cell disease recipients: A survey of transfusion reactions to RBCs. Transfusion
1182 North American laboratories.Arch Pathol 2004;44(1):25-9.
LabMed 2005;129:190-3. 54. Slichter SJ, Kaufman RM, Assmann SF, et aI.
46. Food and Drug Administration. Compliance Dose of prophylactic platelet transfusions and
Program guidance manual. Chapter 42 . Blood prevention of hemorrhage. N Eng!J Med 2010:
and blood components. SilverSpring,MD: FDA, 362(7):600-13.
2013. [Available at https://www.fda.gov/ 55. Tinmouth A, Tannock IF, CrumpM, et aI. Low?
media/84887/download·1 dose prophylactic platelet transfusionsin recipi-
47. 2020 National patient safety goals. Oakbrook
ents of an autologous peripheral blood progeni-
Terrace, IL:TheJoint Commission,2019. [Avail-
able at http://www.jointcommission.orglstan tor cell transplant and patients with acute leuke-
dards_information/npsgs.aspx (accessed No- mia: A randomized controlled trial with a
vember 6, 2019).1 sequential Bayesian design. Transfusion 2004;
48. YoungPP, Cotton BA, Goodnough LT.Massive 44:1711-19.
transfusion protocols for recipients with sub- 56. Heddle NM, Cook RJ,Tinmouth A, et al. A ran-
stantial hemorrhage. Transfus Med Rev 20 II; domized controlled trial comparing standard-
25:293-303. and low-dose strategies for transfusion of plate-
49. Hendrickson JE, Shaz BH, Pereira G, et al. Im- lets (SToP)to patients with thrombocytopenia.
plementation of a pediatric trauma massive Blood2009; 113:1564-73.
transfusion protocol: One institution's experi- 57. Boral LI, Dannemiller FJ, Standard W, et al, A
ence. Transfusion2012;52: 1228-36. guideline for anticipated blood usage during
50. HolcombJB,TilleyBC, Baranluk S, et aI. Trans- elective surgical procedures. Am J Clin Pathol
fusion ofplasma, platelets, and red blood cellsin 1979;71:680-4.
a I: I: 1 vs a 1:1:2 ratio and mortality in recipi- 58. Friedberg RC, Jones BA, Walsh MK. Type and
ents with severe trauma: The PROPPRrandom-
screen completion for scheduled surgicalproce-
ized clinicaltrial.JAMA2015;313:471-82.
dures: A College of American Pathologists 0-
51. Garratty G. Problems associated with passively
transfused blood group ailoantibodies.AmJ Clin Probes study of 8941 type and screen tests in
Pathol 1998;109:169-77. 108 institutions. Arch Pathol Lab Med 2003;
52. Seftel MD, et aI. Universal prestorage leukore- 127:533-40.
duction in Canada decreases platelet alloimmu- 59. Saxena S, Nelson JM, Osby M, et al. Ensuring
nlzation and refractoriness. Blood 2004;103(1): timely completion of type and screen testing
333-9. and the verification of ABO/Rh status for elec-
53. King KE, et aI. Universal leukoreduction de- tive surgical recipients. Arch Pathol Lab Med
creases the incidence of febrile nonhemolytic 2007;131:576-81.
APPENDIX 17-1
Sources of False-Positive Results in Antiglobulin Testing
Cells Agglutinated before Washing
If potent agglutinins are present, agglutinates may not disperse during washing. Observe red cells before the
addition of antihuman globulin (AHG) or use a control tube and substitute saline for AHG. Reactivity before
the addition of AHG or in the saline control invalidates AHG results.
Particles of Contaminants
Dust or dlrt ln glassware may cause clumping (not agglutination) of red cells..Fibrin or precipitates in test
serum may produce red cell clumps that mimic agglutination.
Improper Procedures
Overcentrifugation may pack cells so tightly that they do not easily disperse and they appear to be positive.
Centrifugation of the sample with polyethylene glycol or positively charged polymers beforewashing may
create clumps that do not disperse.
Cells That Have a Positive Direct Antiglobulin Test (OAT) Result
Cells that are positive by OATwill be positive in any indirect antiglobulin test. Procedures for removing IgG
from OAT-positive cells are given in Methods 2-20 and 2-21.
Complement
Complement components, primarily C4, may bind to cells from clots or from citrate-phosphate-dextrose-
adenlne-t donor segments during storage at 4 C and occasionally at higher temperatures. For OAT, use red
cells anticoagulated with EDTA,acid-citrate-dextrose, or citrate-phosphate-dextrose.
Samples collected in tubes containing silicone gel may have spurious complement attachment.'
Complement may attach to red cells in samples collected from infusion lines used to administer dextrose-
containing solutions. Reactions are strongest when large-bore needles are used or sample volume is
<0.5 mL.2
1. GeislandJR, Milam JO. Spuriously positive direct antiglobulin tests caused by silicone gel. Transfusion 1980;20:711-13.
2. Grindon AJ, Wilson MJ. False-positiveDATcaused by variables in sample procurement. Transfusion 1981;21:313-14.
APPENDIX 17-2
Sources of False-Negative Results in Antiglobulin Testing
Neutralization of Antihuman Globulin (AHG) Reagent
Neutralizationof AHGreagentmay result from failure to wash cells adequatelyto removeall serum or plasma.
Filltube at leastthree-quarters full of saline for each wash. Checkvolume dispensed by automated washers.
If increasedserum volumes are used, routine washing may be inadequate.Wash additional times or remove
serum before washing.
The AHG might be contaminated by extraneous protein. Using contaminated droppers or the wrong reagent
dropper can neutralizean entire bottle of AHG. Do not use a finger or hand to cover the tube.
If the concentration of IgG paraproteins in test serum is high, protein may remain evenafter multiple
washes.'
Interruption in Testing
Bound IgG may dissociate from red cells and leavetoo little IgGto detect or neutralizeAHG reagent.
Agglutination of IgG-coated red cells will weaken. Centrifuge and read results immediately.
Improper Reagent Storage
AHG reagent may lose reactivity if it is frozen.
Excessiveheat or repeatedfreezing or thawing may cause loss of reactivity of test serum.
Reagentred cells may lose antigen strength during storage. Other subtle cell changes may causeloss of reac-
tivity.
Improper Procedures
Overcentrifugation may pack red cells so tightly that the agitation required to resuspend red cells breaks up
agglutinates. Undercentrifugation may not be optimal for agglutination.
Failureto add test serum, enhancement medium, or AHGmay causea negativetest result.
Redcell suspensionsthat are too heavymay mask weak agglutination. Suspensionsthat are too light may be
difficult to read.
Improper or insufficient serum:cell ratios can adversely affect results.
Complement
Rareantibodies, notably some antl-Jk" or anti-Jk", may be detected only when polyspecific AHGis used and
active complement is present.
Saline
The low pH of saline solution can decreasethe sensitivity of the test.' The optimal pH of saline wash solution
for most antibodies is 7.0 to 7.2.
Some antibodies may require saline to be at a specific temperature to retain antibody on the cell. Use37 C or
4 C saline.
1. Ylagen ES, Curtis BR, Wildgen ME, et al. Invalidation of antiglobulin tests by a high thermal amplitude cryoglobulin.
2. Rolih S, Thomas R, Fisher E,Talbot J. Antibody detection errors due to acidic or unbuffered saline. Immunohematology
1993;9:15-18.
APPENDIX 17-3
Causes of Positive Pretransfusion Test Results *
Negative Antibody Detection Test Result and Incompatible Immediate-Spin Crossmatch
Donor red cells are ABO incompatible.
Donor red cells are polyagglutinable.
Anti-A1 is in the serum of an individual with A2or A2B.
Other alloantibodies (eg, anti-M) are reactive at room temperature.
Rouleaux haveformed.
Autoantibodies (eg, antl-l) are cold-reactive.
Anti-A or antl-B has been passively acquired.
NegativeAntibody Detection Test Result and Incompatible Antiglobulin Crossmatch
Donor red cells have a positive direct antiglobulin test result.
Antibody is reactive only with red cells having strong expression of a particular antigen (eg, dosage) or varia-
tion in antigen strength (eg, P1).
An antibody to a low-prevalence antigen is present on the donor red cells.
Anti-A or antl-B has been passively acquired.
Positive Antibody Detection Test Result and Compatible Crossmatches
Autoanti-HI (-H) or antl-Le" and non-group-O units are selected.
Antibodies are dependent on the reagent red cell diluent used.
Antibodies demonstrating dosageand donor red cells are from heterozygotes (le, expressing a single dose of
antigen).
Donor unit lacks corresponding antigen.
Positive Antibody Detection Test Result, Incompatible Crossmatches,and NegativeAutocontrol
Alloantibody(ies) are present.
Positive Antibody Detection Test Result, Incompatible Crossmatches,Positive Autocontrol, and Negative
Direct Antiglobulin Test Result
An antibody is present to an ingredient in the enhancement media, or an enhancement-dependentautoanti-
body is present.
Rouleaux haveformed.
Positive Antibody Detection Test Result, Incompatible Crossmatches,Positive Autocontrol, and Positive
Direct Antiglobulin Test Result
Alloantibody is causing a delayed serologic or hemolytic transfusion reaction.
Passively acquired autoantibody (eg, intravenous immune globulin, therapeutic monoclonal antibody) is
present.
Cold- or warm-reactive autoantibody is present.
'Causes depend on serologic methods used.
Administration .of Blood Components
Melanie Jorgenson, RN, BSN, LSSGB
HE SAFEADMINISTRATIONOF BLOOD 5. Pretransfusionsample.

and its components requires a multidisci- 6. Preparation of ordered units for transfusion.
plinary collaboration among clinical and 7. Prophylacticmedications.
ancillary services and clinicians. Policies and 8. Equipment.
procedures must be developed with input from 9. Intravenous (N) access.
transfusionists, the transfusion service, sur- 10. Readinessto transfuse.
geons, anesthesiology care providers, primary-
11. Deliveryof blood components.
care physicians, and transport personnel. The
medical director of the transfusion service or 12. Infusion sets and compatible N solutions.
designee will review and approve the policies 13. Verification of the recipient's identifica-
and procedures on an annual basis. The transfu- tion, and other verifications at the time of
sionist typically provides the last line of defense administration.
in the detection of errors before the transfusion 14. Rates of transfusion.
commences. All personnel involved in prepar- 15. Monitoring during the transfusion.
ing, delivering, and administering a transfusion 16. Suspected transfusion reactions.
must be given appropriate training to ensure the 17. Documentation of the transfusion.
provision of the safest transfusion possible. 18. Unique transfusion settings.
Recipient: Consent
The AABB Standards jor Blood Banks and
Trsnsfusion Services (Standards) states, "The
blood bank or transfusion service medical direc-
Before a transfusion begins, thoughtfui consider- tor shall participate in the development of poli-
ation, planning, and preparation are required. cies, processes, and procedures regarding recipi-
The following areas are discussed in ' detail ent consent for transfusion." l(p49) Recipient
throughout this chapter: informed consent must address indications for;
risks, benefits, and possible side effects of; and
alternatives to transfusions of allogeneic blood
1. The recipient's consent to transfuse, along
components. Some state laws require certain ad-
with transfusion risks, benefits, and alterna- ditional elements in the recipient consent.
tives. The recipient has the right to choose or re-
2. Recipient history and education. fuse a transfusion and must have an opportunity
3. Baseline assessment of the recipient. to ask questions of a learned professional before
4. The provider's orders for blood components providing consent. Documentation of the con-
and administration. sent process must be entered into the recipient's
Melanie Jorgenson, RN, BSN, LSSGB,Client Delivery Lead, Accumen, San Diego, California
The author has disclosed no conflicts of interest.
537
medical record. Some facilities require an 17 for greater detail on pretransfusion process-
institution-approved signed consent form to doc- ing.)
ument that the consent process has occurred The transfusionist must educate the recipient
and that the risks, benefits, and alternatives of about reporting any symptoms that may be in-
transfusion were discussed with the recipient or dicative of a reaction and indicate how long the
legal representative. Each institution must have transfusion will take. The recipient's questions
a process for recording a patient's refusal to re- must be answered before the transfusion is start-
ceive blood or blood components in the patient's ed.
medical record. Institutional policies must iden-
tify health-care providers who are permitted to Baseline Assessment of Recipient
obtain consent and must indicate the length of A baseline physical assessment must include
time and range of recipient care (eg, in- and out- measurement of vital signs, including blood
patient) for which a consent remains valid. pressure, heart rate, temperature, and respirato-
Consent for transfusion must be obtained ry rate. Many institutions also routinely measure
from recipients who have the requisite capacity oxygen saturation using oximetry. The pretrans-
to make such decisions. If a recipient is unable fusion assessment must include symptoms, such
to give consent, a legally authorized representa- as shortness of breath, rash, pruritus, wheezing,
tive or surrogate may do so (depending on local and chills, as a basis for comparison after the
and state laws). If no one is available to provide transfusion is initiated.
consent and the need for transfusion is consid- It is important to consider a patient's baseline
ered a medical emergency, the blood compo- assessment when preparing to transfuse a blood
nent may be administered based on the doctrine component. A recipient with renal or cardiopul-
of implied consent. Individual state and local monary disease may require a slower infusion
laws governing requirements for implied con- rate to prevent transfusion-associatedcirculatory
sent may vary, but the emergent need for trans- overload (TACO).A recipient with an elevated
fusion must be carefully documented in the temperature may destroy cellular components at
medical record.' Informed consent for transfu- an increased rate.' Moreover, if the recipient
sion mayor may not include the name of the presents with an elevated temperature before
health-care provider who obtained consent, and the transfusion, it may be difficult to determine
hospital policy may require the consenter to in- later whether an additional increase in tempera-
clude an entry into the medical record that doc- ture was caused by a transfusion reaction. Ad-
uments the conversation. ministration of an antipyretic may be considered
in such cases.
Recipient History and Education
It is important to collect a history from the recip- Orders to Prepare and to Transfuse
ient before the component is ordered to assess Blood and Blood Components
whether the recipient is at increased risk of a A licensed provider often writes two orders for
transfusion reaction. This history includes previ- the components to be administered. The first or-
ous transfusions and any adverse reactions. If der requests appropriate laboratory testing and
the recipient has had previous reactions to a preparation of the ordered blood component
transfusion, the medical team must determine and notes special processing requirements. The
whether the recipient needs to receive prophy- second order explains to the transfusionist how
lactic medications before transfusion and wheth- to administer the components, including the
er special processing of the component is indi- transfusion rate. Recipient name and another in-
cated to mitigate unnecessary exposure and the dependent identifier (eg, date of birth or medi-
risk of an adverse reaction. Premedication or- cal record number must be included). ideally,
ders should be carefully timed with the antici- orders should include sufficient specificinforma-
pated administration of the unit. (See Chapter tion for clarity, such as:
C HAP T E R 1 8 Administration of Blood Components 539
Component leg, Red Blood Cells (RBCs)or sions, because these components do not require
Apheresis Platelets] to prepare or to adminis- a crossmatch except in rare. cases (eg, unit of
ter. platelets with excessive red cell content). Typi-
" Specialprocessing required (eg, leukocyte reo cally, the sample is obtained within 3 days of
duction, irradiation, or washing). transfusion, with the draw date considered to be
Number of units or volume to administer. day 0.l{p40)
Date and time for the infusion. Institutional policies may vary regarding the
Flow rate or duration for administering the sample outdate. If the recipient has not had a
component. transfusion or been pregnant in the preceding
" Indication for transfusion. 90 days, the sample may be acceptable for Ion"
ger than 3 days for testing purposes. In emer-
It is important to note that it is appropriate gent cases in which the patient's survival is in
for the rate and duration of the transfusion (eg, jeopardy, blood components may be dispensed
not to exceed 4 hours from the time the con" without completing pretransfusion testing, and
tainer is entered until completion]" to be de" retrospective testing may be performed once
scribed in a policy approved by the hospital's sample collection can be achieved.1{p48)
medical staff. The policy might also make con" Containers used for blood specimens must be
siderations for comorbidities (eg, renal disease, labeled in the presence of the recipient.' The
cardiac disease) that impact the rate/duration. sample must be labeled at the recipient's side
When the health-care team is considering with at least two unique identifiers (eg, the reo
transfusion, orders for various laboratory tests cipient's name, date of birth, or identification
(eg, ABO/Rh testing, type and screen, type and number). The identification of the person col"
crossmatch) may be written by the health-care lecting the sample and the date the sample was
provider to prepare for a possibletransfusion. To collected must be traceable.1{p38) Some policles
administer the intended blood components, ad" may also require documentation of the time the
ministration orders are written. sample was collected.
The transfusionist has the responsibility, as All those involved in the pretransfusion sam-
with any other order, to criticallythink through ple collection and recipient verification must be
the order. As with medication orders, the trans" taught that their utmost attention is absolutely
fusionist must determine the orders are for the necessary to avoid mislabeling of samples lead"
right recipient, for the right blood component, ing to ABO mismatches with potentially fatal
for the right reasons, in the right amount, and at outcomes. These types of errors are referred to
the appropriate rate. The ordering provider and as "wrong blood in tube" (WBIT).
the transfusionist must ensure the order does In some institutions, computer-asslsted posi-
not conflict with any faclllty-speclflctransfusion tive recipient identificationand collection of con"
guidelines. If the transfusionist has concern firmatory ABO samples are additional methods
about any of these areas, a conversation with used to further mitigate recipient identification
the ordering provider is essential before carrying errors. Additional information about pretransfu-
out the order. Any deviations from the guide" sion samples may be found in Chapter 17.
lines must be documented in the recipient medi-
cal record by the ordering provider and/or Preparation of Ordered Units for
transfusionist as appropriate. Transfusion
Pretransfusion testing of the recipient's blood
Pretransfusion Sample
sample, including evaluation for unexpected an"
In nonemergent situations, a pretransfusion tibodies to red cell antigens, is described in de"
blood sample is required before all RBCtransfu- tail in Chapters 13 and 17. The time interval
sions. In some hospitals where a historical ABO from sample receipt to availability of the reo
type is known, a pretransfusion sample might quested component can vary greatly. A positive
not be required for plasma and platelet transfu- antibody detection test result requires further
investigation, and it takes time to definitively reduce the incidence or decrease the severity of
identify clinicallysignificantred cell antibodies. future reactions. Corticosteroids may also be
If clinically significant red cell antibodies are useful. Premedication may not prevent an ana-
present, identification of corresponding antigen- phylactic transfusion reaction; therefore, close
negative or crossmatch-compatibleunits may re- observation is required when transfusing pa-
quire additional time, especiallywhen external tients at high risk for anaphylaxis.
suppliers must be consulted to locate an appro- Although antipyretics (eg, acetaminophen)
priate unit. For recipients with multiple or rare are commonly administered to reduce the risk
antibodies, additionalhours and sometimes days of FNHTRs, their effectiveness is limited and
may be required to find a crossmatch-compati- should be discouraged, as they can mask a
ble unit. If the need for transfusionis urgent, the transfusion-related adverse event." For patients
ordering licensed provider must weigh the risks with a history of FNHTR with rigors, meperi-
and benefits of administering least-incompatible dine may be used to premedicate, although its
or uncrossmatched units, ideally in consultation efficacyin this setting has not been studied.
with the transfusion service's medical staff. If premedication is required, it must be ad-
Some components require thawing, pooling, ministered before obtaining the component
relabeling, or other preparation before release. from the transfusion service. Oral premedication
All these factors necessitate timely communica- should be administered 30 minutes before the
tion between transfusion service staff and trans- start of the transfusion. Intravenous medications
fusionists. Components that are pooled or re- should be given 10 minutes before the transfu-
quire thawing may also have a shortened shelf sion is initiated. Corticosteroids require signifi-
life after being prepared (4-24 hours); transfu- cant time to exert their effect, but the optimal
sionists must be made aware when the time timing for their use has not been established in
availableto complete a transfusion of such com- the transfusion settmg.v"
ponents is decreased.I (pp56·64) Nonpharmacologic methods have been
shown to reduce the incidence of common
Prophylactic Medications Given before transfusion reactions. Prestorage leukocyte re-
Transfusion
duction reduces the incidence of febrile reac-
tions. The removal or reduction of plasma pro-
Recent evidence indicates use of premedication teins, by either washing, volume reduction, or
does not minimize transfusion-related adverse dilution with platelet additive solution, can re-
events. In a Cochrane review" of studies on pre- duce the incidence and severity of allergicreac-
medication, the evidence from three random- tions. For anaphylactic reactions, all cellular
ized controlled trials (RCTs) involving 462 components should be washed, and plasma
recipients indicated no pretransfusion medica- lacking the cognate allergen [such as immuno-
tion regimen reduces the risk of allergicreaction globulin A (IgA)-deficientplasma] should be
or febrile nonhemolytic transfusion reaction used. Pooled solvent/detergent-treated plasma
(FNHTR).However, the conclusion is based on may also be used to mitigate the risk of an aller-
the evidence from three trials of low to moder- gic reaction. (Forfurther discussion of noninfec-
ate quality. A better-powered RCT is necessary tious adverse events, see Chapter 22.)
to evaluate the role of pretransfusion medication
in the prevention of allergic reactions and Equipment
FNHTR.
Blood Warmers
Overall, as Duran" states, "in the absence of
definitiveevidence-basedstudies, pretransfusion Infusions of cold components can cause hypo-
medication to prevent transfusion reactions thermia and cardiac complications, increasing
should not be encouraged." Yetfor patients with morbidity and mortality.I I The likelihood of clin-
a history of moderate to severe allergic reac- ically important hypothermia is increased when
tions, premedication with antihistamines (di- blood is transfused through a central venous de-
phenhydramine and/or H2 blockers) may help vice directlyinto the right atrium.
Bloodwarmers are rarely needed during rou- components. If it is not approved, the institution
tine transfusions. However, they are used when must establish a validation plan to confirm the
rapid transfusion of components is required, es- pump will not damage cellular components be-
pecially in trauma or surgery settings. Blood fore using. Most infusion devices require the use
warmers are also advantageous during transfu- of a compatible blood administration set with an
sions to neonates, where hypothermia can cause in-line filter.
serious adverse effects. Opinions vary on the
utility of blood warmers in recipients with cold Syringe Infusion Pumps
agglutintns.P'" Blood warmers are contraindi-
cated for platelet transfusions, but may be used A syringe infusion pump may be usedfor small-
for other blood components; the manufacturer's volume transfusions to neonatal or pediatric re-
suggestions should be followed. cipients. The transfusion service must have es-
AABBStandardsl{p7) states that "warming de- tablished policies for preparing blood in syringes
vices shall be equipped with a temperature- for administration. For more details, see Chapter
sensing device and a warning system to detect 24.
malfunctions and prevent hemolysis or other
damage to blood or blood components." Warm- PressureDevices
ing blood to temperatures >42 G may cause he- The use of an externally applied pneumatic
molysis." The transfusion service must collabo- compression device may achieve flow rates of
rate with departments using blood warmers to
70 to 300 mL per minute, depending on the
ensure the devices are cleared and approved by
pressure applied. The device must have a gauge
the Food and Drug Administration (FDA)for in-
to monitor the pressure, which must be applied
fusion of components. Warming devices must
evenly over the entire bag. Any pressure
be validated, and maintenance, testing of
alarms, and equipment use must be performed >300 mm Hg may cause the seams of the blood
according to the manufacturer's suggestions. As component bag to leak or rupture. When a pres-
with any medical equipment, education and sure device is used, a large-gauge cannula must
competency assessment for the user must be be employed.to prevent hemolysis.
completed and documented.' Blood compo- The application of an external pressure de-
nents must not be warmed by placing them in a vice to the blood bag to expedite the transfusion
microwave, on a heat source, or i11hot water, or of RBCunits causes IlliniIllaldam~get,o. the reci
by using devices that are not approved by the cells and is a safe practice in the majority of re-
FDAspecificallyfor blood warming. cipients." However, the use of pressure devices
has been reported to provide only a small in-
Infusion Systems crease in component flow rates. When rapid in-
fusion is desired, an increase in IV catheter or
Infusion pumps or systems are used to adminis- cannula size typically provides better results.
ter fluids, medications, blood, and blood compo- Pressure devices are contraindicated for platelet
nents through clinically accepted routes of ad- transfusions.
ministration. These devices allow for a
controlled infusion rate over a desired period; Availability of Emergency Equipment
the devices also provide an alarm system to noti-
fy clinicians of problems with the infusion. Con- The transfusionist must be prepared to obtain
sequently, the use of infusion pumps or systems and initiate emergency interventions when
rnay be preferred over simple gravity-based ad- needed. Items used to respond to a transfusion
mihistration. However, there is potential for he- reaction include the following:
molysis of the cellular components infused
through these pumps. The manufacturer of the A new bag of 0.9% sodium chloride IV solu-
pump must be consulted to determine whether tion and a new administration set to keep an
the pump is approved for the infusion of blood N line open.
e> Appropriate medications to treat a reaction, throughout the transfusion in accordance

along with a mechanism to obtain emergen- with the institutional policy.
cy medications prescribed to treat the sequel- 6. The recipient has received any ordered pro-
ae of transfusion reactions. phylactic medications.
'$ A mechanism to activate emergency resusci- 7. The necessary equipment is available and
tation measures in the event of a severe reac- functioning.
tion.
<II Ventilatory assistance and an oxygen source. Occasionally, despite the best attempts at
planning, the blood component arrives at the pa-
Intravenous Access tient's bedside but there is a Significantdelay in
Acceptable IV catheter sizes for use in transfus- the start of transfusion due to unanticipated cir-
ing cellular blood components range from 25 to cumstances. In this situation there must be a
14 gauge.16,17 A 20- to 18-gauge IV catheter is process for prompt return of the blood compo-
suitable for the general adult population and nent to the transfusion service for proper stor-
provides adequate flow rates without excessive age. Transfusionservices must ensure every hos-
discomfort to the recipient. When an infant or a pital department is aware of the requirement for
toddler receives transfusion, a 25- to 24-gauge returning components if a transfusion is de-
IV catheter may be suitable, but a constant flow layed.
rate using an infusion device must be applied.IS
(See Chapter 24.)
When using smaller-gaugecatheters, it is rec-
NEN
ommended that the transfusion rate be slowed. I N ND
The pressure or force used during the transfu-
sion is more likely than the needle gauge to
cause hemolysis of red cells.19 There must be a process to correctly identify the
In some circumstances when IV access can- intended recipient and component and attached
not be achieved, intraosseous infusions may be transfusion record at the time of the request to
warranted. issue the component. To verify the correct unit
is being issued to the correct recipient, transfu-
sion services must allow the issue of only 1 unit
Readiness to Transfuse
at a time unless it is an emergent or large-
After notification that the ordered units are volume transfusion. Upon dispensation, a final
available, in order to reduce the time the blood clerical comparison of transfusion service re-
component is outside of the controlled laborato- cords with each unit or component must be per-
ry environment, the transfusionist should re- formed and documented. Verification must in-
quest components for delivery to the recipient elude' (pp46-47):
location only after the following items have
been addressed: 1. The type of component (red cells, plasma,
platelets, cryoprecipitate,granulocytes,whole
1. The ordered component is available. blood).
2. Informed consent for transfusion has been 2. The intended recipient's two independent
completed and documented. identifiers (name, date of birth, or recipient
3. IV access is available,patent, and appropriate identification number and/or unique identi-
for transfusion. fier given at the time the crossmatch sample
4. The order is appropriate for the clinical situa- is drawn), ABOgroup, and Rh type.
tion of the recipient. 3. The donation identification number (DIN),
5. A transfusionist or an appropriate designee is donor ABOgroup, and, if required, donor Rh
available to properly monitor the recipient type.
4. The interpretation of results of crossmatch Microaggregate Filters

tests, if performed.
Microaggregate filters are not used for routine
5. Special transfusion or blood component pro- blood administration. These second-generation
cessing requirements. filters were originally developed to remove leu-
6. The component's expiration date and, if kocytes and to complement or replace the clot
applicable, time. screen used in the 1970s.20 They have since
7. The date and time of issue. been replaced by more. efficient leukocyte re-
duction filters." Microaggregate filters have a
Before issuing the unit, transfusion service screen filter depth of 20 to 40 microns and re-
personnel must visually inspect it for abnormal tain fibrin strands and clumps of dead cells. Red
appearance (significant color change, cloudi- cells, which are 8 microns in diameter, can flow
ness, clots, clumps, or loss of bag integrity). The through the filters. Microaggregate filters are
component must not be used if abnormal ap- typically used for the reinfusion of shed autolo-
pearance is noted." gous blood collected during or after surgery.
Institutions may use dedicated personnel or
automated delivery systems (eg, validated pneu- Leukocyte Reduction Filters
matic tube systems, validated transport coolers, Leukocyte reduction filters are designed to re-
automated blood delivery robots, or blood dis- duce the number of leukocytes to <5 x 106 per
pensing kiosks in remote sites) to facilitate the
RBCunit, resulting in the removal of >99.9% of
delivery or components to their final destina- the leukocytes. Leukocyte reduction decreases
tion. Provision of RBCs via automated blood the incidence of FNHTRs,risk of HLAallolmmu-
vending machines (remote, automated, computer- nization, and transmission of cytomegalovirus
controlled blood storage and dispensing refriger- (CMV) by cellular blood components.P'" (See
ators) at the point of care may help prevent Chapter 7.) These filters are provided by various
delays in transportation. The use of a remote dis- manufacturers for prestorage use shortly after
pensing solution employs an electronic issuance collection of the units or for poststorage use at
process and requires confirmation of the ab- the recipient's bedside.
sence of clots, clumps, or loss of bag integrity be- Prestorage leukocyte reduction is more effec-
fore stocking the dispensing refrigerator. In all tive than bedside leukocyte reduction, results in
methods of delivery of blood components, insti- lower levels of cytokines in storage, and can pro-
tutions must have a process in place to ensure mote ready access to an adequate inventory of
that the appropriate component is delivered to leukocyte-reduced components." In contrast,
the intended recipient. the use of bedside leukocyte reduction filters
has been associated with dramatic hypotenslon
in some individuals, often in the absence of oth-
IN I N
er symptoms. This happens more frequently
with recipients taking angiotensin-converting
infusion Sets
enzyme inhibitors. The use of components that
Components must be administered throughspe- were filtered in the blood center or transfusion
cial IV tubing with a filter designed to remove service before storage decreases the incidence of
blood clots and particles that are potentially such reactions.f Incorporating prestorage leu-
harmful to the recipient.1(P50) Standard blood ad- kocyte reduction into the blood component
ministration tubing typically has a 170- to 260- manufacturing process greatly reduces the need
micron (macroaggregate) filter, but this particu- for bedside leukocyte reduction.
lar micron size is not mandated or required. The It is important to verify the leukocyte reduc-
tubing can be primed with either 0.9% sodium tion filter used is compatible with the compo-
chloride or the component itself. The manufac- nent transfused (RBCsor platelets) and to note
turer's instructions must be reviewed for proper the maximum number of units that can be ad-
use. ministered through one filter. Filters designed
for RBCs or platelets may not be used inter- Recipient Verification at the Time of
changeably. The manufacturer's instructions Administration
must be followed for priming and administering Proper bedside identification of the recipient is
blood components through the filter. Otherwise, the final step to prevent the administration of an
leukocyte removal may be ineffective or an air incorrect blood component to a recipient. Al-
lock may develop, preventing passage of the though individuals are often concerned about
component through the filter. Leukocyte filters the possibility of exposure to infectious agents
must never be used to administer granulocytes from transfusion, equal concern must focus on
or hematopoietic progenitor cells. the inadvertent transfusion of incompatible
blood. Approximately 1 in every 15,000 to
Compatible IV Solutions 19,000 units of RBCsis transfused to the wrong
No medications or solutions other than 0.9% so- recipient each year; 1 in 76,000 to 80,000
dium chloride injection, USP, must be adminis- transfusions results in an acute hemolytic trans-
fusion reaction; and 1 in 1.8 million units of
tered with blood components through the same
transfused RBCsresults in death from an acute
tubing at the same time. Solutions containing
hemolytic transfusion reaction."
only dextrose may cause red cells to swell and To prevent the potentially fatal consequences
lyse. Lactated Ringer solution or other solutions of misidentification, specific systems have been
containing high levels of calcium may overcome developed and marketed. These include identifi-
the buffering capacity of the citrate anticoagu- cation bracelets with bar codes and/or radio-
lant in the blood preservative solution and cause frequency identification devices, biometric scan-
clotting of the component. 24 If other medica- ning, mechanical or electronic locks that prevent
tions or fluids are used, the tubing must be access to bags assigned to other unintended re-
flushed with 0.9% sodium chloride solution be- cipients, and handheld computers suitable for
fore or after transfusion. transferring blood request and administration
MBB Standards allows exceptions to the data from the recipient's bedside to the transfu-
above restrictions when 1) the drug or solution sion service information system in real time.
has been approved by the FDA for use with Each system provides a method to bring staff to-
blood administration, or 2) there is documenta- ward self-correction during the procedure.26,27
tion available to show the addition is safe and Studiesshow that rates of positiverecipient iden-
does not adversely affect the blood or compo- tificationcan be increased by such systems.How-
nent.1(p50) ever, none of these systems negates the need for
Acceptable solutions according to these crite- goodqualitymanagement, such as standard oper-
ria include ABO-compatible plasma, 5% albu- ating procedures, regular training, periodic com-
min, or plasma protein fraction. Certain solu- petencyassessment, and system monitoring.
tions are compatible with blood or blood
Verifications before the Start of
components as noted in the package inserts re-
Transfusion
viewed by the FDA, including Normosol-R pH
7.4 (Hospira), Plasma-Lyre-Ainjection pH 7.4 Identification of recipient and unit. The
(Baxter Healthcare), and Plasma-Lyte148 injec- transfusionist must verify that the recipient's
tion (Multiple Electrolytes Injection, Type 1, two independent identifiers (eg, name and
USP, Baxter Healthcare). It is important to note identification number) present on the pa-
there are several formulations of Plasma-Lyte tient's armband match the information on
that are not isotonic or that contain calcium; the unit label or attached tag. The require-
package inserts must be checked to confirm ments of the institution for recipient identifi-
their compatibilitywith blood components. cation must be satisfied.
C HAP T E R 1 8 Administration ofB/ood Components 545
~ Donation identification number. The DIN ter as little as 10 mL has been transfused. Poten-
and donor ABO/Rh type on the blood com- tially life-threatening reactions most commonly
ponent label must match the attached tag. occur within 10 to 15 minutes of the start of a
• Blood type. The recipient's ABO group (and transfusion. The recipient must be reassessed,
Rh type if required) must be compatible with and vital signs must be obtained, to evaluate the
that of the unit. Interpretation of any cross- recipient's tolerance of the transfusion."
match tests (ifperformed) must also be veri-
fied. Rates of Transfusion
Ii> Medical order and consent to transfuse. The
After the first 15 minutes of the transfusion of
transfusionist must verify the component
the ordered blood component, if no adverse
matches the provider's order and that any
events are suspected, the rate of transfusion
special processing requested in the order was must be increased to the ordered rate. Transfu-
performed. The consent must be present on
sions of a blood component must be completed
the patient's record. within 4 hours of the start of transfusion." How-
* Expiration date (and time, if applicable). ever, the recipient's size, blood volume, and he-
The transfusion of the unit must start before modynamic condition should be taken into con-
the expiration date or time has passed. sideration in determining the flow rate. (See
Table 18-1.) Special attention must be made to
The transfusion must not be initiated if any ensure the completion of the unit within the 4-
discrepancy or abnormality is found. hour window while avoiding transfusion of the
unit.too rapidly in relation to the recipient's car-
Starting the Transfusion diac and/or respiratory status. If the recipient is
The unit identifiers and compatibility result unable to tolerate completion of the entire trans-
must remain appended to the blood unit until fusion dose in the 4-hour timeframe, a request
the completion of the transfusion. Once the to the transfusion service can be made to issue
identification of the unit and the recipient is ver- an aliquot or a split component of a smaller vol-
ified, the unit is spiked using an aseptic tech- ume to allow for the transfusion dose to be in-
nique. At institutions using Joint Commission fused over two separate transfusions.
hospital accreditation, Joint Commission re- The advantages of using relatively rapid
quirements for the transfusionist (HR.01.02.01) transfusion rates (eg, 240 mLihour) include cor-
apply": "If blood transfusions and intravenous rection of deficiency as rapidly as possible as
medications are administered by staff other than well as reduced recipient and nursing time dedi-
doctors of medicine or osteopathy, the staff cated to transfusion. Disadvantagesinclude the
members must have special training for this du- potential to cause reactions (eg, volume over-
ty." load) or increase the severity of a reaction (eg,
The blood administration tubing must be FNHTRs,septic reactions, or allergic reactions).
primed with either 0.9% sodium chloride or the Many FNHTRs, as well as septic, allergic, and
blood component itself. If any solution or medi- respiratory complications, and even sornehemo-
cation other than 0.9% sodium chloride is in- lytic reactions, may not be evident within the
first 15 minutes. . .
fused before component administration, the tub-
If, at any time, an adverse reaction is suspect-
ing must be flushed with 0.9% sodium chloride
ed, the transfusion must be halted and the trans-
immediately before the blood infusion.
fusion service notified. Saline may be used to
The infusion for all routine (nonemergent)
keep the line open.
administrations of blood components must start
slowly, at approximately 2 mL per minute, for
Monitoring during the Transfusion
the first 15 minutes while the transfusionist re-
mains near the recipient. Some policies may re- The transfusionist must continue to periodically
quire "direct observation of the recipient" monitor the recipient throughout the infusion,
during this time. Severe reactions may occur af- including the IV site and flow rate. If the IVrate
1::0-
Q;)
c:
en o
Cl
c.
c: E
o
<.>
~
en
....c: en
CD ctS
~
Cl ....
o
CD ....
E ::::l
o
CD
c: :)"0
o
;;::: E-§
o Q;)
.5 00
en C'J_
c:
o
'en
:::I ~.... "2
'tn ::::l
::::l
o
c: o
e
I-
..c::
c::::J
§
.- c: E
c: 'E E
CD
c: :::Jo ~g
o E~ EC?o
c.
E
S'lo!£. ,?N
N0
o
c...:>
-c
o
o
iii
ctS
E
en
ctS
a:::
has slowed down, the transfusionist must take common transfusion reactions, including signs
one or more of the following actions: 1) verify and symptoms as well as immediate steps to be
that the N is patent and there are no signs of in- taken or interventions to anticipate.
filtration; 2) raise or elevate the unit; 3) exam- As soon as possible, the transfusion service
ine the filter for air, excessive debris, or clots; 4) must be notlftedof a suspected transfusion reac-
attempt to administer the component through tion, and institutional policy must be followed
an infusion pump; or 5) consider the addition of for returning the component bag and/or to or-
0.9% sodium chloride as a diluent if the unit is der the laboratory studies needed to evaluate
too. viscous. Frequent recipient monitoring the reaction. Documentation of the suspected
during the infusion helps alert the transfusionist transfusion reaction must be completed per in-
to a possible transfusion reaction and allows for stitution policy.
early intervention.
Vital signs must be taken within 15 minutes Completing the Transfusion
of beginning the transfusion and then according
The recipient is assessed at the completion of
to institutional policy. There is littie evidence to
the transfuston, and the followinginformation,is
support a best practice related to the frequency
documented: his or her vital signs and the date,
of vital-sign monitoring other than at baseline,
time, and volume transfused. If the transfusion
soon after the start of the transfusion, and after
was uneventful, the blood unit bag and tubing
transtusion." MBB Standards requires that the are discarded in a biohazard container. The
medical record include vital signs taken before, 0.9% saline bag must be discarded per institu-
during, and after transfusion.!(pSO) Vital signs
tion policy.
must be taken immediately if there is a suspect- Because recipients can experience transfu-
ed transfusion reaction or a change in the clini- sion reactions several hours to days after the
cal condition of the recipient. transfusion is complete, clinical staff must con-
tinue to monitor the recipient periodically after
Suspected Transfusion Reactions the end of the transfusion to detect febrile or
The transfusionistmust be knowledgeable of the pulmonary reactions potentially associated with
signs and symptoms indicative of an adverse re- the blood administration. If the recipient is not
action and be able to act quickly. (See Chapter under direct clinical supervision after a transfu-
22.) Vi~ual observation and recipient reporting sion, clinical staff must provide written instruc-
of any changes must be used to determine if a tions to the recipient and caregiver regarding
reaction has occurred, as the recipient may ex- transfusion reaction signs and symptoms. This
perience symptoms before changes occur in vi- information must also include a contact person
tal signs. If a transfusion reaction is suspected, and phone number to report signsand symptoms.
the transfusion must be stopp~d.The patency of
the N must be maintained with new N tubing
and a new bag of 0.9% saline attached near the CU EN I N F HE
IVinsertion site to prevent infusion of any resid- NSFUSi N
ual blood component to the recipient.
The component unit identification informa- The transfusion must be documented in the re-
tion must be rechecked. Prompt notification of cipient's medical record. At a minimum, MBB
the transfusion service and a licensed provider Standards requires documentation of the fol-
lowing!(pSO):
for treatment of suspected transfusion reactions
is needed. The protocol for collection of a post-
transfusion specimen and evaluation of adverse 1. Transfusionorder.
reaction must be initiated. For serious adverse 2. Recipientconsent.
events, notification of the hospital's rapid re- 3. Component name.
sponse team must be considered. Institutions 4. Donation identificationnumber.
must provide ready access to descriptions of 5. Donor ABO/RH type.
6. Date and time of transfusion. nents can lead to hypothermia, coagulopathy,

7. Vital signs before, during, and after transfu- and electrolyte imbalances. Use of a blood/fluid
sion. warming device can lessen the incidence of hy-
8. Volume transfused. pothermta."
9. Identificationof the transfusionist. Hypocalcemia has been noted with rapid
10. Transfusion-relatedadverse events, if appli- transfusions. This is usually transient and depen-
dent on the amount and rate of citrate infused.
cable.
Calcium replacement may be administered
based on the recipient's ionized serum calcium
It should be noted that although AABBStan- level and the rate of citrate admlnistration."
dards does not specify documentation of start Transfusion-associatedhyperkalemic cardiac ar-
and end times for transfusions, the Circular of rest has been reported with rapid administration
Information for the Use of Human Blood and of RBCs.It may develop with rapid RBCadmin-
Blood Components requires that transfusions istration even with a modest transfusion volume
be completed within 4 hours." Documentation such as 1 unit (in a neonate). Contributing fac-
would be necessary to demonstrate compliance tors are acidosis, hypoglycemia, hypocalcemia,
with this requlrement and is also requlred by and hypothermia at the time of cardiac arrest.3S
the College of American Pathologists (CAP). If If components are urgently needed and a de-
additional units are to be transfused, the institu- lay in transfusion could be detrimental to the
tion's policy and/or manufacturer's recommen- recipient, the transfusion service must have a
dations must be followed to determine whether process to provide components before all pre-
the same blood administration tubing may be transfusion compatibility testing is completed.
used. If there are no contraindications from the In such cases, uncrossmatched units are re-
manufacturer, institutions frequently allow addi- leased with a signed statement from the request-
tional units to be transfused with the same ing provider indicating the clinical situation re-
blood administration set within 4 hours of the quires urgent release before the completion of
start of the initial transfusion. testing. 1(pp47-48)
If components in the transfusion service in-
ventory are not immediately accessible to a trau-
UNIQUE TRANSFUSiON ma unit or operating room, a supply of group 0
SETTINGS red cells may be maintained in an appropriate
remote storage device in these areas. The trans-
See Chapter 19 for information on massive fusion service must ensure proper storage of
transfusion and Chapter 24 for transfusion in components at these satellite storage sites.
neonatal and pediatric recipients.
Out-of-Hospital Transfusion
Rapid Infusions
Transfusion of blood in a non-hospital setting re-
If components need to be administered rapidly, quires a well-planned program incorporating all
the use of rapid-infusion/warming devices, the relevant aspects of the hospital setting and
large-bore administration tubing, and large-bore emphasizing safety considerations."
IV catheters, including central venous or in- Out-of-hospitalsettings for blood transfusion
traosseous access, can decrease the infusion can include dialysis centers, medical transport
time without inducing hemolysis.F'" Some tub- vehicles, skillednursing facilities,outpatient sur-
ing sets with appropriate filters are specifically gery centers, and even recipients' homes. The
designed for rapid blood administration and may proper documentation and maintenance of re-
be used alone or with specific devices. Flow cords is part of a well-designed program. Trans-
rates as fast as 10 to 25 mLisecond (600-1500 fusionists must be competent in performing
mLiminute) have been reported with such tub- blood administration procedures, recipient mon-
ing. Rapid infusion of multiple blood compo- itoring, and recognition and reporting of sus-
pected transfusion reactions .. To optimize the Docurnentation ot prier transfusions with no.
care of these recipients, preper arrangements for history of severe reactions.
treatment of suspected transfusion .reactions ® A way to.properly dispose of medical waste.
must be made. Blood admlntstratlon outside the
hospital must be performed by personnel with
LU
substantial experience in bleed admlnistration
in this setting. Transfusion of blood components and the ere-
Transfusion in the home generally allows atton of bleed administration procedures and
close monitoring of the transfusion event be- pelicies must be recipient centric. Policies and
cause the personnel-to-recipient ratio is 1:1. The procedures must follow best practice and pre-
disadvantage is that no. trained assistant is avail- vide the transfusionist with Information to.com-
able in the event of a severe adverse reaction, Is- petently perform transfusions and recognize and
sues to. consider when preparing fer a transfu- report suspecte~ transfusion reactions. Close
sion in the home include availability of the monitoring and early Intervention when transfu-
following": sion reactions. occur can make a critical differ-
ence in recipient outcomes. Audits or the bleed
<8 A competent adult in the home to assist in administration process to identify areas for im-
recipient identificatien and to.summonmedl- provement, instances of nonconformance, and
cal assistance if needed. analysis of their causes are needed fer optimal
A mechanism to. obtaln immediate provider transfusion safety. Periodical auditing of all seg-
consultation. ments of the bleed administration process is
A telephone to. contact emergency person- strongly suggested for continuous precess im-
nel, and easy access fer emergency vehicles. provement,
9. At the recipient's bedside, verification of recipient and component identification must be per-
formed. The following items must be verified: 1) recipient's two independent identifiers and
ABO/Rh type, 2) DIN and donor ABO group, and if required, Rh type, 3) interpretation of
crossmatch tests, if performed, 4) special transfusion requirements are met, if applicable, and
5) expiration date/time of the component. .
Components must be administered through the appropriate infusion sets and filters. Only com-
patible IV solutions (usually 0.9% sodium chloride injection, USP) may be administered
through the same tubing unless the tubing has been flushed with 0.9% sodium chloride, USP,
immediately before and after the transfusion.
11. The infusion must start slowly at approximately 2 mL per minute for the first 15 minutes.
12. During this time, the transfusionist must remain near the recipient. If no sign of reaction ap-
pears, the infusion rate can be increased. The transfusionist monitors the recipient throughout
the infusion and stops the infusion in the event of an adverse reaction.
13. Infusions must be completed within 4 hours of the start of transfusion. After completion, the
transfusionist takes the recipient's vital signs. If the recipient will not be under direct clinical
supervision after the transfusion, the recipient and caregiver must receive instructions regard-
ing signs and symptoms to report and to whom to report these reactions.
14. The following information, at a minimum, regarding the transfusion must be documented in
the recipient's medical record: 1) the transfusion order, 2) recipient consent for transfusion,
3) name of component, 4) DIN and donor ABO group, and if required, Rh type, 5) date and
time of infusion, 6) vital signs before, during, and after transfusion, 7) volume transfused, 8)
identity of the transfusionist, and 9) any adverse reaction(s).
REFERENCES
1. Gammon R. Standards for blood banks and 7. Duran]. Effectsofleukoreduction and premedi-
transfusion services. 32nd ed. Bethesda, MD: cation with acetaminophen. I Pediatr Oncol
AABB,2020. Nurs 2014;31:223-9.
2. Stowell CP, Sazama K, eds. Informed consent in 8. Ezidiegwu CN, Lauenstein KI, RosalesLG,et al.
blood transfusion and cellular therapies: Pa- Febrile nonhemolytic transfusion reactions.
tients, donors, and research subjects. Bethesda, Management by premedication and cost impli-
MD: AABBPress, 2007. cations in adult patients. Arch Pathol Lab Med
3. Klein H, Anstee D. Mollison's blood transfusion 2004; 128:991-5.
in clinical medicine. 12th ed. Oxford: Wiley- 9. Patterson BI, Freedman I, Blanchette V, et al.
Blackwell,2014. Effect of premedication guidelines and leuko-
4. AABB,American Red Cross, America's Blood reduction on the rate of febrile nonhaemolytic
Centers, Armed Services BloodProgram. Circu- platelet transfusion reactions. Transfus Med
lar of information for the use of human blood 2000; 10:199-206.
and blood components. Bethesda, MD: AABB, 10. Goss IE, Chambers CE, Heupler FA, et al. Sys-
2017. temic anaphylactoid reactions to iodinated con-
5. 2020 National patient safety goals. Oakbrook trast media during cardiac catheterization proce-
Terrace, IL:The Ioint Commission,2019. [Avail- dures: Guidelines for prevention, diagnosis,and
able at http://www.jointcommission.orglstan treatment. Cath Cardiovasc Diagn 1995;34:99-
dards_information/npsgs.aspx (accessed Octo- 104.
ber 24, 2019).J 11. Boyan CP, Howland WS. Cardiac arrest and
6. Marti-Carvajal AI, Sola I, Gonzalez LE, et al. temperature of bank blood. lAMA 1963; 183:
Pharmacological interventions for the preven- 58-60.
tion of allergicand febrile non-haemolytic trans- 12. Donham lA, Denning V. Cold agglutinin syn-
fusion reactions. Cochrane Database Syst Rev drome: Nursingmanagement. Heart Lung 1985;
2010;(6):CD007539. 14:59-67.
13. Iserson KY, Huestis DW. Blood warming: Cur- 24. Dickson DN, Gregory MA. Compatibility of
rent applications and techniques. Transfusion blood with solutions containing calcium. S Afr
1991;31:558-71. MedJ 1980;57:785-7.
14. Hirsch J, Menzebach A, Welters ID, et al. Indi- 25. Vamvakas EC, Blajchman MA. Transfuslon-relat-
cators of erythrocyte damage after microwave ed mortality: The ongoing risks of allogeneic
warming of packed red blood cells. Clin Chern blood transfusion and the available strategies for
2003;49:792-9. their prevention. Blood 2009; 113:3406-17.
15. Frelich R, EllisMH. The effect of external pres- 26. Pagliaro P, Rebulla P. Transfusion recipient iden-
tification. Vox Sang 2006;91 :97-1 01.
sure, catheter gauge, and storage time on he-
27. Koshy R. Navigating the information technology
molysis in RBCtransfusion. Transfusion 2001;
highway: Computer solutions to reduce errors
41:799-802. and enhance patient safety. Transfusion 2005;
16. Stupnyckyj C, Smolarek S, Reeves C, et al. 45(SuppI4): 189S-205S.
Changing blood transfusion policy and practice. 28. Comprehensive accreditation manual for hospi-
AmJ Nurs 2014;114:50-9. tals. Oakbrook Terrace, IL: The Joint Commis-
17. Makic MB, Martin SA, Burns S, et al. Putting sion, 2020.
evidence into nursing practice: Four traditional 29. Bradbury M, CruickshankJP. Blood transfusion:
practices not supported by the evidence. Crit Crucial steps in maintaining safe practice. Br J
Care Nurse 2013;33:28-42. Nurs 2000;9: 134-8.
18. BarcelonaSL,VilichF, Cote CJ. A comparison of 30. Oldham J, Sinclair L, Hendry C. Right patient,
flow rates and warming capabilitiesof the Level right blood, right care: Safe transfusion practice.
1 and Rapid Infusion System with various-size Br J Nurs 2009; 18:312, 314, 316-20.
intravenous catheters. Anesth Analg 2003;97: 31. Davis DT, Iohannigman JA, Pritts TA. New strat-
358-63. egies for massive transfusion in the bleeding
19. Miller MA, Schlueter AJ. Transfusionsvia hand- trauma patient. J Trauma Nurs 2012; 19:69-75.
held syringes and small-gauge needles as risk 32. ACS TOlP massive transfusion in trauma guide-
factors for hyperkalemia. Transfusion 2004;44: lines. Chicago, IL: American College of Sur-
373-81. geons, 2014.
20. Wortham ST,Ortolano GA,Wenz B.A brief his- 33. Shaz B, HillyerC. Massive transfusion. In: Shaz
tory ofblood filtration:Clot screens, microaggre- B, HillyerC, RoshalM, Abrams C, eds. Transfu-
sion medicine and hemostasis. 2nd ed. London:
gate removal, and leukocyte reduction. Transfus
ElsevierScience, 2013.
Med Rev 2003; 17:216-22. 34. Hrovat TM, Passwater M, Palmer RN, for the
21. Lane TA. Leukocyte reduction of cellular blood Scientific Section Coordinating Committee.
components: Effectiveness, benefits, quality Guidelines for the use of blood warming devic-
control, and costs. Arch Pathol Lab Med 1994; es. Bethesda,MD: AABB,2002.
118:392-404. 35. Hayter MA, Pavenski K, BakerJ. Massive trans-
22. Cushing M, Bandarenko N, eds. Blood transfu- fusion in the trauma patient: Continuing profes-
sion therapy: A handbook. 13th ed. Bethesda, sional development. Can J Anaesth 2012:59:
MD: AABB,2020 (in press). 1130-45.
23. loon KC, Jacobson ED, Woodcock J. Hypoten- 36. BensonK. Home is where the heart is: Do blood
sion and bedside leukocyte reduction filters. Int transfusions belong there too? Transfus Med
J Trauma Nurs 1999;5:121-2. Rev2006;20:218-29.
Hemotl1erapy Deci$io ..ns and
Their Outcomes
Nadine Shehata, MD, FRCB and Yunchuan Delores Mo, MD
S WITH ALL MEDICAL INTERVEN- consideration of RBCtransfusion include hemo-

tions, the risks and benefits of blood dynamic instability, chest pain of cardiac origin,
transfusion must be weighed carefully. dyspnea, and tachycardia at rest. In nonbleeding
This chapter provides an overview of the scien- patients, the hemoglobin concentration is used
tific literature supporting the use of transfusion to help guide RBCtransfusion decisions because
therapy in adult patients. 98% of blood oxygen is hemoglobin bound, the
hemoglobin is easy to measure, and no better
physiologicmeasurements to support RBCtrans-
L fusion are currently available. As discussed be-
T low, current RBC transfusion thresholds (ie,
hemoglobin concentrations used for RBCtrans-
Red Blood Cells (RBCs) are transfused to infusion) are lower than those used previously.
crease oxygen-carryingcapacity in patients with
anemia in whom physiologic compensatory Liberal vs Restrictive Transfusion
mechanisms are inadequate to maintain normal Strategies
tissue oxygenation. There are myriad causes of
anemia; one classification scheme is shown in The first high-quality study investigating the
Table 19-1. In patients with chronic, stable ane- clinicaluse of RBCtransfusionswas the Transfu-
mia, RBCtransfusion is often unnecessary. In a sion Requirements in Critical Care (TRICC) tri-
patient with well-compensated anemia from al.' In the TRICC trial, 838 hemodynamically
iron deficiency, for example, replacing iron is stable, critically ill patients with a hemoglobin
the appropriate maneuver to correct the ane- level <9 gI dL were randomly assigned to re-
mia. Conversely, RBC transfusion may be life- ceive RBCtransfusionfor a hemoglobin<10 g/dL
saving in individuals with anemia where physio- (liberal group) or <7 gI dL (restrictive group).
logic compensatory mechanisms are inadequate The primary endpoint, 30-day all-cause mortali-
to maintain tissue oxygenation, such as in indi- ty, did not significantlydifferbetween the study
viduals with trauma-induced hemorrhage. Signs groups. Significantly better survival was ob-
and symptoms of anemia that should prompt served in the restrictive group in younger
Nadine Shehata, MD, FRCP,Associate Professor, Departments of Medicine and Laboratory Medicine and Patho-
biology, University of Toronto, Hematologist, Division of Hematology, and Director, Transfusion Medicine Labora-
tory, Mount Sinai Hospitai, Toronto, Ontario, Canada; and Yunchuan Delores Mo, MD, Associate Medical
Director of Transfusion Medicine, Division of Laboratory Medicine, Children's National Medical Center, and
Assistant Professor, Department of Pathology, George Washington University, Washington, District of Columbia
N. Shehata has grant funding from the Canadian Institute for Health Research and Canadian Blood Services and
chalrs the International Collaboration for Transfusion Medicine. Y. Mo has disclosed no conflicts of interest.
553
TABLE 19·1. Classification of Anemia
Increased Red Cell Destruction (hemolysis)
Extrinsic to Red Cells Microcytic
Immune Iron deficiency

Alloantibody-mediated hemolytic anemia Thalassemia
Warm autoimmune hemolytic anemia Lead poisoning
Cold agglutinin syndrome Anemia of chronic disease
Paroxysmal cold hemoglobinuria Sideroblastic anemia
Paroxysmal nocturnal hemoglobinuria
Drug-related hemolytic anemia Normocytic
Nonimmune Myelophthisic anemia

Mechanical cause (eg, mechanical heart valves) Renal/low erythropoietin
Microangiopathic anemia Anemia of chronic disease
Marrow hypo/dys/aplasia
Intrinsic to Red Cells
Hemoglobin disorders Macrocytic
Membrane defects Megaloblastic

Enzymedefects 812 deficiency
Folatedeficiency
Medication
Nonmegaloblastic
Marrow hypo/dys/aplasia
Alcoholism
Liver disease
Hypothyroidism
patients «55 years old) and in less acutely ill pa- randomized controlled trial (RCT) to examine
tients [Acute Physiology and Chronic Health the clinicalconsequences of adhering to a liberal
Evaluation (APACHE)II score <20]. The 2011 vs restrictive RBC transfusion strategy in adult
trial known as FOCUS2 was the second large patients. In this study, 2016 patients age 50
C HAP T E R 1 9 Hemotherapy Decisionsand TheirOutcomes 555
years or older having hip-fracture surgery with a and the only factor driving the decision to trans-
history of (or risk factors for) cardiovascular fuse RBCs.isthe patient's hemoglobin, then a re-
disease were randomly assigned to receive post- strictive approach should be followed. Second,
operative RBCtransfusion for hemoglobin levels the RCTs conducted to date have almost exclu-
<10 gI dL (liberal group) vs 8 g/dL (restrictive sively included hemodynamically stable, hospi-
group). FOCUSwas designed as a superiority tri- talized, adult patients. Hemoglobin levels may
al; the aim was to determine whether transfus- be of limited utility in patients who are actively
ing RBCsmore liberallywas associated with bet- bleeding during the perioperative period. Also,it
ter functional outcomes following hip-fracture is often appropriate for providers to provide
repair. There was no difference in the primary transfusion to ambulatory outpatients more lib-
endpoint of death or the inability to walk across erally, for logisticalreasons (eg, fewer clinic vis-
a room unassisted at 60 days after randomiza- its). Subjective quality-of-life (OOL) measures
may vary based on a patient's hemoglobin, al-
tion. Smaller trials of liberal vs restrictive post-
though a consistent relationship between hemo-
operative RBCtransfusion in orthopedic surgical
globin and functional activityIOOL has been dif-
patients similarly failed to show a benefit of lib-
ficult to demonstrate."
eral transfusion.3,4
Acute coronary syndromes represent an indi-
RCTsof various sizes comparing liberal vs re- cation distinct from those mentioned. Currently,
strictive RBC transfusion strategies have now the optimal transfusion strategy for patients with
been performed in several populations of hospi- acute myocardial infarction (MI) or unstable an-
talized adult and pediatric patients,"? including gina remains unclear and is the subject of ongo-
cardiac surgery,':" septic shock," acute upper ing study." In the TRICC study, patients with
gastrointestinal bleeding.lv" surgical oncolo- acute coronary syndromes were the only sub-
gy,16postpartum hemorrhage," and traumatic group in which survival was poorer among pa-
brain injury." More RCTs are in various stages tients assigned to the restrictive transfusion
of development. With a few exceptions," these strategy. However, the survival advantage seen
studies fail to demonstrate any clinical benefits in the liberal transfusion group was not statisti-
of a liberal transfusion strategy. A 2016 meta- cally significant.1,23 In cardiac surgery, restrictive
analysis19 evaluated 31 trials comparing liberal transfusion strategies were not found to be asso-
vs restrictive transfusion strategies. A total of ciated with increased adverse events. In the
12,587 patients in various clinical settings were 2015 Transfusion Indication Threshold Reduc-
included. Overall, using a restrictive RBCtrans- tion (TITRe2)trial,'? 2007 adult patients having
fusion threshold (typically hemogloblin of 7.0- elective cardiac surgery were randomly assigned
8.0 g/dL) reduced the proportion of patients ex- postoperatively to a liberal (hemoglobin <9 gI
posed to blood components by 43%, without dL) vs restrictive «7.5 gldL) RBC transfusion
causing either harm or benefit as compared with strategy. There was no significant difference in
a liberal transfusion strategy. On this basis, clini- the primary outcome of serious infection or isch-
cal practice guidelines, including a 2016 AABB emic events (eg, stroke or MI). A secondary
guideline,20 recommended that a restrictive analysis, however, revealed higher 90-day all-
RBCstrategy should be used for hospitalized in- cause mortality among patients in the restrictive
patients. A few points merit emphasis. First, group [4.2% vs 2.6%; hazard ratio (HR), 1.64
clinical practice guidelines are not standards, (1.00-2.67)]. Nonetheless, the Transfusion Re-
nor can they substitute for clinical judgment. quirements in Cardiac Surgery (TRICSIII) trial
The RCTs in this area have tended to simplify did not find a statistically Significantdifference
the decision to transfuse RBCsby basing it on a in the composite primary outcome of mortality
single parameter, the patient's hemoglobin level. from any cause, MI, stroke, and new renal fail-
For individual patients, clinical signs and symp- ure requiring dialysisat discharge or at 28 days
toms, comorbidities, and other factors should be following surgery or after 6 months. TRICSIII
integrated into the transfusion decision. That randomly assigned 5243 patients to a restrictive
said, if an individual patient is clinically stable, RBCtransfusion strategy (hemoglobin <7.5 gldL
from the induction of anesthesia) or a liberal has been examined in several RCTs, including
strategy [<9.5 g/dL in the operating room or ARIPI26 (neonates), ABLE27and TRANSFUSE28
intensive care unit (ICU) or <8.5 g/dL in the (patients in ICU), RECESS29(cardiac surgery pa-
non-lCU ward]. Mortality was 11.4% in the tients), TOTAL30(children with severe anemia,
restrictive group vs 12.5% in the liberal group mainly from malaria), and INFORM31(hospital-
(p <0.001 for noninferiority) at discharge or at ized adult patients). No differences in clinical
28 days after surgery, and 17.4% vs 17.1% re- outcomes were seen in any of these trials. At
spectively (p = 0.006 for noninferiority) after 6 this time, no clinical practice changes based on
months. 11,12A subgroup analysis suggested RBCstorage duration are indicated.
younger patients may benefit from more liberal
thresholds and older patients may benefit from Donor Characteristics and Transfusion
more restrictive thresholds." Future trials need Recipient Outcomes
to confirm or refute these findings. For other pa-
tients such as those with acute coronary syn- The effect of blood donor characteristics on re-
drome, severe thrombocytopenia, or chronic cipient outcomes was not explored for RBCsun-
transfusion-dependent anemia, however, the til recently.32.34In one study, increased risk of
AABB guideline noted that there was insuffi- death was associated with receipt of blood from
cient evidence to recommend a restrictive RBC younger donors (17 to <20 years old) compared
transfusion strategy." Patients with solid tumors to donors 40 to <50 years old [adjusted HR,
with septic shock may also not benefit from re- 1.08; 95% confidence interval (CI), 1.06-
strictive strategies. In 300 such patients ran- 1.10].34Younger donors were found to have a
domly assigned to hemoglobin thresholds of similar effect on recipient death in another data-
<7 g/dL or <9 g/dL only during their ICU stay, base study, but the effect of age was no longer
a trend toward an increased mortality with a re- significant when other confounding factors
strictive strategy (56% vs 45%; p = 0.08) was were considered.P Male recipients of RBCs
observed." from female donors had a higher risk of mortali-
ty,32,34,35
but this risk has not been consistently
RBC Storage Duration described." Because other factors could poten-
tially explain the association, aside from preg-
As described above, clinical trials have typically nancy, and the physiologicmechanisms are not
failed to demonstrate a benefit of RBC transfu- understood, RBCsare not selected according to
sion for most patients with moderate anemia. donor gender or age.
Likely, this reflects the ability of physiologic
compensatory mechanisms to ensure adequate Emergency Transfusion of RBCs
tissue oxygenation at hemoglobin levels in the
range where transfusion is often considered. An RBCtransfusions are typicallymatched for ABO
alternate hypothesis potentially explaining the (Table 19-2) as well as RhD blood group anti-
apparent lack of benefit of RBC transfusion re- gens. In bleeding emergencies, there may be in-
lates to the RBC "storage lesion." In the United sufficient time to complete standard pretransfu-
States, RBCunits may be refrigerated for up to sion testing. Uncrossmatched group 0 RBC
42 days, and various biochemical and morpho- units are used in situations where RBCsmust be
logic changes are known to occur. For example, transfused immediately, before any patient test-
extracellular potassium increases; 2,3-diphos- ing is completed. Group 0, RhD-negativeunits
phoglycerate (2,3-DPG), a key reguiator of oxy- are the component of choice for femaies of
gen offloading, declines; and free hemoglobin childbearing potential. Group 0, RhD-positive
and free iron increase. Observational studies units are used for men and for postmenopausal
suggested that RBCsstored for longer durations women and in some centers when the invento-
might be associated with adverse clinical out- ry cannot support RhD negative units for fe-
comes." The impact of RBCstorage duration on males. Approximately 4% of recipients are ex-
clinical outcomes in various patient populations pected to have one or more non-ABO red cell
C HAP T E R 1 9 Hemotherapy Decisions and Their Outcomes 557
TABLE 19·2. ABO Matching jor is characterized by severe anemia secondary

to ineffective erythropoiesis, and extramedullary
hematopoiesis. Individuals with thalassemia ma-
jor often begin a regular RBC transfusion pro-
gram in childhood if there is poor growth or evi-
dence of extramedullary hematopoiesis resulting
a a A,B,O, AB in bony abnormalities, and/or if the hemoglobin
level is <7 to 9 gldL.37,38 RBC transfusion is
A A,O A,AB used to treatanemia and reduce the risk ofmor-
bidity from extramedullary hematopoiesis. RBCs
B B,O B,AB are provided every 2 to 4 weeks to maintain a
pretransfusion hemoglobin of 9 to 10g/dl.," Al-
AB A,B,O,AB AB loimmunization, which is reported to occur in
RBC= Cell. 20% to 30% of patients with thalassemia, can be
reduced by selecting RBCsmatched for Cc, Ee,
and K antigens, in addition to the usual match-
alloantibodies; nonetheless, in practice, clinical- ing for ABO and RhD in patients who do not
ly Significant hemolytic reactions to uncross- have alloantibodies."
The sickle cell syndromes include hemoglo-
matched type 0 units are rare (0.06%; 95% GI,
bin SS, hemoglobin SC, hemoglobin Sl30and
0.01-0.21%).36 Occasionally, for bleeding emer-
SW. Hemoglobin S results from a single amino
gencies, patients with known red cell alloanti-
acid substitution (valinefor glutamic acid) in po-
bodies may need RBC transfusion before
sition 6 of theB globin protein. Hemoglobin S
antigen-negative units can be identified and
polymerizes in relatively deoxygenated regions
crossmatched. Close communication between
of the circulation, causing abnormal red .cell
the primary service (eg, the emergency room or
morphology, subsequent occlusion of the micro-
operating room staff)and a transfusion medicine
vasculature, and acute and chronic organ dys-
physician is important in such cases. Clinicians
function. Sickled red cells cause microvascula-
may have concerns about transfusing units that
ture occlusion not onlybecause of their rigidity
are not proven to be "fully compatible." Howev-
but also because theytend to adhere to other
er, most non-ABOantibodies will not cause im-
blood cells and the endotheliurn.P'" RBCtrans-
mediate, intravascular hemolysis, as can occur
fusions. in patients with sickle cell disease
with a major-ABO-mismatched transfusion.
decrease .the incidence of acute and chronic
Rather, most non-ABO red cell alloantibodies
complications by reducing the proportion Of
will cause delayed, extravascular hemolysis.
circulating sickle cells. However, allogeneic
ThUS, transfusing units that may be, or are
RBCs are also associated with risk, and RBC
known to be, incompatible is still preferable to
transfusion requires balancing risks and benefits.
exsanguination or life-threatening severe ane-
In patients with sickle cell disease, the preva-
mia. A brief summary of approaches to massive
lence .ofalloimmunization remains approximate-
transfusion is provided toward the end of this
ly 20%.42,43 Alloimmunization rates are high in
chapter.
sickle cell disease partially because of antigen
disparity between donors and sickle cell disease
RBC Transfusion for Thalassemia and
patients and because of variant Rh alleles." Ad-
Sickle Cell Disease ditionally, the inflammatory response that oc-
Two hemoglobin disorders, thalassemia and curs with vase-occlusive crises may predispose
sicklecell disease,are among the most commonly to allotmmunlzatton." In addition to alloimmu-
inherited syndromes. The thalassemiasyndromes nization, the risk of hemolytic transfusion reac-
involve the reduced production of exor 13 globin tions secondary to alloantibodies and the risk of
as a result of gene mutations. l3-thalassemiarna- iron overload and hyperhemolysis secondary to
RBC transfusion also need to be balanced RhD.52,53 Nonetheless, alloimmunization may

against the benefits of transfusion. still occur in these patients despite phenotypic
Hyperhemolysisrefers to the development of matching, due to genetic variants and heteroge-
severe anemia where the hemoglobin level fol- neous epitope expression for any given Rh anti-
lowing transfusion is lower than that before gen. In one study, 38% of alloantibodies oc-
transfusion. Hyperhemolysis may be acute or curred in recipients who phenotypically
delayed. It may be associated with a new alloan- expressed the corresponding Rh antigen.44,54,55
tibody or a previous antibody that was not de- Genotyping for red cell antigens has additional
tected with antibody screening, or it may not be costs; however, the cost of genotyping needs to
associated with an alloantibody. The transfused be balanced against the need to avoid alloimmu-
cells as well as the patient's own cells are hemo- nization in those at high risk who require fre-
lyzed, resulting in a reduction of hemoglobin to quent transfusions. Forpatients who have devel-
levels below the pretransfusion hemoglobin and oped an alloantibody, extended-matched RBCs
characteristic reticulocytopenia.Subsequent RBC (ie, including antigens of the FYand JK systems,
transfusion is also likely to result in hyperhe- and S) are also often used."
molysis.46,47 Transfusion avoidance, intravenous RBCs can be administered to patients with
immune globulin (MG), corticosteroids, and SCD as a simple transfusion, by manual ex-
erythropoiesis-stimulatingagents for anemia and change, or by automated exchange. Automated
reticulocytopenia have been used to treat hyper- exchange transfusion can readily deliver more
hemolysis.48,49 Other interventions that have volume, thereby significantlyreducing hemoglo-
been described include the use of rituximab to bin S levels and reducing the risk of iron over-
prevent subsequent delayed hemolytic transfu- load. RBCs are administered acutely or chron-
sion reactions in the presence of alloantibodies ically as prophylaxis or for various indications,
and potentially eculizumab for the treatment of such as pulmonary hypertension." Clear indica-
hyperhernolysis." tions for the use of RBCs are provided by RCT
Hemoglobin substitutes, or hemoglobin- evidence and, in the absence of RCTs,from clin-
based oxygen carriers (HBOCs),have also been ical guidelines. Table 19-3 summarizes RBC
described in the treatment of sickle cell patients transfusion recommendations in sickle cell dis-
with contraindications to RBC transfusions, in- ease from a recent National Heart, Lung, and
cluding those with rare blood types or extensive Blood Institute (NHLBI)guidellne." The 1998
alloimmunization resulting in widespread do- Stroke Prevention Trial in Sickle Cell Anemia
nor incompatibility, as well as in individuals (STOP trial) showed that chronic RBC transfu-
who refuse blood because of religiousbeliefs.Al- sions significantly reduced the incidence of
though the safety profile of early HBOCs result- stroke in sickle cell patients determined to be at
ed in premature withdrawal of select agents, a high risk based on transcranial Doppler (TCD)
number of second-generation products may be ultrasonography (middle cerebral artery or inter-
better tolerated." Case reports" have demon- nal carotid artery flow velocity of 200 cm/ sec or
strated clinical benefits in recipients, leading to higher]." The subsequent STOP2 trial showed
growing interest in expanding the applications that discontinuing chronic transfusion in this pa-
of HBOCs, especially in the sickle cell popula- tient population results in a reversion to baseline
tion. Currently, the availabilityof such products risk of abnormal flow velocities and stroke." In
in the United States is confined to pharmaceuti- the recent TCD With Transfusions Changing to
cal clinical trials or expanded access (compas- Hydroxyurea (TWITCH) trial, children with
sionate use) granted by the Food and Drug Ad- sickle cell disease and abnormal TCD velocities
ministration (FDA). were randomly assigned to monthly transfusion
To reduce the risk of alloimmunization, pa- or hydroxycarbamide (hydroxyurea) for 1 year.
tients with sickle cell disease, similar to patients Hydroxycarbamide was found to be noninferior
with thalassemia, often receive selected RBC to chronic transfusion for the primary outcome
units (ie, units matched for Cc, Ee, K) in addi- of stroke but was associated with an increased
tion to the usual matching for ABO and risk of vaso-occlusive crises.58 RBCs are not
TABLE19-3,·· The Use of RBC Transfusion for Sickle Cell Disease Oomplications"
Symptomatic severe acute chest syndrome Exchange(strong)

(defined by an oxygen saturation <90% despite
supplemental oxygen)
Acute splenic sequestration and severeanemia Simple (strong)
Acute stroke in children and adults: Initiate a Simple or exchange (strong)

program of monthly transfusions
Hepatic sequestration Simple or exchange (moderate)
Intrahepatic cholestasis Exchangeor simple (consensus)
Multisystem organ failure Exchangeor simple (consensus)
Aplastic crisis Simple (consensus)
Symptomatic anemia Simple (consensus)
Child with transcranial Doppler reading Exchangeor simple (strong)

>200 ern/sec
Adults or children with previous clinically overt Exchangeor simple (moderate)

stroke
* Adaptedfrom Yawn et
generally indicated in an uncomplicated painful IgM autoantibodies; 8%), with the remainder
vaso-occluslve crisis or asymptomatic anemia, producing negative results in the direct antiglob-
Guidance from sickle cell experts is suggested ulin test." The mainstay of therapy for autoim-
for patients with sickle cell diseaserequiring sur- mune hemolytic anemia is immunosuppression,
gery with general anesthesia, as these individu- although RBCtransfusions playa key supportive
als may require simple or exchange transfu- role. Combination therapies and avoidance of
sion." cold is often beneficial for patients with cold au-
toimmune hemolytic anemia. Red cell autoanti-
RBe Transfusion in Autoimmune bodies are often broadly reactive; thus, finding
Hemolytic Anemia compatible RBCs for patients with WAIHA or
Autoimmune hemolytic anemia can be catego- mixed disease may be problematic." Autoanti-
rized as warm autoimmune hemolytic anemia bodies demonstrating in-vitro panreactivity may
(WAIHA) secondary to an immunoglobulin G also mask the presence of one or more clinically
(IgG)autoantibody (60%),cold autoimmune he- significant alloantibodies. AIloantibodies have
molytic anemia secondary to an IgM autoanti- been reported to occur in 20% to 40% of patients
body (30%), or mixed disease (both IgG and with warm autoantibodtes.sl/" For patients who
have not received transfusion in the past 120 RBCTransfusion for Patients Receiving
days, autologous adsorption is the preferred Anti-CD38 and Anti-CD47 Monoclonal
method of removing the autoantibody to permit Antibodies
alloantibody identification. For recent transfu· CD38 is a transmembrane glycoprotein ex-
sion recipients, allogeneic (heterologous) ad- pressed at low levels on lymphoid and myeloid
sorptions may be performed to identify underly- cells, red cells, and some nonhematopoietic tis-
ing alloantibodies. Selecting phenotypically or sues. In patients with multiple myeloma, CD38
genotypically matched RBCsis an option to at· is overexpressed on neoplastic plasma cells/"
tempt to reduce the risks of alloimmunization Daratumumab and other related human mono-
and the need for additional adsorption proce- clonal antibodies are directed against specific
dures in these patients.63,64In some WAIHAcas- CD38 epitopes, and are offered to some patients
with relapsed multiple myeloma and non-
es, the autoantibody will demonstrate relative
Hodgkin lymphoma. Anti-CD38 monoclonal an-
antigenic specificity. For example, the autoanti- tibodies interfere with blood bank compatibility
body may appear to react more strongly in vitro testing due to binding to CD38 expressed on
with Rhli-posltlve units compared to RhD- the surface of reagent red cells. This may result
negative units. In such cases, there may be some in variable, nonspecific reactivity (including
benefit in providing units that are negative for panagglutination) with all serologlc testing
the "mimicking" autoantibody specificityto pro- methods, requiring the use of antihuman globu-
long survival of transfused red cells." However, lin (AHG; eg, antibody screen, AHG cross-
it is more important to avoid exposure to anti- match). The direct antiglobuiin test is typically
negative but may demonstrate reactivity with
gens to which the patient has preformed under-
IgG alone, and ABO/Rh typing is unaffected un-
lying alloantibodies than to provide units less AHG reagents are used for the latter. Signifi-
matched for the relative specificity of the auto- cant hemolysis does not typically occur in
antibody. patients, as daratumamab has been demonstrat-
In many WAIHA cases and in those with ed to result in the loss of CD38 expression from
mixed disease, fully crossmatch-compatibleRBC red cells."
units will never be availableif the patient's auto- CD47-a cellular protein expressed on all
antibody reacts in vitro with all tested red cells. cells, including red cells and platelets, with its li-
However, hemolysis in some cases progresses gand signal regulatory protein ex (SIRPa)-in-
extremely rapidly, and RBCtransfusions should hibits phagocytosts.w" CD47 is overexpressed
in hematologic malignancies and some solid tu-
not be withheld from patients with potentially
mors." Monoclonal antibodies targeting CD47
life-threatening anemia. Clinicians should be re- that are intended to enhance phagocytosis are
assured that even if RBCsare incompatible in vi- currently in clinical trials." Use of one of these
tro, the transfused units may not be destroyed. antibodies, HuSF9-G4, interfered with antibody
Sufficientvolumes of red cells should be trans- screening, reverse typing, and crossmatching of
fused to relieve signs and symptoms of anemia red cells." The direct antiglobulin test is nega-
(eg, air hunger, tachycardia at rest, chest pain). tive, but eluates may be reactive depending on
Such transfusions, which are considered lifesav- the reagent."
ing, should not be avoided, particularly for pa- Options for safelyproviding RBCsfor patients
receiving daratumumab include performing an-
tients who have not received transfusion or
tibody detection testing with reagent cells pre-
been pregnant and thus are highly unlikely to
treated with dithiothreitol (DTTf2 or selection
have alloantibodies. Frequent monitoring of the of phenotypically or genotypically matched
transfusion recipient is warranted, and close RBCs_73,74 DTT is a reducing agent that causes
communication between the transfusion ser- CD38 but not CD47 denaturation on the cell
vice and primary clinicalservice is critical." surface by destruction of disulfidebonds, result-
C HAP T E R 1 9 Hemotherapy Decisionsand Their Outcomes 561
ing in elimination of daratumumab binding and ed in Table 19-4. Because of the tremendous ad-
enabling detection of underlying red cell alloan- vances in the care of patients with cancer since
tibodies (with the exception of antigens also the 1960s, severe bleeding is now extremely ra-
sensitive to DTT such as KELantigens). If RBC re. Consequently, two RCTs challenged the ne-
transfusion is anticipated and pretreatment of cessity of providing prophylactic platelet transfu-
reagent cells with DTT is unavailable, or for sions at al1.85·87In a study by Wandt and
CD47 antibodies, a preliminary type and screen colleagues," 391 patients receiving chemother-
and serologicphenotype should be obtained be- apy for acut~ myelogenous leukemia (AML) or
fore the patient receives the first dose. The ex- undergoing autologous hematopoietic stem cell
tended phenotype can be obtained using molec- transplantation (HSCT) were randomly as-
ular techniques. Close communication between signed to receive or not receive prophylactic
treating physicians and the transfusion service is platelet transfusions for a morning platelet count
essential to ensure safe transfusion practices for at or below 10,000/1lL. Patients in the no-
these patients. prophylaxis arm received platelets only if bleed-
ing occurred. WHO Grade 2 or higher bleeding
was observed among 42% of patients in the no-
PL EL T N FUSI N prophylaxis arm vs 19% of patients receiving
prophylactic platelet transfusions (p <0.0001).
Prophylactic Platelet Transfusions for The risk of bleeding was much higher among
Therapy-lnduced·Hypoproliferative patients receiving chemotherapy for AML com-
Thrombocytopenia pared to the autologous HSCT patients: 27 out
of 28 (96%) Grade 3 or Grade 4 bleeding epi-
Most platelet transfusions are administered to sodes occurred among patients receiving che-
nonbleeding patients with hypoproliferative motherapy for AML. In the Trial of Prophylactic
thrombocytopenia resulting from chemothera- Platelets (TOPPS)study,85,86600 patients receiv-
py or stem cell transplantation. This practice being chemotherapy or autologous HSCT were
gan in the 1960s, at a time when fatal intracere- randomly assigned to platelet prophylaxis or no-
bral hemorrhage was a frequent cause of death prophylaxis for a morning platelet count
among severely thrombocytopenic patients re- <l0,000/)lL. Grade 2 or higher bleeding oc-
ceiving chemotherapy. In the 1990s a seminal curred in 50% of the no-prophylaxispatients vs
study" demonstrated that days of gross hemor- 43% of those receiving prophylaxis. As in the
rhage increased at lower platelet counts, al- study of Wandt and colleagues, the benefits of
though a clear threshold for increased bleeding platelet prophylaxis were much stronger among
risk was not identified. Nevertheless, it became patients receiving chemotherapy compared to
standard practice to transfuse platelets prophy- autologous HSCT recipients. These two RCTs
lactically for a platelet count <20,000/1lL. The were included in a recent meta-analysis that
threshold for platelet prophylaxis was subse- concluded that providing platelet prophylaxis in
quently lowered to 10,000/IlL on the basis of the setting of hypoproliferative thrombocytope-
both observational studies76,77and RCTs,Y8·80 An nia is assoc;:iat~~V/ith asignificant reduction in
observationalstudy suggested that an even low- Grade 2 or h1gnerbleediIlg (odds rauo, O.5~;
erplatelet countthreshold, 5000/IlL, would be 95% CI, 0.32-0.87).88Thus, prophylacticplatelet
sate," but the threshold of 10,000/IlL has been transfusions continue to be standard, although
used most commonly and is currently recom- some individual facilities are contemplating a
mended by severalclinicalpracticeguidelines.82.84 therapeutic-platelet-transfusion-onlystrategy for
Since the year 2000, a number of RCTshave adult autologous HSCTrecipients. It needs to be
been performed to determine the best approach emphasized that the usual platelet transfusion
to platelet transfusion. In many cases, the prima- threshold used, 10,000/IlL, is intended for hos-
ry endpoint used has been bleeding at World pitalized patients only. Outpatients are typically
Health Organization (WHO) Grade 2 score or given platelet transfusions more liberally, for
higher. A summary of the WHO scale is provid- practical reasons (ie, to permit fewer clinic vis-
TABLE 19·4. Summary of WHO Bleeding Scale*
Oropharyngeal bleeding :=;;30minutes in 24 hours
Epistaxis :=;;30minutes in previous 24 hours
Petechiae of oral mucosa or skin
Purpura :=;;1inch in diameter
Positive stool occult blood test
2 Epistaxis >30 minutes in 24 hours
Purpura> 1 inch in diameter
Hemoptysis
Melanotic stool
Grossivisible hematuria
Visible blood in body cavity fluid
Bleeding at invasive sites
3 Bleeding requiring RBC transfusion over routine needs
Bleeding associated with moderate hemodynamic instability
4 Bleeding associated with severe hemodynamic instability
CNS bleeding on imaging study
Fatal bleeding
*Modified from Kaufman et al.82
WHO = World Health Organization;RBC= Red Blood Cell; CNS = central nervous system.
its). Nonetheless, the optimal approach to pro- to be cleared in an age-independent fashion.

phylaxisin outpatients has not yet been formally With this hypothesis in mind, Hersh et al90 pro-
studied. posed that perhaps only a low dose of platelets is
The optimal platelet dose for transfusion has all that is needed for prophylaxis, and published
also been investigated. A provocative 1985 a mathematical model suggestingthat providing
study" suggested that although most platelets low-dose platelets (3 units of platelet concen-
will circulate for their normal life span (approxi- trates vs 6) would, over time, result in an overall
mately 8-10 days), a relativelysmall, fixed num- 22% savings in the number of platelets trans-
ber of platelets, estimated to be approximately fused. Subsequently, a small number of RCTs
7l00/IlL per day, are used to promote vascular examined the question of the optimal dose of
integrity. This population of platelets is thought platelets to transfuse for prophylaxis in the
setting of therapy-related hypoproliferative low- or very-low-qualityevidence.For central ve-

thrombocytopenia." The largest of these was nous catheter placement, AABB suggests that
the PLADQstudy (on platelet dose)," in Which prophylactic platelet transfusion may be consid-
1272 hospitalized hematology-oncologypatients ered for a platelet count <20,000/f.1,L. A prophy-
with hypoproliferative thrombocytopenia were lactic platelet count threshold of 50,000/f.1,L is
randomly assigned to receive low-dose (1.1 x suggested for lumbar puncture and for major
lOll platelets/rrr'), medium-dose (2.2 x lOll elective nonneuraxial surgery. A 2018 American
platelets/nr'), or high-dose platelets (4.4 x 1011 Society of Clinical Oncology (ASCO) Clinical
platelets/rrr') for a morning platelet count of Practice Guideline Update" contained similar
<lO,OOOI]1L. The medium-dose arm was meant recommendations for oncology patients. The
to approximate one apheresis platelet unit, the recommendations included a minimum platelet
current standard at the time. The primary end- count of 40,000 to 50,000/f.1,L for major inva-
point, the proportion of patients in each arm sive procedures and a lower threshold of
with Grade 2 or higher bleeding, was not signifi- 20,000/f.1,L for less-invasiveprocedures, includ-
cantly different (71%, 69%, and 70%, respec- ing central venous catheter insertion or removal
tively), demonstrating that low-dose platelets and marrow aspirations and biopsies. Recent
are a safe alternative to the standard dose. Con- Cochrane Reviews92,93 examining the role of
sistent with the prediction from the Hersh mod- prophylacticplatelet transfusionsbefore invasive
el, fewer overall platelets were transfused in the procedures did not identify significantnew evi-
low-dose arm. However, because patients re- dence from interim RCTs or nonrandomized
ceiving low-dose platelets had a lower incre- studies. Although the platelet count is important
ment, it was necessary to provide them with to consider, it is also relevant to note that this
transfusions more often (averageof five transfu- does not provide information on platelet func-
sions per patient vs three for patients in the me- tion or endothelial dysfunction. Clinical judg-
dium- or high-dose arms.) To date, low-dose ment, rather than a specific platelet count
platelets have not been widely adopted as stan- threshold, is of primary importance when decid-
dard, although low-dose platelets are some- ing whether to transfuse platelets in these set-
times used to stretch inventories during shortag- tings.
es. When low-dose platelets are used, it is
necessary to consider not only the number of Platelet Transfusions to Treat Active
platelets being transfused but also the reclplent's Bleeding
body surface area."
In thrombocytopenic patients who are bleeding,
it is often recommended that platelets should be
Platelet Transfusions as Prophylaxis for
transfused to maintain a platelet count above
Invasive Procedures
50,000/f.1,L. In bleeding patients with qualitative
Platelet transfusions are often administered be- platelet dysfunction (eg, patients taking anti-
fore minor (ie, bedside) and major (ie, surgical) platelet medications or after cardiopulmonary
procedures to try to reduce the bleeding risk in bypass), platelet transfusion has been suggested
patients with quantitative or qualitative platelet even at a normal platelet count. Interestingly, a
deficiencies.The published evidence supporting multicenter RCTexamining the effect of platelet
the use of platelet transfusions in these settings transfusionon 190 patients with acute intracere-
is limited. In 2015, AABBpublished a clinical bral hemorrhage after antiplatelet therapy
practice guideline on platelet transfusion; these found, in fact, an increased risk of death, func-
recommendations are summarized in Table 19- tional disability,or serious adverse outcomes in
5.82 This guideline was based on a systematicre- the transfusion group compared to the control
view of the literature." Except for one recom- group." Although the authors were unable to
mendation (platelet prophylaxis for therapy-re- identify a clear explanation for their findings,
lated hypoproliferative thrombocytopenia), all they hypothesized that platelet transfusion may
other recommendations are weak and based on have led to higher rates of thromboembolic
TABLE19·5. Summary of AABB Recommendationsfor Prophylactic PlateletTransfusion in Adults"
Platelet Transfusion May Be Strength of Quality of

Clinical Setting Indicated for: Recommendation Evidence
Therapy-relatedhypoproliferative Plateletcount :0::1

O,OOO/jJL Strong Moderate
thrombocytopenia
Centralvenous catheter placement Plateletcount <20,OOO/jJL Weak Low
Diagnostic lumbar puncture Plateletcount <50,OOO/jJL* Weak Very low
Major elective nonneuraxialsurgery Plateletcount <50,OOO/jJL Weak Very low
Cardiacsurgery with bypass Perioperativebleedingwith Weak Very low

thrombocytopenia and/or
evidenceof platelet dysfunc-
tion. Routine plateletprophy-
laxis not recommended.
Intracranial hemorrhageon Insufficient evidencefor Uncertain Very low

antiplatelettherapy recommendation
*Clinical judgment should be usedfor patientswith platelet counts between20,000 and 50,000I).tL.
complications or exacerbation of cerebral ische- and refractoriness. Mortality, bleeding, and

mia in cases of misdiagnosed infarction with transfusion reaction rates were not definitively
hemorrhagic conversion. shown to be improved by ABO matching, but
the platelet count increment did increase with
ABO and RhD Matching for Platelets ABO matching. Two controlled trialsl03,104
ABO matching is not an absolute requirement demonstrated a 40% to 60% reduction in refrac-
for platelets (or plasma) as it is for RBCs. But toriness with ABOmatching, but because defini-
platelets do express ABH antigens, often at high tions of refractoriness differed, the absolute
levels.95,96Preformed anti-A or anti-B in the benefit could not be determined. In practice,
recipient may destroy transfused major- ABO-identical or -compatible platelets may be
mismatched platelets (eg, group A donor, transfused whenever permitted by available in-
group 0 recipient).95-98Transfusion of major- ventory, although practices may differby institu-
mismatched platelets usually results in lower tion, patient population, and clinical setting. If
platelet increments.99 Conversely, transfusing ABO-matched units are unavailable, plasma re-
ABO-minor-mismatchedplatelets (eg, 0 donor, duction or selection of units with low anti-A or
B recipient) can cause hemolytic transfusion re- anti-B titers may mitigate the risk of hemolytic
actions, albeit rarely in adults, resulting from transfusion reactions associated with ABO-
passive administration of anti-A or anti-B in the minor-incompatibleplatelets.
plasma."? Several studies have investigated the Platelets do not express Rh antigens.!" but
impact of ABO matching on various out- platelet units do contain minute quantities of
comes,97,99-l03including mortality, bleeding, "contaminating" red cells. As few as 0.03 mL of
transfusion reactions, platelet count increment, RhD-positive red cells are believed to be re-
quired to cause alloimmunization resulting in proportion of platelets needed to maintain vas-

anti-D. Apheresis platelet units now typically cular integrity."
contain only microliter quantities of red cells There has not been agreement on a precise
(0.00043 mL),I°6-108although whole-blood- deflnltion of platelet refractoriness. Published
derived units may contain up to 100-fold-higher studies of platelet transfusion have often used a
amounts (approximately 0.036 mL).109Howev- platelet refractoriness definition of repeated 1-
er, platelets are often administered to immuno- hour corrected count increments (CCls) <7500,
compromised patients, who may be less capable although some groups have used an alternative
of becoming sensitized. Thus, the overall fre- CCI threshold of 5000.113The CCI attempts to
quency of alloimmunlzation from RhD-positive adjust the absolute platelet increment observed
platelet units in both immunocompetent and for the number of platelets transfused (x 10 II)
immunocornpromlsed patients is less than 2%, and the size of the recipient as reflected bybody
as demonstrated in a large, multicenter retro- surface area [BSA(m2)]:
spective review.!" This low risk can essentially
be eliminated by administering Rh Immune Platelet increment x BSA(m2)
Globulin (RhIG)within 72 hours of transfusion Platelets transfused (x 1011)
of RhD-positiveplatelets to an RhD-negativere-
cipient. Nonetheless, the decision to administer Example: A patient with a BSA of 2.0 m2
RhIG to prevent RhD alloimmunization when and a platelet count of 50001l-lL receives a unit
the rate of antibody formation is low should in- of apheresis platelets containing 4 x l O" plate-
clude consideration of risks and benefits and the lets, and the posttransfusion platelet count is
potential clinicalimpact of alloimmunization (ie, 25,00011-lL. The CCI may be calculated as fol-
in females of childbearing potential or before lows:
stem cell transplantation).
CCI = 20,000 x 2.0 10,000
Platelet Refractoriness 4.0
Platelet refractoriness represents a consistent
failure to achieve an appropriate platelet count CCls are not used in routine clinicalpractice,
increment following platelet transfusion. In because the number of platelets transfused is
most cases, platelet refractoriness is thought to usually unavailable. Rather, the unadjusted
have a nonimmune cause such as sepsis, dissem- platelet count increment is used to judge .wheth-
inated intravascular coagulation (DIC), bleed- er the patient's platelet count increased appro-
ing, hypersplenism, drug effects, or other priately. To evaluate a patient forimmune refrac-
platelet-consumptive states. Approximately 20% toriness, platelet counts should be. obtained
of cases of platelet refractoriness are thought to between 10 and 60 minutes after transfusion.
have an immune etiology.I10Somepossible caus- Poor increments (eg, <10,0001I-lL) on at least
es of platelet refractoriness are listed in Chapter two posttransfusion counts may be attributable
15. (See Table 15_3.)111 to immune refractoriness.I 11 Altematively.ffthe
:If a typical apheresis platelet unit (-4 x 1011 platelet count increases appropriately 1 hour af-
platelets) is transfused to an average-size, rela- ter transfusion but then declines to baseline at
tively healthy recipient, the expected l-hour 24 hours, a nonimmune cause of platelet refrac-
posttransfusion increment is ,-30,000 to toriness is likely (eg, consumption).
60,0001 1-lL.112 In severely thrombocytopenic pa- Immune refractoriness is usually caused by
tients receiving platelet prophylaxis, it is com- antibodies that target HLA114, 115 and cause rapid
mon for the observed platelet increment to be clearance of transfused platelets. Less common-
smaller and for the transfused platelets to have a ly, immune refractoriness is attributable to anti-
shorter survival. The prevailing hypothesis ex- bodies directed against platelet-specificglycopro-
plaining this observation is that the lower the teins known as human platelet antigens (HPAs).
pretransfusion platelet count, the higher the Transfusion recipients may become alloimmu-
nized to platelet HLA antigens either by prior review!" examined the efficacy of providing
pregnancy, organ transplantation, or transfusion. HLA-matched platelet units for refractory pa-
Platelets express HLAClass I antigens, but they tients. Most of the existing data come from ob-
are relatively poor immunogens. When immune servational studies performed before 2000, be-
refractoriness occurs in platelet transfusion re- fore the routine use of current HLA antibody
cipients, the HLA antibody response is mainly testing methods. Posttransfusion increments
provoked by "contaminating" white cells in the were the most common outcome reported
unit rather than the platelets thernselves.!" The among immune-refractory patients receiving
Trial to Reduce Alloimmunization to Platelets HLA-matchedplatelets, with varying degrees of
(TRAP)confirmed that leukocyte reduction sig- success. A 2014 Single-center observational
nificantly reduces the risk of HLAalloimmuniza- studyl22 found that providing HLA-matched
tion.!" Pregnancy is by far the most important units was associated with a successful incre-
risk factor for primary HLA sensttization.!" In ment in only 29% of transfusions to refractory
the era of leukocyte-reduced blood components, patients. Although better than providing ran-
immune refractoriness, often reflecting a sec- dom units, transfusing HLA-matched platelets
ondary immune response to HLA antigens, is a was of only limited utility. Studies powered to
particular problem in multiparous women. 118 examine the effect of HLA-selectedplatelets on
Identifying HLA antibodies is a second im- bleeding outcomes have not yet been per-
portant step in approaching immune refractori- formed.
ness. HLAantibody detection is most commonly When HLA-matched platelets are not avail-
performed using flow cytometry of multiantigen- able for immune-refractory patients, prophylac-
coated beads, although other methods (eg, lym- tic transfusions of random units are unlikely to
phocytoxicity assays, enzyme-linked immuno- result in an effective incremental response and
sorbent assays) are also used. Laboratories his- may cause further sensitization to additional
torically reported a panel-reactive antibody HLA antigens. In cases of bleeding complica-
(PRA) score based on the number of reactive tions in such patients, unmatched HLA units
wells observed on cytotoxicity assays to deter- may provide temporary hemostatic benefit and
mine the degree of HLA allotmmuntzation.!" should not be withheld to avoid alloimmuniza-
but this practice has been largelyreplaced by de- tion. IVIGand other therapeutic modalities used
termination of the calculated PRA(cPRA)based to treat immune thrombocytopenia (ITP) have
on antigen frequencies in the United Network not been demonstrated to be effectivein reduc-
for Organ Sharing (UNOS) network.!" There is ing the degree of alloimmunization in both ran-
no standard definition of a meaningful cPRA domized and nonrandomized studies but may
score, and thresholds for defining platelet refrac- be effectivein patients who have ITP secondary
toriness may vary by institution. to their underlying hematologic disorder." Oth-
Maneuvers to mitigate platelet immune re- er measures that may be considered include an-
fractoriness were recently reviewed.118,120 Op- tifibrinolyticagents.
tions to manage immune refractoriness include A minority of refractory patients who do not
providing HLA-matchedplatelets, HLA-antibody have an HLA alloantibody or have a poor re-
avoidance (ie, identifying HLA-antibodyspecific- sponse to HLA-matched platelet transfusions
ity and providing antigen-negativeplatelet units, may harbor alloantibodies directed against
analogous to similar strategies with RBCunits), HPAs.119 In addition to platelet refractoriness,
and platelet crossmatchmg.!" When providing HPA antibodies are also associated with fetal/
HLA-matched platelet units, donor units with neonatal alloimmune thrombocytopenia (FNAlT)
closely matching Class I A and B antigens (see and posttransfusion purpura (PTP). (See Chap-
Chapters 15 and 16 for more information on ters 15 and 23.) Such patients may benefit from
platelet matching) have been demonstrated to additional testing such as HPA antigen typing
result in improved transfusion response, al- and HPA antibody determination. These assays
though a failure to achieve a good increment is may not be available outside of major blood col-
still seen in 20% of cases.!" A recent systematic lection centers or specialized reference laborato-
ries, although a select number of commercial evidence for prophylactic plasma transfusions to
kits are available. Blood suppliers are limited in datel31,132has led to widely divergent practices
their ability to provide a wide variety of HPA- and highlights the need for additional data from
matched platelets due to the relatively small large, multicenter RCTs in order-to establish the
pool of HPAtyped donors. However, donors ideal role of plasma transfusion in this setting.
known to be negative for specific HPA antigens
implicated in FNAIT (eg, HPAla) may be re- Plasma Transfusions to Treat Bleeding
cruited for patients with refractory thrombocy- and Other Conditions
topenia and HPAantibodies.
Plasma transfusion is indicated for bleeding pa-
As pathogen inactivation technology has be-
tients with multiple coagulation factor deficien-
come more widely available, multiple studies
cies (eg, liver disease, DIC). It is also indicated
have reported an increased risk of platelet re-
to manage patients with specificplasma-protein
fractoriness secondary to HLAalloimmunization
deficiencies (eg, Factor V deficiency)for which a
in patients receiving pathogen-reduced platelet
licensed coaguiation factor concentrate is not
concentrates. The majority of observations have
available. Studies examining the impact of plas-
been described with use of the INTERCEPTsys-
ma transfusions in bleeding patients undergoing
tem, although similar findingswere also noted
cardiac surgery have been mostly limited to
in a Mirasol study,123and alloimmunization
non-RCTsI31;the clinical efficacy of plasma
rates continue to be monitored as secondary
transfusion in this setting has yet to be defined.
outcomes in ongoing clinical trials. The exact
A systematic review of plasma transfusion in pa-
etiologyis unclear, although higher rates of allo-
tients with central nervous system hemorrhage
immunization may be related to smaller post-
by an expert consensus panel advised against
transfusion increments and therefore increased
plasma transfusion in the absence of coagulopa-
numbers of platelet transfusions (and donor ex-
thy or vitamin K antagonist therapy, given the
posures) in patients receiving pathogen-reduced
potential for cardiopulmonary complications
vs standard platelets.
and serious adverse reactions following transfu-
sion.!" ...Systematic reviews conducted in
2004134and 2012135 revealed an emphasis on
PlAS A ANSFUSiON
prophylactic rather than therapeutic use of fresh
Plasma Prophylaxis for Invasive plasma in the available literature as well as a
Procedures
paucity of evidence-based plasma transfusion
practices.
Before performing invasive procedures, physi- Therapeutic plasma exchange is considered
cians often transfuse plasma to patients with first-line therapy for treatment of thrombotic
modest abnormalities in coagulation tests leg, thrombocytopenic purpura (TTP).Plasma is the
prothrombin time/international normalized ra- replacement fluid of choice for repletion of
tio (PT/INR) or activated partial thromboplastin ADAMTS13 (A Disintegrin And Metalloprotein-
time (aPTT)], with the goal of reducing the ase with a ThromboSpondin type 1 motif, mem-
bleeding risk. In most cases, this practice expos- ber 13) in order to restorevon Willebrand factor
es patients to all of the risks of plasma transfu- (vWF)-cleavingactivity.Although the efficacyof
sion without providing true benefits, because 1) plasma exchange has been clearly demonstrat-
mild-to-moderate coaguiation marker abnormal- ed for treatment of TTP,the optimal type of plas-
ities fail to predict bleeding in nonbleeding indi- ma component (ie, Fresh Frozen Plasma vs
viduals"; 2) modest elevations in PT/INR are Plasma Cryoprecipitate Reduced or solvent!
usually not corrected into the normal range by detergent-treated plasma) has not been defini-
plasma transfusion alone!"; and 3) RCTs and tively established; all appear efficacious.!" Plas-
observational studies have failed to demonstrate ma may also be used in conjunction with (or in
that prophylactic plasma transfusions affect place of) albumin as replacement fluid for pa-
bleeding outcomes.P'!" The lack of conclusive tients with underlying coaguiopathy or bleeding
(eg, diffuse alveolar hemorrhage) who require rent vitamin K administration is also recom-
therapeutic exchange. mended during reversal to ensure a sustained ef-
fect. Selection of the route of administration for
Vitamin K Antagonist Reversal supplemental vitamin K is dependent on clini-
cal urgency, because intravenous formulations
During clot formation, several coagulation fac-
take effect within 6 to 12 hours, compared to
tors such as Factors II, VII, IX, and X associate 12 to 24 hours for oral formulations.143
with the surface of activated platelets via hydro-
Plasma may be used if PCCs are contraindi-
phobic protein domains called gamma-carboxy-
cated, such as in patients who have had heparin-
glutamic acid (Gla) domains. Gla domains help
induced thrombocytopenia (because some PCCs
ensure that, when activated, coagulation factors
contain heparin) or in situations where PCCs
localize where they are needed to provide he-
may be unavailable. Given the transient effect of
mostatic function. To form Gla domains, specific
plasma and the relatively large dose required
glutamic acid (Glu) residues must undergo post-
(15-20 mLlkg) for effective reversal, patients
translational gamma-carboxylation. The reduced
should be closely monitored for evidence of flu-
form of vitamin K is required to contribute elec-
id overload, especially those with particular vol-
trons to these carboxylation reactions. In the
ume sensitivity (eg, cardiac and/or renal insuffi-
process, vitamin K becomes oxidized. Enzymes
ciency).
called vitamin K epoxide reductases serve to re-
In the absence of specific reversal agents for
cycle vitamin K back to its "useful" reduced
the direct oral anticoagulants (as in the case of
form so that it can participate in subsequent
idarucizumab for dabigatran or andexanet alfa
gamma-carboxylation reactions. Warfarin and
for Factor Xa inhibitors), four-factor PCCs have
other types of vitamin K antagonists (VKAs) are been found to be superior to plasma for reversal
structurally similar to vitamin K, resulting in
of Factor Xa inhibitors, although the evidence is
competitive inhibition of epoxide reductases limited to a small number of studies.144
when present. ThUS,VKAintake results in defi-
ciency of reduced vitamin K, which in turn
Types of Plasma
leads to decreased functional activity of Factors
II (thrombin), VII, IX, and X, as well as anti- Several varieties of plasma are available for
thrombotic factors protein C and protein S.137 transfusion, including Fresh Frozen Plasma
There are several methods available to re- (FFP), Plasma Frozen Within 24 Hours After
verse the effect of VKAs.For patients in whom Phlebotomy (PF24), Plasma Cryoprecipitate Re-
urgent reversal is needed (eg, bleeding or emer- duced, and solvent/detergent-treated plasma
gent indication for invasive surgery), the treat- (SDplasma). By definition, FFP is frozen within
ment of choice is a four-factorprothrombin com- 8 hours of collection and transfused within 24
plex concentrate (PCC). PCCs contain high hours of thawing, to preserve levels of the most
levels of Factors II, VII, IX, and X in the nonacti- heat-labile coagulation factors, Factors V and
vated state, as well as proteins C and S. Three- VIII. Many transfusion services provide Thawed
factor PCCs containing smaller quantities of Fac- Plasma, which is plasma that has been thawed
tor VII are also available and were used off-label and maintained in a closed system at 1 to 6 C
before FDA approval of four-factor PCCs specifi- for up to 5 days. Thawed Plasma is not currently
cally for VKAreversal.138Several studies directly regulated by the FDAbut is included in both the
comparing three-factor to four-factor PCCs have Circular of Information for the Use of Blood
demonstrated a slight advantage of four-factor and Blood Components+ and AABBStandards
over three-factor products.138,139 A recent RCT for Blood Banks and Transfusion Services (Stan-
demonstrated that reversal was more rapid and derdsi'" Advantages of Thawed Plasma in-
reliable in bleeding patients taking warfarin who clude immediate availability in bleeding emer-
received a PCC compared to those who re- gencies and reduced wastage due to its longer
ceived plasma.140,141 Both products had a similar shelf life. The activity of individual coagulation
profile for thromboembolic events.l" Concur- factors, such as Factor VIII, may decline at vari-
able rates over time, but overall factor activities brinogen in patients who are bleeding or having
have been shown to remain within the normal invasive procedures.
range during 5-day refrigerated storage.!" Dif- Pregnancy is associated with an increase in fi-
ferences in clinical outcome among recipients of brinogen concentration above the laboratory
various types of plasma have not been demon- levels of normal (approximately 6 giL in the
strated, and many transfusion services currently third trimester compared to 2-4 giL in the non-
use Thawed Plasma from any original source pregnant state).153A low fibrinogen concentra-
(FFP, PF24, SD plasma, etc) interchangeably tion among women with postpartum hemor-
with freshly thawed FFP. SD plasma provides an rhage has been reported to be independently
extra measure of safety with respect to transmis- associated with severe bleeding.154,155 To restore
sion of enveloped viruses and other pathogens'" hemostasis, fibrinogen replacement has been ad-
but is Significantlymore expensive than.untreat- vocated.P? Plasma, cryoprecipitate, and fibrino-
ed plasma components, as it is a pooled, treated gen concentrate are all sources of fibrinogen.
product. Primary indications include severe al- However, the volume of plasma needed is con-
lergic transfusion reactions or the need for re- siderably larger than that of cryoprecipitate to
duced risk of infectious disease transmission in achieve the same replacement dose of fibrino-
those requiring large-volume plasma transfu- gen (eg, 300-400 mg of fibrinogen can be re-
sions on a chronic basis (eg, TTP).136Although placed with 250 mL of plasma or with only 10-
15 mL of cryoprecipitate). Fibrinogen concen-
the first-generation SD plasma was noted to
trate is also a low-volume option, and it has the
have 20% to 30% lower Factor V levels com-
additional advantages of being pathogen re-
pared to FFP,second-generation Octaplas LGap-
duced and requiring no thawing time. To date,
pears to contain levels similar to other plasma
cryoprecipitate and fibrinogen concentrate have
products and may therefore be used inter-
not been directly compared in RCTsin adults. In
changeably for treatment of congenital or ac- a smallretrospective study of women with post-
quired Factor V deficiency.!" partum hemorrhage, similar outcomes were
achieved among patients receiving either cryo-
precipitate or fibrinogen concentrate. lSI In the
C
2015 Fibrinogen Concentrate as Initial Treat-
T ment for Postpartum Haemorrhage (FIB-PPH)
trtal,!" women with postpartum hemorrhage
Cryoprecipitate is a plasma derivative that is rel- were randomly assigned to receive 2 g of fibrin-
atively enriched for fibrinogen, FactorVlII, vWF, ogen concentrate or placebo. No difference in
fibronectin, and Factor XIII. There are limited clinical outcomes was observed (20% of patients
indications for cryoprecipitate, as there are in the fibrinogen concentrate group received
pathogen-reduced and recombinant products RBCsvs 22% of controls). However, only a small
available for several of the .indications where proportion of patients had severe postpartum
cryoprecipitate was used previously. Cryoprecip- hemorrhage in this trial. In two other small trials
itate is suggested for fibrinogen replacement for in patients with acquired hypofibrinogenemia
acquired hypofibrinogenemic conditions such as (ie, through surgery or bleeding), mortality,
liver transplantation and postpartum hemor- bleeding, and transfusion requirements did not
rhage.150-152Pathogen-reduced concentrates are differ when using cryoprecipitate or fibrinogen
standard-of-care to treat congenital hypofibrino- concentrate. 158
genemia, dysfibrinogenemia, and von Wille- Cardiac surgery is associated with acquired
brand disease. Congenital Factor XIIIdeficiency, hemostatic defects secondary to cardiopulmo-
associated with a delayed bleeding phenotype, is nary bypass that lead to excess bleeding.159-161
extremely rare, and there is now a recombinant Individuals undergoing cardiac surgery are at
Factor XIII concentrate available. Fibronectin is high risk of receiving blood transfusion.l= The
not currently used as a therapeutic agent. Thus, use of prophylactic fibrinogen is intended to re-
cryoprecipitate is used primarily to replace ft- duce the risk of bleeding and consequently the
exposure to blood components, because a lower per MBB Standards.146(p60) Studies have shown
fibrinogen concentration following cardiac sur- that granulocytes may maintain functional activ-
gery was associated with a higher bleeding ity for up to 48 hours if stored at 10 C, although
risk.163 In a small, single-center, placebo- the percentage of recovered cells decreases by
controlled, double-blinded RCT, prophylactic fi- approximately 50% after the first 24 hours.169,170
brinogen concentrate administered after prota- Because of their short shelf life, granulocyte
mine administration reduced blood component components may be released from the collection
usage among nonanemic cardiac surgery pa- facilitybefore results of donor infectious disease
tients. Patients randomly assigned to fibrinogen tests are complete. Therefore, granulocyte do-
concentrate vs placebo were significantly less nors are often selected from a pool of pre-
likely to receive any allogeneic blood compo- screened or frequent, long-term apheresis do-
nent (67% vs 45%; p = 0.015). Postoperative nors who have undergone recent testing.
bleeding was significantly (although modestly) Granulocyte components contain a significant
lower in the fibrinogen concentrate group (me- number of red cells and require crossmatching
dian blood loss of 300 mL vs 355 mL; p = unless they contain <2 mL of red cells.146(p42) To
0.042).164Nonetheless, further studies are need- prevent acute hemolytic reactions, granulocyte
ed to define the role of fibrinogen replacement components are matched to the recipient ac-
and its effectson bleeding, mortality, blood com- cording to ABO-matchingrules for RBCs (Table
ponent utilization, and adverse events.l'" specif- 19-2). Alternatively, red cells may be removed
ically in comparison to cryopredpitate.l'" A from ABO-incompatible units with methods
more recent placebo-controlled multicenter such as gravity sedimentation for donations, us-
RCT comparing fibrinogen concentrate to place- ing hydroxyethyl starch (HES).171Similarly, the
bo in postoperative cardiac surgery patients recipient should receive HLA-matched compo-
showed no benefit, and patients receiving fibrin- nents when HLA alloantibodies are present.
ogen concentrate actually received significantly Consideration may also be given to the cyto-
more allogeneic blood components.!" Fibrino- megalovirus (CMV)status of the donor if the re-
gen replacement using either cryoprecipitate or cipient is CMVnegative, although there is limit-
fibrinogen concentrate has not been associated ed evidence for proven transmission of CMV to
with increased mortality or thromboembolic seronegative recipients after granulocyte trans-
events. fusion.172,173Furthermore, CMV-seronegative
granulocytes may not always be available in a
timely manner, particularly if the recipient has
stringent ABO or HLA antigen restrictions or if
the donor population has a high prevalence of
CMVinfection.174In these cases, the clinical ur-
Prolonged severe neutropenia with intensive gency for granulocyte therapy should be
chemotherapy for hematologic malignancies or weighed against the risks of a CMV infection
in the setting of HSCT (defined as an absolute and/or transfusing a mismatched component.
neutrophil count of <500/f!L) predisposes pa- All granulocyte components must be irradiated
tients to life-threatening bacterial and fungal in- to prevent transfusion-associated graft-vs-host
fections despite aggressive antimicrobial thera- disease.!" Leukocyte reduction filters should
py.168Granulocyte transfusions are believed to never be used for administration. Recipients
reduce the risk of morbidity and mortality asso- should be closely monitored during granulocyte
ciated with such infections. Donors are stimu- infusions, as febrile cytokine reactions, volume
lated with corticosteroids and/or granulocyte overload, and pulmonary toxicity due to local-
colony-stimulating factor (G-CSF), allowing ization of granulocytes to the lungs may occur.
large numbers of granulocytes to be collected by However, a survival benefit of transfusing
apheresis. Granulocyte components are stored granulocytes to septic, neutropenic patients has
at room temperature and should be transfused not been definitively demonstrated.l" This lack
as soon as possiblewithin 24 hours of collection of benefit may be due to the inability to provide
C HAP T E R 1 9 Hemotherapy Decisions and Their Outcomes 571
an adequate number of granulocytesfor transfu- the past, trauma patients with substantial blood
sion.168,177 In a systematic review of 10 RCTs, loss were typically treated with RBC transfu-
prophylactic granulocyte transfusionwas not as- sions plus crystalloid, with hemostatic blood
sociated with a difference in morbidity or components such as platelets, plasma, and cryo-
mortality secondary to infections; however, precipitate administered based on laboratory
intermediate-dose granulocyte transfusions (1-4 test results. In recent years, this approach has
x 1010 granulocytes per day) were associated been largely superseded by a more aggressive
with a reduction in the number of patients with and empiric approach, whereby the initial resus-
infections after 30 days [relativerisk (RR),0.4; citation of trauma patients is focused on early
95% CI, 0.26-0.63] and the number of patients transfusion with plasma, platelets, and RBCsin
with bacteremia and fungemia (RR,0.45; 95% a fixed ratio (eg, 1:1:1; note: for platelets, the
CI, 0.30-0.65).177Trials of therapeutic granulo- "1" refers to a single whole-blood-derivedplate-
cyte transfusion in septic immunocompromised let concentrate and not 1 apheresis platelet
patients have also shown inconsistent results.l'" unit). These fixed ratios are intended to approxi-
For example, in the multicenter ResolvingInfec- mate the transfusion of whole blood through a
tion in Neutropenia with Granulocytes (RING) combination of components in order to prevent
trial, 176 neutropenic patients with proven or like- dilutional coagulopathy. The fixed ratio or
ly infection were randomly assigned to receive "formula-based" approach was devised by mili-
standard antimicrobial therapy or standard anti- tary physicians during the Iraq and Afghanistan
microbial therapy plus granulocytes collected wars of the 2000s. The publication credited
from donors stimulated with G-CSFand dexa- with sparking interest in this approach'" de-
methasone. Overall, no benefit was observed scribed 246 injured soldiers in Iraq who were
from transfusing granulocytes. Nonetheless, the retrospectivelygrouped by the ratio of plasma to
RINGstudy was underpowered, having enrolled RBCs received. Patients in the low plasma-to-
only 50% of the predefined sample size required RBCgroup (median of 1 unit ofplasma for every
to detect a difference in the primary composite 8 RBCunits) had a 65% mortality rate, as com-
outcome of survival and microbial clearance af- pared with a 19%mortality rate among patients
ter 42 days. Also, the target dose of >4 x 1010 in the high plasma-to-RBCgroup (median of 1
granulocytes per transfusion (0.6 x 109 cells/kg) unit of plasma for every 1.4 RBC units). Al-
was achieved in only 70% of transfusions.178 A though striking, this study and multiple other
secondary analysis suggested that patients who subsequent retrospective studies were highly
received higher doses of granulocytes tended to confounded. Trauma patients who die from
have better outcomes than patients who re- their injuries due to blood loss tend to do so
ceived lower doses.!" Currently, the role of very early (ie, often in less than an hour after
granulocyte transfusion remains undefined, and hospital arrivall.!" It was unclear whether early
clinicai judgment is requlred. If granulocyte and aggressiveplasma transfusion led to better
transfusions are used, higher doses of granulo- survival, or whether plasma transfusion was
cytes, as typically obtained from donors stimu- available for the less-severely injured patients
lated with G-CSF (or a combination of G-CSF who survived (ie, there was time to thaw and
and corticosteroids) compared to corticoster- transfuse frozen plasma in cases where patients
oids alone, may be more effective.!" did not die rapidly on arrival).183,184
Two recent multicenter studies examined
transfusion management of massively bleeding
trauma patients. The Prospective, Observation-
al, Multicenter, Major Trauma Transfusion
(PROMMTT)study185 was a prospective obser-
"Massive transfusion" is most often defined as vational study of adult trauma patients treated
transfusion of 10 or more RBC units in a 24- at 1 of 10 civilian trauma centers in the United
hour period in adults, although other definitions States. Study staff performed direct bedside
are also used (eg, 4 RBCunits in 1 houri.!" In observation as patients were resuscitated. To
reduce potential survivor bias, patients dying strategy.192,193The PrehospitalAir Medical Plas-
within the first 30 minutes of arrival were ex- ma (PAMPer) trial randomly assigned trauma
cluded. Patients who received plasma to RBCs patients at risk of hemorrhagic shock to either 2
in a 1:1 ratio had significantlybetter o-hour sur- units of thawed plasma (group AB or group A
vival than patients receiving a lower ratio of with a low anti-B antibody titer) or standard
plasma to RBCs.However, survivalat later time trauma care during air medical transport.192The
points did not differ significantly.A subsequent 30-day mortality rate was lower in the 230 pa-
RCT, called the Pragmatic Randomized Optimal tients who received plasma (23.2%) compared
Platelet and Plasma Ratios (PROPPR) trial.!" with that of the 271 patients (33%) who re-
compared outcomes among 680 adult civilian ceived standard trauma care (p = 0.03).192The
trauma patients who were randomly assigned to Control of Major Bleeding After Trauma Trial
be resuscitated using a 1:1:1 vs 1:1:2 ratio of (COMBAT)randomly assigned trauma patients
plasma to platelets to RBCs. The primary out- to 2 units of frozen AB plasma (75 patients) or
comes, 24-hour and 30-day survival,did not sig- saline (69 patients) and did not find a statistical-
nificantly differbetween the study groups. Cur- ly significant difference in 28-day mortality
rently, it is common for blood banks to (15% in the plasma group and 10%in the con-
incorporate fixed ratios of blood components (le, trol group; p = 0.37) and was thus terminated
1:1:1 or 1:1:2) into their local massive transfu- early (144 of 150 patients had been enrolled}!"
sion protocols (MTPs).Although it is difficultto The difference in outcomes may be due to the
judge the effectivenessof this approach from the injury severity of the trauma patients and pre-
published data, it does improve the speed and hospitalization use of red cells and ventilation
simplicity of the initial response. Laboratory- in the PAMPer trial. Future studies are needed
based, targeted transfusion of specific compo- to clarify the role of early resuscitation with
nents is often used after the patient has stabi- plasma.
lized. It is important to note that although much Whole blood resuscitation in the absence of
of the data on MTPs relates to trauma, in civil- blood component availability is currently em-
ian hospitals, massive transfusions are actually ployed in military settings and is regaining ac-
more likely to occur among other patient popu- ceptance for civilian trauma. Concerns regard-
lations (eg, solid-organ transplantation patients ing the functionality of platelets stored in the
and cardiac surgery patients).187,188 cold as part of refrigerated whole blood, and
Group ABplasma is the preferred blood com- plasma incompatibility have largely been
ponent in trauma MTPs before blood group de- surmounted. Cold-storage platelets arguably
termination because it lacks both anti-A and demonstrate superior hemostatic effect (includ-
anti-B. However, because AB plasma is in short ing increased adhesion, aggregation, and clot
supply due to the low prevalence of type AB do- strength), albeit inferior in-vivo survival and
nors (-4%), group A plasma has been used in posttransfusion recovery, compared with plate-
several centers as an alternate to AB plasma. lets stored at room temperature.!" MBB Stan-
The use of relatively plentiful group A units al- dards146 now permits transfusion of ABO-
lows for routine availabilityof a prethawed incompatible whole blood rather than requiring
ventory for immediate use without wastage of a ABO-identicalwhole blood, facilitating the use
precious resource (le, group AB plasma) if of low-titer group 0 whole blood for civilian
thawed components are not used during their 5- trauma resuscitation. Thus far, early studies
day shelf life.189Studies of group B and AB trau- have not demonstrated a significant risk of he-
ma patients who have received group A plasma molysts'" or evidence of inferior clinical out-
support the safety of this practice.190,191 comes compared to traditional component ther-
Earlyuse of plasma has been advocated to re- apy.!" However, determination of specific titer
duce the risk of trauma-associatedcoagulopathy. thresholds and other selection criteria for whole
Two recent RCTsexamined the use of prehospi- blood components have not yet been defined,
tal plasma resuscitation for civilian trauma and and current practices are highly variable from
yielded divergent results for the benefit of this institution to institution. Whole blood has the
advantages of smaller volumes than the combi- single product."? Experience is gaining in the
nation of RBCs,plasma, and platelets; early plas- use of whole blood for civilian trauma, but
ma and platelet resuscitation for effective man- whole blood has not yet become standard for
agement of coagulopathy; and fewer donor these patients nor for other patients with mas-
exposures by combining all components into a sive hemorrhage.
REFERENCES
1. Hebert PC, Wells G, BJajchman MA, et al. A 2. Carson JL, Terrin ML, Noveck H, et al. Liberal
multicenter, randomized, controlled clinical trial or restrictive transfusionin high-riskpatients after
of transfusion requirements in critical care. hip surgery. N Eng!J Med 2011;365:2453-62.
Transfusion Requirements in Critical Care 3. Grover M, Talwalkar S, Casbard A, et al. Silent
Investigators, Canadian Critical Care Trials myocardial ischaemia and haemoglobin concen-
Group. N EnglJ Med 1999;340:409-17. tration: A randomized controlled trial of transfu-
sion strategy in lower limb arthroplasty. Vox 16. de Almeida JP, Vlncent l-L, Galas FRBG,et al.
Sang 2006;90: 105-12. Transfusion requirements in surgical oncology
4. So-OsmanC, Nelissen R, Te Slaa R, et al. A ran- patients: A prospective, randomized controlled
domized comparison of transfusion triggers in trial. Anesthesiology2015;122:29-38.
elective orthopaedic surgery using leucocyte-de- 17. Prick BW, Jansen A, Steegers E, et al. Transfu-
pleted red blood cells.Vox Sang2010;98:56-64. sion policy after severe postpartum haemor-
5. Carson JL, Carless PA, Hebert PC. Transfusion rhage: A randomised non-inferioritytrial. BJOG
thresholds and other strategies for guiding allo- 2014;121: 1005-14.
geneic red blood cell transfusion. Cochrane Da- 18. Robertson CS, Hannay HJ, YamalJ-M, et al. Ef-
tabase SystRev 2012;(4):CD002042. fect of erythropoietin and transfusion threshold
6. Holst LB, Petersen MW, Haase N, et al. Restric- on neurological recovery after traumatic brain
tive versus liberal transfusion strategy for red injury. JAMA2014;312:36-47.
blood cell transfusion: Systematicreview of ran- 19. Carson JL, Stanworth SJ, Roubinian N, et al.
domised trials with meta-analysis and trial se- Transfusion thresholds and other strategies for
quential analysis.BMJ20 15;350:h1354. guiding allogeneic red blood cell transfusion.
7. Lacroix J, Hebert PC, Hutchison IS, et ali TRI- Cochrane Database Syst Rev 2016;10:
PICU Investigators; Canadian Critical Care Tri- CD002042.
als Group; Pediatric Acute Lung Injury and Sep- 20. Carson JL, Guyatt G, Heddle NM, et al. Clinical
sis InvestigatorsNetwork. Transfusionstrategies practice guidelines from the AABB:Red Blood
for patients in pediatric intensive care units. N Cell transfusion thresholds and storage. JAMA
EnglJ Med 2007;356(16):1609-19. 2016;316:2025-35.
8. Bracey AW, Radovancevic R, Riggs SA, et al. 21. So-OsmanC, Nelissen R, Brand R, et al. Postop-
Lowering the hemoglobin threshold for transfu- erative anemia after joint replacement surgery is
sion in coronary artery bypassprocedures: Effect not related to quality of life during the first two
weeks postoperatively.Transfusion2011 ;51:71-
on patient outcome. Transfusion 1999;39: 1070-
81.
7.
22. Carson JL, BrooksMM, Abbott JD, et al. Liberal
9. Hajjar LA,Vincent JL, Galas FR, et al. Transfu-
versus restrictive transfusion thresholds for pa-
sion requirements after cardiac surgery: The
tients with symptomatic coronary artery disease.
TRACSrandomized controlledtrial.JAMA2010;
Am HeartJ 2013;165(6):964-71.e1.
304:1559-67. 23. Hebert PC, Yetisir E, Martin C, et al. Is a low
10. Murphy GJ, Pike K, RogersCA, et al. Liberalor transfusion threshold safe in criticallyill patients
restrictive transfusion after cardiac surgery. with cardiovascular diseases? Crit Care Med
N EnglJ Med 2015;372:997-1008. 2001 ;29:227-34.
11. Mazer CD, Whitlock RP, Fergusson DA, et ali 24. BergarninFSI, AlmeidaJP, Landoni G, et al. Lib-
TRICSInvestigators and Perioperative Anesthe- eral versus restrictive transfusion strategy in crlt-
sia Clinical Trials Group. Restrictive or liberal ically ill oncologicpatients: The Transfusion Re-
red-cell transfusion for cardiac surgery. N EnglJ quirements in Critically IIIOncologic Patients
Med 2017;377:2133-44. Randomized Controlled Trial. Crit Care Med
12. Mazer CD, Whitlock RP, Shehata N. Restrictive 2017;45(5):766-73.
versus liberal transfusion for cardiac surgery. 25. Koch CG, LiL, SesslerDI, et al. Duration of red-
N EnglJ Med 2018;379(26):2576-7. cell storage and complications after cardiac sur-
13. Holst LB, Haase N, Wetterslev J, et al. Lower gery. N EnglJ Med 2008;358:1229-39.
versus higher hemoglobin threshold for transfu- 26. Fergusson DA, Hebert P, Hogan DL, et al. Effect
sion in septic shock. N Engl] Med 2014;371: of fresh red blood cell transfusions on clinical
1381-91. outcomes in premature, very low-birth-weight
14. VillanuevaC, ColomoA, BoschA, et al. Transfu- infants: The ARIPI randomized trial. JAMA
sion strategies for acute upper gastrointestinal 2012;308: 1443-51.
bleeding. N EnglJ Med 2013;368:11-21. 27. LacroixJ, Hebert PC, Fergusson DA, et al. Age
15. Iairath V, Kahan BC, Gray A, et al. Restrictive of transfused blood in critically ill adults. N Engl
versus liberal blood transfusion for acute upper J Med 2015;372:1410-18.
gastrointestinal bleeding (TRIGGER):A prag- 28. Cooper DJ, McQuilten ZK, Nichol A, et ali
matic, open-label, cluster randornised feasibility TRANSFUSEInvestigators and the Australian
trial. Lancet 2015;386:137-44. and New Zealand Intensive Care Society Cltni-
cal Trials Group. Age of red cells for transfusion 41. Yawn BP, Buchanan GR, Afenyi-Annan AN, et
and outcomes in critically ill adults. N Engl J al. Management of sickle cell disease: Summary
Med 2017;377:1858-67. of the 2014 evidence-based report by expert
29. Steiner ME, Ness PM, Assmann SF, et al. Effects panel members. JAMA2014;312: 1033-48.
of red-cell storage duration on patients undergo- 42. RosseWF, Gallagher D, Kinney TR, et al. Trans-
ing cardiac surgery. N Engl J Med 2015;372: fusion and alloimmunization in sickle cell dis-
1419-29. ease. The Cooperative Study of Sickle Cell Dis-
30. Dhabangi A, Ainomugisha B, Cserti-Gazdewich ease. Blood 1990;76:1431-7.
C, et al. Effect of transfusion of Red Blood Cells 43. Yazdanbakhsh K, Ware RE, Noizat-Pirenne F.
with longer vs shorter storage duration on ele- Red blood cell alloimmunization in sickle cell
vated blood lactate levels in children with se- disease: Pathophysiology, risk factors, and trans-
vere anemia. JAMA2015;314:2514-23. fusion management. Blood2012;120:528-37.
31. Heddle NM, Cook RJ,Arnold DM, et al. Effect 44. Chou ST, Jackson T, Vege S, et al. High preva-
of short-term vs. long-term blood storage on lence of red blood cell alloimmunization in sick-
mortality after transfusion. N Engl J Med 2016; le cell disease despite transfusion from Rh-
375:1937-45. matched minority donors. Blood 2013; 122:
32. Chasse M, McIntyre L, English SW, et al. Effect 1062-71.
of blood donor characteristics on transfusion 45. Fasano RM, Booth GS,Miles M, et al. Red blood
outcomes: A systematic review and meta-analy- cell alloimmunization is influenced by recipient
sis. Transfus Med Rev 2016;30:69-80. inflammatory state at time of transfusion in pa-
33. Edgren G, Ullum H, Rostgaard K, et al. Associa- tients with Sickle cell disease. Br J Haematol
tion of donor age and sex with survival of pa- 2015;168:291-300.
tients receiving transfusions. JAMA Intern Med 46. Danaee A, Inusa B, Howard J, Robinson S. Hy-
2017;177:854-60. perhemolysis in patients with hemoglobinopa-
34. Chasse M, Tinmouth A, English SW, et al. Asso- thies: A single-center experience and review of
ciation of blood donor age and sex with recipi- the literature. Transfus Med Rev 2015;29:220-
ent survival after Red Blood Cell transfusion. 30.
JAMAIntern Med 2016;176:1307-14. 47. Win N. Hyperhemolysis syndrome in sickle cell
35. Caram-Deelder C, Kreuger AL, Evers 0, et al. disease. Expert Rev Hemato12009;2:111-15.
Association of blood transfusion from female do- 48. Win N, Sinha S, Lee E, Mills W. Treatment with
nors with and without a history of pregnancy intravenous immunoglobulin and steroids may
with mortality among male and female transfu- correct severe anemia in hyperhemolytk trans-
sion recipients. JAMA2017;318:1471-8. fusion reactions: Case report and literature re-
36. Fiorellino J, Elahie AL, Warkentin TE. Acute view. Transfus Med Rev 2010;24:64-7.
haemo1ysis, DIC and renal fal1ureafter transfu- 49. Pirenne F, Yazdanbakhsh K. How I safely trans-
sion of uncross-matched blood during trauma fuse patients with sickle-celldisease and manage
resuscitation: Illustrative case and literature re- delayed hemolytic transfusion reactions. Blood
view. Transfus Med 2018;28(4):319-25. 2018;131:2773-81.
37. Rachmilewitz EA, Giardina Pl. How I treat thal- 50. Aiayash AI. Hemoglobin-based blood substitutes
assemia. Blood 2011; 118:3479-88. and the treatment of sickle cell disease: More
38. Goss C, Giardina P, Degtyaryova D, et al. Red harm than help? Biomolecules 2017;7(1). pii:
blood cell transfusions for thalassemia: Results E2.
of a survey assessing current practice and pro- 51. Davis JM, El-Haj N, Shah NN, et al. Use of the
posal of evidence-based guidelines. Transfusion blood substitute HBOC-201 in critically ill pa-
2014;54:1773-81. tients during sickle cell crisis: A three-case se-
39. Compernolle V, Chow ST,Tanael S, et al for the ries. Transfusion 2018;58:132-7.
International Collaboration for Transfusion 52. Kacker S, Ness PM, SavageWJ, et al. Cost-effec-
Medicine Guidelines (ICTMG). Red cell specifi- tiveness of prospective red blood cell antigen
cations for patients with hemoglobinopathies: A matching to prevent alloimmunization among
systematic review and guideline. Transfusion sickle cell patients. Transfusion 2014;54:86-97.
2018;58: 1555-66. 53. Vichinsky.EP, Luban NL, Wright E, et al. Pro-
40. Bunn HF. Pathogenesis and treatment of sickle spective RBC phenotype matching in a stroke-
cell disease. N EnglJ Med 1997;337:762-9. prevention trial in sickle cell anemia: A multi-
center transfusion trial. Transfusion 2001 ;41: myeloma and other hematologicaltumors. J Im-
1086-92. munol 2011;186:1840-8.
54. Tournamille C, Meunier-Costes N, Costes B, et 67. SullivanHC, Gerner-SmidtC, Nooka AK,et al.
al. Partial C antigen in sickle cell disease pa- Daratumumab (anti-CD38) induces loss of
tients: Clinicalrelevance and prevention of allo- CD38 on red blood cells. Blood 2017;129:
immunization. Transfusion2010;50: 13-19. 3033-7.
55. The American Societyof Hematology2020 clin- 68. Oldenborg PA, Zheleznyak A, Fang YF, et al.
ical practice guideline for sickle cell disease. Role of CD47 as a marker of self on red blood
Blood2020 (in press). cells. Science2000;288(5473):2051-4.
56. Adams RJ, McKieVC, Hsu L, et al. Prevention 69. Olsson M, Bruhns P, FrazierWA, et al. Platelet
of a first stroke by transfusions in children with homeostasis is regulated by platelet expression
sickle cell anemia and abnormal results on tran- of CD47 under normal conditions and in pas-
scranial Doppler ultrasonography.N EnglJ Med sive immune thrombocytopenia. Blood 2005;
1998;339:5-11. 105(9):3577-82.
57. Adams RJ, BrambillaD; STOP2 Trial investiga- 70. Russ A, Hua AB, Montfort WR, et al. BlockIng
tors. Discontinuing prophylactic transfusions "don't eat me" signal of CD47-SIRPain hema-
used to prevent stroke in sickle cell disease. tological malignancies, an in-depth review.
N EngIJ Med 2005;353:2769-78. BloodRev2018;32(6):480-9.
58. Ware RE, Davis BR, Schultz WH, et al. Hy- 71. Velliquette RW,AeschlimannJ, KirkegaardJ, et
droxycarbamide versus chronic transfusion for al. Monoclonal anti-CD47 interference in red
maintenance of transcranial doppler flow veloci- cell and platelet testing. Transfusion 2019;
ties in children with sickle cell anaemia- TCD 59(2):730-7.
With Transfusions Changing to Hydroxyurea 72. Chapuy CI, NicholsonRT,AguadMD, et al. Re-
(TWiTCH):A multicentre, open-label,phase 3, solving the daratumumab interference with
non-inferioritytrial. Lancet 2016;387:661-70.
blood compatibility testing. Transfusion 2015;
59. BarcelliniW, Fattizzo B, ZaninoniA, et al. Clini-
55(6 Pt 2):1545-54.
cal heterogeneity and predictors of outcome in
73. Hannon JL, Clarke G. Transfusionmanagement
primary autoimmune hemolytic anemia: A GI-
of patients receiving daratumumab therapy for
MEMA study of 308 patients. Blood 2014;
advanced plasma cell myeloma. Transfusion
124(19):2930-6.
2015;55(11):2770.
60. Petz LD. A physician's guide to transfusion in
74. Mitigating the anti-CD38 interference with se-
autoimmune haemolytic anaemia. BrJ Haema-
toI2004;124:712-16. rologic testing. Association Bulletin, #16-02.
61. Issitt PD, Combs MR, Bumgarner DJ, et al. Bethesda, MD: AABB,2016.
Studies of antibodiesin the sera of patients who 75. Gaydos LA, Freireich EJ,Mantel N. The quanti-
have made red cell autoantibodies. Transfusion tative relation between platelet count and hem-
1996;36:481-6. orrhage in patients with acute leukemia. N EngI
62. Laine ML, Beattie KM. Frequency of alloanti- J Med 1962;266:905-9.
bodies accompanying autoantibodies. Transfu- 76. Wandt H, Frank M, Ehninger G, et al. Safety
sion 1985;25:545-6. and cost effectivenessof a 10 x 10(9)/L trigger
63. ShireyRS,BoydJS,ParwaniAV,et al. Prophylac- for prophylactic platelet transfusions compared
tic antigen-matched donor blood for patients with the traditional 20 x 10(9)/L trigger:A pro-
with warm autoantibodies: An algorithm for spective comparative trial in 105 patients with
transfusion management. Transfusion2002;42: acute myeloid leukemia. Blood 1998;91:3601-
1435-41. 6.
64. EI Kenz H, Efira A, Le PO, et al. Transfusion 77. Slichter SJ, Harker LA. Thrombocytopenia:
support of autoimmune hemolytic anemia: How Mechanisms and management of defects in
could the blood group genotyping help? Transl platelet production. Clin HaematoI1978;7:523-
Res2014;163:36-42. 39.
65. Petz LD, Garratty G. Immune hemolytic anemi- 78. Heckman KD, Weiner GJ, Davis CS, et al. Ran-
as. Philadelphia:Churchill Livingstone,2004. domized study of prophylactic platelet transfu-
66. de Weers M, TalYT,van der Veer MS, et al. Da- sion threshold during induction therapy for
ratumumab, a novel therapeutic human CD38 adult acute leukemia: 1O,OOO/microL versus
monoclonal antibody,induces killingof multiple 20,000/microL. J Clin Onco11997;15:1143-9.
79. Rebulla P, Finazzi G, Marangoni F, et al. The ly platelet transfusion-dependent patient. Trans-
threshold for prophylactic platelet transfusions fusion 1998;38:637-44.
in adults with acute myeloid leukemia. Gruppo 91. Stichter SJ, Kaufman RM, Assmann SF, et al.
Italiano Malattie Ematologiche Maligne Dose of prophylactic platelet transfusions and
dell'Adulto. N EngiJ Med 1997;337: 1870-5. prevention of hemorrhage. N Engi J Med 2010;
80. Zumberg MS, del RosarioMLU, Nejame CF, et 362:600-13.
al. A prospective randomized trial of prophylac- 92. Estcourt LJ, Malouf R, Hopewell S, et al. Use of
tic platelet transfusion and bleeding incidence in platelet transfusions prior to lumbar punctures
hematopoietic stem cell transplant recipients: or epidural anaesthesia for the prevention of
1O,OOO/L versus 20,000/microL trigger. BioI complicationsin people with thrombocytopenia.
BloodMarrow Transplant 2002;8:569-76. Cochrane Database Syst Rev 2018;4:
81. Gmur J, BurgerJ, Schanz U, et al. Safetyof strin- CDOl1980.
gent prophylactic platelet transfusion policy for 93. Estcourt LJ,Malouf R, Doree C, et al. Prophylac-
patients with acute leukaemia. The Lancet tic platelet transfusions prior to surgery for peo-
1991;338: 1223-6. ple with a low platelet count. Cochrane Data-
82. KaufmanRM, DjulbegovicB, Gernsheimer T, et base SystRev 2018;9:CDOI2779.
al. Platelet transfusion: A clinicalpractice guide- 94. Baharoglu MI, Cordonnier C, Salman RA, et al.
line from the AABB.Ann Intern Med 2015;162: Platelet transfusion versus standard of care after
205-13. acute stroke due to spontaneous cerebral haem-
83. Nahirniak S, Slichter SJ, Tanael S, et al. Guid- orrhage associated with antiplatelet therapy
ance on platelet transfusion for patients with hy- (PATCH):A randomized, open-label, phase 3 tri-
poproliferativethrombocytopenia. TransfusMed al. Lancet 2016;387:2605-13.
Rev2015;29:3-13. 95. Cooling L. ABOand platelet transfusion therapy.
84. SchifferCA, Bohlke K, DelaneyM, et al. Platelet Immunohematology 2007;23:20-33.
transfusion for patients with cancer: American 96. Kelton JG, Hamid C, Aker S, Blajchman MA.
Society of Clinical Oncology clinical practice The amount of blood group A substance on
guideline update. J Clin Oncol2018;36(3):283- platelets is proportional to the amount in the
99. plasma. Blood 1982;59:980-5.
85. Stanworth SJ, Estcourt LJ, Llewelyn CA, et al. 97. Iulmy F, Ammann RA, Taleghani BM, et al.
Impact of prophylactic platelet transfusions on Transfusion efficacyof ABO major-mismatched
bleeding events in patients with hematologic platelets (PLTs)in children is inferior to that of
malignancies: A subgroup analysis of a random- ABO-identical PLTs. Transfusion 2009;49:21-
ized trial. Transfusion 2014;54:2385-93. 33.
86. Stanworth SJ, Estcourt LJ,Powter G, et al. A no- 98. Aster RH. Effect of anticoagulant and ABO in-
prophylaxis platelet-transfusion strategy for he- compatibility on recovery of transfused human
matologic cancers. N Engl J Med 2013;368: platelets. Blood 1965;26:732-43.
1771-80. 99. Lee EJ, Schiffer CA. ABO compatibility can
87. Wandt H, Schaefer-EckartK, Wendelin K, et al. influence the results of platelet transfusion.
Therapeutic platelet transfusion versus routine Resultsof a randomized trial. Transfusion 1989;
prophylactic transfusion in patients with haerna- 29:384-9.
tological malignancies: An open-label, multi- 100. KaufmanRM. Platelet ABOmatters. Transfusion
centre, randomised study. Lancet 2012;380: 2009;49:5-7.
1309-16. 101. Triulzi DJ, Assmann SF, Strauss RG, et al. The
88. Kumar A, Mhaskar R, Grossman BJ, et al. Plate- impact of platelet transfusion characteristics on
let transfusion: A systematic review of the clini- posttransfusion platelet increments and clinical
cal evidence. Transfusion 2014;55:1116-27. bleeding in patients with hypo proliferative
89. Hanson SR, Slichter SJ. Platelet kinetics in pa- thrombocytopenia. Blood 2012; 119:5553-62.
tients with bone marrow hypoplasia: Evidence 102. Kaufman RM, Assmann SF, Triulzi DJ, et al.
for a fixed platelet requirement. Blood 1985;66: Transfusion-related adverse events in the Plate-
1105-9. let Dose study. Transfusion2015;55: 144-53.
90. Hersh JK, Hom EG, Brecher ME. Mathematical 103. Heal J, Rowe J, McMican A, et al. The role of
modeling of platelet survival with implications ABOmatching in platelet transfusion. Eur J Hae-
for optimal transfusion practice in the chronical- matol1993;50:1l0-17.
104. Carr R, Hutton J, Jenkins J, et al. Transfusion of 117. Triulzi DJ, Kleinman S, Kakaiya RM, et al. The
ABO-mismatched platelets leads to early platelet effect of previous pregnancy and transfusion on
refractoriness. BrJ HaematoI1990;75:408-13. HLAal1oimmunizationin blood donors: Implica-
105. Dunstan RA, Simpson MB, Rosse WF. Erythro- tions for a transfusion-related acute lung injury
cyte antigens on human platelets: Absence of risk reduction strategy. Transfusion 2009;49:
Rh, Duffy, Kell, Kidd, and Lutheran antigens. 1825-35.
Transfusion 1984;24:243-6. 118. Pavenski K, Rebulla P, Duquesnoy R, et al. Effi-
106. Molnar R, Johnson R, Geiger TL. Absence of D cacy of HLA-matched platelet transfusions for
alloimmunization in D- pediatric oncology pa- patients with hypoproliferative thrombocytope-
tients receiving D-incompatible single-donor nia: A systematic review. Transfusion 2013;53:
platelets. Transfusion 2002;42: 177-82. 2230-42.
107. Culibrk B, Stone E, Levin E, et al. Application of 119. Kopko PM, Warner P, Kresie L, Pancoska C.
the ADVIA cerebrospinal fluid assay to count re- Methods for the selection of platelet products
sidual red blood cells in blood components. Vox for allotmmune-refractory patients. Transfusion
Sang 2012;103: 186-93. 2015;55:235-44.
108. Santana JM, Dumont LJ, A flow cytometric 120. Stanworth SJ, Navarrete C, Estcourt L, Marsh J.
method for detection and enumeration of low- Platelet refractoriness - practical approaches and
level, residual red blood cells in platelets and ongoing dilemmas in patient management. Br J
mononuclear cell products. Transfusion 2006; Haemato12015; 171:297-305.
46:966-72. 121. Moroff G, Garratty G, Heal JM, et al. Selection
109. Cid J, Lozano M, Ziman A, et al. Low frequency of platelets for refractory patients by HLA
of anti-D alloimmunization following D+ plate- matching and prospective crossmatching. Trans-
let transfusion: The Anti-D Alloimmunization fusion 1992;32:633-40.
after D-incompatible Platelet Transfusions 122. Rioux-Masse B, Cohn C, Lindgren B, et al. Utili-
(ADAPT)study. Br J HaematoI2015;168:598- zation of cross-matched or HLA-matched plate-
603. lets for patients refractory to platelet transfusion.
110. Doughty HA, Murphy MF, Metcalfe P, et al. Rel- Transfusion 2014;54:3080-7.
ative importance of immune and non-immune 123. Escourt LJ, Malouf R, Hopewell S, et al. Patho-
causes of platelet refractoriness. Vox Sang 1994; gen-reduced platelets for the prevention of
66:200-5. bleeding. Cochrane Database Syst Rev 2017;
111. Hod E, Schwartz J. Platelet transfusion refracto- 30(7):CD009072.
riness. BrJ Haemato12008;142:348-60. 124. Holland L, Sarode R. Should plasma be trans-
112. Aster RH. Pooling of platelets in the spleen: Role fused prophylactically before invasive proce-
in the pathogenesis of "hypersplenic" thrombo- dures? Curr Opin Hematol2006; 13:447-51.
cytopenia. J Clin Invest 1966;45(5):645-57. 125. Abde1-Wahab 0, Healy B, Dzik W. Effect of
113. Slichter SJ. Leukocyte reduction and ultraviolet fresh-frozen plasma transfusion on prothrombin
B irradiation of platelets to prevent allolmmuni- time and bleeding in patients with mild coagula-
zation and refractoriness to platelet transfusions. tion abnormalities. Transfusion 2006;46: 1279-
The Trial to Reduce Alloimmunization to Plate- 85.
lets Study Group. N Engl J Med 1997;337: 126. Karam 0, TucciM, Combescure C, etal. Plasma
1861-9. transfusion strategies for critically ill patients.
114. Yankee RA, Grumet FC, Rogentine GN. Platelet Cochrane Database Syst Rev 2013; 12:
transfusion: The selection of compatible platelet CD010654.
donors for refractory patients by lymphocyte 127. Murad MH, Stubbs JR, Gandhi MJ, et al. The ef-
HL-Atyping. N Eng!J Med 1969;281: 1208-12. fect of plasma transfusion on morbidity and
115. Slichter SJ, Factors affecting posttransfusion mortality: A systematic review and meta-analy-
platelet increments, platelet refractoriness, and sis. Transfusion 2010;50: 1370-83.
platelet transfusion intervals in thrombocytope- 128. SegalJB, Dzik WHoPaucity of studies to support
nic patients. Blood 2005;105:4106-14. that abnormal coagulation test results predict
116. Claas FH, Smeenk RJ, Schmidt R, et al. Alloim- bleeding in the setting of invasive procedures:
munization against the MHC antigens after An evidence-based review. Transfusion 2005;
platelet transfusions is due to contaminating leu- 45:1413-25.
kocytes in the platelet suspension. Exp Hematol 129. Yang L, Stanworth S, Hopewell S, et al. Is fresh-
1981;9:84-9. frozen plasma clinically effective? An update of
a systematic review of randomized controlled versus plasma for rapid vitamin K antagonist re-
trials. Transfusion2012;52: 1673-86. versal in patients needing urgent surgical or in-
130. Ita 0, Brown MJ, CliffordL, et al. Prophylactic vasive interventions: A phase 3b, open-label,
plasma transfusion for surgicalpatients with ab- non-inferiority, randomised trial. Lancet 2015;
normal preoperative coagulation tests: A single- 385:2077-87.
institution propensity-adiusted cohort study. 142. Milling TJ Jr, RefaaiMA, Sarode R, et al. Safety
Lancet Haematol20 16;3:e139-48. of a four-factor prothrombin complex concen-
131. Desborough M, Sandu R, Brunskill SJ, et al. trate versus plasma for vitamin K antagonist re-
Fresh frozen plasma for cardiovascular surgery. versal: An integrated analysis of two Phase IIIb
Cochrane Database Syst Rev 2015;7:CDOO clinical trials. Acad Emerg Med 2016;23(4):
7614. 466-75.
132. Huber J, Stanworth SJ, Doree C, et al. Prophy- 143. Meehan R, Tavares M, Sweeney J. Clinical ex-
lactic plasma transfusion for patients undergoing perience with oral versus intravenous vitamin K
non-cardiac surgery (protocol). Cochrane Data- for warfarin reversal. Transfusion 2013;53:491-
base SystRev 2017;8:CDO12745. 8.
133. Shander A, Michelson EA,Sarani B,et al. Use of 144. Tornkvist M, Smith JG, Labaf A. Current evi-
plasma in the management of central nervous dence of oral anticoagulant reversal. Thromb
system bleeding: Evidence-basedconsensus rec- Res 2018;162:22-31.
ommendations. Adv Ther 2014;31 :66-90. 145. MBB, American Red Cross, America's Blood
134. Stanworth SJ, Brunskill SJ, Hyde CJ, et al. Is Centers, Armed ServicesBlood Program Office.
fresh frozen plasma clinically effective? A sys- Circular for the use of human blood and blood
tematic review of randomized controlled trials. components. Bethesda,MD: MBB, 2017.
BrJ Haemato12004; 126:139-52. 146. Gammon R, ed. Standards for blood banks and
135. Yang L, Stanworth S, Hopewell S, et al. Is fresh transfusion services. 32nd ed. Bethesda, MD:
frozen plasma clinicallyeffective? An update of MBB,2020.
a systematic review of randomized controlled 147. Downes KA,Wilson E, YomtovianR, Sarode R.
trials. Transfusion 2012;52:1673-86. Serialmeasurement of clottingfactors in thawed
136. O'Shaughnessy DF, Atterbury C, Bolton-Maggs plasma stored for 5 days. Transfusion 2001 ;41:
P, et al. Guidelines for the use of fresh-frozen 570.
plasma, cryoprecipitate, and cryosupernatant. Br 148. Benjamin RJ, McLaughlin LS. Plasma compo-
J Haematol2004; 126:11-28. nents: Properties, differences,and uses. Transfu-
137. Presnell SR, Stafford DW. The vitamin K- sion 2012;52(Suppll ):9S-19S.
dependent carboxylase.Thromb Haemost 2002; 149. Cushing MM, AsmisL, CalabiaC, et al. Efficacy
87:937-46. of solvent/detergent plasma after storage at
138. Jones GM, Erdman MJ, Smetana KS, et al. 3- 2-8 °C for 5 days in comparison to other plasma
Factor versus 4-factor prothrombin complex products to improve factor V levels in factor V
concentrate for warfarin reversal in severe deficient plasma. Transfus Apher Sci 2016;
bleeding: A multicenter, retrospective, propensi- 55:114-19.
ty-matched pilot study. J Thromb Thrombolysis 150. LevyJH, Goodnough LT. How I use fibrinogen
2016;42: 19-26. replacement therapy in acquired bleeding.
139. Al-Majzoub0, RybakE, Reardon DP, et al. Eval- Blood2015; 125:1387-93.
uation of warfarin reversal with 4-factor pro- 151. Ahmed S, Harrity C, Johnson S, et al. The effica-
thrombin complex concentrate compared to cy of fibrinogen concentrate compared with
3-factor prothrombin complex concentrate at a cryoprecipitate in major obstetric haemor-
tertiary academic medical center. J Emerg Med rhage-an observational study. Transfus Med
2016;50(1 ):7-13. 2012;22:344-9.
140. Sarode R, Milling TJ, RefaaiMA, et al. Efficacy 152. Pavord S, Maybury H. How I treat postpartum
and safety of a four-factorprothrombin complex hemorrhage. Blood2015;125:2759-70.
concentrate (4F-PCC)in patients on vitamin K 153. Reger B, PeterfalviA, Litter I, et aI. Challenges
antagonists presenting with major bleeding: A in the evaluation of D-dimer and fibrinogen lev-
randomized, plasma-controlled, Phase IIIb els in pregnant women. Thromb Res 2013;131:
study. Circulation 2013;128:1234-43. eI83-7.
141. Goldstein IN, Refaai MA, MillIng TJ Ir, et al. 154. Charbit B, Mandelbrot L, Samain E, et al. The
Four-factor prothrombin complex concentrate decrease of fibrinogen is an early predictor
of the severity of postpartum hemorrhage. double-blindPhase III study of haemostatic ther-

J Thromb Haemost 2007;5:266-73. apy. BrJ Anaesth 2016;117:41-51.
155. Cortet M, Deneux-Tharaux C, Dupont C, et al. 168. Strauss RG. Role of granulocyte/neutrophil
Associationbetween fibrinogenlevel and severi- transfusions for haematology/oncology patients
ty ofpostpartum haemorrhage: Secondaryanaly- in the modern era. Br J Haematol 2012;158:
sis of a prospective trial. Br J Anaesth 2012; 299-306.
108:984-9. 169. Hubel K, Rodger E, GavirlaJM, et al. Effective
156. Abdul-KadirR, McLintock C, DucloyA-S,et al. storage of granulocytes collected by centrifuga-
Evaluation and management of postpartum tion leukapheresis from donors stimulated with
hemorrhage: Consensus from an international granulocyte-colony-stimulatingfactor. Transfu-
expert panel. Transfusion2014;54: 1756-68. sion 2005;45(12):1876-89.
157. WikkelsoAJ, Edwards HM, AfshariA, et al. Pre- 170. Drewniak A, BoelensJJ, VrielinkH, et al. Gran-
emptive treatment with fibrinogen concentrate ulocyte concentrates: Prolonged functional ca-
for postpartum haemorrhage: Randomized con- pacity during storage in the presence of pheno-
trolled trial. BrJ Anaesth 2015;114:623-33. typic changes. Haematologica 2008;93(7):
158. Jensen NH, Stensballe J, Afshari A. Comparing 1058-67.
efficacyand safety of fibrinogen concentrate to 171. Bryant BJ,Yau YY,Byrne PJ, et al. Gravity sedi-
cryoprecipitate in bleeding patients: A systemat- mentation of granulocytapheresis concentrates
ic review. Acta Anaesthesiol Scand 2016;60: with hydroxyethyl starch efficiently removes
1033-42. red blood cells and retains neutrophils. Transfu-
159. Besser MW, KleinM. The coagulopathy of car- sion 2010;50: 1203-9.
diopulmonarybypass.Crit RevClin LabSci2010; 172. Narvios A, Pena E, Han X, Lichtiger B. Cyto-
megalovirus infection in cancer patients receiv-
47:197-212.
160. BesserMW, Ortmann E, KleinM. Haemostatic ing granulocyte transfusions. Blood 2002;99:
management of cardiac surgical haemorrhage. 390-1.
173. NicholsWG, Price T, BoeckhM. Cytomegalovi-
Anaesthesia 2014;70:87-95.
rus infectionsin cancer patients receiving granu-
161. Woodman R, Harker LA. Bleeding complica-
locyte transfusions (comment). Blood 2002;
tions associated with cardiopulmonary bypass.
99(9):3483-4.
Blood2003;76: 1680-97.
174. Diaz R, Soundar E, Hartman SK,et al. Granulo-
162. Bennett-Guerrero E, Zhao Y, O'Brien SM, et al.
cyte transfusions for children with infections
Variationin use of blood transfusion in coronary
and neutropenia or granulocytedysfunction. Pe-
artery bypass graft surgery. JAMA 2010;304: diatr Hematol Onco12014;31(5):425-34.
1568-75. 175. West K, Gea-Banac1ocheJ, Stroncek D, Kadri
163. KindoM, Hoang Minh T, Gerell!S, et al. Plasma SS. Granulocyte transfusions in the manage-
fibrinogen level on admission to the intensive ment of invasive fungal infections. BrJ Haema-
care unit is a powerful predictor ofpostoperative to12017; 177:357-74.
bleeding after cardiac surgerywith cardiopulmo- 176. Price TH, Boeckh M, Harrison RW, et al.
nary bypass.Thromb Res2014;134:360-8. Efficacyof transfusion with granulocytes from
164. Ranucci M, BaryshnikovaE, Crapelli GB, et al. G-CSF/dexamethasone-treated donors in neu-
Randomized, double-blinded, placebo-con- tropenic patients with infection. Blood 2015;
trolled trial of fibrinogen concentrate supple- 126:2153-61.
mentation after complex cardiac surgery. J Am 177. Estcourt LJ,Stanworth S, Doree C, et al. Granu-
Heart Assoc 2015;4:e002066. locyte transfusions for preventing infections in
165. WikkelsoA, Lunde J, Johansen M, et al. Fibrino- people with neutropenia or neutrophil dysfunc-
gen concentrate in bleeding patients. Cochrane tion. Cochrane Database Syst Rev 2015;6:
Database SystRev 2013;8:CD008864. CD005341.
166. Maeda T, Miyata S, UsuiA, et al. Safetyoffibrin- 178. Cancelas JA. Granulocyte transfusion: Ques-
ogen concentrate and cryoprecipitate in cardio- tions remain. Blood2015; 126:2082-3.
vascular surgery: Multicenter database study. 179. Marfin M, Price TH. Granulocyte transfusion
J Cardiothorac VascAnesth 2019;33:321-7. therapy. J lnt Care Med 2015;30(2):79-88.
167. Rahe-MeyerN, LevyJH, Mazer CD, et al. Ran- 180. Moren AM, Harnptom D, DiggsB, et al. Recur-
domized evaluation of fibrinogen vs placebo in sive partitioning identifies greater than 4 U of
complex cardiovascular surgery (REPLACE):A packed red blood cells per hour as an improved
massive transfusion definition. J Trauma Acute in the United States. Transfusion 2016;56: 125-
Care Surg 2015;79:920-4. 9.
181. Borgman M, Spinella P, PerkinsJ, et al. The ra- 190. Chhibber V, Greene M, Vauthrin M, et al. Is
tio of blood products transfused affectsmortality group A thawed plasma suitable as the first op-
in patients receiving massive transfusions at a tion for emergency release transfusion?Transfu-
combat support hospital. J Trauma 2007;63: sion 2014;54: 1751-5.
805·l3. 191. Dunbar NM and YazerMH. Safetyof the use of
182. AcostaJA, YangJC, WinchellRJ, etal. Lethal in, group A plasma in trauma: The STAT study.
juries and time to death in a level I trauma cen- Transfusion2017;57: 1879-84.
ter. J Am Coll Surg 1998;186:528-33. 192. SperryJL, Guyette FX,BrownJB, et al; PAMPer
183. Stansbury LG, Dutton RP, Stein OM, et al. Con- Study Group. Prehospital plasma during air
troversy in trauma resuscitation: Do ratios of medical transport in trauma patients at risk for
plasma to red blood cells matter? Transfus Med hemorrhagic shock. N Eng1 J Med 2018;
Rev 2009;23:255-65. 379(4):315-26.
184. CallumJL, Nascimento B,Tien H, RizoliS. "For- 193. Moore HB,Moore EE, Chapman MP, et al. Plas-
mula-driven" versus "lab-driven"massive trans- ma-first resuscitation to treat haemorrhagic
fusion protocols: At a state of clinical equipoise shock during emergency ground transportation
(editorial).Transfus Med Rev2009;23:247-54. in an urban area: A randomised trial. Lancet
185. Holcomb JB, del Junco OJ, Fox EE, et al. The 2018;392(10144):283-91.
Prospective, Observational, Multicenter, Major 194. StubbsJR, Tran SA,Emery RL,et al. Cold plate-
Trauma Transfusion (PROMMTT)study. JAMA lets for trauma-associated bleeding: Regulatory
Surgery 2013;148:127-36. approval, accreditation approval, and practice
186. HolcombJB, Tilley BC, Baraniuk S, et al. Trans- implementation - just the "tip of the iceberg."
fusion of plasma, platelets, and red blood cells in Transfusion2017;57:2836-44.
a 1:1:1 vs a 1:1:2 ratio and mortality in patients 195. Seheult JN, Bahr M, Anto V, et al. Safetyprofile
with severe trauma: The PROPPRrandomized of uncrossmatched, cold-stored,low-titer, group
clinicaltrial.JAMA2015;313:471-82. 0+ whole blood in civilian trauma patients.
187. Dzik WS, Ziman A, Cohen C, et al. Survivalaf- Transfusion2018;58(10):2280-8.
ter ultramassive transfusion: A review of 1360 196. Seheult IN, Anto V, Alarcon AH, et al. Clinical
cases. Transfusion 2016;56:558-63. outcomes among low-titer group 0 whole blood
188. Johnson OJ, Scott AV, BarodkaVM, et al. Mor- recipients compared to recipients of convention-
bidity and mortality after high-dose transfusion. al components in civilian trauma resuscitation.
Anesthesiology2016; 124:387-95. Transfusion2018;58: 1838-45.
189. Dunbar NM, Yazer MH. A possible new para- 197. YazerMH, Cap AP, SpinellaPC, et al. How do I
digm? A survey-based assessment of the use of implement a whole blood program for massively
thawed group A plasma for trauma resuscitation bleeding patients? Transfusion2018;58:622-8.
Patient Blood anagement
Steven M. Frank, MD, and Nicole R. Guinn, MD
LOOD TRANSFUSIONS CAN BE LIFE- based transfusion guidelines are fundamental,

saving but are also associated with risks PBM goes beyond appropriate blood component
and complications.' Growing attention by utilization. It encompasses the entire course of
professional associations and health-care organi- care, from before the patient enters the hospital
zations to the wide variation in transfusion prac- to after treatment is completed. The primary
tice2 and overuse of blood" has intensified the aim of PBM is to improve patient safety and clin-
need for more appropriate utilization. Further- ical outcomes by appropriately managing the pa-
more, efforts to lower health-care costs and im- tient's own blood. PBM entails a proactive ap-
prove quality of care and patient safety have led proach in which clinicians strive to reduce or
to increased focus on reducing unnecessary avoidunnecessary transfusionsby using evidence-
transfusions. based pharmacologic, medical, and surgical mo-
Accordingly, many institutions have imple- dalities to manage anemia, optimize hemostasis,
mented patient blood management (PBM)strat- and minimize blood loss in a patient-specific
egies and related activities. The 2017 National manner. In addition, when a transfusion is the
Blood Collection and Utilization Survey"report- only appropriate intervention, PBM requires
ed that more than one-third of hospitals had a that transfusion decisions be based on best prac-
PBM program and that even more facilities had tices and that the amount given be the appropri-
implemented one or more interventions to im- ate amount to improve patient outcome.
prove care and reduce unnecessary transfusions. Although the term "patient blood manage-
These initiatives have resulted in a steady de- ment" may be fairly new, the concepts have de-
cline in blood and blood component transfu- veloped over time in parallel with medical ad-
sions in the United States.' vances. A full discussion of the history is beyond
the scope of this chapter. However, three exam-
ples Illustrate the shared drivers of blood trans-
fusion and PBM. One example is the influence
of wartime injuries. The discoveries of blood
groups, crossmatching, and blood storage in the
early years of the 20th century enabled the use
PBM is an evidence-based, multidisciplinary ap- of banked blood to treat exsanguinating wounds
proach to optimizing the care of patients who suffered in armed conflicts. Yet, transporting
might need a transfusion. Although evidence- blood was difficult, and battlefield surgeons de-
Steven M. Frank, MD, Professor, Department of Anesthesiology/Critical Care Medicine, Medical Director, Johns
Hopkins Health Systems Blood Management Program, and Faculty, Armstrong Institute for Patient Safety and
Quality, The Johns Hopkins Medical Institutlons, Baltimore, Maryland; and Nicole R. Guinn, MD, Associate Pro-
fessor, Department of Anesthesiology, and Medical Director, Center for Blood Conservation, Duke University
Medical Center, Durham, North Carolina
S. Frank has disclosed a financial relationship with Haemonetics, Medtronic, and Baxter International. N. Guinn
has disclosed no conflicts of interest.
583
veloped techniques to treat casualties when no pending on the baseline degree of transfusion
transfusion was possible. A second example is overuse, a successful PBM program can often
the influence of Jehovah's Witnesses, who cite pay for itself several times over by reducing
biblical passages as the foundation for not ac- transfusion-related costs." Given that the bun-
cepting blood transfusions. By the middle of the dled payment system makes reimbursement for
20th century, when blood transfusion had be- blood poor to nonexistent in the United States,
come universally accepted as a medical treat- and that the overhead costs of transfusion are
ment for a wide range of indications, Jehovah's approximately three- to fourfold the blood acqui-
Witness patients had to seek out those few phy- sition costs," a solid PBM program will easily be
sicians and hospitals that would provide medical self-supporting.
and surgical care without transfusions. These The first step toward developing a successful
surgeons and hospitals became increasingly program is to generate a business plan that sets
used by Jehovah's Witnesses and others who financial goals and engages the hospital's admin-
wished to avoid transfusions, and blood conser- istration to gain financial support. The program's
vation programs began to flourish. Third, the leaders will need salary support to buy time to
high risk of viral transmission for human immu- get the work done. Personnel cannot develop a
nodeficiency virus (HIV) and hepatitis B and C successful program on nights and weekends, or
peaked in the early 1980s, along with an increas- in their spare time. Depending on the size of the
ing awareness of, and concern about, transfu- hospital, some portion of time should be sup-
sion-transmitted infectious disease. With an em- ported for a medical director (physician), a full-
phasis on an individualized, patient-centric time nurse coordinator or transfusion safety offi-
approach, these issues led the way toward the cer, administrative support, and a data manager.
PBM movement seen today. For smaller hospitals, it is entirely possible that
Although the focus is frequently on surgical 5% of blood acquisition costs will be needed to
patients, PBM encompasses the entire scope of support the personnel to operate the program.
any patient's hospital experience. It includes: For larger hospitals or health systems, perhaps
2% to 3% of the annual blood acquisition budget
1. Efforts to identify and manage anemia and will be needed to support the program. To justi-
bleeding risks before any treatment begins. fy this expense, one must consider the potential
2. Use of blood-sparing surgical techniques and return on investment from the reduction in
intraoperative blood recovery methods. blood acquisition cost, which has been reported
3. Adjunctive strategies during intensive care to be as much as 400%-indicating five dollars
unit (ICU) stays and postoperative care that saved for every one dollar spent." Not to be ig-
nored are the substantial time and effort re-
decrease the need for transfusion.
quired from the information technology team to
4. Blood utilization review for transfusion guide- collect and analyze data on compliance with
line compliance and feedback to ordering transfusion guidelines, format dashboards, and
physicians. report results. Such activities are critical to im-
5. PBM education for all health-care providers proving performance. Of course, electronic re-
involved in patient care. cords help in the data collection process, but
skilled programmers are often in high demand,
and these individuals are essential to the success
u R of the PBM efforts. Without dedicated support,
the programmers will likely not put PBM high
on their priority list.
One of the most appealing concepts of PBM is The best PBM programs have representation
that it can reduce risk, improve outcomes, and from many departments within the hospital and
save money, all at the same time. Furthermore, thus are truly multidisciplinary. For example,
by improving quality and reducing costs, PBM members should be included from hospital ad-
increases the value of health care delivered. De- ministration, nursing, hospitalists, surgical ser-
C HAP T E R 2 0 Patient Blood Management 585
vices, quality and safety, the blood bank, infor- vancement of Blood Management (SABM).12
mation technology, hematology, critical care, These standards include similar requirements to
anesthesiology, pharrnacy.. and finance. Presen- operate a high-quality PBMprogram.
tation of the business plan to the top administra-
tors is helpful to gain financial support and buy-
in from leadership.
Support for PBM often falls into the category
of safety and quality efforts. If the institution has
a safety and quality institute or division, one can Education
make the case that, given the risks and expenses
of transfusion and the potential return on invest- The general methods used to implement and
ment in reduced blood-acquisition costs, PBMis sustain a strong PBM program are outlined in
an important effort to support. If one considers Table 20-2. It is intuitive that education is per-
the medical, legal, and financial implications of haps the most important component of any
givingthe wrong unit to a patient and the associ- quality improvement effort. Even educated clini-
ated risks of a life-threatening hemolytic reac- cians are unlikely to be familiarwith each of the
tion, the initiative is even more likely to attract nine large randomized trials that have been pub-
support from the safety and quality division. lished, most in the past decade,13'21all in sup-
PBM can also fall into the category of "supply port of arestrictive (Red Blood Cell) RBCtrans-
chain management," which is designed to opti- fusion strategy. Each of these studies supports a
mize and reduce spending on equipment and hemoglobin threshold lower than those tradi-
supplies-in this case the blood-acquisition cost tionally used (Table 20-3). In fact, it is likely that
and other activities related to transfusion. more high-profile,prospective clinical trials have
been conducted on hemoglobin transfusion
thresholds than on almost any other practice in
medicine. All nine of these studies were pub-
D lished in high-impact journals. Each of the trials
supports a threshold of 7 g/dl; even for critical-
ly ill patients,15.18,20
or 7.5 to 8 g/dl, for those
The AABBStandards jor a Patient Blood Man- with cardiovascular disease.13,14,19,21 When these
agement Program (PBM Standards) was created restrictive thresholds were compared to the lib-
and is updated by a consensus panel of experts eral thresholds of 9 to 10 g/ dL, the clinical out-
in the field.11The PBM Standards forms the ba- comes (including both primary and secondary
sis for a PBM certification process for hospitals outcomes) were the same in five of the stud-
that is jointly offered by AABB and The JOint ies,13,14,16,19,20
meaning that the extra blood was
Commission. Certification is provided .on three not helpful. In the.otherfo1.1I'studies, outcomes
levels, according to the extent of PBM-related were worse with liberal thresholds (either over-
activity at a facility. As indicated in Table 20-1, all or in some subgroups),15,17,18,21 meaning that
level 3 status is achieved when the top 17 re- the. extra blood was actually harmful. Through
quirements are met; level 2 with the top 20; simple educational efforts, clinicians can be
and level 1 with all 24. To receive level 1 certifi- made aware of these landmark studies. As a re-
cation, hospitals must have the capacity to per- sult, blood utilization will be reduced and pa-
form autologous blood recovery (use of a cell- tient outcomes improved with restrictive thresh-
saver device); have a formal program to render olds.
quality care to those patients who will not ac- It should be noted that the randomized trials
cept transfusion (also called a bloodless pro- emphasize the hemoglobin "threshold," which
gram); and use methods to identify and manage is the level before transfusion, rather than the
preoperative anemia in both medical and surgi- hemoglobin "target," a term that has been used
cal patients. Other standards have been released to refer to the level achieved after the transfu-
by other societies, such as the Society for Ad- slon." In Table 20-3, it is clear that in these
TABLE 20-1. PBM Program Levels"!'
Item
Evidenceof institutional support for the patient blood manage- X X X
ment program at the hospital administration level.
2 Metrics regardingtransfusion appropriatenessin accordance X X X
with transfusion guidelines.
3 Documentation of transfusion including patient consent, X X X
observation, adverseevents and outcomes.
4 Budgetingto the level of care required by the implementation X X X
of the AABB Standards for a Patient Blood ManagementPro-
gram
5 Pretransfusion patient testing and evaluation. X X X
6 Patient- or case-specificassessmentof potential blood usage. X X X
7 Preproceduralblood ordering including completion of type X X X
and antibody testing before procedure start time with a plan
for antibody-positive patients.
8 Preprocedureoptimization of patient coagulationfunction. X X X
9 Monitoring of blood componentswastageand cause. X X X
10 Minimize blood loss due to laboratorytesting. X X X
11 Processfor managingthe blood needsof unidentified patients X X X
and resolving their identification.
12 Processesto identify, prior to or upon admission, patients X X X
who may declinetransfusion under any circumstances with
notification to the appropriate individuals.
13 Massivetransfusion protocol useand compliance. X X X
14 Transfusion care and anemia managementof pre-term, neo- X X X
nate, infant, and pediatric critical care patients, if applicable.
15 PBM for obstetric patients including postpartum hemorrhage X X X
protocol with evidenceof its use, plan(s) for patients with
known high bleeding risk (eg, placentalabnormalities), and
plans for patients where blood is not an option.
16 Single-unit transfusion strategiesfor defined patient popula- X X X
tion(s).
17 Manageacquired coagulopathy. X X X
18 Blood conservation strategiesfor high blood usageservice X X N/A
lines.
19 Processesand/or equipmentto facilitate rapid decision mak- X X N/A
ing with regard to anemiaand coagulation management.
TABLE20-1. PBM Program Levels"!' (Continued)
Evaluating and managing iron and micronutrient deficiencies

in defined patients with Red Blood Cells ordered in the outpa-
tient setting.
21 Evaluation and management of anemia in nonoperative x
patients.
22 Program to care for patients who decline use of blood or x
blood-derived products.
23 Identification and management of presurgical anemia before x
elective procedures for which type and screen or type and
crossmatch is recommended.
24 Use of perioperative techniques consistent with current AABB x
Standards for Perioperative Autologous Blood Collection and
Administration.
* A patient blood managementprogram can be designatedas a program activity level 1, 2, or :3 program. To be designateda
specific activity level, the program shall be responsible for or have direct involvement with oversight and monitoring of the
activities listed above. [Usedwith permission from AABB Standards for a Patient Blood Management Program.11(PP3-5)j
RBCs= Red Blood Cells.
trials, the target is about 1 g/dl. higher than the newsletters, and e-mailscan be easilyignored or
threshold, and this difference should be consid- deleted and therefore are likely to be less effec-
ered when defining evidence-based transfusion tive than an in-person lecture. With regard to
practice. The actual dose of blood is the primary blood components other than RBCs, the evi-
determinant of the target; thus, a popular PBM dence is less complete. For plasma, platelets,
campaign called "Why give 2 when 1 will do?" and cryoprecipitate, the primary outcome of
was created to encourage single-unit RBCtrans- concern is bleeding, which is difficult to mea-
fu~ions in nonbleeding, hemodynamically sta- sure and thus difficultto study. However, guide-
ble patients. In fact, eight of the nine clinical tri- lin~s do exist for plasma" and. platelets" that
als mentioned above specifically called for the promote what evidence base there is, in terms
use of single-unit RBCtransfusions, followed by of reducing unnecessary transfusions. But many
reassessment, before administration of addltlon- of these recommendations are based on evi-
al units. The Choosing Wisely campaign, de- dence rated as weak or very weak because ade-
signed to reduce unnecessary tests and proce- quate randomized trials on indications for these
dures, includes six different societies that have components are lacking.
aims to reduce unnecessary transfusions. MBB, The Johns Hopkins Health System has creat-
for example, emphasizes the evidence-basedhe- ed an electronic tutorial to educate clinical pro-
moglobin threshold of 7 to 8 g/dL and the im- viders on blood management principles. The tu-
portance of giving single-unit RBC transfusions torial covers everything from the definitions for
in its Choosing Wisely aims." type and screen and crossmatch and who needs
The various venues for education are many. them preoperatively, to the hospital guidelines
A well-delivered lecture showing evidence from for transfusion thresholds for RBCs,plasma, and
the randomized trials is probably the most effec- platelets. It also covers transfusion reactions,
tive method of education. Online tutorials, how to properly label specimens for blood
TABLE 20-2. Methods for Implementing a Patient Blood Management Program*

1. Obtain support from health system leadership (business plan).
2. Assemble a multidisciplinary team of stakeholders.
3. Educate(with emphasis on the nine RCTssupporting restrictive transtuslonj.P"
4. Harmonizetransfusion guidelines.
5. Add decision support for computerized provider order entry (with best practice advisories).
6. Implement data acquisitionianalytics.
7. Create dashboards."
8. Provide transfusion guideline compliance audits with feedback (reports) to providers.
g. Implement methods to improve blood utilization:
Evidence-basedtransfusion thresholds.
"Why give 2 when 1 will do" ChoosingWisely campaignfor RBCs.23
~ Preoperativeanemia management."
Antifibrinolytics (eg, aminocaproic acid,tranexamicacid).
Intraoperativeautologous blood recovery(Cellsavert instrument).
Anestheticmanagement(autologousnormovolemichemodilution,controlled hypotension,normothermia).
Surgical methods (newer cautery methods,topical hemostatics,and sealants).
Reductionof phlebotomy blood loss (smaller tubes, eliminate unnecessarytesting).
Point-of-caretesting (eg, thromboelastography).
'Modified from Franket al."
'Haernonetlcs Corp.
RCT= randomized controlled trial; RBCs = Red Blood Celis.
preparation orders, and a short review of the scribed above, before being granted generalized
ABO and Rh blood groups. It introduces the privileges to practice. Extremely few hospitals,
maximum surgical blood order schedule however, have specific privileges for ordering a
(MSBOS),30identifies where this list of surgical transfusion. Instead of stipulating specific cre-
procedures can be found, and explains how it is dentials for transfusion, the PBM Standards"
used to determine appropriate preoperative states that "individuals who order and/or trans-
blood orders.
fuse blood shall meet facility defined require-
Qualification/Competence
ments for education related to patient blood
management." 11(p8) This statement implies that
An area of debate is the use of qualifications or hospitals should have some requisite education
credentials to grant hospital privilegesto individ- forproviders who order transfusions.
ual providers for administering blood to pa-
tients. It has been said that historically, all one Preoperative Strategies
needed was a pen and paper, or now a computer
and keyboard, to order blood for a patient, with Although PBM methods apply to both medical
little or no specifictraining. Some hospitals have and surgical patients, for the latter they can be
required that new staffbe trained in PBM, as de- categorized according to their relevance in the
CHAPTER 20 Patient Blood Management 589
:c
Q)
:::J
It)
co
(V)N m en Cf) Cf) N -er c::
en "<t m en Cf)
C! "<I: :;:::;
0 0 0 c::i 0 z z z 0 0
c::
0
2-
;:,!? ;:,!? ;:,!?

;:,!?
c ;:,!?
c ;:,!? ;:,!?
o c ;:,!?
c
e
C'!
o
CD "#. ;:,!?
e
c
0
0 CD
LO CD It) It) r--.:
N r--.:
en l!)
(V) -e-
.~
, ca
t:::
c.>
, Q.)
en
"C "C
c:: 0'- c:: :::l
Q.) E ij5 Q.) ro
.E:l
'en
>,0>
ro .2 .:.::
Q.)
:t:::
en
:r.
cap
o
0.
D"'C-U
, .... 0
oro=
o
c. ..... >, =
ro ro
"Ct:::
E (V)(.)en E c:: , 0
o o 0
(.) (.) 0. ~ E
(J)
"'0
o
..c=
(J)
e
.s:
l-
e
o
'w
=:l
"iii It)
e
~
f-
U
co
o
o
.....
-
o
o
o
+-'
C!
o
.s
O"l
o-i
o
+-'
en
a:
e
o
(J)
LO
~
.;::
LO ..- o-i N LO
ex) o-i o-i -:
f- o
..... o
..... a o
..... o
"'0
Q) r-, co o r--..
+-'
r--..
N ex)
'E
o
"'0
e
C1:S
a:
Q)
>
:g
Q)
c..
(J)
e
0..
Q)
e'
C1:S
...J
00 Cl
N co C') (/)
ci z ci z
~
o
LO
N
::c-
Cl)
::s
e::
+=i
e::
o
S
en ":
"C
"0
.s: ~
=
CI
e a:=
.
en
"CI
Q)
.s::::
l-
e::
o
-e
"w
::s
en
e::
l- o
+-'
00
ci
o
+-'
Cl
.s
Cl
.s
Cl LO
e...:> LO LO 0) 0)
co 0) 0)
c:::
e::
o
en
ctl
"C t- t- 0) 0)
I- o) eo .s .s
"C
Cl) .s o
+-' LO LO
N t- r-; r-..: r-..:
"E
o
"C
e::
~
c:::
Cl)
>
:g
Cl)
C-
en
o
.....
0... "lD.o
Cl) ",:I:
e>
~
c::
_J ~~
~
LO
U"J"O
E g
II ~2i5
"0"0
'"
\;::cc
:a
'"II
0(.;)
:2 co
• a::
C HAP T E R 20 Patient Blood Management 591
pre-, intra-, and postoperative periods, as out- with caution in patients who have a history of
lined in the sections below. thrombosis, ischemic stroke, uncontrolled hy-
pertension, seizures, or cancer." Hence, the
Preoperative Anemia Diagnosis and risk/benefit ratio must be carefully considered
Treatment when prescribing ESAs. Additionally, because
ESAscause functional iron deficiency, adminis-
One important and especially challenging area tration of concomitant iron therapy will help
in PBM is the timely diagnosis and treatment of minimize the lowest effectivetotal dose of ESAs
preoperative anemia. Treatment is especiallyim- needed to see an adequate response."
portant for patients who are undergoing elective One of the most challenging aspects of diag-
surgery, as bringing such patients to the operat- nosing and treating preoperative anemia is hav-
ing room with untreated anemia represents sub- ing enough time before surgery to achieve the
optimal careY-33Anemia has been shown to be goals. Often, preoperative laboratory tests are
an independent predictor of increased perioper- ordered 3 days before the surgery date, leaving
ative morbidity and mortality" and should be little or no time to correct anemia. Of course, for
considered a modifiable risk factor. Thus, elec- semi-urgent surgeries, the options are limited,
tive surgery should be delayed when possible to but for truly elective cases, the preoperative
allow for diagnosis and adequate treatment. tests should be performed 4 or more weeks be-
Some centers have set up preoperative anemia fore the surgery date, giving adequate time for
clinics that are designed to optimize the condi- proper diagnosis and treatment of anemia." Ad-
tion of patients before surgery. Their goal is to ditionally, when appropriate, providers should
improve outcomes and reduce overallcosts." rule out other medical causes of anemia, such as
First, it is important to determine the cause a gastrointesttnal malignancy, when iron defi-
of anemia. Simpleiron deficiencycan be treated ciency anemia is detected. Most patients will re-
with either oral or intravenous iron. Patient spond well to intravenous iron within 3 to '4
compliance with oral iron is low because of sig- weeks, but they will respond even more dramat-
nificant gastrointestinal side effects. Moreover, ically and rapidly when ESAs are given along
oral iron is both poorly and slowly absorbed.
with the iron."
Therefore, many experts in the field advocate
for intravenous iron therapy to allow for more
Maximum Surgical Blood Order Schedule
rapid resolution of anemia.35,36The newer com-
pounds can be administered in one or two high One of the key components of a strong PBM
doses for complete iron repletion.37,38 They also program is a data-driven protocol for determin-
have a lower incidence of adverseevents than the ing which patients need preoperative blood
older iron compounds such as high-molecular- orders. The concept of the MSBOSwas first de-
weight iron dextran. The specific formulation veloped in the mid-1970s to prevent the over-
and dose should be determined by the patient's ordering of blood before surgery-hence, the
total iron deficit, institutional formulary, and term "maximum" surgical blood order sched-
also cost and coverage by insurance plans. For ule." The main concern is that many institu-
specific patients, erythropoiesis-stimulating tions have an outdated MSBOSthat is based on
agents (ESAs)may be indicated for treating pre- consensus opinions rather than on actual blood
operative anemia. Two concerns regarding the utilization data for specific surgical procedures.
use of ESAs are difficulty with reimbursement Now, with the popularity of electronic anesthe-
and the Food and Drug Administration (FDA) sia records, actual institution-specifictransfusion
"black box" warning about thrombotic. events data can be used to generate a more accurate list
and promotion of tumor growth.3~,40. Use of of recommended preoperative blood orders. Us-
pharmacologic therapy such as low-molecular- ing an algorithm that includes three variables-
weight heparin for venous thrombosis prophy- the percentage of patients who received transfu-
laxis may. decrease the risk of thrombosis in sion, the median estimated blood loss, and the
postoperative patients, but ESAsshould be used averagenumber of units transfused per patient-
methods have been described for creating an Optimizing Coagulation

institution-specific MSBOS derived from elec-
An important way to reduce blood loss and un-
tronic data." The actual document for that necessary transfusions is to optimize coagulation
MSBOS, created as a guide for preoperative before surgery. For example, P2Y12 inhibitors
blood orders, includes 135 types of surgical pro- such as clopidogrel should be discontinued, if
cedures and the associated recommended blood possible,in time for their effect to subside before
order for each (Fig 20-1). Of course, these rec- elective surgery. Often, a cardiac surgery patient
ommendations can be modified-for example, needs 2 to 5 days offthe medication for coagula-
in patients with preoperative anemia or in those tion to normalize. Tests such as the Verify Now
with red cell antibodies for whom compatible assay (Instrumentation Laboratories]" can de-
units may be difficultto find. tect residual P2Y12 inhibition, enabling the pro-
It has been shown that a data-driven MSBOS vider to determine the optimal time for surgery.
not only improves the blood ordering process Because the return to normal coagulation has
but can also decrease costs by reducing unnec- significant variability when these drugs are dis-
essary blood orders ($150,000-$300,000/ continued, the test is important. Additionally,
year].' The crossmatch-to-transfusion ratio, a several over-the-counter herbal supplements,
classicmeasure of blood-ordering efficiency,can such as garlic, ginseng, and ginkgo, have been
be improved (decreased) by using an accurate shown to affect coagulation and should be dis-
MSBOS.7 For those procedures in which blood continued before elective surgery."
is rarely or never transfused, the authors specify
that no preoperative blood orders are needed. In Preoperative Autologous Blood Donation
the case of unexpected bleeding, the backup Historically, preoperative autologous blood do-
plan is emergency-release, uncrossmatched nation (PAD) was used in an attempt to avoid al-
blood, which is much safer than many clinicians logeneic blood. However, over the last decade,
believe." there has been a significant downward trend in
Having an up-to-date MSBOShas other bene- the number of autologous units collected in the
fits as well. First, blood units will not be set United States. In 2017, only 10,000 units were
aside unnecessarily for cases that have a low collected, representing approximately 0.08% of
likelihood of transfusion. Overordering of preop- the total allogeneic RBC/whole blood collection
erative crossmatches and setting aside RBC and 62% fewer units than were collected in
units leads to potential outdating and wastage. 2015.4,49 Major factors contributing to this de-
On the other extreme, patients who truly need cline include the increased safety of and public
blood prepared are more likely to have blood confidence in the blood supply, adoption of in-
units ready when they are needed. When cases traoperative blood-conserving techniques, high
are identified that clearly should have blood wastage of PAD blood (>45% discarded), and a
higher risk for preoperative anemia after dona-
ready to transfuse, the process of type and
tion.49,SO Studies showed that although patients
screen or type and crossmatch is best completed
participating in a PADprogram had lower expo-
before the day of surgery, thereby decreasing the
sure to allogeneic transfusions than did patients
risk that surgery will begin before the blood is who did not participate, they had a higher likeli-
ready. The Joint Commission has recognized hood of receiving any transfusion (allogeneic
this particular problem as a potential perfor- and/or autologous) as a result of donation-
mance measure;" and use of an MSBOShelps induced anemia.49,S! Errors related to produc-
to reduce the problem by specifyingwhich pation and handling, delays in receipt of the units
tients need blood prepared ahead of time. Many at the designated hospital, and increasing acqui-
centers now use the 30-day time limit for expi- sition costs also added to the decrease in PAD.52
ration of the type and screen or crossmatch, as The patient also may accrue additional cost in
long as the patient has not been pregnant or re- the form of lost wages if work time is required
ceived transfusion within the last 90 days. for the donation.
CHAPTER 20 Patient Blood Management 593
SURGICAL BLOOD ORDER SCHEDULE

Cardiac Surgery Obstetrics Thoracic Surgery
Case Category Rec Case Category Rec
Heart or lung transplant T/C4U
Case Category Rec Esophageal open T/C2U
Complex Cesarean
Minimally invasive valve T/C4U T/C4U Sternal procedure T/C2U
(Accreta, Percrera, Previa, etc.I
Revision sternotomy T/C4U Repeat Cesarean T/C2U Chest wall T/C2U
CABG/valve T/C4U Routine Primary Cesarean Thoracotomy T/C2U
T/S
Valve T/C2U Pectus repair T/C2U
Vaginal Delivery T/S
Assist device T/C4U VATS T/S
D&CjD&E!Genetic Termination T/S
Cardiac/major vascular T/C4U Mediastinoscopy T/S
Tubal ligation No Sample
Open ventricle T/C4U EGO/fOB No Sample
CABG T/C2U Cerlage No Sample Central venous access No Sample
. Cardiac wound surgery T/C2U Orthopedic Surgery
Percutaneous cardiac T/C2U Urology
Case Category Rec
Pericardium T/C2U Thoracic/lumbar/Sacral fusion T/C4U Case Category Rec
lead extraction T/C4U Pelvic orthopedic T/C4U cystoprostatectomy T/C2U
AICO/pacemaker placement T/S Open hip T/C2U Urology open T/C2U

Femur open (fracture) T/C2U Nephrectomy open T/C 2U
General Surgery Above/below knee amputation T/C2U
Rec ; lap/Robotic kidney/adrenal T/S
CaseCategory Total hlp arthroplasty T/C2U
AP resection T/C2U RRP T/S
Humerus open T/S
Intra-abdominal GI T/C2U Percutaneous nephrolithotomy T/S
Fasciotomy T/S
Whipple or pancreatic T/C2U Robotic RRP
Shoulder Incision & Drainage T/S No Sample
Liver resection T/C2U
Tibial/fibular T/S External genitalla/Penife No Sample
Retroperitoneal T/C2U
Total knee replacement T/S TURP No Sample
Substernal T/C2U
Shoulder open T/S
Liver resection minor T/S Cysto/ureter/urethra No Sample
Knee open T/S
Bone marrow harvest T/S TURBT No Sample
Thigh soft tissue No Sample
Hernia - ventral/Inctstoriat T/S
Ortho external fixation No S~Il1p'I~
-'. Vascular/Transplant Surgery
Hernia -InguInal/Umbilical No Sample
Peripheral nerve/tendon No Sample Case Category Rec
Appendectomy No.Sample
lower extremity 1&0 No Sample
Abdomen/chest/soft tissue No Sample liver transplant T/C6U
Hand orthopedic No Sample
Lap..or open cholecystectomy No Sample Thoraccabdomlnal aortic T/C 12U
Upper extremity arthroscopy No Sample
Thyroid/parathyroid No Sample Major liver resection T/C4U
Upper extremity open No Sample
Central venous access No Sample Major vascular
Podiatry/Foot No Sample T/C4U
Any Breast - except w/flaps No Sample
Hip dosed/percutaneous No Sample Exploratory lap. vascular T/C4U
Gynecological Surgery lower extremity arthroscopic No Sample Kidney pancreas transplant T/C 2U
Case Category Rec Shoulder dosed No Sample
Major endovasculer T/C2U
Uterus open (radical) T/C2U Tibial/fibular dosed No Sample
Open pelvic T/C2U Above/below knee amputation T/S
Uterus/ovary open T/S Otolaryngology Surgery Nephrectomy/kidney transplant T/C2U
Total vaginal hysterectomy T/S Case Category Rec Organ procurement T/C2U
Hysterectomy robot/lap T/S Laryngectomy T/C2U
Peripheral vascular T/C 2U
Cystectomy robotic assisted T/S Facial reconstruction T/C2U
Cranial surgery T/C2U Vascular wound I and D T/C2U
Cystoscopy No Sample
Bxternetgenttella No Sample Radical neck dissection T/S Carotid vascular T/S
GYN cervix No Sample Carotid body tumor T/C2U AVfjstula T/S
Hysteroscopy No Sample Mandibular surgery T/S Peripheral endovasculat Tis
Superficial wound No Sample Neck dissection T/S No Sample
Angio/Arterlogram
Mastoidectomy No Sample
Neurosurgery Peripheral wound I&D No Sample
Parotidectomy NoSampJe
Case Category Rec No Sample 1st rib resection/thoracic outlet No Sample
Facial plastic
Thoracic/lumbar/Sacral fuslcn T/C4U No Sample
Oral surgery No Sample Superflc1alorskln
Spine tumor T/C2U No Sample
Sinus surgery Foot/toe amputation/debride No Sample
Posterior cervical spine fusion T/C2U Thyroid/parathyroidectomy NoSampte
Central venous access NoSampie
Spine Incision and Drainage T/C2U Suspension laryngoscopy No Sample
Intracranial tumor / aneurysm T/C2U Bronchoscopy No Sample
Laminectomy/discectomy T/S Cochlear Implant No Sample
Spine hardware removal/biopsy EGO No Sample
T/S
External ear No Sample
ACDF T/S
Inner ear No Sample
Extrecranlal Nc Semple
Tonsillectomy/adenoidectomy No Sample
Nerveprocedure No Sample No Sample
Tympanomastoid
.CSF/shunt procedure No Sample
Updated Sept.2018
FIGURE 20-1. An institution-specific maximum surgical blood order schedule (MSBOS) derived by collecting
blood utilization data from an anesthesia information management system." This list specifies recommended
preoperative blood orders for different types of surgical procedures. (Modified from Frank et aL3D)
TIC = type and crossmatch; TIS = type and screen; U = unit.
Despite these concerns about efficacy, PAD duce contamination risk significantly,however,
can be a reasonable option for patients with rare and currently avallable literature does not show
blood types or multiple red cell alloantibodies. worse outcomes when recovered blood is used
In these situations, advance planning and pain these cases." Another limitation is that some
tient evaluation are crucial before PAD is at- smaller hospitals do not have personnel on-site
tempted. To mitigate the anemia induced by to operate the Cell Savers. Such institutions of-
PADand to avoid allogeneic transfusions, efforts ten need to call in an outside contractor, which
should focus on 1) timing of collections to allow requires advance planning and increased costs.
3 to 4 weeks between the last donation and One workaround for this limitation is to "collect
planned surgery, 2) collecting the minimal only" with a collection reservoir and anticoagu-
amount, and 3) prescribing iron-replacement lant (Citrateor heparinl'"; then, when the quali-
therapy with or without an ESA before dona- fied personnel are avallable, the shed blood can
tion. be processed through the machine. In some hos-
pitals, the anesthesia or nursing staff are trained
Intraoperative Strategies to operate the machines, but in others the perfu-
Autologous Blood Recovery sionists who operate the extracorporeal bypass
equipment for cardiac surgery are responsible
Intraoperative autologous blood recovery, one of for processing recovered blood.
the earliest blood conservation methods, be- The use of autologous blood recovery has
came well established in the late 1970s.53 The several advantages. Evidence shows that when
Cell Saver (Haemonetics) uses a specialized one or more units are returned to the patient,
bowl to collect and wash debris from the shed intraoperative blood recovery adds economic
blood. This method became popular in the value.57,58 Additionally, recovered red cells are
1980s, when patients strongly preferred to have likely to be higher quality than stored (banked)
their own blood rather than banked blood, pri- RBCs.59 Because recovered red cells do not suf-
marily to avoid HN. fer from "storage lesions," red cell membrane
In this procedure, shed blood is washed in a deforrnabillty'? and 2,3-diphosphOglyceratelev-
cone-shaped or cylindrical centrifuge bowl that els are near normal," whereas both of these pa-
concentrates the red cells, which are then trans- rameters are decreased in stored blood. Use of
fused back to the patient. The resulting product autologous blood also eliminates the risk for vi-
has a hematocrit similar to that of RBCs from ral transmission and alloimmunization. For all of
the blood bank and is devoid of plasma and these reasons, recovered red cells are usually
platelets. If enough recovered blood is adminis- preferred over allogeneic stored RBCs.More de-
tered (approximately 5 or more units), the pa- tails on intraoperative blood recovery appear in
tient will begin to develop a dilutional coagulop- AABB Standards for Perioperative Autologous
athy. Nonetheless, for some vascular, transplant, Blood Collection and Administration.62
orthopedic, and cardiac procedures, use of au-
tologous recovered blood has become a stan- Reducing Intraoperative Blood Loss
dard of care for blood conservatlon.t':" A prima-
ry limitation of blood recovery is that a cavity Another strategy for reducing transfusions cen-
where blood will pool is needed for optimal col- ters on decreasing intraoperative blood loss. Re-
lection and recovery of shed blood. Additionally, ducing intraoperative blood loss begins with me-
blood collection can be inefficient when more ticulous surgical technique, but a number of
than one suction source is used in the surgical other strategies to reduce bleeding have been
field, causing blood to be routed to the waste developed (see Table 20-2). Simplymaintaining
suction and not the Cell Saver. Concerns also normothermia by warming intravenous fluids
exist for potential contamination of the shed and the patient directly (eg, forced-airwarming)
blood with microbes, tumor cells, or amniotic will reduce bleeding. Even mild hypothermia
fluid during cesarean section. Washing and leu- (35 C) increases bleeding by approximately 20%
kocyte reduction filters have been shown to re- by inhibiting platelet function and the clotting
cascade." Another simple method of reducing great enough to make a difference. Because
intraoperative blood loss is controlled hypoten- these three requirements are not always pres-
sion, which is especially effective in orthopedic ent, whether the ANH technique reliably reduc-
and spine surgery. By increasing anesthetic es the need for allogeneic transfusion remains
depth and/or administering potent vasodilators, controversial. A recent meta-analysisof 63·ran-
the provider can carefully reduce blood pressure domized trials showed that although ANH de-
while maintaining vital organ perfusion with a creased the likelihood of transfusionby 26% and
mean arterial blood pressure above the autoreg- decreased the volume of allogeneicblood trans-
ulation threshold. Excess crystalloid should be fused by about 1 unit, possible publication bias
avoided because the resulting hemodilution may have Jed to an overestimate of the bene-
leads to decreased hemoglobin levels. Excess fits." Many small trials were included, and
crystalloid can be avoided by administering col- many lacked a transfusion protocol or threshold,
loid to expand intravascular volume (eg, albu- which in these unblinded studies can lead to bi-
min) and/or low-dose vasoconstrictors (eg, as. The authors of an editorial concluded that
phenylephrine) to treat anesthetic-induced hy- ANH may be beneficialfor specificcases that in-
potension. Topical hemostatic agents, such as fi- corporate all three of the above-mentionedcrite-
brin, thrombin, gelatin, collagen, and bone wax ria and that patients might benefit most by re-
have been shown to aid in hemostasis." Com- ceiving fresh coagulation factorsand platelets in
mercially available combination products such the whole blood that is reinfused.Thus, it might
as Floseal (Baxter Healthcare Corp) contain bo- be most useful for patients undergoing major
vine gelatin and human thrombin in the appro- procedures, such as cardiac surgery, because it
priate ratio to optimize hemostasis. Newer cau- prevents the whole blood from being subjected
tery methods, such as a saline-irrigated bipolar to cooling and prevents potential damage to
cautery, or the harmonic scalpel, which cauter- platelets by the cardiopulmonary bypass rna-
izes vessels as it cuts, can also effectively reduce chine."
intraoperative bleeding.t" Some evidence sug-
gests that neuraxial anesthesia(spinal or epidur- Minimally Invasive Surgical Approaches
al) may reduce bleeding by about 20%, perhaps In the past two decades, new surgicalapproach-
by reducing venous and/or arterial blood pres- es have been introduced} including laparoscop-
sures." tc, robotic, and endovascular techniques, that
have dramatically reduced blood use. For in-
Acute Normovolemic Hemodilution stance, Johns Hopkins researchers found that
Acute normovolemic hemodilution (ANH) is a only 1 in 800 patients undergoing robotic pros-
technique that involves phlebotomy of 1-4 units tatectomy received a transfusion," whereas his-
of whole blood before the blood-lossportion of torically, ..the vast majority of patients ..•
who
the surgery, and intentional hemodilution with underwent open prostatectomy received trans-
crystalloid and/or colloid.67,6B Because this pro- fusion.66,71 Similar impact has also beep recog-
cess creates a state of intraoperative anemia, the nized withIaparoscopic and robotic methods of
shed blood contains fewer red cells. Near the gynecologicsurgery,for..example with hysterec-
end of the procedure, when most of-the antici- tomyand myolllectomyprocedures. Initially,ro-
botic and other minimally invasive procedures
pated blood loss is complete, the phlebotomized
were marketed for reducing pain and length of
blood is reinfused. For ANH to effectively re-
stay and enabling earlier return to work. How-
duce transfusion requirements, all three of the
ever, these ~pproacheshave also dramaticallyre-
followingconditions must be met: the preopera-
duced the need for transfusion.
tive hematocrit must be sufficiently high that
the patient can tolerate the phlebotomy and he-
Antifibrinolytic Medications
modilution; the blood loss during surgery must
be substantial enough to obtain the benefits; and Antifibrinolytic drugs such as tranexamic acid
the volume of phlebotomized blood must be and aminocaproic acid were introduced almost
50 years ago, but only in the past decade have However, the ideal dose has not yet been deter-
they gained popularity for reducing periopera- mined. Recent findings suggest that 3 to 5 mg/
tive blood loss and transfusion. Tranexamic ac- kg/h may provide a steady state and efficacious
id, especially, has been called a "game changer" therapeutic level.81,82The contraindications to
at some of the national orthopedic and blood systemic tranexamic acid include uncontrolled
management meetings. It is quickly becoming seizures or an active thrombotic event, but a
the standard of care for certain procedures, de- history of these conditions is not thought to be a
spite its use to reduce surgical bleeding being contraindication. Some centers are applying
considered "off-label." Multiple studies have tranexamic acid topically into the joint capsule
shown that antifibrinolytic drugs reduce bleed- in such patients undergoing hip and knee ar-
ing, transfusion, and cost for spine surgery, hip throplasty to minimize systemic levels of the
and knee arthroplasty, and cardiac surgery.72-75 drug. Efficacywith topical use has been demon-
Overall, compared to placebo, the studies collec- strated."
tively show that tranexamic acid reduces blood
loss and transfusion requirements by approxi- Point-at-Care Testing
mately 30%. The drug is thought to stabilize clot
that has already formed, by preventing its break- When turnaround times for laboratory tests are
down (fibrinolysis). The risk of deep venous long, clinicians often decide to administer trans-
thrombotic events does not appear to increase fusion to the patient before receiving the test re-
with these drugs, even in the three largest clini- sults. This is especially true with plasma or
cal trials with placebo control groupS.76-78 In platelets because laboratory tests for coagulation
studies of hemorrhaging trauma patients (the and platelet counts take longer than hemo-
CRASH-2Trialf9 and postpartum hemorrhage globlin measurements. With point-of-care test-
(the WOMAN Trial)," each with 20,000 pa- ing, which has a rapid turnaround, the clinicians
tients, investigators found a reduction in mortal- will not have to make blind decisions about
ity when tranexamic acid was administered whether or not to administer blood compo-
within 3 hours of bleeding onset; however, no nents. An example of point-of-care testing is
benefit was observed after this 3-hour win- thromboelastography (TEG) (Haemonetics) or
dow." The CRASH-2trial showed a 9% reduc- rotational thromboelastometry (ROTEM) (In-
tion in overall mortality and a 15%reduction in strumentation Laboratories), which can give
mortality from hemorrhage. In the WOMAN tri- meaningful results on coagulation function
al, overall mortality decreased by 19%,and mor- within 10 to 15 minutes, or even fasterwith the
tality from hemorrhage decreased by 31%. rapid-TEG. The results from these tests reflect
These studies included many hospitals in devel- not only the number of platelets but also their
oping nations, and one limitation to consider is functional integrity, which is more clinicallyrel-
that blood is sometimes unsafe or unavailable in evant than a platelet count alone. Undoubtedly,
such areas, perhaps increasing the impact of point-of-caretesting is an important component
tranexamic acid on mortality. Some centers use in a PBM program and can reduce unnecessary
antifibrinolytics only when hyperfibrinolysis is transfusions."
evident based on data from viscoelastic testing.
This practice seems appropriate but can delay Postoperative Strategies
treatment beyond the important 3-hour window Postoperative Blood Recovery
of effectiveness. Dosing tranexamic acid is also
somewhat controversial. A l-g loading dose is Postoperative blood recovery involves collecting
now common for adult patients undergoing to- and reinfusing blood from surgical drains and/
tal joint replacement surgeries and was the dose or wounds. Adequate amounts of blood need to
used in the two above-mentioned trials. For lon- be collected and processed for this strategy to be
ger surgeries, such as some spine procedures, effective. Thus, it is used mainly in trauma and
this loading dose is often followed by a continu- vascular, cardiac, and complex orthopedic surgi-
ous infusion that ranges from 1 to 10 mg/kg/h. cal cases for which the shed blood volume can
be substantial (::::500 mL). Blood recovered post- useful for reducing blood loss, and in-line devic-
operatively can be unwashed or washed. When es that return the wasted discard in a sterile
unwashed recovered blood is used, shed blood fashion are also helpful. One of the IeUs, the
is collected and filtered until a sufficient volume neurocritical care unit, was able to cut blood
is reached; then it is transferred to an infusion loss in half by using an in-line return device (Fig
bag for reinfusion. Alternatively, once sufficient 20-2). Simplyeliminating unnecessary laborato-
shed blood is collected, it can be processed by ry tests is also important, as for some patients,
washing and then transferred to a bag for reinfu- tests are ordered routinely with little or no clear
sion. rationale. In some cardiac surgery patients, up
In the past, reinfusion of unwashed shed to 1 or 2 units of blood can be lost just from
blood was popular as a blood conservation tech- phlebotomy during longer leU stays."
nique in joint replacement surgery. However,
the growing use of antifibrinolyticshas led to a Transfusion Thresholds
decline in surgical bleeding, making postopera-
tive blood recovery unnecessary.t-" In addition, A patient's response to anemia is highly individ-
unwashed shed blood is less desirable because it ualized and depends on ability to maintain ade-
quate oxygen delivery to tissues. Tolerance
has a hematocrit of 20% to 30% and contains ac-
depends on the patient's volume status, physio-
tivated clotting and complement factors, inflam-
matory mediators, cytokines, and fat particles logic reserve (including cardiac, pulmonary, and
renal function), and dynamics of the anemia. Pa-
that can increase the risk for febrlle reac-
tions.87,88 For patients with substantial postoper- tients with chronic anemia caused by chronic
ative blood loss, improved product quality and renal failure, slow gastrointestinal bleeding, or
menorrhagia often adapt physiologically to a
safety (eg, hematocrit of 60% to 80% with re-
lower hemogloblin level by increasing cardiac
moval of contaminants) can be achieved with
devices that wash and concentrate the postoper- output, heart rate, or stroke volume. However,
ative wound-drainage blood. Maintaining the rapid blood loss from surgical bleeding or trau-
competency of nursing staff and the higher cost ma often results in hemodynamic instability,
for these devices may be disadvantageous for shock, and other symptoms that require more
rapid volume replacement. Evidence suggests
some hospitals. However, for those centers en-
gaged in complex, high-risksurgical cases, using that the change in hemoglobin level (delta he-
postoperative autologous washed blood may re- moglobin) is a better predictor of adverse out-
comes than the absolute nadir hemoglobin
duce the need for allogeneicblood.
during a hospital stay.92Thus, chronic anemia
seems to be much better tolerated than acute
Reducing Phlebotomy Blood Loss
anemia.
It is well recognized that patients are prone to As discussed previously and according to the
iatrogenic blood loss from laboratory testing, es- studies shown in Table 20-3, strong evidence
pecially when they are in the leU, where fre- supports using a restrictive transfusion strategy,
quent laboratory tests are ordered.89,90 In addi- with lower hemoglobin thresholds than were
tion, the easy access to blood draws when used historically (7-8 g/dl, instead of 10 g/dL).
arterial and central venous catheters are present However, the literature lacks evidence for apply-
predisposes these leU patients to lose approxi- ing a restrictive threshold to patients with sub-
mately 1% or more of their circulating blood vol- stantial bleeding, those with cardiac or cerebral
ume per day to testing. Approximately half the ischemia, and patients with hematologic malig-
blood is lost when lines are cleared to draw an nancies undergoing cheInotherapy atld.stem cell
undiluted sample, and the rest is sent to the lab- transplantation. More importantly, an arbitrary
oratory. Figure 20-2 shows the average amount hemoglobin level or threshold should not be the
of blood lost per day from laboratory testing for sole driver for transfusion. Transfusion deci-
leU patients in five differentadult IeUs atlohns sions should be individualized and based not
Hopkins Hospital. Smallerphlebotomy tubes are oniy on the hemoglobin level but also on the
70
60
- blood for lab tests
50
- wasted discard
mL 40
day 30
20
10
o
NCCU SICU MICU CSICU WICU
FIGURE 20-2. The bar graph illustrates the averagevolume of blood that patients lose as a result of laboratory
testing in the five different adult intensive care units (ICUs) at the Johns Hopkins Hospital. The typical ICU
patient loses approximately 60 mLlday, which is just over 1% of total blood volume in an average-sizepatient,
or 2% of total blood volume for a small adult patient. In the NCCU,an in-line device is usedto return the blood
drawn to clearthe saline from the lines. This device has reducedtotal blood loss by approximately 50%.
NCCU= neurocritical care unit; SICU = surgical intensivecare unit; MICU = medical intensive care unit; CSICU=
cardiac surgical intensive care unit; WICU = Weinberg intensive care unit (primarily surgical patients).
patient's clinical signs and symptoms of anemia unit of blood can have varying effects on hemo-
and ability to tolerate and compensate for the globin and hematocrit, depending on the pa-
anemia." For example, there is some evidence tient's total blood volume and fluid shifts. Often,
that bleeding is more likelyin thrombocytopenic a single RBCunit provides an adequate response
patients when the hematocrit is ::0;25%,94al- and relieves symptoms. In 2014, when the
though this has never been specificallystudied AABB Choosing Wisely campaign" was
in a randomized trial. Intravascular volume launched, the first aim was "Don't transfuse
should also be carefully considered, as dilutional more units of blood than absolutely necessary."
anemia from excess intravenous fluids can lead This aim included a recommendation that physi-
to unnecessary transfusions,and volume-depleted cians administer single-unit RBCtransfusions to
or actively bleeding patients may need transfu- nonbleeding patients and then perform a clinical
sions at higher thresholds. More simply stated, reassessment before giving additional units. Im-
the recommendation is to treat the whole pa- plementing a single-unit transfusion policy can
tient and not just his or her laboratoryvalues. have a significant impact and may reduce over-
all blood utilization more than monitoring he-
Single-Unit RBe Transfusions
moglobin thresholds." The Johns Hopkins
Health System launched a "Why give 2 when 1
Traditionally, physicians were strongly encour- will do?" campaign" that resulted in a 50% de-
aged to order 2-unit RBCtransfusions, a practice crease in double-unit RBC transfusion orders
that evolved several decades ago when single- and an overall 20% decrease in RBCutilization."
unit transfusions were extensively criticized." A Even in leukemia patients who required multi-
ple RBC transfusions, a single-unit policy was cryoprecipitate transfusions that are endorsed by
safe and effective in a small randomized pilot the hospital's transfusion committee and medi-
study." One effective method of encouraging cal executive committee. Concurrent reminders
single-unit RBC orders was a custom-designed at the point of transfusion decision, such as pre-
screen-saver image displayed on all health sys- transfusion checklists or order sets with the in-
tem computer workstations (Fig20-3). stitution's transfusion guidelines and indica-
tions, can improve practice. Using computerized
Transfusion Guidelines and Clinical provider order entry (CPOE) systems with the
Decision Support capacity for clinicaldecision support (CDS),and
requiring physiciansto document the indication
Evidence-based transfusion guidelines are the when outside of guidelines, facilitates adoption
basis for improving transfusion practice. Key and improves transfusion practtce.v" An inter-
physicians from various specialties should be in- ruptive "pop-up alert," also known as a best
volved in formulating these guidelines. For suc- practice advisory, can be built into the order set
cessfulimplementation, the guidelines must be with logic to retnevethe most recent laboratory
combined with committed educational efforts. value. For example, the Johns Hopkins system
Issuing a memo of the transfusion guidelines to advocates a hemoglobinthreshold of 7 to 8 gI dL
physicians without follow-up typically fails to and single-unitRBCtransfusions in patients who
reduce blood usage. Ideally the guidelines in- are hemodynamically stable..and not actively
clude indications for RBC,plasma, platelet, and bleeding· (Fig 20-4). It has been shown that
.
gl
Single ,UnitRBC Transfusion
FIGURE 20-3. Image.usedfor a "Why give 2 when 1 will do?" campilign toemphasize the importance of single-
unit Red Blood Cell (RBC) transfusions in hemodynamically stable, nonbleeding patients. The image was
dispJayedas a screen saver onworkstations across.the health system. Thisrecommendation is backed by the
AAB,B Choosing Wisely guidelines.27 (Reprinted from Sadanaet a1.25)
combining this alert with education is most effi- cent measured pretransfusion hemoglobin level
cacious. Embedding the published evidence into and comparing time stamps of the laboratory
the alert with hyperlinks to the largest random- tests and transfusion orders placed in the CPOE
ized trials is also important.v" One limitation to system. This rank order bar graph with an
electronic alerts is "alarm fatigue," meaning that evidence-based, color-coded depiction of guide-
clinicians become overwhelmed with these pop- line compliance is easy to interpret and has im-
ups and are more likely to ignore them." proved practice by reducing unnecessary trans-
fusions. The other feature of this type of data
Audits with Provider Feedback presentation is the length of each bar, which in-
dicates the total number of transfused units for
Auditing or monitoring physician practice is the l-month period.
another useful intervention that reinforces Audits should also be designed to detect "un-
evidence-based transfusion guidelines. Prospec- dertransfusion." With aggressive PBM pro-
tive or "real-time" audits (approval before issu- grams, there is a possibilitythat some providers
ing the component), in which the review is per- will avoid transfusions even when patients
formed manually by laboratory staff, may be the would indeed benefit. Providing feedback to
most effective, but such reviews can be labor- providers on patients with very low postopera-
intensive and time-consuming, and can create tive hemoglobin levels is helpful for preventing
animosity. This process could also delay transfu- undertransfusion.
sions in hemorrhagic emergencies, and a pro-
cess should be in place to avoid such delays.
In the authors' experience, monthly reports L N
that compare blood utilization and transfusion
guideline compliance rates of providers to those Blood Utilization
of peers in their own department are very effec-
tive. One method of presenting data to provid- In addition to the data described above, some
ers is the rank order bar graph, as shown in Fig metrics commonly collected to assess blood uti-
20-5.2 When providers are compared directly to lization include the following:
peers within their own specialty service, the in-
dividuals transfusing outside the evidence-based 1. Average number of units transfused per
range will be compelled to change their prac- patient or, alternatively, units transfused per
tice. One can present the data with physician 1000 patient days, which provides some
codes or with names, but the greatest impact oc- degree of adjustment for case complexity.This
curs with names. In the authors' experience, cli- measure is most classically associated with
nicians, especially surgeons, are more receptive blood utilization as it correlates with transfu-
to data presented as hemoglobin thresholds and sion cost adjusted for an institution's patient
single-unit transfusion rates than as percentage volume. This metric can be used to quantify
of patients receiving transfusion or average the overall success of a PBM program, but it
number of units transfused per patient. For ex- cannot be used to compare providers to one
ample, after the hemoglobin threshold rank or- another unless they are performing similar
der bar graph (Fig 20-5) was sent to surgeon
procedures on similarpatients.
#44, who had the highest threshold for transfu-
2. Average number of units per transfusion
sion, this surgeon's blood utilization in average
number of units per patient decreased by more recipient. This value is less useful than item
than 55%. 1 because it does not include the patient pop-
Another option for presenting the data is il- ulation that avoided transfusion, possiblyas a
lustrated in Fig 20-6, in which the proportion of result of successful blood conservation meth-
RBCorders is shown according to the hemoglo- ods.
bin threshold level. These hemoglobin thresh- 3. Percentage of patients who received transfu-
olds were collected by looking for the most re- sion. This metric may be useful for compar-
C HAP T E R 20 Patient Blood Management 601
A.
2. Carson JL et 31. N Eng J Mod 2011 365 2453-62
Pleasechoose an appropriate indication to proceed with transfusion.

OR
Check the box below to DISCONTINUEorder.
B.
FIGURE 20-4. (A) The best practiceadvisory (BPA) shown is designedto display when a Red Blood Cell (RBC)
order is placedfor a patient whose precedinghemoglobin level is :?: 7 g/dL or whose hemoglobin has not been
measured in the past 24 hours. This BPA is used to remind clinicians to follow evidence-basedtransfusion
guidelinesin patientswho are not actively bleedingand are hemodynamicallystable. (B) The reasonsto override
the BPAand proceed with the RBCtransfusion order are shown. One must be chosen to finalize the order.
(Reprintedfrom Franket a1.9)
N
rl
rl
rl
<',
it ii,
~www~wwww~ww~w~w~~~ww~w~www~w~~~ww~www~w~~w
5SSSSSSsssssssss££sssssssssssssssssssssssss
N
rl
I I I 00
.\.
w~~~~wwww~~~~~~w~www~~w~www~w~~wwww~~~w~w~w~
-cc-----------------------------------------
ell II
•••••••••••••••
cccc:cc:ccccccccccccc:ccccccc:cc:ccc:c:c:cCC:CC:ccc
~~;~~~~oom~~~~~~~~~~~N~~~~~~~~~M~~~~~~~~~~~~~
# uoa3ms
""0
o
...c::
U)
Q)
I.-
.s:
I-
..Q
:J:
>-
..Q
U)
---
o ....J Vl
a..
Q)
....J
-0
....J :~
-0 tlO -0 C
0::: eo CJ) tlO ~ ;:)
r-, r--.
I
00 U
Q) V r--. AI CO
>
Q)
..0
:c
..0
:c
..0
:c
0:::::
""I-
.....J 0
c: e
l.-
OJ
ro M
...c
.u
..,_ E
U) :::::s
>-
.s: Z
a...
I ~
ing providers to other providers who perform the occurrence of anyone of these morbid
similar procedures on a similar population of events.
patients. Again, it is less useful than item I The in-hospital mortality rate is usually ob-
because it disregards the number of units a tainable from the electronic medical record
patient receives; that is, whether a patient (EMR)(unlike 30-day or l-year mortality) and is
an important clinical outcome to measure in
receives I or IO units, the patient counts as
PBM programs. It has been recognized that,
only one transfusion recipient. compared with nonrecipients of transfusion,
transfusion recipients have about a threefold-
Clinical Outcomes higher morbid event rate and about a ninefold-
Clinical outcome data are important for deter- higher in-hospital mortality. On further investi-
mining how a PBM program is influencing the gation, however, it becomes apparent that
quality of patient care. Unfortunately, such data transfusion recipients have more comorbidities
tend to be the most difficultto capture. Morbid and undergo more complex procedures, thus
events are commonly assessed in PBM pro- demonstrating the need to risk-adjustretrospec-
grams, and examples of these are given in Table tively collected transfusion data. Otherwise ,
20-4. Often, grouping these morbid events into transfusion is so strongly associated with severi-
larger categories (as shown in the table) can ty of illness and complexity of procedure that
ease interpretation. Programs can also use a confounding by indication is commonplace in
composite morbid event rate, which includes retrospective studies. 100
One particularly worthwhile method of as-
sessing outcomes is to use the ICD-9 or lCD-I0
(International Classification of Diseases) codes
TABLE 20-4. Clinical Outcomes to Follow in a from the billing database.l'"!" However, this
Patient Blood Management Program method is dependent on the clinician charting
the patient records accurately and the hospital
Thrombotic events coding team coding the charts correctly. None-
Disseminatedintravascular coagulation theless, these codes can be very useful. Registry
Deepvenous thrombosis data can also be useful if an institution partici-
pates in programs such as the National Surgical
Pulmonary embolus Quality Improvement Program or the Societyfor
Hospital-acquiredinfections Thoracic Surgeons database.
Surgical-site infection
Benchmark Data
Drug-resistant infection
The ideal benchmark data allow the comparison
Sepsis of a hospital to other hospitals that perform simi-
Clostridiumdifficile lar procedures on similar patients. This compari-
son is particularly helpful for blood utilization
Ischemic events metrics such as average number of units per pa-
Myocardial infarction tient undergoing a given procedure. This bench-
Transient ischemic attack
mark is useful for standardized procedures, such
as a total hip arthroplasty, but less useful for cas-
Cerebralvascular accident es such as spinal fusion, because the number of
Respiratory insufficiency/failure spinal levelsis an important variable that often is
not accounted for. When comparing blood use
Renalinsufficiency/failure among hospitals, or even among providers (sur-
Mortality geons), one method of risk-adjustment that can
Length of stay
be used is the case mix index. The case mix in-
dex has been shown to correlate directly to
blood utilization, not only for RBCs but also for as well as the MSBOScategory of "no sample"
plasma and platelets. 104The All Patients Refined- "type and screen," or "type and crossmatch" to '
Diagnosis Related Groups (APR-DRG) weighted indicate low-, medium-, and high-blood-losssur-
index (also called case mix index) is a good risk gical procedures, respectively. Providers also ac-
index .that can be used to adjust for differences count for the patient's body mass, which reflects
in procedure complexity and severity of illness circulating blood volume.IIO,111 For example, the
for a group of patients or for an entire institu- optimal preoperative hemogloblin for a 40,kg
tion. These case mix index numbers are publi- patient would be substantially higher than that
cally available on the Centers for Medicare and for an 80-legpatient, merely because the larger
Medicaid Services website. 105 patient has a greater circulating blood volume
and thus will tolerate a larger blood loss. A phy-
sician who is experienced in treating anemia
EXTR NSFU N with intravenous iron and ESAs can be very
valuable tQ a bloodless program. Having autolo-
Bloodless Care gous bloodrecovery equipment readily available
A good PBM program will ensure optimal care and personnel to operate the machines is critical
for patients who do not accept transfusion of al- to Safelymanaging patients who will not accept
logeneic blood components for religious or other transfusion. One should also use rigorous meth-
reasons. These patients receive what has been ods to minimize blood loss and optimize hemo-
called "extreme blood management," whereby globin levels and coagulation. For example, it is
all the methods of blood conservation described important to use small phlebotomy tubes and re-
above are used. Often, such care is delivered duce the frequency of blood testing. Often, sim-
through a coordinated "bloodless program."!" ply tolerating lower-than-usual hemoglobin lev-
First, such patients need to be identified early in els is necessary when patients will not accept
their hospital stay, or early in the preoperative transfusions. Providing supplemental oxygen
period before surgery. Each individual patient and minimizing myocardial demand may help
may have different beliefs with regard to accep- decrease risk of ischemia. In cases of severe ane-
tance of fractionated blood components (minor mia, use of artificial oxygen carriers may be con-
fractions such as albumin, cryoprecipitate, and sidered for emergency use through the ,FDA's
Expanded Access (formerly called Compassion-
clotting factors) as well as procedures involving
ate Use) program. liZ In summary, rendering care
their own blood when kept in circuit (blood re-
for these bloodless-care patients entails all the
covery, ANH). A knowledgeable provider should
best practices of PBM, often taking them to the
thoroughly discuss these options with the pa- extreme. 102,109
tient. Then the patient's decision should be well
It has been shown that when a carefully co-
documented in the medical record and made
ordinated program implemented all of the
available..for all members of the health-care
above-mentioned PBM methods, the .clinlcal
team, perhaps through use of alerts or a best
outcomes for patients who did not accept trans-
practice advisory in the EMR.A transfusion con-
fusion were .as good as ot.better than those of a
sent form may further assistwith thls process.107
carefully matched control patient group who did
Protocols to diagnose and aggressively treat
accept transtuslon.!" In addition, .the hospital
preoperative anemia are necessary for patients
costs incurred to care for the bloodless-care pa-
undergoing procedures with substantial blood
tients were lower than those. for the patients
loss and should include intravenous iron and
ESAswhen needed.108,109 For this process, the who accepted transfusion (total costs were 12%
lower, an<:l direct costs were 18%10wer).102
institutional MSBOS can be used to determine
the likelihood of significant blood loss and use
Massive Transfusions
this information to choose a hemoglobin target
before elective surgery. Taken into account are Providing optimal care for patients who receive
the particular surgeon performing the procedure massive transfusion is one goal of a good PBM
program. Although the optimal ratio of blood vigilant to prevent, diagnose, and treat infec-
components is somewhat controversial, recent tions and thrombosis in massive transfusion re-
evidence suggests that a 1:1:1 ratio of RBCsto cipients in order to optimize their outcomes.
plasma to platelets is ideal for those who require Dzik et al11S reported a similar series of "ultra-
transfusion of an entire blood volume or massive" transfusions (>20 RBC units over 48
more.'!' A hospital should have a system in hours), in which trauma patients had the high-
place to thaw frozen plasma rapidly, and plate- est mortality, followed by medicine patients;
lets and cryoprecipitate should be readily avail- however, organ transplant, general, and cardiac
able to support massive transfusions. Some insti- surgery patients had lower mortality.
tutions are using viscoelastic testing (TEG or
ROTEM) to guide the ratio of blood compo-
nents. As discussed above, good evidence sup-
ports the use of antifibrinolytic drugs for reduc-
ing mortality in patients with trauma and Successfully implementing a PBM program re-
postpartum hemorrhage. The rapid delivery of a quires planning, education, and teamwork. A
balanced combination of blood components is PBM program is an important patient safety and
important, and, as with any massive transfusion quality measure that can also lead to cost sav-
scenario, the provider should aggressivelypre- ings. To implement a successfulprogram, hospi-
vent hypothermia, which will increase bleeding. tals need sufficientresources to support the peo-
Massive transfusion can lead to citrate toxicity, ple who can accomplish the methods and
which results in hypocalcemia that in turn caus- techniques that are discussed above. However,
es hypotension from decreased cardiac contrac- the return on investment can be 400%.9Institu-
tility and vasomotor tone. Therefore, replacing tions also need sufficient information technolo-
calcium with calcium chloride or gluconate is gy support to obtain the data required to im-
important for patients with massive transfu- prove practice. Education, evidence-based
sions. Ionized calcium should be measured fre- transfusion guidelines, and best practice adviso-
quentlyand maintained above 1.0 mmollL. It is ries effectively encourage good practice and re-
important to recognize that plasma and platelets duce unnecessary transfusions.
have approximately fivefold more citrate than The specific activities performed by a PBM
RBCs for a given transfused volume. Conse- program can vary by institution and are defined
quently, when giving a 1:1:1 ratio of these com- by the executive management team of the facili-
ponents, the patient receives a large amount of ty. Guidance is available from various profes-
citrate.!" sional organizations, including AABB and
Recently, clinical outcomes were reported af- SABM. The latter offers Administrative and
ter massive transfusion of patients, including Clinical Standards jor Patient Blood Manage-
those who received very high transfusion doses ment Programs,which outlines 13 standards re-
(up to and over 75 units of RBCs).It was found lated to the activities of a PBMprogram. For the
that for every 10 additional RBCunits adminis- AABB/Joint Commission certification, a PBM
tered during the hospitalization, mortality in- program can be designated as an activity levell,
creased by 10%, and after 50 RBCunits, mortal- 2, or 3 program, as delineated in AABB'sStan-
ity reached 50% (the 50150 rule).'?' Perhaps dards jor a Patient Blood Management Pro-
more importantly, hospital-acquired infections gram. Information on this program and many
and thrombotic events were found to be four- to other PBM resources may be found on the
fivefold more common than renal, respiratory, AABBwebsite (aabb.orglpbm). With a success-
or ischemic events in the massive transfusion re- ful PBMprogram, providers can reduce risk, im-
cipients, and increases in the frequency of these prove outcomes, and reduce cost, which in
events were also dose dependent. The research- combination increases the value of health care
ers concluded that providers should be more they deliver.
A N 19th edition of the Technical Manual, and por-

tions of her chapter were used in the current
The authors would like to thank Kathleen E. version. The authors also wish to thank Claire F.
Puca, MD, MT(ASCP)SBB,Medical Director, Levine, MS, ELS, Scientific Editor, Department
BloodCenter of Wisconsin, and Associate Profes- of Anesthesiology/Critical Care Medicine, Johns
sor, Medical College ofWisconsin. Dr. Puca con- Hopkins Medicine, Baltimore, Maryland, for ed-
tributed as the author of the PBM chapter in the itorial assistance.
REFERENCES
1. Carson JL, Triulzi DJ, Ness PM. Indications for www.jointcommission. org/ overuse_summi tI
and adverse effects of red-cell transfusion. (accessedNovember 29, 2019).J
N EnglJ Med 2017;377:1261-72. 4. Jones JM, Sapiano MRP, Savinkina M, et al.
2. Frank SM, SavageWJ, RothschildJA, et al. Vari- Slowing decline in blood collection and transfu-
abilityin blood and blood component utilization sion in the United States-2017. Transfusion
as assessed by an anesthesia information man- 2020;60(S2):Sl-S9.
agement system. Anesthesiology 2012; 117:99- 5. The 2016 MBB Blood Survey Fact Sheet.
106. Bethesda, MD: AABB, 2019. [Available at:
3. Proceedingsfrom the National Summit on Over- http://www.aabb.org/research/hemovigi
use (July 8,2013). Oakbrook Terrace, IL: The lance/blood survey /Pages/ default. aspx (ac-
Joint Commission, 2013. [Availableat https:// cessed March 26, 2020).J
6. Zuckerberg GS, ScottAV,WaseyJO, et al. Effica- 18. VillanuevaC, ColomoA, BoschA, et al. Transfu-
cy of education followed by computerized pro- sion strategies for acute upper gastrointestinal
vider order entry with clinician decision support bleeding. N EnglJ Med 2013;368:11-21.
to reduce red blood cell utilization. Transfusion 19. Murphy GJ, Pike K, RogersCA, et al. Liberalor
2015;55:1628-36. restrictive transfusion after cardiac surgery.
7. Frank SM, Oleyar MJ, Ness PM, Tobian AA. Re- N EnglJ Med 2015;372:997-1008.
ducing unnecessary preoperative blood orders 20. Holst LB, Haase N, Wetterslev J, et al. Lower
and costs by implementing an updated institu- versus higher hemoglobin threshold for transfu-
tion-specific maximum surgical blood order sion in septic shock. N Engl J Med 2014;371:
schedule and a remote electronic blood release 1381-91.
system. Anesthesiology2014; 121:501-9. 21. Mazer CD, Whitlock RP, Fergusson DA, et al.
8. Goodnough LT, Shieh L, Hadhazy E, et al. Im- Restrictive or liberal red-cell transfusion for car-
proved blood utilization using real-time clinical diac surgery. N EnglJ Med 2017;377:2133-44.
decision support. Transfusion 2014;54: 1358- 22. Wintermeyer TL, Liu J, Lee KH, et al. Interac-
65. tive dashboards to support a patient blood man-
9. Frank SM, Thakkar RN, Podlasek SJ, et al. Im- agement program across a multi-institutional
plementing a health system-wide patient blood healthcare system. Transfusion 2016;56: 1480-
management program with a clinical communi- 1.
ty approach. Anesthesiology2017;127:754-64. 23. Podlasek SJ, Thakkar RN, RotelloLC, et al. Im-
10. Shander A, Hofmann A, Ozawa S, et al. Activi- plementing a "Why give 2 when I will do?"
ty-based costs of blood transfusions in surgical Choosing Wisely campaign. Transfusion 2016;
patients at four hospitals. Transfusion 2010; 56:2164.
50:753-65. 24. Gulnn NR, Guercio JR, Hopkins TJ, et al. How
II. Frey K, ed. Standards for a patient blood man- do we develop and implement a preoperative
agement program. 3rd ed. Bethesda, MD: anemia clinic designed to improve perioperative
AABB,2020. outcomes and reduce cost? Transfusion 2016;
12. SABMadministrative and clinical standards for 56:297-303.
patient blood management programs. 5th ed. 25. Sadana D, Pratzer A, Scher LJ,et al. Promoting
Englewood, NJ: Societyfor the Advancement of high-value practice by reducing unnecessary
BloodManagement, 2019. [Availableat https:// transfusions with a patient blood management
www.sabm.org/publications/ (accessed No- program.JAMAInternMed 2018;178:116-22.
vember 29,2019).] 26. Frank SM, Resar LM, Rothschild JA, et al. A
13. Carson JL, Terrin ML, Noveck H, et al. Liberal novel method of data analysis for utilization of
or restrictive transfusion in high-riskpatients af- red blood cell transfusion. Transfusion 2013;
ter hip surgery. N Engl] Med 2011;365:2453- 53:3052-9.
62. 27. CallumJL,Waters JH, Shaz BH,et al. The AABB
14. Hajjar LA, Vincent JL, Galas FR, et al. Transfu- recommendations for the ChoosingWiselycam-
sion requirements after cardiac surgery: The paign of the American Board of Internal Medi-
TRACS randomized controlled trial. JAMA cine. Transfusion2014;54:2344-52.
2010;304:1559-67. 28. Roback JD, Caldwell S, Carson J, et al. Evi-
15. Hebert PC, Wells G, Blajchman MA, et al. A dence-based practice guidelines for plasma
multicenter, randomized, controlled clinicaltrial transfusion. Transfusion2010;50: 1227-39.
of transfusion requirements in critical care. 29. KaufmanRM, DjulbegovicB, Gernsheimer T, et
N EnglJ Med 1999;340:409-17. al. Platelet transfusion: A clinicalpractice guide-
16. LacroixJ, Hebert PC, Hutchison JS, et al. Trans- line from theAABB.Ann Intern Med 2015;162:
fusion strategies for patients in pediatric inten- 205-13.
sive care units. N Engl] Med 2007;356:1609- 30. Frank SM, RothschildJA, Masear CG, et al. Op-
19. timizing preoperative blood ordering with data
17. Robertson CS, Hannay HJ, YamalJM, et al. Ef- acquired from an anesthesia information man-
fect of erythropoietin and transfusion threshold agement system. Anesthesiology 2013;118:
on neurological recovery after traumatic brain 1286-97.
injury: A randomized clinical trial. JAMA2014; 31. Shander A. Preoperative anemia and its manage-
312:36-47. ment. TransfusApher Sci2014;50: 13-15.
32. Elhenawy AM, Meyer SR, Bagshaw SM, et aI. 46. Gammon HM, Waters JH, Watt A, et al. Devel-
Role of preoperative intravenous iron therapy to oping performance measures for patient blood
correct anemia before major surgery: Study pro- management. Transfusion 2011 ;51:2500-9.
tocol for systematic review and meta-analysis. 47. Can MM, Tanboga IH, Turkyilmaz E, et al. The
Syst Rev 2015;4:29. risk of false results in the assessment of platelet
33. Karkouti K, Wijeysundera DN, Beattie WS. Risk function in the absence of antiplatelet medica-
associated with preoperative anemia in cardiac tion: Comparison of the PFA-l 00, multiplate
surgery: A multicenter cohort study. Circulation electrical impedance aggregometry and verify
2008;117:478-84. now assays.Thromb Res 2010;125:eI32-7.
34. Mantilla CB, Wass CT, Goodrich KA, et aI. Risk 48. Wang Cl, Moss J, Yuan CS. Commonly used di-
for perioperative myocardial infarction and mor- etary supplements on coagulation function
tality in patients undergoing hip or knee arthro- during surgery. Medicines 2015;2:157-85.
49. VassalloR, Goldman M, Germain M, Lozano M,
plasty: The role of anemia. Transfusion 2011;
for the BESTCollaborative. Preoperative autolo-
51:82-91. gous blood donation: Waning indications in an
35. Auerbach M. Oral or N iron in inflammatory
era of improved blood safety. Transfus Med Rev
bowel disease. Am J Gastroenterol 2012; 107: 2015;29:268-75.
950-1. 50. Lee GC, Cushner FD. The effects of preopera-
36. Auerbach M, Macdougall IC. Safety of intrave- tive autologous donations on perioperative
nous iron formulations: Facts and folklore. blood levels. J Knee Surg 2007;20:205-9.
~lood Transfus 2014;12:296-300. 51. Kennedy C, Leonard M, Devitt A, et aI. Efficacy
37. Auerbach M, Pappadakis JA, Bahrain H, et aI. of preoperative autologous blood donation for
Safety and efficacy of rapidlyadministered (one elective posterior lumbar spinal surgery. Spine
hour) one gram of low molecular weight iron 2011;36:EI736-43.
dextran (INFeD) for the treatment of iron defi- 52. Goldman M, Remy-Prince S, Trepanier A, De-
cient anemia. Am J Hematol2011 ;86:860-2. cary F. Autologous donation error rates in Cana-
38. Auerbach M. Intravenous iron in the periopera- da. Transfusion 1997;37:523-7.
tive setting. AmJ HematoI2014;89:933. 53. Waters JH. Indications and contralndications of
39. Fishbane S, Nissenson AR. The new FDA label cell salvage. Transfusion 2004;44:40S-4S.
for erythropoietin treatment: How does it affect 54. Waters JH. Optimization of recovering shed
hemoglobin target? KidneyInt 2007;72:806-13. blood when performing blood salvage. Anesth
40. Fox J1. FDAlikely to further restrict erythropoi- Analg 2009; 108: 1714-15.
etin use for cancer patients. Nature Biotech 55. Wang G, Bainbridge D, Martin J, Cheng D. The
2007;25:607-8. efficacy of an intraoperative cell saver during
41. Robles NR. The safety of erythropoiesis-stimulat- cardiac surgery: A meta-analysis of randomized
ing agents for the treatment of anemia resulting trials. Anesth Analg 2009; 109:320-30.
from chronic kidney disease. CUnDrug Investig 56. Esper SA, Waters JH. Intra-operative cell sal-
2016;36:421-31. vage: A fresh look at the indications and contra-
42. Roger SD, Tio M, Park HC, et al, Intravenous indications. BloodTransfus 2011;9:139-47.
57. WatersJH, DygaRM, WatersJF, YazerMH. The
iron and erythropoiesis-stimulating agents in
volume of returned red blood cells in a large
haemodialysis: A systematic review and meta-
blood salvage program: Where does it all go?
analysis. Nephrology 2017;22;969-76.
Transfusion 2011;51:2126-32.
43. Goodnough H, Skikne B, Brugnara C. Erythro- 58. Frank SM. Who benefits from red blood cell sal-
poietin, iron, and erythropoiesis. Blood 2000; vage?-Utmty and value of intraoperative autol-
96:823-33. ogous transfusion. Transfusion 2011 ;51 :2058-
44. Friedman BA, Oberman HA, Chadwick AR, 60.
Kingdon K1.The maximum surgical blood order 59. Salaria ON, Barodka VM, Hogue CW, et aI. Im-
schedule and surgical blood use in the United paired red blood cell deformability after transfu-
States. Transfusion 1976;16:380-7. sion of stored allogeneic blood but not autolo-
45. Dutton RP, Shih D, Edelman BB, et aI. Safety of gous salvaged blood in cardiac surgery patients.
uncrossmatched type-O red cells for resuscita- Anesth Analg 2014; 118: 1179-87.
tionfromhemorrhagic shock. J Trauma 2005; 60. Frank SM, Abazyan B, Ono M, et aI. Decreased
59:1445-9. erythrocyte deformability after transfusion and
the effects of erythrocyte storage duration. Orthopaed Traumatol Surg Res 2012;98:477-
Anesth Analg 2013; 116:975-81. 83.
61. Scott AV, Nagababu E, Johnson DJ, et al. 2,3- 74. Zufferey PJ, LanoiseleeJ, Chapelle C, et al. In-
Diphosphoglycerate concentrations in autolo- travenous tranexamic acid bolus plus infusion is
gous salvagedversus stored red blood cells and not more effectivethan a singlebolus in primary
in surgicalpatients after transfusion.Anesth An- hip arthroplasty: A randomized controlled trial.
alg 2016;122:616-23. Anesthesiology2017;127:413-22.
62. BergMP, ed. Standards for perioperative autolo- 75. Myles PS, SmithJA, ForbesA, et al. Tranexamic
gous blood collection and administration. 8th acid in patients undergoing coronary-artery sur-
ed. Bethesda, MD: AABB,2018. gery. N EnglJ Med 2017;376: 136-48.
63. Schmied H, KurzA, SesslerDI, et al. Mild hypo- 76. Huang F, Wu D, Ma G, et al. The use of
thermia increases blood loss and transfusion re- tranexamic acid to reduce blood loss and trans-
quirements during total hip arthroplasty. Lancet fusion in major orthopedic surgery: A meta-
1996;347:289-92. analysis.J Surg Res2014; 186:318-27.
64. Achneck HE, Sileshi B, IamiolkowskiRM, et al. 77. Gillette BP, DeSimone LJ, Trousdale RT, et al.
A comprehensive review of topical hemostatic Low risk of thromboembolic complicationswith
agents: Efficacyand recommendations for use. tranexamic acid after primary total hip and knee
Ann Surg 2010;251:217-28. arthroplasty. Clin Orthop Re1atRes 2013;471:
65. Frank SM,WaseyJO, Dwyer IM, et al. Radiofre- 150-4.
quency bipolar hemostatic sealer reduces blood 78. WOMAN Trial Collaborators. Effect of early
loss, transfusion requirements, and cost for pa- tranexamic acid administration on mortality,
tients undergoing multilevel spinal fusion sur- hysterectomy, and other morbidities in women
gery: A case control study. J Orthop Surg Res with post-partum haemorrhage (WOMAN):An
2014;9:50. international, randomised, double-blind, place-
66. Shir Y, Raja SN, Frank SM, Brendler CB. Intra-
bo-controlledtrial. Lancet 2017;389:2105-16.
operative blood loss during radical retropubic 79. Roberts I, Shakur H, AfolabiA, et al. The impor-
prostatectomy: Epidural versus general anesthe-
tance of early treatment with tranexamic acid in
sia. Urology 1995;45:993-9.
bleeding trauma patients: An exploratory analy-
67. Goodnough LT.Acute normovolemichemodilu-
sis of the CRASH-2randomised controlled trial.
tion. Vox Sang 2002;83(Suppl 1):211-15.
68. Shander A, Rijhwani TS. Acute normovo1emic Lancet 2011 ;377:1096-101.
80. Gayet-AgeronA, Prieto-Merino D, Ker K, et al.
hemodilution. Transfusion 2004;44(12 Sup-
Effect of treatment delay on the effectiveness
pl):26S-34S.
69. Zhou X, Zhang C, Wang Y, et al. Preoperative and safety of antifibrinolytics in acute severe
acute normovolemic hemodilution for minimiz- haemorrhage: A meta-analysis of individual pa-
ing allogeneic blood transfusion: A meta-analy- tient-level data from 40 138 bleeding patients.
sis.Anesth Analg 2015; 121:1443-55. Lancet 2018;391: 125-32.
70. Grant MC, Resar LM, Frank SM. The efficacy 81. Johnson DJ, Johnson CC, Goobie SM, et al.
and utility of acute normovolemic hemodilu- High-dose versus low-dose tranexamic acid to
tion. Anesth Analg2015;121: 1412-14. reduce transfusion requirements in pediatric
71. Monk TG, Goodnough LT, Brecher ME, et al. scoliosis surgery. J Pediatr Orthop 2017;37:
Acute normovo1emichemodilution can replace e552-e7.
preoperative autologous blood donation as a 82. GoobieSM,Meier PM, Sethna NF, et al. Popula-
standard of care for autologous blood procure- tion pharmacokinetics of tranexamic acid in
ment in radical prostatectomy. Anesth Analg paediatric patients undergoing craniosynostosis
1997;85:953-8. surgery. Clin Pharmacokinet 2013;52:267-76.
72. Benoni G, Fredin H, Knebel R, Nilsson P. Blood 83. Konig G, Hamlin BR, Waters JH. Topical
conservation with tranexamic acid in total hip tranexamic acid reduces blood loss and transfu-
arthroplasty: A randomized, double-blind study sion rates in total hip and total knee arthroplas-
in 40 primary operations. Acta Orthop Scand ty. J Arthroplasty2013;28:1473-6.
2001 ;72:442-8. 84. Shore-LessersonL, Manspeizer HE, DePerio M,
73. Irisson E, Hernon Y, Pauly V, et al. Tranexamic et al. Thromboelastography-guided transfusion
acid reduces blood loss and financial cost in pri- algorithm reduces transfusions in complex car-
mary total hip and knee replacement surgery. diac surgery. Anesth Analg 1999;88:312·19.
85. Oremus K, Sostaric S, Trkulja V, Haspl M. Influ- sion practice: A systematic review. Transfus
ence of tranexamic acid on postoperative autol- Med Rev 2015;29: 14-23.
ogous blood retransfusion in primary total hip 99. Mitka M. Joint Commissionwarns of alarm fa-
and knee arthroplasty: A randomized controlled tigue: Multitude of alarms from monitoring de-
trial. Transfusion 2014;54:31-41. vices problematic.JAMA2013;309:2315-6.
86. Springer BD, Odum SM, Fehring TK. What is 100.Carson JL, Hebert PC. Here we go again-
the benefit of tranexamic acid vs reinfusion blood transfusion kills patients?: Comment on
drains in total [oint arthroplasty? J Arthroplasty "Associationof blood transfusion with increased
2016;31:76-80. mortality in myocardialinfarction:A meta-analy-
87. Munoz M, Garcia-Vallejo JJ, Ruiz MD, et al. sis and diversity-adjustedstudy sequential analy-
Transfusion of post-operative shed blood: Labo-
sis."JAMAIntern Med 2013;173:139-41.
ratory characteristics and clinical utility. Eur
101. Kim Y, Spolverato G, Lucas DJ, et al. Red cell
Spine I 2004;13:S107-13.
transfusion triggers and postoperative outcomes
88. Sinardi D, Marino A, Chillemi S, et al. Composi-
after major surgery. J Gastrointest Surg 2015;
tion of the blood sampled from surgical drainage
after joint arthroplasty: Quality of return. Trans-
19:2062-73.
102. Frank SM, Wick EC, Dezern AE, et al. Risk-ad-
fusion 2005;45:202-7.
89. Thavendiranathan P, Bagai A, Ebidia A, et al. Do
justed clinicaloutcomes in patients enrolled in a
blood tests cause anemia in hospitalized pa- bloodless program. Transfusion 2014;54:2668-
tients? The effect of diagnostic phlebotomy on 77.
hemoglobin and hematocrit levels. J Gen Intern 103. Johnson DJ, Scott AV, BarodkaVM, et al. Mor-
Med 2005;20:520-4. bidity and mortality after high-dose transfusion.
90. Chant C, Wilson G, FriedrichJO. Anemia, trans- Anesthesiology2016;124:387-95.
fusion, and phlebotomy practices in criticaily ill 104. Stonemetz JL,Allen PX,WaseyJ, et al. Develop-
patients with prolonged ICU length of stay: A ment of a risk-adjustedblood utilization metric.
cohort study. Crit Care 2006; 10:R140. Transfusion2014;54:2716-23.
91. Koch CG, Reineks EZ, Tang AS, et al. Contem- 105. Case mix index. Baltimore, MD: Centers for
porary bloodletting in cardiac surgical care. Ann Medicare and Medicaid Services, 2019. [Avail-
Thorac Surg 2015;99: 779-84. able at https://www.cms.gov/Medicare/Medi
92. SpolveratoG, KimY,EjazA, et al. Effectof rela- care- Fee-for-Service- Payment/ AcuteInpa
tive decrease in blood hemoglobin concentra- tientPPS/ Acute-Inpatient-Files-for-Download-
tions on postoperative morbidity in patients ltems/CMS022630.html (accessed November
who undergo major gastrointestinal surgery. 29,2019).J
JAMASurg 2015; 150:949-56. 106. Resar LM, Wick EC, Almasri TN, et al. Blood-
93. CarsonJL, Guyatt G, Heddle NM, et ai. Clinical less medicine: Current strategies and emerging
practice guidelines from the AABB:Red Blood treatment paradigms. Transfusion 2016;56:
Cell transfusion thresholds and storage. JAMA 2637-47.
2016;316:2025-35. 107. Jorgenson TD, GolbabaB, Guinn NR, Smith CEo
94. Uhl L,Assmann SF,HamzaTH, et al. Laboratory When blood is not an option: The case for a
predictors of bleeding and the effect of platelet
standardized blood transfusion consent form.
and RBCtransfusions on bleeding outcomes in
ASAMonitor 2017;81:48-50.
the PLADOtrial. Blood2017; 130:1247-58.
108. Guinn NR, Roberson RS,White W, et al. Costs
95. Crispen JF. The single-unit transfusion. A con-
and outcomes after cardiac surgery in patients
tinuing problem. Pa Med 1966;69:44-8.
96. YangWW, Thakkar RN, Gehrie EA,et al. Single- refusing transfusion compared with those who
unit transfusions and hemoglobin trigger: Rela- do not: A case-matched study. Transfusion
tive impact on red cell utilization. Transfusion 2015;55:2791-8.
2017;57:1163-70. 109. ResarLM, Frank SM. Bloodlessmedicine: What
97. DeZern AE, Williams K, Zahurak M, et al. Red to do when you can't transfuse. Hematology Am
blood cell transfusion triggersin acute leukemia: Soc Hematol Educ Program 2014;2014:553-8.
A randomized pilot study. Transfusion 2016; 110. Shander A, Rijhwani TS. Clinical outcomes in
56:1750-7. cardiac surgery: Conventional surgery versus
98. HibbsSP, Nielsen ND, BrunskillS, et al. The im- bloodless surgery. Anesthesiol Clin North Am
pact of electronic decision support on transtu- 2005;23:327-45.
111. IassarAS, Ford PA, Haber HL,et al. Cardiac sur- a 1:1:1 vs a 1:1:2 ratio and mortality in patients
gery in Jehovah's Witness patients: Ten-year ex- with severe trauma: The PROPPRrandomized
perience. Ann Thorac Surg 2012;93: 19-25. cllnicaltrial. JAMA2015;313:471-82.
112. Tan GM, Guinn NR, Frank SM, Shander A. Pro- 114. Dzik WH, Kirkley SA. Citrate toxicity during
ceedings from the Society for Advancement of massive blood transfusion. Transfus Med Rev
Blood Management Annual Meeting 2017: 1988;2:76-94.
Management dilemmas of the surgical patient- 115. Dzik WS, Ziman A, Cohen C, et al for the Bio-
when blood is not an option. Anesth Ana1g medical Excellencefor SaferTransfusionCollab-
2019; 128(1):144-51. orative. Survival after ultrarnassive transfusion:
113. HolcombJB, Tilley BC, Baraniuk S, et al. Trans- A review of 1360 cases. Transfusion 2016;56:
fusion of plasma, platelets, and red blood cells in 558-63.
Approaches to B.lood
Utilization Auditing
Ira A. Shu/man, MD; Jay Hudgins, DO; and Irina Maramica, MD, PhD, MBA
LOOD TRANSFUSIONSARE AMONG To curtail inappropriate transfusions, several

the most common procedures performed accreditation organizations have endorsed pa-
in hospitals in the United States but are tient blood management (PBM) programs,
also associated with significant risk for the pa- which are based on multidisciplinary approach-
tients. With millions of units of blood compo- es to optimize the safety and outcomes in pa-
nents transfused annually, quality organizations tients who are candidates for transfusion. By in-
have focused on appropriate blood manage- tegrating evidence-based steps to reduce the
ment as an area of opportunity to improve clini- probability of transfusions, PBM programs can
cal outcomes through evidence-based standard- reduce health-care costs while ensuring that
ization. Transfusion of blood components has blood components are availablefor patients who
also been identified as one of the most overused need them. 14
therapeutic interventions performed during pa- However, successful implementation of a
tient hospitalizations.!" Several clinical trials comprehensive PBM program requires the sup-
provide evidence that patient outcomes associat- port of administrative and clinical leadership,
ed with a restrictive transfusion strategy are sim- who can remove obstacles to achieve interde-
ilar to, if not better, than patient outcomes asso- partmental consensus regarding existing gaps
ciated with more liberal transfusion strategies.S-7 and goals of the program. This support is best
Specifically, patients receiving fewer transfu- achieved by developing a clear business. case
sions have a shorter length of stay, a lower inci- and enlisting clinical champions to highlight pa-
dence of infection, and lower readmission rates tient care benefits of the program. Central to
for postoperative complications+" With a grow- this effort is the ability to audit transfusion prac-
ing evidence base linking transfusions with ad- tices and demonstrate measurable improve-
verse clinical outcomes, a significant proportion ments in blood component utilization. 15 Fur-
of transfusions may be unwarranted. 11-13 In addi- thermore, hospitals are required by accrediting
tion, the optimal use of blood components means agencies, such as The Joint Commission and
not only avoiding overtransfusion or inappropri- MBB, to have blood utilization audit programs
ate transfusions but also making better transfu- that. monitor transfusions of all types of blood
sion decisionsthat would avoidundertransfusion. components.
Ira A. Shulman, MD, Director, USC Transfusion Medicine Services Group, Department of Pathology, University
of Southern California, LosAngeles, California;Jay Hudgins, DO, Medical Director of Transfusion Medicine Ser-
vices, Los Angeles County+USC Medical Center, LosAngeles, California; and Irina Mararnica, MD, PhD, MBA,
CLlA Laboratory Medical Director, Ouest Diagnostics Nichols Institute, San Juan Capistrano, California
U I ~N the hospital's quality assurance department can

be trained to process transfusion data and gener-
Hospitals are allowed flexibilityin designing the ate reports. Alternatively, with the help of the
scope of the audit process; however, to satisfy information systems department, hospitals can
this requirement, the review must be based on develop a blood transfusion dashboard to cap-
objective guidelines for assessing blood utiliza- ture specific trends in transfusion practices and
tion and transfusion effectiveness. Institutional allow benchmarking with less effort and cost.
transfusion committees can provide oversight of These data should be further analyzed by mem-
PBM programs and take the first step in the de- bers of the transfusion committee for inappropri-
velopment of the blood utilization review pro- ate trends, with the goal of identifying areas for
cess by developing or adopting evidence-based
improvement. Aspects that may be useful for
guidelines for blood component use. This com-
monitoring by the transfusion committee in-
mittee can also create auditing criteria for de-
tecting outliers and targeting those practices re- clude the appropriateness of blood component
quiring further evaluation. Audit criteria are ordering, specific quality indicators related to
designed to flag potentially inappropriate or handling and dispensing blood components in
questionable transfusion decisions. Such criteria both blood banks and satellite blood storage ar-
are often institution-specificand may differfrom eas (including storage in or near the operating
clinical transfusion guidelines. However, the in- rooms), steps in blood component administra-
stitutional guidelines are often based on national tion, and adverse events related to transfu-
guidance. The 2013 AABB Blood Collection, sions.18,19 Metrics should also monitor
Utilization, and Patient BloodManagement Sur- undertransfusion of patients." Hospital-based
vey reported that the majority of hospitals (93%) transfusion safety officers (TSOs) are used by
use transfusion guidelines, most commonly de- some institutions. The TSO is a key person who
veloped by AABB(73%), the College of Ameri- supports the PBMprogram outside the laborato-
can Pathologists (32%), or the American Red ry through education, active surveillance of
Cross (12%).16 transfusion practices, and blood utilization re-
Monitoring of blood component ordering, view."
transfusion, return, and wastage provides data In 2011, The Joint Commission published a
for assessingan institution's transfusion efficien- performance measure set for PBM that focuses
cy.17,18Data must be collected, tabulated, and on transfusion appropriateness and clinical
analyzed at regular, specified times. Blood utili- decision-making regarding blood component
zation review ideally should be performed for all transfusion, as well as optimization of patient
services that use blood, but the greatest impact clinical status to reduce blood transfusion re-
of such a review is likelyto be seen when audit- quirernents.f Hospitals are encouraged to per-
ing clinical services with high transfusion vol-
form objective gap-analysisof Joint Commission
umes (eg, surgery) and/or providing transfu-
PBM measures and identify areas for improve-
sion support to high-risk patients (eg, trauma
victims, liver transplant recipients). Metrics re- ments." AABBand The Joint Commission offer
flecting the ordering and transfusion of all types a joint PBM certification program that is based
of blood components used within the institution on the AABB Standards for a Patient Blood
should form the basis for review of blood utiliza- Management Program (PBM Standardsj.24,25
tion. Data analysis should include trends in Certifiedprograms are required to obtain and re-
blood use throughout the institution, as well as view at least quarterly (unless noted) the data
by department; by patient population; by the summarized in Table 21-1.
protocol used, such as massive transfusion pro- In addition to any combined programs be-
tocol (MTP);and by physician. tween AABBand The Joint Commission,institu-
Data sources include electronic patient re- tions may adopt one or more of the following
cords, transfusion service records, and reports specific quality improvement objectives under
from the blood supplier(s). Clerical staff from the rubric of blood component stewardship:
C HAP T E R 2 1 Approaches to Blood Utilization Auditing 615
Promote reduced wastage of blood compo- NIN

nents that have been dispensed to the patient
care area. Before implementing an audit process, it is es-
Promote reduced wastage of blood compo- sential to define the criteria to which the audit
nents that are in the hospital inventory but should adhere. The PBM Standards provides a
never dispensed. broad overview of areas to be monitored reg,
@ Identify, develop, and promote the imple- crossmatch/transfusion (C/T) ratio or MTP ef-
mentation of PBM to improve appropriate fectiveness] but the metrics mentioned are not
use of blood and blood components by necessarily the most effective ones to improve
health providers. blood utilization. Programs like the joint certifi-
Improve clinical outcomes and reduce the cation program mentioned above provide a base
risk of adverse events from blood transfu- by which an institution can begin the process of
sion. evaluating transfusion practices, and this pro-
cess should be customized, either by expanding
or narrowing criteria based on the practice at
each medical center.
Defining an institution's audit criteria should,
at a minimum, adhere to language similar to the
TABLE21-1. Review Items Required by AABB previously mentioned quality improvement ob-
and The Joint Commission for the PBM jectives for blood component stewardship, but
Certification Program the implementation of these recommendations
must be institution specific. Other PBM pro-
gram resources, such as the Patient Blood Man-
agement Guidelines from the National Blood
Bloodcomponentuse Authority of Australia, give point-by-point rec-
ommendations for transfusion practices, cover-
Bloodcomponentwastageand productexpira- ing MTPs, pediatrics, critical care, obstetrics,
tion and perioperative care. Although these guide-
lines do not provide instruction on the optimal
@ Crossmatch/transfusion(CIT)ratio way to initiate auditing, they organize and pres-
ent recommendations, practice points, and ex-
Deviationfromtransfusionpracticesand pert opinions based on a graded, evidence-based
protocols review in each area. Considering material in this
manner can aid in the selection of areas for im-
Transfusionreactions provement at individual instltutions.f
The impetus for PBM programs is primarily
Intraoperativebloodrecoveryuseand quality the improvement of patient outcomes, but they
control also aim to consistently and durably amend phy-
sician .behaviors and improve adherence to
e Informedconsentfor bloodtransfusiondocu- guidelines." Typically this is achieved by focus-
mentation ing on at least one of the following categories:
education, feedback, cost awareness, rationing,
• Massivetransfusionprotocoleffectiveness or financial incentives. For the purposes of
PBM, peer-to-peer education and. feedback are
Bloodinfusionequipmentand warmersmainte- likely to be the best options. From this point, the
nanceprogram(annually) development and implementation of a multidis-
ciplinary transfusion/blood utilization commit-
Externalassessment results(biannually) tee is the first and likely most important step to
help define and develop institutional guidelines,
*Quarterly except as noted. by establishing physician champions who can
support, defend, and help enforce guideline ad- ic guidelines, and provide feedback based on
herence at an institutlon. subspecialty practices or appropriate documen-
In 2015, Yeh et al published the findings of tation.
their prospective interventional study that used In 2017, Cauldwell et al performed a study
a multimodal intervention as a strategy to re- to evaluate whether professional guidance pro-
duce unnecessary transfusionsin surgicalinten- moting a policy of restrictive blood transfusion
sive care unit (ICU) patients." Authors used was being followed. They performed a retro-
peer-to-peer feedback and monthly audits to in- spective analysisof postdeliverytransfusion data
crease adherence to restrictive transfusion from 17 maternity units in the United Kingdom
guidelinesdefined in the TRICCtrial. Specifical- and the United States from 1988 to 2000. They
ly, during the 6-month intervention period, if a also performed an audit of women receiving
patient received transfusion outside of restric- transfusion at three centers 6 to 24 hours after
tive guidelines, the clinicians were notified by deliveryfrom 2013 to 2016. They found that in
e-mail, and education from a surgical colleague both time frames the rate of 2-unit transfusion
was provided within 72 hours of the transfu- in the postpartum period remained above 90%,
sion. They examined transfusion thresholds of though the median estimated blood loss re-
hemoglobin levels greater than 8.0 g/dL and mained the same between women who re-
the rate of overtransfusion, defined as posttrans- ceived 1 unit vs 2 units. In all institutions in the
fusion hemoglobin greater than 10.0 g/dl. in 2013-2016 period >85% of women receiving2-
the pre- and postintervention periods. As a re- unit transfusions had recorded pretransfusion
sult of these interventions, they saw a statistical- hemoglobin >7.0 gldL.29 In this case, the find-
ly significant decrease in both transfusions for ings for the obstetrics groups at these hospitals
hemoglobin greater than 8.0 g/dL (from25% to fall outside of accepted practice recommenda-
2%) and the rate of overtransfusion (from 11% tions from the American College of Obstetri-
to 3%).Monthly audits performed 6 months af- cians and Gynecologists, which endorse the
ter the end of the intervention demonstrated du- view of the American Board of Internal Medi-
rable improvements in the mean pretransfusion cine Foundation's Choosing Wisely Campaign
threshold and the overtransfusion rate. This that transfusion is not required if a patient is sta-
study shows that peer-to-peerfeedback,particu- ble with a hemoglobin >7.0 gldL.3o
larly when received from someone of the same Another example of specialty-specificaudit-
subspecialty, can have a long-lastingeffect on ing was described by Lieberman and col-
transfusion practices, which further highlights leagues." In this prospective study, they re-
the need for diverse representation and partici- viewed transfusions in the neonatal and
pation with PBM. pediatric ICU (NICU and PICU) populations,
Setting audit criteria is often a daunting pro- specificallyin the use of plasma, cryoprecipitate,
cess. Because of the specializednature of medi- and recombinant Factor VIla. The appropriate-
cine, each subspecialty often has specificguide- ness of each order was independently evaluated
lines to influence the way physicians conduct using predefined transfusion criteria by two he-
their practice. In contrast, transfusion/blood uti- matologists. The transfusion criteria for the use
lization committees often implement global of plasma was developed based on a tool devel-
transfusion guidelines for an institution, at- oped by Tinmouth et al," while those for cryo-
tempting to craft guidelines that would address precipitate and recombinant Factor VIla were
the requirements of all clinical services in the developedby a team offivephysicianstrained in
hospital. These multidisciplinary committees either transfusionmedicine or hematology.They
may be better served by examining transfusion found that plasma was ordered as the fluid re-
practices at their institutions and targeting cer- placement in 7% of cases, to correct abnormal
tain areas, services, or specific documentation coagulation tests in 11% of cases, and for pa-
errors/omissions associatedwith transfusion.By tients with minor or no bleeding in 46% of cas-
narrowing focus, they limit the number of es. Overall, they determined that 19%ofplasma
physicians, possibly address subspecialty-specif- orders, 21% of cryoprecipitate orders, and 91%
C HAP T E R 2 1 Approaches to Blood UtilizationAuditing 617
of recombinant Factor VIla orders were inappro- there is consensus, each hospital should exam-
priate. By performing an audit in this manner, ine its scope of practice to determine what
they could identify areas of possible improve- guidelines are most appropriate for the patient
ment that could be applied to a high-riskpatient population. Although several strategies can be
population. employed to accomplish this goal, the following
When attempting to create audit criteria, pro- principles have emerged as integral to success:
spective analysisDf subspecialtypractice can be
an effective tool for change, particularly when Identify resources that can provide basic met-
physician "buy-in" is gained. In medical centers rics to initiate a PBMprogram.
with many subspecialty practices, this approach When forming a transfusion committee,
may not be feasiblebecause of the time or man- identify physician champions from the clini-
power necessary to perform the analysis. In cal services to provide feedback while devel-
such institutions, a more globai approach may oping institutional guidelines.
be needed that can overcome these limitations, Make small changes at first and tailor toward
such as harnessing automated electronic sys- a subspecialtypractice.
tems to monitor transfusion practices. One such Employ a mechanism for feedback from a
system was described in 2006 by Grey and col- peer of the same subspecialty.
leagues, who developed a system to electroni- When possible, use LIS/ERR systems to help
cally extract and collate large amounts of data guide the development of institution-specific
from two laboratory information systems (LIS), guidelines and possibly automate the audit
ULTRAversion 3.2 and TM, using standard in- selection process.
formation technology (IT)scripts." The filefrom
ULTRAcaptured hemoglobin values, and the file
from the TM module extracted blood groups, T
antibody screens, crossmatches, and transfu- U
sion. Pre- and posttransfusion hemoglobin val-
ues were monitored 48 hours after the red cell AABB guidelines describe three methods of
transfusions. These data were imported into the blood utilization review, which can be used in-
MS ACCESSdatabase, where they were linked dividually or in combination depending on the
with ICD-10 (International Classificationof Dis- institution's needs: prospective, concurrent, and
eases) clinical codes for surgical cases. This tool retrospective." Each method has advantages
allowed monitoring of transfusion practices over and shortcomings, which are summarized in Ta-
an extended period, which helped define insti- ble 21-2.
tution-specific transfusion guidelines and subse-
quently identified cases for peer review. In the Prospective Review
years since this article was published, there have
been advances in hospital electronic health re- A prospectlve review of individual blood compo-
cords (ERRs)and LIS,and some information sys- nent orders occurs in real time before the com-
tems can serve as both clinical ERR and LIS. ponent is distributed from the blood bank and
The use of an approach described here is predi- immediately before transfusion. This proactive
cated on having the resources availableto devel- approach to blood utilization review provides an
op these tools, but with the current goals of im- opportunity to intercept unnecessary transfusion
proving quality of care throughout medicine, it and modify both component selection, as appro-
becomes more difficult to deny that methods priate, and timing of administration. In academ-
such as these promote knowledge, transparency, ic institutions, a prospective review is often per-
and accountability in the era of data-driven qual- formed by clinicalpathology residents under the
ity assurance. supervision of a transfusion medicine physician.
In defining audit criteria, institutions are not It requires vigilance by the blood bank staff to
beholden to a homogenous approach to promot- bring to the attention of residents all orders for
ing adherence to transfusion guidelines. Until blood components that fall outside established
TABLE 21-2. Types of Blood Utilization Review
Prospective Real-time Residents Proactive Labor-intensive

Medical director III! Improves patient III! Potential to
Blood bank staff care and saves cause delays in
CPOE blood in real issuing blood
time
Concurrent Within 12-24 hours Residents Consultative Labor-intensive

TSO opportunity Ineffective if no
Training tool for physician
residents involvement
Retrospective Days to weeks Qualityassur- III! Easiestapproach Nonstandard-

internal ance personnel Provides datafor ized review
Medical director trending and Difficult to use
Clinical peers benchmarking data for direct
physician com-
parison
Retrospective Days to weeks Network of Objective, thor- III! Up-front

external trained peer ough, and stan- expensethat can
reviewers dardized review be offset by sav-
Produces ings in reduced
directly compa- blood usage
rable data
CPOE= computerized provider (physician) order entry; TSO = transfusion safety officer.
transfusion guidelines. Ideally, this review incor- Examples of orders subject to prospective re-
porates both laboratory values and clinical data views include the following: requests for multi-
obtained from patient records and/or direct ple doses of platelets without checking post-
communication with the clinical team. Prospec- transfusion platelet counts in patients without
tive review improves patient care and saves acute hemorrhage, orders of inadequate doses of
blood components but is labor-intensive and can plasma components or cryoprecipitate, and or-
cause delays in situations requiring urgent trans- ders for cytomegalovirus (CMV)-negativeor irra-
fusion support. Furthermore, increasing use of diated components. In case of a disagreement
between the ordering and reviewing physicians
point-of-care testing in operating rooms allows
about the appropriateness of a blood order, the
clinicians to make real-time transfusion deci-
usual practice is to defer to ordering physicians,
sions based on results that are usually not avail- as they are more familiarwith the patient's clini-
able to transfusion services at the time of the re- cal condition, as long as the transfusion of the
quest for blood components. Because of these requested component does not have the poten-
considerations, most hospitals exempt the emer- tial for an immediate adverse effect on the pa-
gency department, labor and delivery depart- tient. However, questionable blood component
ments, and operating rooms from a prospective requests such as these may be subsequently re-
review. ferred to the transfusion committee for further
review. Sometimes, the transfusion service phy- evidence-based guidelines for plasma transfu-
sician may request additional testing to be per- sion and evaluating additional information, such
formed before subsequent transfusions (eg, as the degree of abnormal result correction by
CMV testing for requests for CMV-negative plasma transfusion and the. presence of drugs
blood). that may interfere with the coagulation sys-
tem.35,36For example, cryoprecipitate audits can
Concurrent Review be based on fibrinogen levels, presence of bleed-
The concurrent review process involves the re- ing, or Factor XIIIdeficiency.
view of transfusions administered within the However, in nonacademic institutions, if
previous 12 to 24 hours, thus avoiding the risk there is no TSO, a concurrent review might be
of delaying patient care. This allows the review- too labor.intensive for the blood bank staff and
er to judge the appropriateness of blood compo- allow only for review based on laboratory trans-
nent use based on all pertinent laboratory and fusion thresholds without the ability to evaluate
clinical data. If the transfusion episode appears the patient's Clinicalcondition. Moreover, a Con-
to be out of line with good clinical practice, the current review is effective only when the trans-
review leads to an interaction with the transfus- fusion service physician or TSO is directly in-
ing clinician, who is still likely .to remember volved, resulting in a Significant reduction in
events surrounding the transfusion. Concurrent transrusions.":" In contrast, concurrent. review
review is a consultative opportunity, and the without physician involvement has failed,to re-
consultant should not be perceived as a gate- duce blood usage."
keeper. Rather, if properly conducted, this inter-
action can promote cooperative relationships be- Retrospective Review
tween the transfusion service and medical staff,
and influence transfusion practices.'? Further- A retrospective review can be performed by
more, performing these audits in academic insti- those internal to, or external to, the organiza-
tutions represents an excellent training tool for tion. Such reviews are performed days to weeks
clinical pathology residents. Residents should be following the transfusion event, usually at speci-
encouraged to document these reviews, evalu- fied time intervals, using preset audit criteria
ate them for efficacy,and present them in an ed- that might differ from transfusion guidelines.
ucational format to other residents.
An example of a concurrent review that can Internal Retrospective Review
result in consultative opportunity is a daily re- Quality assurance personnel can perform the ini-
view of platelet transfusions along with pre- and
tial review. Cases that fall outside audit criteria
posttransfustcn platelet counts. In their audits of
can be referred to a transfusion medicine physi-
transfusion events that fall outside audit criteria
cian or a clinical peer of the physician whose
for acceptable practice, residents should evalu-
ate patient records for the presence of bleeding, transfusion decision is under review; this allows
deficiencies in platelet function, and the pres- the closer evaluation to determine whether the
ence of medications known to affect platelet patient's cllnical condition justified the compo-
number or function. This review allows the nent request. This is the easiest approach to uti-
identification of not only patients who may be lization review, providing valuable information
receiving platelets outside of transfusion guide- about overall institutional transfusion practices
lines but also those who might be refractory. 34 and data for trending and benchmarking. How-
Similarly, residents can perform plasma audits ever, one limitation of internal retrospective re-
from daily reports that include pre- and post- view is that patient data are often generated in a
transfusion prothrombin time/international nor- nonstandardized and inconsistent manner; thus,
malized ratio (PT/INR) and activated partial direct comparison of care provided by any two
thromboplastin time (aPTT) levels, using physicians can be difficult.
External Retrospective Review tients and that early, aggressive,and ratio-based

blood component resuscitation provides im-
This process uses an external third party that
proved outcomes." This is best achieved by
provides a network of trained physicians to
protocol-driven care, now recommended by the
serve as peer reviewers. Patient charts and
National Partnership for Maternal Safety, Na-
transfusion decisions are evaluated using objec-
tional Institute for Health and Care Excellence
tive criteria in a thorough, critical, and standard-
policy, and Trauma Quality Improvement Pro-
ized way, to produce directly comparable data.
gram guidelines.P'" Development of MTPs that
To be successful, it must allow review of all
include point-of-caretesting and hemostatic re-
medical records, including all forms of paper
suscitation requires careful planning and cooper-
and electronic data, and not require additional
work on the part of the hospitals. Because a net- ation between various clinical teams, including
work of trained physicians provides the anony- surgery, anesthesia, and transfusion medicine.
mous review of colleagues whose identity is Review of transfusionsis especiallyimportant in
blinded to the reviewers, the reviews are less this high-risk patient population, as it offers an
likely to be biased; physicians often find it chal- opportunity for continuous improvement of
lenging to perform unbiased critical reviews of both the protocol and the delivery of care by the
colleagues with whom they have social, eco- multidisciplinary team. Such reviews can yield
nomic, and politicalrelationships. important results as measured by improved pa-
Although retrospective reviews cannot tient outcomes and improved blood utilization.
change blood component usage in real time or Because of the acuity of transfusion require-
effectively involve transfusing clinicians in the ments, performing a prospective review of mas-
review process, they are excellent models of sive transfusions in real time can lead to delays
peer review." They provide opportunities to in patient care and is often not feasible.This re-
mentor physicianson proper transfusionpractic- view is best achieved by concurrent or retro-
es, such as correction of anemia, blood loss con- spective review of data on MTP availability,cri-
trol, iron supplementation, evaluation of vital teria for activation, timeliness of provision of
sign monitoring, and avoiding 2-unit transfu- components, and protocol effectiveness.2s,43,44
sions without checking the response." One ef-
fective indicator for retrospective monitoring of
transfusion appropriateness is the transfusion re-
cipient's discharge hemoglobin value." Further-
more, if trends are identified in certain depart-
ments indicating frequent transfusion outside of
transfusion guidelines, letters can be sent to cli-
nicians and department heads, and representa- Applicationof health-care IT to transfusion med-
tives can be invited to transfusion committee icine provides PBMprogramswith new tools for
meetings. blood utilization review and for influencingphy-
sician ordering practices. Blood component or-
der entry screens can incorporate clinicalguide-
lines designed to guide physiciansregarding the
appropriateness of component orders." Imple-
mentation of computerized provider order entry
(CPOE)allows the creation of electronic reports
Uncontrolled hemorrhage requmng massive that can be used for order auditing, assessment
transfusion in patients with trauma or postpar- of blood utilization, and compliance monitor-
tum hemorrhage remains one of the leading ing.' In the 2013 AABBsurvey, 77.5%ofinstitu-
causes of preventable death. It is now recog- tions with transfusion guidelines incorporated
nized that severe trauma-induced coagulopathy them into paper or electronic order sets. Sixty-
increases morbidity and mortality in these pa- five percent reported that their CPOEsystem in-
eluded transfusion guidelines. Of these, approxi- atic reviews of the impact of an electronic CDSS
mately half had guidelines with "hard stops," found that it is a useful educational tool and its
where the physician/provider must select a implementation improves RBCusage, increases
transfusion rationale before the order is accept- provider compliance with guidelines, and re-
ed. Only 45.9% of these CPOEsystems included sults in cost savings.48.50Additional studies are
an algorithm orclinical decision support to warn needed to assess the effectiveness of the CDSS
or alert the physician/provider when the blood on ordering practices for other components and
component order is outside of guidelines. Most patient outcomes.Y"
reporting hospitals (71.9%) required that a phy-
sician document the reason .or clinical justifica-
tion for transfusion in the medical record based
on guidelines developed by the hospital transfu- N
sion or quality committee. 16
In some hospitals, guidelines incorporated
into the CPOE system do not result in hard
stops, but rather, the indication for a blood com- The advent of the EHRis opening up the field of
ponent order is captured electronically and "big data" in health care. Data collection and
stored for later review. If the most recent labora- data mining through the linkage of different data
tory value does not match the indication, an au- sets within the EHRcan provide a plethora of in-
tomatic message is provided to the clinician, re- formation, which can drive improvements in pa-
questing an indication for the "override." tient outcomes.l'P Examples of using big data
Component orders with overrides are collated in transfusion medicine are described below.
electronically with other relevant information
and are reviewed by transfusion committee Benchmarking
members.
CPOE alone does not always improve blood Comparisons of blood usage between institu-
use if only laboratory parameters are used for tions and/or countries, performed with the aim
determining transfusion appropriateness, or if of identifying best practices in transfusion medi-
clinical users record inaccurate reasons for cine, were recently lntroduced.P Major chal-
transfusion. Therefore, retrospective review re- lenges in developing meaningful benchmarking
mains important, even with CPOE, to evaluate processes include.definingbest practices and de-
transfusion appropriateness by reviewing clini- veloping cost-effectivedata .collection methods.
cal information. Additional functionality is Often, hospitals do not have adequate IT sup-
achieved by incorporating a clinical decision port to develop .data mining tools that enable
support system (CDSS) into CPOE to guide transfusion. services to collect uniform data
transfusion ordering by providing clinicianswith across different hospitals.In the.20 13AABBsur-
tailored treatment recommendations based on vey, 41.9% of responding hospitals participated
individual laboratory and clinical information, in performance benchmarking programs relating
along with local transfusion guidelines.1,46,47 For to transfusion medldne."
example, based on the clinical reason for trans- Implementing benchmarking principles in
fusion selected by the clinician from a menu of transfusion medicine has improved practices on
choices, and the most recent laboratory values, individual and organizational levels. Bench-
the system could provide adaptive alerts, allow- marking can identify existing gaps that can be-
ing the clinician to modify orders that do not fall come PBM program improvement targets and
within guidelines. This system is particularly represents an instrument of learning, as it relies
useful if it requires physicians to enter the rea- on communication, collaboration, and sharing
son for overriding the system's recommenda- of experiences. Benchmarking can identify
tion. Subsequent analysis of reports, with rea- opportunities for savings, which can be high-
sons for the override, can be used to audit lighted in presentations to hospital leadership.
individual physician ordering practices. System- Furthermore, after the implementation of new
practices, performance should be reevaluated, tality data, and observations or supportive state-
preferablyby continuous data collection. ments from key stakeholders." Several bench-
Three possible models of benchmarking in marking databases exist that allow comparisons
transfusion medicine have been described by of PBM metrics with member hospitals nation-
Apelseth et al.52 The ideal model according to wide or with other Similarlysized hospitals/"
these authors would be a regional benchmark- Commonly used metrics are summarized in Ta-
ing model, which requires a central coordinator ble 21-3. In the 2013 MBB survey, the two
who facilitates communication between institu- most common metrics used by institutions to
tions and manages the information flow to par- measure transfusion rate were 1) the total num-
ticipants. In a sentinel model, data from a limit- ber of RBCs, platelets, and plasma units trans-
ed number of sites are collected into a central fused (71.5%), and 2) the total number of com-
database, preferably by web-based reporting. ponents transfused (57.3%).16 Other measures
This model is less costly and easier to implement included the percentage of patients who re-
and can be applied nationally and international- ceived transfusion per hospital admission; the
ly. These two models gather information into a number of RBCunits per transfusion recipient;
centralized database, whereas in a third, institu- transfused RBC units per 100 hospital admis-
tional model, all data are collected for the insti- sions or discharges, per 1000 patient days, per
tution initiating the benchmarking process. The adjusted patient days, or per adjusted or case
latter is least likely to succeed because it re- mix index (CMI)-adjusted discharges; transfu-
quires individual initiative, volunteer collabora- sions per surgical case; and transfusions per
tion, and resources to be successful. medical/surgical admission. For a meaningful
Benchmarking data can come from various comparison of transfusion rates between differ-
sources, including medical record reviews, ent institutions, it is important to use metrics
third-party gap assessments, blood bank invento- that capture transfusions in both inpatient and
ry management, financial systems (eg, patient outpatient settings; also, both the number of
billing and budget), patient morbidity and mor- patients (eg, admissions or discharges) and
TABLE 21-3. PBM Metrics Used for Benchmarking
€I Transfusion rate compared with that of comparably sized hospitals measuredas one of the following:
- Overall number of components transfused
- Percentageof patients per admission who receivedtransfusion
- RBGunits per transfusion recipient
- Transfused RBGunits per 100 hospital admissions or discharges
- Transfused RBGunits per 1000 patient days
- Transfused RBGunits per adjusted patient days
- Transfused RBGunits per adjusted or GMI-adjusted discharges
- Transfusions per surgical case
- Transfusions per medical/surgical admission
Percentageof transfusions that fall outside of hospital or professional transfusion guidelines
€I Transfusion administration compliance rates
Reviews of wasted blood components
III Budget for transfusion services
RBC= Red Blood Cell; CMI = case mix index.

complexity of care (eg,iCMI-adjusted discharg- staff knowledge and compliance with .reporting
es) affect the normalization of data for correla- of adverse events. 14
tion between institutions.
Transfusion benchmarking described in the Monitoring Patterns of Blood Use by
2014-2015 MBB survey is another example of Procedure over Time
the sentinel model." This benchmark report col-
lated data sourced from 435 hospitals respond- Linkingtransfusion episode data with diagnostic
ing to the blood utilization survey. Hospitals and procedure codes allows monitoring of trans-
were stratified into eight categories based on fusion rates for specific types of cases (eg, hip re-
their 2014-2015 annual patient surgical vol- placement, cardiac surgery). These rates can be
ume. Transfusion and blood component expira- compared to national data or data in the litera-
tion data were presented for each component as ture. In the MBB survey, some hospitals report-
mean and median values for all categories. Us- ed using the percentage of patients who re-
ing this report, hospitals can compare their data ceived transfusion per selected diagnosis code
to other hospitals with similar annual inpatient (eg, lCD-I0), or average blood component use
surgical volumes. Benchmarking can contribute per surgical case category (eg, coronary artery
to evidence-based guidelines, especially when bypass graft only or knee/hip replacement). 16
randomized controlled trials may be difficult or
costly to perform. Examples of countries that re- Definition of Blood Order Schedules
duced blood utilization using national PBM and
benchmarking programs include Finland, Scot- A maximum surgical blood order schedule
land, Spain, the Netherlands, Germany, the (MSBOS)for surgery can be developed by using
United Kingdom, Canada, and Australia. data held within the anesthesia information
management system.56 Creating a data-driven
Active Surveillance of Transfusion- MSBOS is useful in optimizing transfusion-or-
Related Complications dering practices.57 One metric used for assessing
and benchmarking transfusion ordering practic-
An example of active surveillance of transfusion- es is the CIT ratio, which may indicate overor-
related complications is searching large Medi-
dering of crossmatched blood compared to actu-
care databases and linking the procedure code
al transfusion when the ratio exceeds 2.0.58
for transfusion with diagnostic codes to deter-
Duplicate type-and-screen procedures and un-
mine the rate of a recognized complication such
as posttransfusion purpura, or using an automat- necessary crossmatches for patients undergoing
ed electronic algorithm to detect transfusion- elective surgical procedures can be wasteful of
associated circulatory overload or transfusion- transfusion service resources. 59 However, it is
related acute lung injury and linking this to the important to emphasize that should a patient
transfusion episode.51,55 Transfusion reaction have an.active order for type and screen but be
tracking can help identify underrecognition or brought to the operating room for elective sur-
underreporting of transfusion-related adverse gery before completing the ordered testing, such
events. Once a problem is identified, education a practice could present both potential safety
can be developed and implemented to improve and quality issues in transfusion medicine.
3. Transfusion auditing can be performed as a prospective, concurrent, or retrospective review.

Retrospective reviews may use either internal or external resources.
4. Different metrics have evolved over time to assess various aspects of transfusions and allow
benchmarking of transfusion practices at different levels.
S. Various national organizations have developed educational resources and campaigns to pro-
mote the wise use of blood components. AABBand The Joint Commission, for example, have
joined forces to produce a PBM certification program. These tools enable physicians to elimi-
nate unnecessary blood transfusions and optimize transfusions in high-risk patients while re-
ducing health-care costs and improving patient outcomes.
REFERENCES
1. Goodnough LT, Shah N. The next chapter in pa- 10. Rohde JM, Dimcheff DE, Blumberg N, et al.
tient blood management: Real-timeclinical deci- Health care-associated infection after red blood
sion support. Am J Clin Patho12014;142:741-7. cell transfusion: A systematic review and meta-
2. Morton J, Anastassopoulos KP, Patel ST, et al. analysis.JAMA2014;311: 1317-26.
Frequency and outcomes of blood products 11. Goodnough LT,Verbrugge D, Vizmeg K, Riddell
transfusion across procedures and clinical condi- J. Identifying elective orthopedic surgical pa-
tions warranting inpatient care: An analysis of tients transfused with amounts of blood in ex-
the 2004 healthcare cost and utilization project cess of need: The transfusion trigger revisited.
nationwide inpatient sample database. Am J Transfusion 1992;32:648-53.
Med Qual 201 0;25:289-96. 12. Shander A, Fink A, Iavidroozi M, et al. Appro-
3. Ferraris VA, Davenport DL, Saha SP, et al. Surgi- priateness of allogeneic red blood cell transfu-
cal outcomes and transfusion of minimal sion: The international consensus conference on
amounts of blood in the operating room. Arch transfusion outcomes. Transfus Med Rev 2011;
Surg 2012;147:49-55. 25:232-46.
4. Carson JL, Carless PA, Hebert PC. Outcomes us- 13. Spahn DR, Shander A, Hofmann A. The chiasm:
ing lower vs higher hemoglobin thresholds for Transfusion practice versus patient blood man-
red blood cell transfusion. JAMA 2013;309:83- agement. Best Pract Res Clin Anaesthesiol2013;
4. 27:37-42.
5. Goodnough LT, Shah N. Is there a "magic" he- 14. Building a better patient blood management
moglobin number? Clinical decision support program: Identifying tools, solving problems and
promoting restrictive blood transfusion practic- promoting patient safety (white paper). Bethes-
es. AmJ Hematol2015;90:927-33. da, MD: AABB, 2015. [Available at http://
6. Hebert PC, Wells G, Blajchman MA, et al. A www.aabb.org/pbm/Documents/AABB-PBM-
multicenter, randomized, controlled clinical trial Whitepaper.pdf (accessed December 1,2019).1
of transfusion requirements in critical care. 15. Murphy MF, Yazer MH. Measuring and moni-
Transfusion requirements in critical care investi- toring blood utilization. Transfusion 2013;53:
gators, Canadian Critical Care Trials Group. 3025-8.
N EnglJ Med 1999;340:409-17. 16. Whitaker BI, Rajbhandary S, Harris A. The 2013
7. Carson JL, Brooks MM, Abbott JD, et al. Liberal AABBblood collection, utilization, and patient
or restrictive transfusion in high-risk patients blood management survey report. Bethesda,
after hip surgery. N Engl I Med 2011;365:2453- MD: AABB, 2015. [Available at http://www.
62. aabb.org/research/hemovigilance/bloodsur
8. Salpeter SR, Buckley JS, Chatterjee S. Impact of vey/Pages/ default.aspx (accessed December 1,
more restrictive blood transfusion strategies on 2019)·1
clinical outcomes: A meta-analysis and systemat- 17. Wagner J, AuBuchon JP, Saxena S, Shulman lA,
ic review. AmJ Med 2014;127:124-31. for the Clinical Transfusion Medicine Commit-
9. Goodnough LT, Maggio P, Hadhazy E, et al. Re- tee. Guidelines for the quality assessment of
strictive blood transfusion practices are associat- transfusion. Bethesda, MD: AABB,2006.
ed with improved patient outcomes. Transfu- 18. Becker J, Shaz B, for the Clinical Transfusion
sion 2014;54(Pt 2):2753-9. Medicine Committee and the TransfusionMedi-
cine Section Coordinating Committee. Guide- About_ACOGI ACOG_Departments/Patient_

lines for patient blood management and blood Safety_and_Quality_Improvement! Choosing;
utilization. Bethesda, MD: AABB,2011. Wisely (accessedDecember 1,2019).J
19. Saxena S, ed. The transfusion committee: Put- 31. Lieberman L, Lin Y, Cserti-Gazdewich C, et al,
ting patient safety first. 2nd ed. Bethesda, MD: for the QUEST-Quality in Utilization Educa-
AABBPress, 2013. tion and Safety in Transfusion=Research Col-
20. Mair B, Agosti SJ, Foulis PR, et al. Monitoring laborative. Utilization of frozen plasma, cryopre-
for undertransfusion. Transfusion 1996;36:533- cipitate, and recombinant factor VIla for
5. children with hemostatic impairments: An audit
21. Dunbar NM, SzczepiorkowskiZM. How do we of transfusion appropriateness. Pediatr Blood
utilize a transfusion safety officer? Transfusion Cancer 2018:65(4).
2015;55:2064-8. 32. Tinmouth A, Thompson T, Arnold DM, et al.
22. Implementation guide for The Joint Commis- Utilization of frozen plasma in Ontario: A prov-
sion Patient Blood Management Performance ince wide audit reveals a high rate of inappropri-
Measures 2011. Oakbrook Terrace, IL: The ate transfusions. Transfusion 2013;53(10):
Joint Commission, 2011. [Availableat https:11 2222-9.
www.jointcommission.org/ assets/l/6/pbm_ 33. Grey DE, Smith V, Villanueva G, et al. The utili-
implementation_guide_20 11 0624. pdf (ac- ty of an automated electronic system to monitor
cessed December 12, 2019).J and audit transfusion practice. Vox Sang 2006:
23. De Leon EM, SzallasiA. "Transfusionindication 90;316-24.
RBC (PBM-02)": Gap analysis of a Joint Com- 34. Kaufman RM, DjulbegovicB, Gernsheimer T, et
mission Patient Blood Management Perfor- al. Platelet transfusion:A clinical practice guide-
mance Measure at a community hospital. Blood line from the AABB.Ann Intern Med 2015;162:
Transfus 2014;12(Suppl1):187-90. 205-13.
24. Frey K, ed. Standards for a patient blood man- 35. Roback JD, Caldwell S, Carson J, et al.
agement program. 3rd ed. Bethesda, MD: Evidence-based practice guidelines for plasma
AABB,2020. transfusion. Transfusion 2010;50: 1227-39.
25. AABB, The Joint Commission. Patient blood 36. Murad MH, SlubbsJR, Gandhi MJ, et al. The ef-
management certification review process guide fect of plasma transfusion on morbidity and
for health care organizations 2017. Oakbrook mortality: A systematic review and meta-analy-
Terrace, IL:The Joint Commission,2017. [Avail- sis. Transfusion 2010;50: 1370-83.
able at https:llwww.jointcommission.org/ 37. Silver H, Tahha HR, Anderson J, et al. A non-
assets/1/1812017 _PBM_Org_RPG.pdf (ac- computer-dependent prospective review of
cessed December 1, 2019).J blood and blood component utilization. Transfu-
26. National Blood Authority Australia. Patient sion 1992;32:260-5.
blood management guidelines. Canberra, Aus- 38. Simpson MB. Prospective-concurrent audits and
tralia: National Blood Authority, 2019. [Avail- medical consultation for platelet transfusions.
able at https:llwww.blood.gov.au/pbm-guide Transfusion 1987;27: 192-5.
lines (accessed December 1, 2019).J 39. Hawkins TE, Carter JM, Hunter PM. Can man-
27. Smith WR. Evidence for the effectiveness of datory pretransfusion approval programmes be
techniques to change physician behavior. Chest improved? TransfusMed 1994;4:45-50.
2000;118(2 Suppl):8SYI7S. 40. Lam HT, Schweitzer SO, Petz L, et al. Effective-
28. Yeh DD, Naraghi L, LarentzakisA, et al. Peer-to- ness of a prospective physician self-audit
peer physician feedback improves adherence to transfusion-monitoring system. Transfusion
blood transfusion guidelines in the surgical in- 1997;37:577-84.
tensive care unit. J Trauma Acute Care Surg 41. Paone G, Brewer R, LikoskyDS, et al. Transfu-
2015;79:65-70. sion rate as a quality metric: Is blood conserva-
29. Cauldwell M, Shamshirasz A, Wong TY, et al. tion a learnable skill? Ann Thorac Surg 2013;
Retrospective surveys of obstetric red cell 96: 1279-86.
transfusion practice in the UK and USA. Int J 42. Edwards J, Morrison C, Mohluddin M, et al. Pa-
Gyenecol Obstet 2017; 139:342-5. tient blood transfusion management: Discharge
30. Choosing Wisely. Washington, DC: American hemoglobin level as a surrogate marker for red
College of Obstetricians and Gynecologists, blood cell utilization appropriateness. Transfu-
2019. [Available at https:llwww.acog.org/ sion 2012;52:2445-51.
43. Stephens CT, Gumbert S, HolcombJB. Trauma- 52. Apelseth TO, Molnar L, Arnold E, Heddle NM.
associated bleeding: Management of massive Benchmarking:Applicationsto transfusion med-
transfusion. Curr Opin Anaesthesiol 2016;29: icine. TransfusMed Rev2012;26:321-32.
250-5. 53. Carson JL, Guyatt G, Heddle NM, et al. Clinical
44. Kacmar RM, Mhyre JM, Scavone BM, et al. The practice guidelines from the AABB:Red blood
use of postpartum hemorrhage protocols in cell transfusion thresholds and storage. JAMA
United States academic obstetric anesthesia 2016;316:2025-35.
54. Rajbhandary S, Whitaker B, Perez GE. The
units. Anesth Analg 2014;119:906-10.
2014-2015 AABBblood collection and utiliza-
45. Dzik S. Use of a computer-assisted system for
tion survey report. Bethesda,MD: AABB,2018.
blood utilization review. Transfusion2007;47(2 [Availableat http://www.aabb.org/research/
Suppl):142S-4S. hemovigilance/bloodsurvey /Pages/ default.
46. Goodnough LT, Shieh L, Hadhazy E, et al. Im- aspx (accessedDecember 1, 2019).]
proved blood utilization using real-time clinical 55. CliffordL, SinghA, Wilson GA,Toy P. Electron-
decision support. Transfusion 2014;54: 1358- ic health record surveillancealgorithmsfacilitate
65. the detection of transfusion-related pulmonary
47. RothschildJM, McGurk S, Honour M, et al. As- complications.Transfusion2013;53: 1205-16.
sessment of education and computerized deci- 56. Frank SM, RothschildJA,Masear CG, RiversRJ.
sion support interventions for improving trans- Optimizing preoperative blood ordering with
fusion practice. Transfusion2007;47:228-39. data acquired from an anesthesia information
48. Hibbs SP, Nielsen ND, BrunskillS, et al. The im- management system. Anesthesiology2013; 118:
pact of electronic decision support on transfu- 1286-97.
sion practice: A systematic review. Transfus 57. Cheng CK, Trethewey D, Brousseau P, Sadek I.
Creation of a maximum surgical blood ordering
Med Rev2015;29: 14-23.
schedule via novel low-overhead database meth-
49. Cohn CS, Welbig J, Bowman R, et al. A data-
od. Transfusion2008;48:2268-9.
driven approach to patient blood management.
58. Dexter F, Ledolter J, Davis E, Witkowski TA.
Transfusion 2014;54:316-22. Systematic criteria for type and screen based on
50. Dunbar NM, Szczepiorkowski ZM. Hardwiring procedure's probability of erythrocyte transfu-
patient blood management: Harnessinginforma- sion. Anesthesiology2012; 116:768-78.
tion technology to optimize transfusion practice. 59. Compton ML, SzklarskiPC, Booth G. Duplicate
Curr Opin Hematol2014;21:515-20. type and screen testing: Waste in the clinical
51. Pendry K. The use of big data in transfusion laboratory.Arch Pathol LabMed 2018;142;358-
medicine. TransfusMed 2015;25:129-37. 63.
CHAPTER
Non.infectious Complications of
Blood Transfusion
Eldad A. Hod, MD, and Richard O. Francis, MD, PhD
TATISTICALLY,THE GREATESTRISKOF
morbidity and mortality from transfusion is
from the noninfectious complications. In
F N
fact, transfusion-associated circulatory overload
(TACO), transfusion-related acute lung injury
(TRALI), and hemolytic transfusion reactions
Identification of a Transfusion Readion
(HTRs) are the three most commonly reported
causes of transfusion-related mortality. 1 As with many medical therapies, adverse effects
of transfusion usually cannot be accurately pre-
dicted or the risks completely avoided. Trans-
H fusing clinicians should be aware of such risks
when discussing the need for transfusion with a
Hemovigilance involves systematic surveillance patient. Informed consent for transfusion may
for the complications of transfusion, analysis of
include a discussion of the risks of infectious
these data, and subsequent data-driven improve-
disease and serious noninfectious complica-
ments in transfusion practices. One of the main
purposes of a hemovigilance program is to im- tions, such as TACO, TRALI, and HTRs. Fur-
prove reporting of transfusion-related adverse thermore, medical staff administering blood
events. It is widely believed that the major non- components should be well aware of the signs
infectious complications are both underrecog- and symptoms of possible reactions. Staff
nized and underreported. should be prepared to mitigate any immediate
The National Healthcare Safety Network episodes through appropriate procedures as
Hemovigilance Module was created via a collab- well as to prevent future similar reactions
oration between government and nongovern- whenever possible.
ment organizations to implement a national sur- Many common clinical signs and symptoms
veillance of transfusion-related adverse events are associated with more than one type of ad-
aimed at improving patient safety. Definitions verse reaction. (See Table 22-1.) Early recogni-
and classificationschemes are detailed in the ap- tion, prompt cessation of the transfusion, and
pendices of the Hemovigilance Module Surveil- further evaluation are key to a successful out-
lance Protocol.' come. The signs and symptoms that may be
Eldad A. Hod, MD, Associate Professor; and Richard O. Francis, MD, PhD, Assistant Professor, Department of
Pathology and Cell Biology, Columbia University Medical Center, New York, New York
627
.i!l
e:
CD
cn e:
E o
CD
c.
e: E
o o
c. (..)
E
o
..... (..)
.E (..)
e ~
55 ~
e E
cn
CD
a.. ::c
CD
.::.:;
zs
"(j5
(..) cn
CD
.c: o
(..) c.
CD
e: "_£'
~
"0:0
I
"0
'"
CD
.c:
fi
o Ol
Ole:
~~
== 0
>
.c:
(..)_
,_ .c:
- CD
I- CD (..)
"Q5
~ ..f '"
"0
c:::
'"
en
c::: ....
a o
~ e:
~ "S,
o
'"
a:>
0:
CI
C;
o
+-'
"0
.s>-
c::: >-cn
a
"Cii =EE
C
"0(.)
..8 co
"0
o
.c
-:::J
en
c:::
~ 1
c
~:s
«
:;::;
«
e:
I- CI
a:>
ea:> ~ca
I
G) cn
>
"0
a: "E o
o
c
«
-a
en
a:>
CI
"iii
=
';,i
c
.c:£
cr.1;i
°E
co
«
o
c5
"':":
"C
a 1'!!
C) t-
a:>
1<; ~
(.) =
CI
(..)
:g
..c
o
Q
E
~ :g CD
s:
(5 e:
oS E o
=
~ CD e:
< ::c
CHAPTER 22 Noninfectious Complications of Blood Transfusion 629
Cl>
c::
'C
.x:
Cl.
Cl>
c::
'0..
u.J
,
<:(
~ -
e
Cl>
u.J
c:: Cl >
c:: E en
'-'
0; 0 .....
'Ci)
::::> C'd
Cl.
»
Q3
c:: '-' Cl> .....

c:: en
f5
en :c -e
0 Cl c::
en
- -
0
rJ
'C
0,7
0 +=
c:: ~ -c
Cl>
Cl. Cl. C'd c:: en ::::>
0 C'd
..... .r:: en '-' 't;
Cl>
Cl.-c c:: c:: c::
Cl.c::
C'd C'd
.:
0
+= '-'
0
-
~
o
o
o
ci
If?
.,....
6
o
o
ci
'.'!
b
"~ 0
(Jl 0
"=::-E
o C1)
E.c ~
C1) .... c.
-
<=
C1)
.c,E
,
<=
:c :c
o
C1)
::::l o
o
<=
+='
<=
E
C1)
o .c
S
-
•<=
C1)
E
C1)
.~"'
::::l
<=
zs
o
c:n Oro
§ "E
ro o
<=
ro C1)
2 :::c
"C
o
o
15
,.. C1)
§
(Jl
Ow
N
N E~
.....
W
a:I
E E
"_ C1)
§.c
;:: 2:
CHAPTER 22 Noninfectious Complications of Blood Transfusion 631
en
Q) c::
+-' 0
e °Ui
=>
c:: c:: 'Iii -0
0 0 0
°Ui ~ c:: 0
=> +-' ~ zs
.....
'Iii c::
c:::Q)
i::' -0
Q)
e
+-'
E
Q)
<tS
en
en
(.>
=>
Q) 0.. -0
(.>
=>
0.
=>
Q)
(.>
Q)
e,
-0 en c::: Q)
.... E
Q)
'<,
c::
=>
>,
(.>
Q) =>
en "C:5 .-0 0
=> °0 ..x:
=>
<tS Ci'i > ...J
Q)
a... u «
c::
0 c::
Q)
Ci'i
2: E e
(.>
c:::
Q)
Q)
+-'
oS:: 0> e en
» e
(.>
E I-;- 0 -0
0
en
=> >-
°u a ~<tS .Q
:;::; -0
0
Cil -0
Q)
(.> c:: .Q
(.> <tS
-0
0> l:: +-'
:;::;
c::
Q) c::: +-'
(.>
Q)
<tS
0
~ «
N Q)
02:
~
e
a...
03 ...J
0:: :c
,
..c:::
>,
....
..c:::
.... en
""
:>;
en
Q)
c:::
c::: 0;::
~
.... .....
0
(.>
c<f e
°Ui
Q)
1i3
....
..c::: ~
1;) <tS
e
<tS
°E
03
1<1
a... 0::
c::
c::: 0 E en
0 °Ui en °Ui
-c:::i=0 e
-
°Ui :::::l
=> 'lii 0 Q)
.s c:: 0
....~ zs ~
.Q
x:
o.~
en
::s
Q)
"" ;S
'r2 >
Q)
-0
Q) E as ~
=> -0 en
...... l'! -01<1 -0 c::: ~
°u °Uien ..... Q) .... Q) <tS
'zs g·E 0....
- - en 03
(.>
-0 <tS
00.. .§_ Q) U 1<1
<tS 0 -0 0 0. co 0..
cc 3:
,
oS::
C3
c:::
....c:::
0 Ol
c:::
E ~
Q) Q) 0
,...
c::: en
-0
~
gs_"§ 0
Q) 0- ''':'":
Cl
i::" c::-
=>
°u ~
0 en
C'C
Ci'i
°u ~N c:::
°E (.> 'x
0 02:
Q)
0>
Q)
-0 :;::;
(.> Q)
+-' :::s c:::
Cil .t::!
Q)
E <tS
l'!
(.) .. t::
o 0 E «
...J
0. 0- :t= :§ :c
».~ '-'
:c «
'"
's,
CI
'E
=E=
E
'=
CI
1
=
CI
~ca
a:>
cr::
=
CI
'in
=
1;;
=
~
'E'
=
CI
.c:
~
~
'C
~
ca
a;
Q
C HAP T E R 22 NoninfectiousComplicationsof Blood Transfusion 633
indicators of a transfusion reaction include the Component-Focused Steps

following:
1. Contact the transfusion service for directions
'" Fever, generally defined as a 21 C rise in on investigating and documenting the poten-
temperature to 238 C (the most common tial causes of the reaction.
sign of an acute HTR). 2. Obtain instructions concerning the return of
~ Chillswith or without rigors. any remaining component, associated intra-
'* Respiratory distress, including wheezing, venous fluid bags, and tubing,
coughing, hypoxia, and dyspnea. 3. The transfusion service determines whether
@ Hyper-or hypotension.
the blood donor center should be notified of
" Abdominal, chest, flank, or back pain.
 Pain at the infusion site. the acute transfusion reaction. The Foodand
'* Skin manifestations, including rash, flush- Drug Administration (FDA)requires report-
ing, urticaria, pruritus, and localized edema. ing to the blood center when the component
• Jaundice or hemoglobinuria. is at fault for causing the reaction (eg, sus-
$I Nausea/vomiting. pectedproblem with labelingor manufactur-
• Abnormal bleeding. ing or suspected bacterial contamination of
* Oliguria/anuria. the component). The FDA must be notified
"as soon as possible" when a complication of
Clinical Evaluation and Management of transfusion is confirmed to be fatal [Code of
a Transfusion Reaction Federal Regulations (CFR) Title 21, Part
The evaluation of a suspected transfusion reac- 606.l70(b)].
tion involves a two-pronged investigation com-
bining clinicalevaluation of the patient with lab- Standard laboratory Investigation of ill
oratory verification and testing. The clinical Transfusion Reaction
team should discontinue the transfusion of the When the laboratory receives notice of a possi-
implicated component, and contact the blood ble transfusion reaction, the technologist should
bank for directions on the investigation. When perform several steps:
an acute transfusion reaction is suspected, sever-
al steps must be taken immediately (below). 1. Request return of any remaining component,
associated intravenous fluid bags, and tubing
Patient-Focused Steps for possible bacteria culture or Gram stain-
ing.
1. Stop the transfusion immediately but keep 2. Request collection of a posttransfusion blood
the line open with normal saline, and pre- sample.
serve any residual component and the infu- 3. Perform a clerical check of the component
sion set for further evaluation. bag, label, paperwork, and patient sample.
2. Document the clerical recheck between the 4. Repeat ABO testing on the posttransfusion
patient and the component. The labels on sample.
the component, on patient records, and: on 5. Perform a visual check of pre- and posttrans-
patient identification should be examined for fusiO)).?a,tllples.for .:eVidence of l1eW()~y~is
identification errors. Transfusing facilities (which may not be vi.s.ible.if <50 mg/dl, of
may require repeat ABOand Rh typing of the hemoglobin is present).
patient using a new sample. 6. Perform a direct antiglobulin test (DAT)on a
3. Consult the clinical team for a plan of care. posttransfusion sample.
4. Identify the appropriate additional diagnostic 7. Report findings to the blood bank supervisor
tests to work up the case. or transfusion service medical staff,who may
request additional studies or tests, request symptom is fever with or without accompanying
quarantine of co-components generated from chills or rigors. A patient with a mild reaction
the same donor collection, or impose transfu- may have abdominal, chest, flank, or back pain.
sion restrictions/instructions. If a patient has a severe AHTR, hypotension,
dyspnea, and flank pain may be present and, in
some cases, progress to shock with or without
The transfusion service must retain any pa-
accompanying disseminated intravascuiar coag-
tient records that are related to transfusion reac-
ulation (DIC).Red or dark urine may be the first
tions, clinically significant antibodies, or special sign of intravascular hemolysis, particularly in
transfusion requirements. Transfusion services an anesthetized or unconscious patient, who
are able to share medical information on patient may also present with oliguria or, in rare cases,
transfusion history. When a patient is cared for DIC. The severity of the symptoms of this reac-
by different transfusion services, medical warn- tion is related to the amount of incompatible
ing bracelets or wallet identification cards may blood transfused, antigen density, and antibody
benefit patients with red cell alloantibodies. characteristics (eg, concentration, class, sub-
class). Prompt recognition of the reaction and
Specialized Laboratory Investigations immediate cessation of the transfusion can pre-
for Selected Reactions vent grave consequences.
Additional laboratory evaluation may be re-
quired for investigations of some nonhemolytic Differential Diagnosis
transfusion reactions, such as anaphylaxis, sep- Many of the signs and symptoms of an immune-
sis, or TRALI, as described in their respective mediated AHTRalso occur in other acute trans-
sections below. fusion reactions. Fever with or without chills
and accompanied by hypotension may also de-
velop in transfusion-related sepsis and TRALI.
UT However, hemolysis is not associated with
AN TRALI,and respiratory difficultyis not typically
a symptom of AHTRs. Fever or chills are more
Acute or immediate transfusion reactions occur commonly caused by an FNHTRbut cannot be
within 24 hours of the administration of a com- distinguished initially from the more serious
ponent and often during the transfusion. Acute AHTRwithout an assessment of hemolysis. The
transfusion reactions include immune and patient's underlying disease process can also
nonimrnune-medlated hemolysis, transfusion- make the diagnosis of an AHTR difficult.Acute
related sepsis, TRALI,allergic reactions, TACO, hemolysis may result from nonimmune mecha-
sequelae of massive transfusion, air embolism, nisms, as describedin the "Nonimmune-Mediated
hypotensive reactions, febrile nonhemolytic Hemolysis"section below.
transfusion reactions (FNHTRs),and hypother-
mia. The clinical significance of an acute trans- Pathophysiology
fusion reaction often cannot be determined by The interaction of preformed antibodies with
the patient's clinical history or signs and symp- red cell antigens is the immunologic basis for
toms alone but requires laboratory evaluation. AHTRs.The most severe reactions are associat-
ed with transfusions of red cells that are ABOin-
Immune-Mediated Acute Hemolytic compatible with the recipient, resulting in acute
Transfusion Reactions intravascular destruction of the transfused cells.
Presentation Transfusion of ABO-incompatibleantibodies, as
can happen with minor-incompatible apheresis
Rapid hemolysis of as little as 10 mL of incom- platelets or intravenous immune globulin (IVIG)
patible blood can produce symptoms of acute infuslon, may also cause hemolysis. The most
HTR (AHTR). The most common presenting common circumstance for hemolysis following
C HAP T E R 2 2 Noninfectious Complications of Blood Transfusion 635
platelet transfusion is when group 0 platelets vation, including release of anaphylatoxins and
from donors with high titers of anti-A are trans- opsonization of red cells, may still have adverse
fused to group A recipients.' Although these effects. Furthermore, it has been demonstrated
forms.of acute hemolysis are not usually clinical- in a hemolytic transfusion reaction mouse mod-
ly significant or characterized by typical hemo- el that extravascular hemolysis can lead to cyto-
lytic symptoms, they can be severe if the trans- kine release, which may playa role in producing
fused components have high titers of ABO the effects of acute hemolysis."
antibodies. It should be noted that there are cas- Coagulation abnormalities that are associated
es of hemolysis with low-titer antibodies, and with AHTRs may be caused by various mecha-
that not all high-titer antibodies result in hemol- nisms. The intrinsic pathway of the clotting cas"
ysis, suggesting that donor/recipient character- cade may be activated by antigen-antibody inter-
istics may influence hemolysis risk. action, resulting in activation of Factor XII, also
When preformed immunoglobulin M (IgM) known as Hageman factor. Activation of the
or IgG antibodies recognize corresponding red Hageman factor can result in hypotension
cell antigens, complement activation may oc- through its effect on the kinin system. The kinin
cur, resulting in intravascular hemolysis, hemo- system produces bradykinin, which in turn in-
globinemia, and hemoglobinuria. IgM antlbod- creases vascular permeability and causes vasodi-
ies are strong activators of complement, and IgG lation," Activated complement, tumor necrosis
antibodies, when present at sufficient concen- factor alpha (TNFa), and interleukin 1 (IL-1)
trations and of the relevant subclass, may acti- may increase the expression of tissue factor. Tis-
vate complement as well. sue factor can activate the extrinsic pathway
Complement activation involves C3 cleavage and is associated with the development of DIC.
with the ensuing production of C3a, an anaphyl- DIC is a potentially life-threatening consumptive
atoxin thatis released into the plasma, and C3b, coagulopathy. Its characteristics include micro-
which coats the red cells. If complement activa- vascular thrombus formation with ischemic or-
tion proceeds to completion, membrane attack gan and tissue damage; consumption of plate-
complexes are assembled on the red cell sur- lets, fibrinogen, and coagulation factors; and
face, and intravascular lysis occurs. C5a, an ana- activation of fibrinolysis with production of fi-
phylatoxin that is 100 times more potent than brin degradation products. The end result of
C3a, is produced as part of this hemolysis. C3a these activations can vary from generalized ooz-
and C5a promote the release of histamine and ing to uncontrolled bleeding.
serotonin from mast cells, leading to vasodila- Shock may also accompany AHTRs. Hypo-
tion and smooth-muscle contraction, particular- tension, caused by the release of vasoactive
ly of bronchial and intestinal muscles. C3a and amines, kinins, and other mediators, produces a
C5a are recognized by many other cell types compensatory vasoconstrictive response that
and are involved in the production and release further aggravates organ and tissue damage. Re-
of cytokines, leukotrlenes, free radicals, and ni- nal failure may occur as well. Free hemoglobin
tric oxide." The end result may include wheez- impairs renal function, but compromised renal
ing, flushing, chest pain or tightness, and gastro- cortical blood supply is thought to be the major
intestinal symptoms. These symptoms may also contributing factor in renal failure. In addition,
be caused by release of bradykinin and norepi- antigen-antibody complex deposition, vasocon-
nephrine resulting from antigen-antibody com- striction, and thrombi formation contribute to
plex stimulation. the development of renal vascular compromise.
If complement activation does not proceed to
completion, which is usually the case with non"
Frequency
ABO antibodies, the red cells can undergo extra-
vascular hemolysis, where cells coated with The frequency of AHTRs is not easy to deter-
C3b and/or IgG are rapidly removed from the mine. The authors of a review article based on
circulation by phagocytes.' In extravascular he- data from several surveillance systems estimated
molysis, the consequences of complement acti- the risk of clinical or laboratory evidence of ABO
HTRto be 1 in 76,000 to 80,000 and the risk of lytic transfusion reaction. A single case report on
a fatal ABO HTR to be 1:1.8 mtllion." Of the the use of eculizumab, a monoclonal antibody
transfusion-related fatalitiesreported to the FDA that blocks the cleavage of complement compo-
from 2012 to 2016, 8% and 10% were caused nent C5, suggests that this may be a useful strat-
by ABOand non-ABOHTRs,respectively.1 egy for preventing hemolysis of incompatible
red cells."
Treatment Prompt initiation of therapy to aggressively
Prompt recognition of an AHTRand immediate manage hypotension, renal blood flow, and DIC
cessation of the transfusion are crucial. The unit provides the greatest chance of a successfulout-
of blood should be returned to the blood bank come. Furthermore, consultation with appropri-
for investigation. Saline should be infused to ate medical specialists early in the course of
maintain venous access, treat hypotension, and treatment will ensure that the patient receives
maintain renal blood flow, with a goal of a urine hemodialysis, cardiac monitoring, and mechani-
flow rate of> 1 mLlkglhour. Consultation with cal ventilation when needed.
transfusion medicine, critical care, renal, and
hematology experts should be considered. Prevention
The addition of the diuretic furosemide pro-
motes increased urine output and further en- Clerical and human errors involving patient,
hances renal cortical blood flow. If urine output sample, and blood unit identification are the
remains diminished after a liter of saline has most common causes of mistransfusion and,
been infused, acute tubular necrosis may have therefore, AHTRs. Reported estimates place the
occurred, and the patient may be at risk of de- risk of a near-miss at 1:1000, wrong blood given
veloping pulmonary edema. Oliguric renal fail- at 1:15,000-19,000, ABO-incompatibletransfu-
ure may be complicated by hyperkalemia and sion at 1:40,000, and error that results in harm
subsequent cardiac arrest. Metabolic acidosis at 1:4500.8•10 Institutional policies and proce-
and uremia often necessitate the institution of dures must be in place to minimize the likeli-
dialysis. hood of such errors, and corrective and preven-
DIC is an equally serious component of an tive action programs should target continual
AHTR. DIC is difficult to treat and may be the reduction of such errors. However, no one
first indication that hemolysishas occurred in an method for reducing the number of errors is
anuric or anesthetized patient. Traditional thera- foolproof.11 Products available to increase pa-
py for DIC includes treating or removing the un- tient safety include technology-based solutions,
derlying cause and providing supportive care via such as radiofrequency identification chips,
the administration of platelets, plasma, and cryo- handheld bar-code scanners, and "smart" refrig-
precipitate.
erators similar to systems used for pharmacolog-
Unconscious or anesthetized patients may re-
ic agents.
ceive multiple units of incompatible blood be-
The prevention of potential hemolysis from
fore acute hemolysis is recognized. Because the
the administration of mtnor-Abo-incompatible
severity of an AHTRis related to the amount of
incompatible red cells transfused, exchange platelets remains a challenge with constrained
transfusion may be considered. Some severe re- platelet inventories. A number of options, in-
actions to a single unit of strongly incompatible cluding anti-A or anti-B titration of the compo-
blood may require exchange transfusion as well. nent, limiting the total amount of incompatible
Antigen-negativeblood must be used for the red plasma transfused from platelets, and volume re-
cell exchange. Likewise, plasma and platelets duction may offer some benefit." The use of
that will not contribute to hemolysis should be platelet additive solutions for reducing minor-
chosen. incompatible hemolysis risk has not been clini-
Finally, inhibiting the complement cascade cally studied, although it decreases the titers of
may be beneficial, especially early in the hemo- anti-Aand anti-Bin the components.
C HAP T E R 22 Noninfectious Complications of Blood Transfusion 637
Nonimmune-Mediated Hemolysis Treatment
Transfusion-associated hemolysis can also result Hemolysis of nonimmune etiology may cause
from several nonimmune causes. Longer dura- symptoms whose severity. depends on the de-
tion of storage before issue is associated with in- gree of hemolysis and amount of component
creased hemolysis of storage-damaged red transfused. In all cases, the transfusion should
cells." Furthermore, improper shipping or stor- be discontinued and appropriate care should be
age temperatures, as well as incomplete deglyc- administered. (See the earlier section on the
erolization of frozen red cells, can lead to hemol- treatment of AHTRsfor details on managing hy-
ysis. At the time of transfusion, using a needle potension and declining renal function.)
with an inappropriately small bore size or em-
ploying a rapid-pressure infuser can cause me- Prevention
chanical hemolysis, which may be related to the Written procedures for all aspects of the manu-
use of roller pumps. Improper use of blood facture and transfusion of blood and compo-
warmers or the use of microwave ovens or hot nents should always be followed. Prompt recog-
waterbaths can cause temperature-related he- nition of nonimmune hemolysis and robust root
molysis. AABB Standards for Blood Banks and cause analysis may prevent additional occur-
Transfusion Services (Standards) allows for rences.
0.9% sodium chloride to be added to blood
during infusion.14(p50) Other fluids may be trans- Transfusion-Related Sepsis
fused with blood if "they have been approved Presentation
for this use by the FDA" or "there is documenta-
tion available to show that the addition is safe Fever (particularly a temperature of ;?:38.SC or
and does not adversely affect the blood or blood 101 F), chills, rigors, and hypotension during or
component." Infusion of red cells simultaneous- shortly after transfusion are the most frequently
presenting symptoms of transfusion-related sep-
ly through the same tubing with hypotonic solu-
sis. Gram-negative bacteria typically cause more
tions or with certain pharmacologic agents may
severe symptoms, including shock, renal failure,
cause osmotic hemolysis. For safe administra- and DIC. Gram-positive organisms may present
tion, red cells and these solutions or agents with isolated fever in the patient and may pres-
should be given via alternate venous access loca- ent hours after the completion of transfusion.
tions. In rare cases, hemolysis may be caused by This reaction is observed most commonly with
bacterial contamination of the Red Blood Cell transfusion of platelets stored at room tempera-
(RBC)unit. Patients may also experience hemol- ture.
ysis as part of their underlying disease process.
Although a negative DAT result usually indi- Differential Diagnosis
cates no evidence of an immune-mediated cause
The abruptness of onset and severity of the signs
of hemolysis, complete destruction of incompat-
and symptoms associatedwith transfusion-related
ible transfused red cells may be associated with
sepsis may be very similar to those of AHTRs.
a negative DATresult. Mild cases may be confused with FNHTRs. Fe-
When both immune and nonlmmune causes ver or bacteremia unrelated to transfusion may
of hemolysis have been excluded, the possibili- confound the diagnosis. The key to diagnosing
ty of an intrinsic red cell membrane defect in transfusion-related sepsis is culturing the same
the recipient or even in the transfused cells organism from both the patient and the remain-
should be considered. Cells with these defects, der of the component. The returned component
such as G6PD deficiency," are present in the should be visually examined in suspected cases
blood supply; they have increased fragilitywhen of posttransfusion sepsis. The bag should be
challenged with particular stressors and may un- sampled aseptically from the residual compo-
dergo coincidental hemolysis. nent rather than the tubing to avoid retrograde
contamination and false-positivecultures. Atten- laboratory results that help distinguish them
tion should be paid to any color changes, espe- from an FNHTRonce an investigation is begun.
cially brown or purple discoloration in an RBC Hemolysisand sepsis must be ruled out in a pa-
component and bubbles/frothiness in a platelet tient who experiences fever associated with
component. A Gram stain should be performed transfusion. Fever may commonly occur as a
on the returned component. manifestation of a patient's underlying illness. In
a patient who has been experiencing spiking fe-
Treatment vers during the course of admission, it may be
If transfusion-related sepsis is suspected, the difficult to rule out an FNHTR. Bacterial con-
transfusion should be stopped immediately, and tamination of the component must always be
supportive care and antibiotics should be initiat- considered to be a potential cause of a febrile re-
ed. action.
Prevention Pathophysiology
Suspected septic transfusion reactions should be FNHTRsmay be the result of accumulated cyto-
immediately reported to the blood collector so kines in a cellular blood component. This mech-
that co-components from the same donation can anism may be particularly relevant in reactions
be intercepted to avoid exposing other patients that occur after the transfusion of platelets."
to contaminated blood components." Any co- Whether passively transfused or generated by
components or aliquots from the same donation leukocytes in the recipient, pyrogenic cytokine
that are in the hospital's inventory should be im- release is the common event leading to symp-
mediately quarantined, pending investigationre- toms of FNHTR. Recipient leukocyte antibodies
sults. Bacteria detection tests and pathogen inac- may also cause febrile transfusion reactions. 18
tivation technologiesare discussedin Chapter 7. HLAantibodies in particular can react with cog-
nate antigens on transfused lymphocytes, granu-
Febrile Nonhemolytic Transfusion locytes, or platelets.
Reactions
Presentation Treatment
An FNHTRis usually defined as the occurrence When an FNHTR is suspected, the transfusion
of a ;:::1 C rise in temperature ;:::38C that is asso- should be discontinued and a transfusion reac-
ciated with transfusion and for which no other tion workup initiated. Antipyretics (eg, acet-
cause is identifiable. Accompanying symptoms aminophen) may be administered. For more se-
may include rigors, chills, and respiratory chang- vere reactions that include rigors, meperidine
es. In some instances, the patient may be afe- may be administered, although its efficacy has
brile but have the remaining constellation of not been rigorouslystudied.
symptoms. Symptoms usually occur during When fever develops during transfusion, the
transfusion but may occur up to 4 hours after- remainder of the implicated component should
ward. Although FNHTRs are self-limited, they not be transfused. Among the few situations in
may cause significantdiscomfort. which transfusion of the remainder of the com-
ponent should be considered is when the com-
Differential Diagnosis ponent is a medically indicated rare unit. If a
Recognition of an FNHTRrequires diagnosis by portion of the component remains, the laborato-
exclusion. The symptoms associated with an ry workup to exclude hemolysis must be com-
FNHTRmay be present in several other types of pleted and a discussion with the patient's clini·
transfusion reactions, the most serious of which cal care team regarding the likelihood of
are HTRs,sepsis, and TRALI.Each of these oth- transfusion-transmitted sepsis must be held be-
er reactions has signs, symptoms, and associated fore the transfusion is resumed.
Prevention anaphylactic shock are vasovagal and hypoten-

sive reactions. Urticaria, angioedema, pruritus,
Prestorage leukocyte reduction decreases the
and respiratory symptoms, such as wheezing or
frequency. of FNHTRs.19,20Preil'ledication with
stridor, are symptoms of anaphylaxis that do not
acetaminophen reduces the overall incidence of
occur in vasovagal or hypotensive reactions. The
febrilereactions, without impairing the ability to respiratory symptoms of anaphylaxis may be
detect other complications of transfusion such suggestive of asthma exacerbations or TRAlI.
as acute hemolysis, sepsis, and TRALI.21Plasma
However, the classic symptoms of allergy, in"
reduction reduces the rate of FNHTRsto plate" cluding urticaria, angioedema, and pr.uritus do
lets,zz .'
not occur in asthma or TRAlI. Fever, a prorni-
nent symptom of HTR and bacterial contamina-
Allergic Reactions tion, is not a feature of anaphylaxis. Patients
Presentation who take angiotensin-converting enzyme (ACE)
inhibitors and undergo plasma exchange some"
Most allergic transfusion reactions (ATRs) are times develop hypotensive reactions that mimic
mild, but their spectrum can range from a sim- anaphylaxis.
pie allergic reaction (urticaria or hives) to life"
threatening anaphylaxis. Symptoms generally Pathophysiology
occur within minutes after the start of the trans"
fusion. If symptoms do not appear until >4 The pathophysiology of nearly all ATRs is not
hours later, they may represent an allergic reac- understood. Extrapolating from a small number
tion that is unrelated to the blood transfusion. of case reports, ATRsare attributed to hypersen-
Hives are usually associated with itching sitivity reactions to allergens in the component
(pruritus) but may also burn or sting. Hives can caused by preformed IgE antibody in the recipi-
appear anywhere on the body and can coalesce ent. A combination of recipient and donor fac-
over large areas. They can last from hours to tors is likelyinvolved, as recipients with an atop"
several days before fading, but most respond ic predisposition appear to have a higher rate of
quickly to treatment with antihistamines. More ATRs, and certain donors' platelets are .associat-
extensive cases may be accompanied by: an" ed with an increased risk.23,24Clinical transfu-
gioedema, which is a deep tissue swelling, often sion parameters (eg.Infuston rate or volume,
around the eyes and lips. Angioedema can in" ABO mismatc~, ,component age, premedication)
volve the throat, tongue, or lungs, causing respi- have not been associatedwith ATRs.25
ratory distress, although feeling throat fullness Selective protein. defiCiency, classically 19A
or dyspnea without respiratory insufficiency is deficiency, is a rare cause of ATRs. These reac-
more common. tions are caused by antHgA in the recipient."
Anaphylactic transfusion reactions can be Although 19Adeficiency is present in ~ppr9xi"
generally defined as the presentation of mucocu- mately I :700 people of European ancestry, only
taneous signs of urticaria and angioedema in a small percentage of these people ever make
combination with other organ system involve" antibodies against 19A.People with absolute 19A
ment (cardiovascular, respiratory, gastrointesti- deficiency «0.05 mg/dl) may form class-speciflc
nal)." Manifestationsof anaphylaxisare hypoten- antibodies that are associated with anaphylactic
sion, loss of consciousness, dyspnea, wheezing, reactions. Those with decreased but detectable
stridor, abdominal pain, and vomiting. amounts of 19A,or relative 19Adeflciency, can
form subclass-spedfic antibodies (eg, anti-Igal
or anti"IgA2)that typically result in less severe
Differential Diagnosis
reactions."
Initially it may be difficultto distinguish anaphy- Although precautions should be taken when
laxis from other reactions characterized by hy- transfusing an 19A"deficientpatient, it must be
potension, dyspnea, and/or loss -of .conscious" kept in mind that the majority of ATRs' are
ness. Reactions that may be mistaken for caused by substances other than 19A.28Most of
these allergens are not identified but rarely may transfusion is needed; however, most cases of
include haptoglobin" or the complement pro- anaphylaxis are idiosyncratic to a specific unit
tein C4.30Transfusion can passivelysensitize pa- and not the result of selectiveprotein deficiency.
tients to donor antibodies, resulting in transient Tolerance of prior plasma and platelet transfu-
allergy (eg, to peanuts)." sions is a reasonable indicator that subsequent
plasma transfusions may be acceptable from un-
Frequency selected donors. Pooled solvent! detergent-
treated plasma has a lower risk of allergic reac-
ATRsare common, with an overall frequency of tions, but its value for prevention of recurrent
approximately 2% with traditional platelet and anaphylactic reactions has not been established.
plasma components." The incidence is about At some institutions, the plasma from 19A-
l Ofold lower with RBCs.33Platelet additive deficient donors is preferred for transfusions to
solution platelets have lower rates of ATRs.34 patients with a history of severe ATRs and
Life-threatening anaphylactic reactions account diagnosed 19Adeficiency «0.05 mg/dL). If 19A-
for <1% of all ATRs. Of the transfusion- deficient plasma is not available, desensitization
associated fatalities reported to the FDA from with 19A-containingplasma may be possible."
2012 to 2016,6% (11 of 176) were caused by Cellular components (RBCs and platelets) can
anaphylaxis, and most were not caused by 19A be depleted of plasma proteins through wash-
deficiency.1,28 ing. 19Adeficiencywithout the presence of anti-
19A or without a history of an ATR does not
Treatment clearlywarrant the use of Iga-deflcientor plasma-
Urticaria is the only transfusion reaction in depleted components.
which the administration of the component may
be routinely resumed after prompt treatment. Transfusion-Related Acute Lung Injury
When a patient develops symptoms, the transfu- Presentation
sion should be paused so that an antihistamine
may be administered. Once the symptoms have TRALIis clinically similar to acute respiratory
disstpated, the transfusion may be resumed, and distress syndrome (ARDS)but usually resolves
a laboratoryworkup need not be initiated. within 96 hours. Clinicalsigns and symptoms of
Severe urticarial reactions may be treated TRALIinclude fever, chills, dyspnea, cyanosis,
with HI and H2 receptor antagonists and corti- hypotension, hypoxia, and new-onset or wors-
costeroids. Epinephrine is the consensus first- ening bilateral pulmonary edema." A dramatic
line treatment for anaphylaxis; the dose may be transient neutropenia or leukopenia may also be
repeated up to every 5 minutes.35 observed." Symptoms arise within 6 hours of
transfusion, with most cases becoming evident
Prevention within 1 to 2 hours.
All plasma-containing components, includ-
Evidence does not support routine premedica- ing whole blood, RBCs, platelets, cryoprecipi-
tion to prevent ATRs.36Antihistamines (diphen- tate, and Fresh Frozen Plasma (FFP),have been
hydramine or a nonsedating antihistamine, eg, implicated in TRALI' Transfusion volumes as
cetirizine) are useful to treat allergic symptoms small as 15 mL have led to TRALI'
once they arise. If HI receptor antihistamines TRALIis a form of ARDS.The Berlin defini-
are not sufficient,an H2 receptor antagonist (eg, tion" delineated various categories of ARDS,all
ranitidine) or corticosteroids may be considered. of which were characterized by bilateral pulmo-
For patients whose reactions are severe or recur- nary edema on frontal chest radiograph but dif-
rent, washed RBCsor platelets, platelet-additive- fered based on the degree of hypoxemia: mild
solution platelets, or pooled solvent!detergent- defined as acute hypoxemia with a PaOZ/Fi02
treated plasma may be considered. (partial pressure of oxygen in arterial blood/
Redpient 19Aor haptoglobin concentrations inspired oxygen concentration) ratio between
are often not available by the time a subsequent 200 mm Hg and :;::;300mm Hg, moderate as be-
tween 100 mm Hg and ::;;200 mm Hg, and se- classified TRALIinto Type I, with no temporal
vere as ::;;100 mm Hg, In recognition of the diffi- relationship with an alternative risk factor for
culty in determining whether an underlying ARDS,and Type II, with.stable respiratory status
ARDSrisk factor or the transfusion is the cause in the 12 hours before transfusion, but with risk
of the acute lung injury, a panel of experts re- factors for ARDS or with. mild ARDS,40but
cently proposed revision of the consensus defini- whose respiratory status deteriorates and is
tion for TRALI.41The revised diagnostic criteria judged to be due to transfusion.
for TRAU are as follows (see Table 22-2): 1) Although the lung injury in ARDSis often ir-
acute onset of symptoms within 6 hours of reversible, the lung injury in TRALIis most of-
transfusion, 2) ARDS with hypoxemia and ten transient. Approximately 80% of patients
Pa02/Fi02 ::;;300 mm Hg or Sp02 (blood oxygen with TRALIimprove within 48 to 96 hours. The
saturation) <90% on room air, 3) clear evidence remaining 20% of patients who do not improve
of bilateral pulmonary edema on imaging, and rapidly have either a protracted clinical course
4) no evidence of left atrial hypertension (LAH) or a fatal outcome. In one TRALIstudy, 100% of
or, if present, deemed to not be the main con- the patients required oxygen support and 72%
tributor to hypoxemia. The panel further sub- required mechanical ventilation."
TABLE 22·2. Consensus TRALI Definition*
Patients who have no risk factors for ARDS and meet the following criteria:
1. Symptoms:
a. Acute onset
b. Hypoxemia (P/F ::;;300t or Sp02 <90% on room air)
c. Clear evidence of bilateral pulmonary edema on imaging (eg, chest radiograph, chest CT,or ultrasound)
d. No evidence of LAH* or, if LAH is present, it is judged to not bethe main contributor to the hypoxemia
2. Onset during or within 6 hours of transtuslont
3. No temporal relationship to an alternative risk factor for ARDS
Patients who have risk factors for ARDS (but who have not been diagnosed with ARDS) or who have existing
mild ARDS (P/F of 200-300), but whose respiratory status deteriorates" and is judged to be due to transfu-
sion based on:
1. Findings as described in categories 1 and 2 of TRALI Type I, and
2. Stable respiratory status in the 12 hours before transfusion
*Modified with permission from Vlaaret al.
"lf altitude is higher than 1000 m, the correction factor should be calculatedasfollows: [(P/F) x (barometric pressure/760)].
*Use objective evaluation when LAH is suspected (imaging, eg, echocardiography; or invasive measurement using, eg,
pulmonary artery catheter).
§ Onset of pulmonary symptoms (eg, hypoxemia-lower P/F ratio or Sp02) should be within 6 hours of end of transfusion.
The additional findings neededto diagnoseTRALI (pulmonary edemaon a lung imaging study and determination of lack of
substantial LAH)would ideally beavailableat the sametime but could be documentedup to 24 hours after TRALI onset.
°Use P/F ratio deterioration along with other respiratory parametersand clinical judgment to determine progression from
mild to moderateor severeARDS.Seeconversion table (Vlaar et a141)to convert nasal02 supplementationto Fi02.
TRALI = transfusion-related acute lung injury; ARDS = acute respiratory distress syndrome; P/F = PaO/Fi02 (partial
pressure of oxygen in arterial blood/inspired oxygen concentration); CT = computerized tomography; LAH = left atrial
hypertension.
Differential Diagnosis lungs, leading to leakage of protein-rich fluid

into the alveolar space.
The main complications that need to be distin- A two-event model of the mechanism of
guished from TRALIare 1) anaphylactic transfu- TRALI has been hypothesized." In the first
sion reactions, 2) TACO, 3) TRALI/TACO, 4) event, generation of biologically active com-
ARDS from underlying illness, 5) transfusion- pounds activates pulmonary vascular endotheli-
associated dyspnea (TAD), and 6) transfusion- al cells and primes neutrophils, resulting in se-
related sepsis. In anaphylactic transfusion reac- questration of neutrophils in the pulmonary
tions, bronchospasm, laryngeal edema, severe microvasculature. This first event predisposes
hypotension, erythema (often confluent), and the recipient to develop TRALIand can result
urticaria are prominent symptoms. Fever and from a variety of physiologicstressors, including
pulmonary edema are not associated with ana- sepsis, surgery, and massive transfusion. The in-
phylactic reactions. The clinical presentation of fusion of BRMs or antibodies is the second
TACO is very similar to that of TRALI,with re- event. BRMs consist of a mixture of lysophos-
spiratory distress, tachypnea, and cyanosis as phatidylcholines that accumulate in some cellu-
the most prominent features. Key distinctions lar components during storage. The transfused
between TACO and TRALIare that the pulmo- antibodies may be to HLAClass I, HLAClass II,
nary edema in TACO is cardiogenic and respon- or human neutrophil antigens (HNAs). These
sive to diuretics, whereas it is noncardiogenic stimuli are hypothesized to activate the primed
and not responsive to diuretics in TRALI.The neutrophils in the pulmonary microvasculature,
classificationof TRALI/TACO is reserved for sit- resulting in pulmonary endothelial damage, cap-
uations compatible with TRALIand TACO and/ illaryleakage, and pulmonary edema.
or with lack of data to establish whether or not
significant LAB is present. Cases that meet the Frequency
criteria for TRALIbut without stable pulmonary
TRALIoccurs in about 1 in 10,000 units trans-
status in the previous 12 hours should be con-
fused." TRALI was the leading cause of
sidered ARDS,and if there is no documentation transfusion-related mortality reported to the
that Pa02/Fi02 ::::300mm Hg, then cases should FDAuntil 2016, when TACOsupplanted It.' Be-
be classified as TAD. High fever with hypoten- cause HLA or HNA antibodies are more com-
sion and vascular collapse are prominent fea- mon in multiparous women, efforts to reduce
tures of transfusion-related sepsis. Respiratory the risk of transfusing plasma from potentially al-
distress and ARDS are infrequently associated loimmunized female donors have been associat-
with septic transfusion reactions. Finally, other ed with decreased TRALIfatalities." In 2006,
possible causes of ARDS should be considered, the year before many blood centers implement-
such as coincident sepsis, myocardial infarction, ed measures to reduce the risk of TRALIfrom
and pulmonary embolus. plasma transfusions (eg, donor selection criteria
and HLA-antibodyscreening of selected donors),
Pathophysiology 35 fatal cases of TRALI,22 of which were asso-
Several mechanisms for the pulmonary manifes- ciated with transfusion of FFP,were reported to
tations of TRALIhave been proposed. The pri- the FDA.Since 2008, the year after many blood
mary effector cell is the neutrophil, with lung centers implemented such measures, the num-
histology from fatal cases showing predominant- bers of TRALIfatalities have decreased by over
half.'
ly neutrophilic infiltrates and alveolar edema."
TRALIhas been associated with the infusion of
Treatment
antibodies to leukocyte antigens and of biologic
response modifiers (BRMs).44Infusions of these Treatment of TRALIconsists of respiratory and
antibodies or BRMsare thought to initiate cellu- circulatory support. Oxygen supplementation
lar activation and neutrophil-mediated damage with or without mechanical ventilation is re-
of the endothelial basement membrane in the quired in almost all cases. Pressor agents may be
C HAP T E R 2 2 NoninfectiousComplicationsof Blood Transfusion 643
needed to support blood pressure. Because cardiogram, elevated serum troponin, .and ele-
TRALIis not associated with volume overload, vated brain natriuretic peptide (BNP).48 Radio-
diuretics are typicallynot indicated and may in- graphs may show a widened cardiothoracic ratio
crease the risk of hypotension. Administration of in addition to pulmonary edema. Hemovigilance
corticosteroids has not been shown to improve systems have reported cases of TACO beyond 6
clinical outcome in TRALIor ARDS.47 hours of transfuston." Additionally, it was real-
ized that some reactions accepted clinically as
Prevention TACO did not meet the 2011 surveillance defi-
nition from the International Society of Blood
There is no method to predict which patients
Transfusion (I~.ElT)/InternationalHemovigilance
will develop TRALI.Donors whose collections
Network (IHN)/MBB. An international revi-
are linked to cases of TRALIare permanently de-
sion group composed of clinical, laboratory,
ferred. Although approximately 10% of blood
blood bank, and hemovigilance experts ..pub-
donations contain HLAand/or HNA antibodies,
lished a revised and validated TACO definition
TRALIis much rarer. Nevertheless, an important
in 2019.50 The revised surveillance definition
mitigation strategy is to collect plasma compo-
(see Table 22·3) classifiescases as TACO during
nents, whole blood, and platelets from male do-
or up to 12 hours after transfusion when they
nors, never-pregnant female donors, or females
meet a total of three or more of the following
who have been tested since their last pregnancy
criteria, with at least one from the required cri-
and found to be negative for HLA antibodies.
teria (l and 2): 1) acuteor worsening respirato-
Although these measures reduce the risk of
ry compromise, 2) evidence of pulmonary ede-
TRALI,it is important to recognize that they do
ma, 3) evidence of cardiovascular system
not eliminate TRALI completely because they
changes not explained by the patient's underly-
do not address the risk of TRALIfrom RBC or
ing medical condition, 4) evidence of fluid over-
cryoprecipitate components, and donor testing
of previously pregnant females does not screen load, and 5) supportive result of a relevant bio-
marker leg, an increase of BNP or N-terminal-
for HNAantibodies and BRMs.
propeptide BNP (NT-proBNp)].51
Tran.sfusion-Assodated Circulatory
Overload
TACO is frequently confused with TRALI be-
Presentation
cause both types of reactions produce pulmo-
It is well known that transfusion can precipitate nary edema. It is also possible for TACO and
acute pulmonary edema caused by volume over- TRALIto occur concurrently in the same pa-
load. Patients >70 years and infants are at great- tient. The timelines and the clinicalpresentation
est risk, as well as patients with compromised are similar, but hypertension is a more typical
ability.to regulate fluid balance (eg, those with feature of TACO, whereas it is only an infre-
congestive heart failure and end-stage renal dis- quent and transient manifestation of IRALI.
ease), although all transfusion recipients are sus- Furthermore, rapid improvement with diuresis
ceptible to some degree. Wher.eas.1Wgevolumes is consistent with TACO. . ...
of components and nonblood fluids We most.fre- In congestive heart failure, BNP levels are el-
quently implicated, modest volumes can also evated. A posttransfusion-to-pretransfusionBNP
precipitate TACO in susceptiblepatients. A high ratio of 1.5 with a posttransfusion leyel ofat
flow rate is frequently a cofactor. least 100 pg/mL as a cutoff yields a sensitivity
TACO has no pathognomonic signs or symp- and specificity >80% in TACO.48However, in
toms. Within 1 to 2 hours of transfusion, pa- the intensive care setting, BNPis only of moder-
tients may develop any or all of the following: ate value for distinguishingTACOfrom TRALI.52
S3 gallop, jugular venous distension, elevated Some clinical laboratories measure the NT-
central venous pressure, dyspnea,· orthopnea, proBNP, which has a longer half-lifethan BNP.
new Sf-segment and T-wavechanges on electro- NT-proBNP·is also predictive of TACO.53With
TABLE 22-3. TACO Reporting Criterla'"
Patientsclassified with TACO(surveillance diagnosis) should haveacute or worsening respiratory compro-

mise and/or evidenceof pulmonary edema(A and/or B below) during or up to 12 hours after transfusion and
presenceof a total of 3 or more of the criteria below:
A. Acute or worsening respiratory compromise.
B. Evidenceof acute or worsening pulmonary edema basedon:
@ Clinical physicalexamination,and/or
.. Radiographicchest imaging and/or other noninvasiveassessmentof cardiacfunction (eg, echccarclo-

gram).
C. Evidencefor cardiovascular system changes not explained by the patient's underlying medical condition,
including development of tachycardia, hypertension, widened pulse pressure, jugular venous distension,
enlarged cardiac silhouette, and/or peripheral oedema.
D. Evidenceof fluid overload, including any of the following: a positive fluid balance;responseto diuretic
therapy, eg, from diuretic therapy or dialysis combined with clinical improvement; and change in the
patient's weight in the peritransfusion period.
E. Supportive result of a relevant biomarker, eg, an increaseof B-type natriuretic peptide level (eg, BNPor
NT-proBNP)abovethe age-group-specific reference rangeand greaterthan 1.5 times the pretranstuslon
value. A normal posttransfusion NP level is not consistent with a diagnosis of TACO;serial testing of NP
levels in the peritransfusion period may be helpful in identifying TACO.
'Modified from ISBTet al.51
tThese criteria establish a surveillance case definition based on a complete description of an event, including information
that becomesavailablewell after onset. This is for reporting and tracking purposesand the criteria do not constitute clinical
diagnosis for the purpose of real-time clinical interventions. If a case could be TACOaccording to clinical judgment but
fewer than three criteria are met basedon availableinformation, the listed criteria can guide collection of additional details,
eg, from case notes or discussion with clinical staff.
TACO= transfusion-associated circulatory overload; NT-proBNP= N-terminal-propeptlde BNP.
rapid onset of respiratory distress, possible caus- ber (32%) of transfusion-associated fatalities re-
es of ARDS, such as coincident myocardial in- ported to the FDA (59 patients) were a conse-
farction, pulmonary embolus, and others, quence of TACO, with TACO-related fatalities
should be considered in addition to TACO and surpassing those from TRALIafter 2015.1 Plate-
TRALI. let and plasma components are associated with
a TACOincidence of approximately 1%/5.56 and
Frequency RBCs,with an incidence of up to 2.7%.54
TACO is an underreported adverse reaction to
Treatment
transfusion, and hemovigilance and retrospec-
tive studies underestimate the incidence. Fur- As soon as symptoms suggest TACO, the trans-
thermore, the incidence differs across patient fusion should be stopped. The symptoms should
populations with different comorbldlties." In be treated by placing the patient in a seated posi-
aggregate, from 2013 to 2017, the highest num- tion (if possible), providing supplementary oxy-
gen, and reducing the intravascular volume (eg, flushing, angioedema, urticaria). Septic-shock
with diuretics. If symptoms persist in confirmed from transfusion is often accompanied by fever.
TACO, administration of additional diuretics or AHTRs .are .associated with hemoglobinuria,
therapeutic phlebotomy is appropriate. pain, and fever. A minority of TRALIcases in-
volve clmtcally significant hypotension, but
Prevention TRALIcases.lJ.aveacute. pulmonary insufficien-
In the absence of ongoing and rapid blood loss, cy, a finding that is atypicalfor HyTRs.
components should be administered slowly, par-
ticularly in patients at risk of TACO (ie, pediatric Pathophysiology
and elderly patients, patients with severe ane- Bradykinin is thought to have a causal role in
mia, and patients with congestive heart failure). HyTRs.59 Bradykininis a vasoactive peptide gen-
Rates of 2 to 4 mLiminute and 1 mLikg of body erated via activation of the kinin- kallikrein sys-
weight per hour are the most frequently cited, tem from its precursor, high-molecular-weight
despite a paucity of data on appropriate infusion kininogen. Factors that increase concentrations
rates. Total fluid input and output must be mon- of bradykinin in blood components include stor-
itored. age, filtration, and ACE activity in donors and
recipients. ACE activity can be affected by ACE-
Hypotensive Reactions inhibitor medication and cardiopulmonary by-
Presentation pass circuits, because the lungs are a primary
site of ACE actlvity/" Recent prostatectomy may
Hypotensive transfusion reactions (HyTRs)are increase bradykinin concentrations in patients
defined as the sudden and unexpected onset of as a result of kalllkreins released from the pros-
clinicallysignificanthypotension associatedwith tate.
the transfusion of blood or blood components
that resolves shortly after .the transfusion is Treatment
stopped. Systolic blood pressure (SBP)declines
by >30 mm Hg or to <80 mm Hg in adults. In The primary treatment intervention is to stop
children, an SBPdecrease of >25% is consistent the transfusion. Bloodpressure usually increases
with a HyTR.Another characteristicis that HyTRs within minutes after cessation of transfusion,
usually start within the first 15 minutes of trans- but circulatory support with intravenous fluids
fusion. All types of patients can be affected, with and vasopressors may be needed. With acute
any blood component." The HyTR rate per onset of hypotension, the etiology is often not
blood component has been reported as 0.019% immediately clear. Treatments for anaphylaxis
for platelets, 0.015% for RBCs,and 0.006% for and sepsis may be initiated as the clinical pre-
plasma." Another study found 1% of platelet sentation unfolds.
transfusions can be complicated by HyTRs.58
Prevention
If a patient who experienced a HyTR.is taking
Hypotension can be a primary manifestation ACE-inhibitor medication. that is.not discontin-
of HyTR, anaphylaxis, septic transfusion reac- ued, subsequent transfusions should be given as
tion, AHTR, TRALI,or an underlying disease/ slowly asis feasibleto prevent a recurrence. Be-
medication. All may present within the first 15 cause HyTRsare typicallyidiosyncratic to a spe-
minutes of transfusion, but the absence of con- cific unit, patients without risk factors for a
comitant signs and symptoms and the prompt HyTR typically tolerate subsequent transfusions
resolution of hypotension upon discontinuation well. Washing cellular blood components will
of transfusion distinguish HyTRsfrom other re- reduce accumulated bradykinin, but washed
actions. Anaphylactic shock usually is accompa- component protocols are seldom needed, as
nied by mucocutaneous allergic manifestations most cases are not recurrent.
Complications of Massive Transfusion venous calcium replacement with calcium glu-

conate or calcium chloride should be considered
The potential complications of massive transfu-
early, as hypocalcemia contributes to a hypoco-
sion, usually defined as the receipt of >10 RBC
agulablestate in massive transfusion.f
units within 24 hours, include metabolic and
hemostatic abnormalities, immune hemolysis,
Hyperkalemia and Hypokalemia
and air embolism. Metabolic abnormalities can
depress cardiac function. Hypothermia from re- Pathophysiology. When red cells are stored at
frigerated blood, citrate toxicity, and lactic aci- I to 6 C, the intracellular potassium gradually
dosis from underperfusion and tissue ischemia, leaks into the supernatant plasma or additive
which are often complicated by hyperkalemia, solution. Although the concentration in the su-
can contribute to this effect.Although metabolic pernatant may be high, because of the smallvol-
alkalosis caused by citrate metabolism may oc- ume, the total extracellular potassium load is
cur, it is not likelyto be clinicallysignificant.Pa- <0.5 mEq for fresh RBC units and only 5 to
tients who lose blood rapidly may have preexist- 7 mEq for units at expiration. These potassium
ing or coexisting hemostatic abnormalities or concentrations rarely cause problems in the re-
develop them during resuscitation. Hemo- cipient because rapid dilution, redistribution
static abnormalities may include dilutional co- into cells, and excretion blunt the effect." How-
agulopathy, DIC, and liver and platelet dysfunc- ever, irradiation followed by prolonged storage
tton." may increase supernatant potassium to unac-
ceptable levels. Furthermore, hyperkalemia can
Citrate Toxicity be a problem in patients with renal failure, in
Pathophysiology and Manifestations. When premature infants, and in newborns receiving
large volumes of citrated blood components are large transfusions, such as in cardiac surgery or
transfused rapidly,particularlyin the presence of exchange transfusion; otherwise, hyperkalemia
liver disease, plasma citrate levels may rise, is typically a transient effect during very rapid
binding calcium, resulting in hypocalcemia. In transfusions."
patients with a normally functioning liver, ci- Hypokalemia occurs more frequently than
trate is rapidly metabolized; thus, these symp- hyperkalemiaafter transfusionbecausepotassium-
toms are transtent." Hypocalcemiais more like- depleted donor red cells reaccumulate this ion
ly to cause manifestations in patients who are intracellularly, and citrate metabolism causes
hypothermic or in shock. further movement of potassium into the cells in
A decrease in ionized calcium levels increas- response to the consumption of protons. Cate-
es neuronal excitability,which in the conscious cholamine release and aldosterone-induced uri-
patient leads to symptoms of perioral and pe- nary loss can also trigger hypokalemiain the set-
ripheral tingling, shivering, and lightheaded- ting of massivetransfusion."
ness, followed by a diffuse sense of vibration, Treatment and Prevention. Usually, no
muscle cramps, fasciculations, spasm, and nau- treatment or preventive strategy for hypokale-
sea. In the central nervous system, hypocalcemia and hyperkalemia, aside from appropriate
mia is thought to increase the respiratory center monitoring, is necessary provided that the pa-
sensitivity to carbon dioxide, causing hyperven- tient is adequately resuscitated from the under-
tilation. Because myocardial contraction is de- lying condition that required massive transfu-
pendent on the intracellular movement of ion- sion. For infants receiving routine transfusions,
ized calcium, hypocalcemia depresses cardiac units infused up to 0.5 cc/kg/minute may be
function. used safelyuntil the expiration date." Although
Treatment and Prevention. Unless the pa- washing of RBCunits results in very low levels
tient has a predisposing condition that hinders of potassium, there is no evidence that this is in-
citrate metabolism, hypocalcemia caused by ci- dicated for routine RBCtransfusions, even in pa-
trate overload during massive transfusion can tients with impaired renal function," and it may
usually be treated by slowing the infusion. Intra- even cause increased hemolysis and higher po-
tassium levels in the component with time after gree of platelet and clotting abnormalities cor-
washing." relates with the length of time that the patient is
hypotensive, suggesting that shock is the most
Hemostatic Abnormalities in Massive important cause oLDIC. In aggregate, hypoper-
Transfusion fusion is a major risk factor for coagulopathy in
recipients of heavy transfusion."
Pathophysiology. Coagulopathy can occur in These data may not be generalizable to pa-
massive transfusion, particularly when. the lost tients undergoing massive transfusion in the
blood is initially replaced with RBCsand crystal- controlled setting of the operating room, where
loids. Coagulopathy in massive transfusion is fre- hypotension caused by volume loss. is prevent-
quently ascribed to the dilution of platelets and ed. In this context, coagulation factor levels may
clotting factors as patients lose hemostatically be more significant than platelet problems. Mur-
active blood, and enzymatic activity is reduced ray and colleagues documented that excessive
as the core body temperature lowers if a blood bleeding in elective surgery patients who re-
warmer is not used. Mortality rates associated ceived > 1 blood volume (RBCsand crystalloid)
with hemostatic abnormalities range from 20% corresponded to a prolongation in PT and PTT
to 50%.69The high rate of mortality results from compared to patients with normal hemostasis."
hypothermia, metabolic acidosis, and coagulopa- Treatment and Prevention. The dilutional
thy." model of coagulopathy in massive transfusion
Studies of military and civilian trauma pa- suggests thatprophylactic replacement of hemo-
tients, before wide acceptance of transfusions static components based on the volume of RBCs
with a more balanced ratio of RBCs to platelets or whole blood transfused prevents the develop-
to plasma, demonstrated a progressive increase ment of a bleeding diathesis. No specific regi-
in the incidence of microvascular bleeding men has yet been: shown to be superior to any
(MVB)characteristic of a coagulopathy with in- other in prospective studies. Although there was
creasing transfusion volumes that typically oc- no statistically significant difference in mortality,
curs after replacement of 2 to 3 blood volumes the Pragmatic Randomized Optimal Platelet and
(20 to 30 units]." Although platelet counts, co- Plasma Ratios (PROPPR)trial indicated improve-
agulation parameters, and levels of selected clot- ment in hemorrhage control using a 1: 1: 1 plasmal
ting parameters correlate with the volume trans- plateletlRBC unit ratio vs a 1:1:2 ratio." Fur-
fused, contrary to expectations from a simple thermore, in injured patients at risk for hemor-
dilutional model, the relationship is marked by rhagic shock, early administration of thawed
tremendous variability. Moreover, there is fre- plasma results in lower mortality."
quently discordance between the laboratory as- Although the optimal transfusion ratio in
sessment and the clinical evidence of bleeding. trauma resuscitation is still debated, institutions
MVB increases with platelet counts below should develop a massive transfusion protocol.
-50,000/1lL; however, no simple relationship Replacement of platelets and coagulation fac-
can be determined between a patient's coagula- tors in the surgical or trauma patient receiving
tion test results and the onset of bleeding. The massive transfusion should be based on the
etiology of bleeding (elective surgery vs massive identification of a specific abnormality using
trauma) may playa role as well.72 platelet counts, PT, activated PTT, and fibrino-
Subsequent studies have refined these obser- gen levels, as clinicallyindicated. Frequent mon-
vations. Significant platelet dysfunction has itoring of these laboratory values serves to avoid
been demonstrated in recipients of massive overuse of platelets and plasma components by
transfuslon.? Low fibrinogen and platelet anticipating the specific components needed
counts are better predictors of hemostatic failure while avoiding dilutional coagulopathy. It is im-
than elevations of prothrombin time (PT) and perative that the laboratory provide results of
partial thromboplastin time (PTT), suggesting these tests rapidly. Intraoperative and postopera-
that consumptive coagulopathy is an important tive laboratory testing, such as thromboelastog-
factor in MVB in addition to dilution." The de- raphy, may be useful.
Antifibrinolytics have a role in controlling than an AHTRand typicallydoes not precipitate

massive bleeding from trauma. The Clinical the acute signs and symptoms of an AHTR, al-
Randomization of an Antifibrinolytic in Signifi- though some patients may develop jaundice and
cant Hemorrhage 2 (CRASH-2)and other stud- leukocytosis. In a DHTR, the hemolysis is pri-
ies conclude that tranexamic acid should be giv- marily extravascular, so although hemoglobin-
en as early as possible in trauma patients." uria may occur in rare cases, acute renal failure
Activated clotting factors do not have a defined and DIC are not generally present. In some cas-
role in massive transfusion. es, the hemolysis occurs without causing clini-
cal symptoms. These patients present with un-
Air Embolism explained anemia or do not experience the
expected sustained increase in hemoglobin con-
Air embolism can occur if blood in an open sys- centrations followingtransfusion.
tem is infused under pressure or if air enters a
central catheter while containers or blood ad- Differential Diagnosis
ministration sets are being changed. Air embo-
lism has been reported in association with intra- Fever with hemolysis may also occur well after
operative and perioperative blood recovery transfusion when the component has been con-
systems that allow air into the blood infusion taminated with an intracellular red cell parasite,
bag. The minimum volume of air embolism that such as malaria or babesia. Fever without hemol-
is potentially fatal for an adult is approximately ysis may be an indication of graft-vs-hostdisease
100 mL.79 Symptoms include cough, dyspnea, (GVHD)[describedin the section on transfusion-
chest pain, and shock. If air embolism is suspect- associated (TA)·GVHD,below] or transfusion-
ed, the patient should be placed on the left side transmitted viral disease (eg, cytomegalovirus).
with the head down to displace the air bubble Hemolysis resulting from antibody production
from the pulmonic valve. Aspiration of the air is by donor passenger lymphocytes may occur af-
sometimes attempted." ter transplantation of a minor-ABO-incompatible
organ (eg, transplantation of a group 0 liver in a
group A patient).
Hypothermia
In a DHTR/DSTR, antibodies may be found
Blood warmers may be used to prevent hypo- in the serum, on transfused red cells, or both.
thermia. Proper procedures for the use of blood Diagnosisby routine antibody screening and an-
warmers should be followed because overheat- tibody identification should be possible. If trans-
ing may induce hemolysis and serious transfu- fused red cells are still present in the patient's
sion reactions, including fatalities. circulation, the DAT result may be positive.
When the DAT result is positive, an eluate
should be performed and the antibody identi-
N S~ON fied. If a segment from the unit is available, anti-
gen typing may confirm the diagnosis.
Delayed Hemolytic Transfusion Pathophysiology

Reactions After transfusion, transplantation, or pregnancy,
Presentation a patient may make an antibody to a red cell an-
tigen that he or she lacks. Red cell antibodies,
Development of an alloantibody followingtrans- missed during pretransfusion testing due to
fusion can result in an asymptomatic delayed se- waning of their peak levels with time (ie, eva-
rologic transfusion reaction (DSTR)or in a de- nescence), may cause a delayed transfusion re-
layed HTR(DHTR).Fever and anemia occurring action if the patient subsequently receives a unit
days to weeks after transfusion of an RBCcom- of blood expressing the corresponding red
ponent are characteristic of a DHTR.The hemol- cell antigen. Primary aIloimmunization occurs
ysis associated with a DHTR is more protracted anywhere from days to months after a transfu-
sion of antigen-positivered cells, depending on common as DHTRs.83These reactions are likely

the immunogenicity and dose of the antigen. to be greatly underrecognized because most pa-
Approximately 1% to 1.6% of RBC transfu- tients do not undergo red cell antibody screen"
sions are associated with antibody formation, ing followingtransfusion."
excluding .antibodies to antigens in the Rh sys-
tern. D"negativeblood is usually transfused to D" Treatment
negative patients, so the frequency attributable
to anti-D is relatively low. Newly formed alloan- The treatment of DHTRsconsists of monitoring
tibodies are routinely detected during pretrans- the patient and providing supportive care. Cor"
fusion screening. (See Chapters 13 and 17.) Re- rection .of the anemia by transfusing antigen"
cipients of recent transfusion or pregnant negative RBCs may be needed. In addition a
patients must have samples drawn for compati- case report has shown some benefit of ecull-
bility testing within 3 days of the scheduled zumab in the treatment of DHTRs.85
transfusion to ensure identification of any poten-
tial new alloantibodies. A 5"year retrospective Prevention
study of alloimmunization showed that 11 of DHTRs/DSTRs caused bylmown antibody spec"
2932 patients (0.4%) had developed new anti" ificities can be prevented by the transfusion of
bodies, including antl-E, anti-K, and antl-lk",
antigen-negative RBCs. It is essential to obtain
within 3 days after transfusion."
prior transfusion records for the recipient be"
DHTRsand DSTRsrarely, if ever, occur as a
cause of antibody evanescence. Many Institu-
result of primary immunization and are general"
ly associated with subsequent transfusions. Be" tions have programs to provide prophylactic,
cause of evanescence (described aoove)," as partial phenotype-matching of blood for patients
many as 30% to 60% of"alloantibodies become with sickle cell disease or other chronic anemias
undetectable over months to years. Antibodies to prevent alloimmunization. Patients with sick"
agalnst antigens of some blood group systems, le cell disease may develop a life"threatening
such as the JK system, frequently exhibit this be" complication known as hyperhemolysis, even af-
havior. Subsequent transfusion of an antigen" ter crossmatch-compatlble RBC transfusion, in
positive unit triggers an anamnestic response, which autologous and allogeneic cells are de"
with production of antibody occurring over the stroyed. (See Chapter 19.)
next several days to weeks after the transfusion.
The rapidity of antibody production and the he" Transfusion-Associated Graft-vs-Host
molytic potential of the antibody both combine Disease
to influence the clinical presentation. Blood
Presentation
group antibodies most commonly associated
with DHTRs/DSTRsinclude those targeting an" TAGVHD occurs in <1 per million transfusions.
tigens of the JK, FY,KEL,and MNS systems. The clinical manifestations of TA"GVHDtypical"
ly begin 8 to 10 days after transfusion, although
Frequency symptoms can occur as early as 3 days and as
As with AHTRs, the estimated rate of DHTRs late as 30 days. Signs and symptoms include
varies widely from study to studyrSomeof this rash, fever, enterocolitis with watery diarrhea,
variation is the result of the practice of consider" elevated liver function test results, and pancyto-
ing DSTRsand DHTRsas one category.Also,im- penia. The rash begins on the trunk and pro"
provements in laboratory techniques have con" gresses to the extremities. In severe ..cases,
tributed to an increased number of DSTRs bullae may develop." Unlike GVHDafter allege-
detected. Nonetheless, delayed reactions occur neic hematopoietic stem cell transplantation,
much more frequently than AHTRs, with esti- TAGVHD leads to profound marrow aplasia,
mates of approximately 1:2500 for either type of with a mortality rate higher than 90%. The time
delayed reaction, and DSTRs being twice as course of the reaction is very rapid; death
typically occurs within 1 to 3 weeks of the first foreign. The transfused lymphocytes are able to
symptoms. recognize the host cells as foreign and are able
to mount an immunologic attack on the host. As
Differential Diagnosis mentioned, the degree of genetic diversity in
Because the clinical manifestations of TA-GVHD a population affects the risk of developing TA-
appear several days after a transfusion, it may be GVHD.
difficultto associate the patient's symptoms with
the transfusion. The symptoms can easily be at- Treatment
tributed to other conditions, including drug re- Treatment of TA-GVHD has been attempted
actions and viral illness. In cases of TA-GVHD,a with a variety of immunosuppressive agents.
skin biopsy reveals a superficial perivascular Unfortunately, the disorder is almost uniformly
lymphocytic infiltrate, necrotic keratinocytes, fatal; oniy rare cases of successful treatment,
compact orthokeratosis, and bullae formation.
many involving stem cell transplantation, have
Molecular techniques, including HLA typing,
been reported. Therefore, emphasis is placed on
cytogenetics, and chimerism assessment, can be
used to make the diagnosis. prevention of the disorder.
Prevention
Pathophysiology
There are three requirements for GVHD to de- TA-GVHDcan be prevented by irradiation of
velop in a patient. First, there must be differenc- cellular blood components. MBB Standards re-
es in the HLA antigens expressed between the quires a minimum dose of 25 Gy (2500 cGy)de-
donor and the recipient. Second, immunocom- livered to the central portion of the container
petent cells must be present in the component. and a minimum of 15 Gy (1500 cGy) else-
Finally, the host must be incapable of rejecting where. 14(p27) The Standards requires blood banks
the immunocompetent cells. The three primary and transfusion services to apply methods
factors that determine the risk of developing known to prevent TA-GVHD,including irradia-
TA-GVHDare the degree of recipient cellular tion, to cellular blood components when 1) the
immunodeficiency, number of viable T lympho- patient is identified as being at risk of TA-GVHD,
cytes in the transfusion, and degree of a popula- 2) the donor is a blood relative of the recipient,
tion's genetic diversity. The number of viable and 3) the donor is selected for HLAcompatibili-
lymphocytes in a transfusion can be affected by ty by typing or crossmatching.14(PP44-4S) These
the age, leukocyte reduction status, and irradia- standards are minimum requirements for irradi-
tion status of the component." Although cur- ation of cellular blood components, and institu-
rent leukocyte-reduction technologies signifi- tions may choose to administer irradiated com-
cantly reduce the number of lymphocytes in a ponents to other categories of patients. (See
component, leukocyte reduction does not elimi-
Table 22-4.) Finally,pathogen inactivation tech-
nate the risk of TA-GVHD.
nologies are also effective against proliferatingT
Clinicalrisk factors for developing TA-GVHD
include leukemia, lymphoma, use of immuno- cells and offeran alternative to lrradlation."
suppressive drugs administered for transplanta-
tion or myeloablative chemotherapy, congenital Posttransfusion Purpura
immunodeficiency disorders, and neonatal age, Presentation
although immunodeficiency is not required for
TA-GVHD.88TA-GVHDcan occur after a trans- Posttransfusion purpura (PTP)is an uncommon
fusion from a donor who is homozygous for an complication of transfusion, and its true inci-
HLAhaplotype to a heterozygous recipient (one- dence is therefore difficultto estimate. Nonethe-
way haplotype match). In this circumstance, the less, >200 cases have been reported in the liter-
recipient's immune system does not reject the ature, and data from the Serious Hazards of
HLA-homozygous transfused lymphocytes as Transfusion (SHOT) program in the United
TABLE22·4. Well-Documented Indications for Irradiated Components

Intrauterine transfusions
Prematurity, low birthweight, or erythroblastosis fetalis in newborns
Congenital immunodeficiencies
Hematologic malignancies or solid tumors (neuroblastoma, sarcoma, Hodgkin disease)
Peripheral blood stem cell/marrow transplantation
Components that are crossmatched or HLA matched, or donations from family members (blood relatives)
Fludarabinetherapy
Granulocyte components
Kingdom suggest that the disorder may be more Pathophysiology

common than previouslybelieved." The pathogenesis of PTP is related to the pres-
Patients typically present with wet purpura
ence of platelet-specific alloantibodies in a pa-
and thrombocytopenia within 2 weeks after a tient who has previously been exposed to plate-
transfusion. The thrombocytopenia is often pro- let antigens via pregnancy or transfusion.:The
found, with platelet counts of <10,000/]1L.91 female-to-maleratio of affected patients is 5 to
Bleeding from mucous membranes and the gas- l. Antibodies against human platelet antigen 1a
trointestinal and urinary tract is common. Mor- (HPA-1a), located on glycoprotein lIla, are iden-
tality rates in large case series range up to 16%, tified in about 70% of PTP cases. Antibodies to
primarily due to intracranial hemorrhage." HPA-1b, other platelet antigens, and HLA anti-
PTP has most commonly been associated gens have also been implicated in PTP.93
with transfusions of RBCsor whole blood; how- The reason for the concomitant destruction
ever, the disorder has also been associated with of autologous platelets in this disorder is un-
transfusions of platelets or plasma. known. The theory that the platelet alloanti-
body has autoreactivity that develops when
Differential Diagnosis a .patiel1t is .re-exposed to a foreign pIM~let-
specificantigen currently has the most support.
The differential diagnosis of PTP includes alter-
native causes of thrombocytopenia, such as Treatment
autoimmune thrombocytopenic purpura, throm-
Because the duration of thrombocytopenia in
botic thrombocytopenic purpura, heparin-in-
untreated patients is about 2 weeks, it can be
duced thrombocytopenia, DIC, and drug-
difficult to assess the effectiveness of therapies
induced thrombocytopenia. Although the for PTP. Steroids, whole-blood exchange, and
diagnosis of PTP can be obvious in patients with plasma exchange have all been used to treat
previously normal platelet counts and no other PTP. The current treatment of choice for PTP is
Significant medical abnormalities, it can be a MG.94 Patients respond within 4 days, on aver-
challenge in patients with multlple medical age, and some respond within hours. HPA-l a-
problems. Platelet serology studies may aid in negative platelet transfusions have been useful
the diagnosis. in some cases."
Prevention and endocrine organs, tissue damage leading to

heart failure, liver failure, diabetes, and hypo-
The use of prestorage leukocyte reduction may
thyroidism may occur. Patients who chronically
decrease the incidence of PTP. In the 3 years be-
receive transfusion for diseases such as thalas-
fore the implementation of universal leukocyte
semia, sickle cell disease, and other chronic ane-
reduction in the United Kingdom, there were
mias are at greatest risk for iron overload. Cu-
10.3 cases of PTP per year, compared to 2.3 cas-
mulative doses of as few as 20 or more RBC
es per year after universal leukocyte reduction
units are associated with increased morbidity
(p <0.001).90
and mortality.97,98Preventing the accumulation
PTP typicallydoes not recur with subsequent
of toxic iron levels by reducing iron stores
transfusions. However, there are case reports of
PTP recurrence. Therefore, for patients with through the use of exchange transfusions, iron
previously documented PTP, efforts should be chelators, or therapeutic phlebotomy can reduce
made to obtain components from antigen- these complications.
matched donors." Autologous donations and di-
rected donations from antigen-matched donors i.rrv R
and family members may also be appropriate.
Because PTPhas also occurred after transfusions
of deglycerolized,rejuvenated, or washed RBCs, When the death of a patient results from a reac-
such manipulations are not indicated to prevent tion to or complication of a transfusion, current
recurrence. regulations require that the fatality be reported
to the FDA by the facility that performed the
Iron Overload
compatibility testing. The director of the Office
A unit of RBCs contains approximately 200 to of Compliance and BiologicsQuality at the FDA
250 mg of iron. Because humans lack a physio- Center for Biologics Evaluation and Research
logic means to excrete excess iron, persistent in- must be notified as soon as possible, followed by
crease in iron influx from transfusions can result the submission of a written report within 7
in iron overload. When the accumulation ofiron days. Table 22-5 lists the contact information for
overwhelms the capacity for safe storage, tissue the FDA. The report should contain the perti-
damage can ensue. As iron accumulates in the nent medical record, including laboratory re-
reticuloendothelial system, liver, heart, spleen, ports, and the autopsy results when available.
TABLE 22-5. How to Contact the FDA99
E-mail fatalities2@fda.hhs.gov
Telephone/voicemail 240-402-9160
Fax 301-827-0333, Attn: CBERFatality Program Manager
Express mail US Foodand Drug Administration

CBEROffice of Complianceand Biologics Quality
Document Control Center
10903 New Hampshire Avenue
W071, G112
Silver Spring, MD 20993-0002
FDA = Food and Drug Administration; GBER= Genterfor Biologics Evaluation and Research.
The patient's underlying illness may make deter- associated fatalities are caused by acute hemoly-
mination of the cause of death difficult. If there sis, TRALI,or TACO. Investigations of these cas-
is any clinicalsuspicion that the transfusion may es must attempt to rule out errors made in the
have contributed to the patient's death, that pos- areas of the laboratory, the transfusion service,
sibility should be investigated. Most transfusion- or during blood administration.
REFERENCES
1. Food and Drug Administration. Transfusion/ do- vision of Healthcare Quality Promotion, Nation-
nation fatalities (annual summaries); Silver al Center for Emerging and Zoonotic Infectious
Spring, MD: CBEROffice of Communication, Diseases, 2018. [Available at https://www.
Outreach, and Development, 2019. [Available cdc.gov/ nhsn/pdfs/biovigilance/bv- hv-proto
at https://www.fda.gov/vaccines-blood-biolog
col-current.pdt]
ics/report-prob1em-center-biologics-evaluation-
3. Berseus 0, Boman K, Nessen SC, Westerberg
research/transfusiondonation-fatalities.]
2. Centers for Disease Control and Prevention. Na- LA.Risksof hemolysis due to anti-A and anti-B
tional Healthcare Safety Network manual: Bio- caused by the transfusion of blood or blood
vigilance Component v2.S.2 Hemovigilance components containing ABO-incompatibleplas-
Module surveillance protocol. Atlanta, GA: Di- ma. Transfusion2013;S3(Supp11):114S-23S.
4. Stowell SR,Winkler AM, Maier CL, et al. Initia- 17. Heddle NM, KlamaL, Singer1, et al. The role of
tion and regulation of complement during he- the plasma from platelet concentrates in transfu-
molytic transfusion reactions. Clin Dev Immu- sion reactions. N EngJ Med 1994;331 :625·8.
nol20 12;2012:307093. 18. Brubaker DB. Clinical significanceof white cell
5. Brodsky RA. Complement in hemolytic anemia. antibodies in febrile nonhemolytic transfusion
Blood2015; 126:2459-65. reactions. Transfusion 1990;30:733-7.
6. Hod EA, Cadwell CM, LiepkalnsJS, et al. Cyto- 19. KingKE,ShireyRS,Thoman SK,et al. Universal
kine storm in a mouse model of IgG-mediated leukoreduction decreases the incidence of fe-
hemolytic transfusion reactions. Blood 2008; brile nonhemolytlc transfusion reactions to
112:891-4. RBCs.Transfusion2004;44:25-9.
7. LongAT, Kenne E, lung R, et al. Contact system 20. PaglinoJC, Pomper GJ, Fisch GS, et al. Reduc-
revisited: An interface between inflammation, tion of febrile but not allergicreactions to RBCs
coagulation, and innate immunity. J Thromb and platelets after conversion to universal
Haemost 2016; 14:427-37. prestorage leukoreduction. Transfusion 2004;
8. Vamvakas EC, Blajchman MA. Transfusion- 44:16-24.
related mortality: The ongoing risks of allogene- 21. Ezidiegwu CN, Lauensteln Kl, RosalesLG,et al.
ic blood transfusion and the available strategies Febrile nonhemolytic transfusion reactions.
for their prevention. Blood2009; 113:3406-17. Management by premedication and cost impli-
9. Weinstock C, Mohle R, Dorn C, et al. Successful cations in adult patients. Arch Pathol Lab Med
use of eculizumab for treatment of an acute he- 2004; 128:991-5.
molytic reaction after ABO-incompatible red 22. Heddle NM, Blajchrnan MA, Meyer RM, et al.
blood cell transfusion. Transfusion 2015;55: A randomized controlled trial comparing the fre-
quency of acute reactions to plasma-removed
605-10.
platelets and prestorage WBC-reducedplatelets.
10. Maskens C, Downie H, Wendt A, et al. Hospital-
Transfusion2002;42:556-66.
based transfusion error tracking from 2005 to
23. SavageWJ, Tobian M, FullerAK, et al. Allergic
2010: Identifyingthe key errors threatening pa-
transfusion reactions to platelets are associated
tient transfusion safety. Transfusion 2014;54:
more with recipient and donor factors than with
66-73; quiz, 65.
product attributes. Transfusion 2011;51:1716-
11. Heddle NM, Fung M, HervigT, et al. Challeng-
22.
es and opportunities to prevent transfusion er- 24. SavageWJ, Hamilton RG, Tobian M, et al. De-
rors: A Qualitative Evaluation for Safer Transfu- fining risk factors and presentations of allergic
sion (QUEST).Transfusion2012;52: 1687-95. reactions to platelet transfusion. J Allergy Clin
12. Dunbar NM, Ornstein DL, Dumont LJ.ABOin- Immunol20 14;133:1772-5.e9.
compatible platelets: Risksversus benefit. Curr 25. SavageWJ, TobianM, SavageJH, et al. Transfu-
Opin HematoI2012;19:475-9. sion and component characteristics are not asso-
13. RapidoF, Brittenham GM, BandyopadhyayS, et ciated with allergic transfusion reactions to
al. Prolonged red cell storage before transfusion apheresis platelets. Transfusion 2015;55:296-
increases extravascular hemolysis. 1 Clin Invest 300.
2017;127:375-82. 26. Vyas GN, Fudenberg HH. Isoimmune anti-IgA
14. Gammon R, ed. Standards for blood banks and causing anaphylactoid transfusion reactions.
transfusion services. 32nd ed. Bethesda, MD: N EngiJ Med 1969;280: 1073-4.
MBB,2020. 27. Vyas GN, Holmdahl L, Perkins HA, Fudenberg
15. Francis RO, Jhang J, Hendrickson JE, et al. Fre- HH. Serologicspecificityof human anti-lgAand
quency of glucose-o-phosphate dehydrogenase- its significance in transfusion. Blood 1969;34:
deficiencyred blood cell units in a metropolitan 573-81.
transfusion service. Transfusion 2013;53:606- 28. Sandler SG, Eder AF, Goldman M, Winters JL.
11. The entity of immunoglobulin Arelated anaphy-
16. Eder AF, Meena-Leist CE, Hapip CA, et al. Clos- lactic transfusion reactions is not evidence
tridium perfringens in apheresis platelets: An based. Transfusion2015;55: 199-204.
unusual contaminant underscores the impor- 29. Shimada E, Tadokoro K, Watanabe Y, et al. Ana-
tance of clinical vigilance for septic transfusion phylactic transfusion reactions in haptoglobin-
reactions (CME).Transfusion 2014;54:857-62; deficient patients with IgE and IgG haptoglobin
quiz, 6. antibodies. Transfusion2002;42:766- 73.
30. Westhoff CM, Sipherd BD, Wylie DE, Toalson 45. Silliman CC, Boshkov LK, Mehdizadehkashi Z,
LD. Severe anaphylactic reactions following et al. Transfusion-related acute lung injury: Epi-
transfusions of platelets to a patient with anti- demiology and a prospective analysis of etiologic
Ch. Transfusion 1992;32:576-9. factors. Blood 2003;101:454-62.
31. Poisson JL, Riedo FX, AuBuchon IP; Acquired 46. Eder AP, Dy BA, Perez JM, et al. The residual
peanut hypersensitivity after transfusion. Trans- risk of transfusion-related acute lung injury at
fusion 2014;54:256-7. the American Red Cross (2008-2011): Limita-
32. Kaufman RM, Assmann SF, Triulzi DJ, et al. tions of a predominantly male-donor plasma
Transfusion-related adverse events in the Plate- mitigation strategy. Transfusion 2013;53: 1442-
let Dose study. Transfusion2015;55:144-53. 9.
33. Kleinman S, Chan P, Robillard P. Risksassociat- 47. Steinberg KP, Hudson LD, Goodman RB, et al.
ed with transfusion of cellular blood compo-
Efficacy and safety of corticosteroids for per-
nents in Canada. Transfus Med RiOV2003; 17:
sistent acute respiratory distress syndrome.
120-62.
34. Tobian M, Fuller AK,UglikK, et al.The impact N EnglJ Med 2006;354: 1671-84.
48. Zhou L, Giacherio D, Cooling L, Davenport RD.
of platelet additive solution apheresis platelets
on allergic transfusion reactions and corrected Use of B-natriureticpeptide as a diagnosticmark-
count increment. Transfusion2014;54: 1523-9. er in the differential diagnosis of transfusion-
35. Kemp SF, Lockey RF,SimonsFE. Epinephrine: associated circulatory overload. Transfusion
The drug of choice for anaphylaxiS.A statement 2005;45: 1056-63.
of the World Allergy Organization. Allergy 49. Bolton-MaggsP, Poles D, et al for the Serious
2008;63:1061-70. . .' " Hazards of Transfusion (SHOT)Steering Group.
36. Tobian M, KingKE,Ness PM. Transfusion pre- The 2017 Annual SHOT Report. Manchester,
medications: A growing practice not based on UK: SHOT, 2018. [Available at https:11
evidence. Transfusion 2007;47:1089-96. www.shotuk.org/shot-reports/.]
37. Kiani-AlikhanS, YongPF, Grosse-KreulD, et al. 50. Wiersum-OsseltonJC, Whitaker B, Grey S, et al.
Successful desensitization to immunoglobulin A Revised international surveillance case defini-
in a case of transfusion-related anaphylaxis. tion of transfusion-associated circulatory over-
Transfusion 2010;50: 1897-901. load: A classification agreement validation
38. Popovsky MA, Haley NR. Further characteriza- study. Lancet HaematoI2019;6(7):e350-e358.
tion of transfusion-related acute lung injury: De- 51. International Societyof BloodTransfusionWork-
mographics, clinicaland Iaboratoryfeatures, and ing Party on Haemovigilance, International
morbidity. Immunohematology 2000; 16:157-9. Haemovigilance Network, MBB. Transfusion-
39. Nakagawa M, Toy P. Acute and transient de- associated circulatory overload (TACO) defini-
crease in neutrophil countin transfusion-related tion (2018). [Available at http://www.aabb.
acute lung injury: Cases at one hospital. Transfu- org/research/hemovigilance/Documentsl
sion 2004;44: 1689-94. TACO-2018-Definition.pdf(accessed December
40. Ferguson ND, Fan E, Camporota L, et al. The
3,2019).]
Berlin definition of ARDS: An expanded ratio-
52. Li G, Daniels CE, KojicicM, et al. The accuracy
nale, justification, and supplementary material.
of natriuretic peptides (brain natriuretic peptide
Intensive Care Med 2012;38: 1573-82.
41. VlaarAP], Toy P, Fung M, et al. A consensus re- and Ntermmal pro-brainnatriuretic) in the dif-
definition of transfusion-related acute lung inju- ferentiation between traTlsfusion-relatiOdacute
ry.Transfusion 2019;59:2465-76. lung Injury and transfusion-related circulatory
42. PopovskyMA, Moore SB.Diagnostic and patho- overload in the critically ill. Transfusion 2009;
genetic considerations in transfusion-related 49:13-20.
acute lung injury. Transfusion 1985;25:573-7. 53. Tobian A, Sokoll L, Tisch D, et al. N-terminal
43. Cherry T, Steciuk M, Reddy W, Marques MB. pro-brain natriuretic peptide is a useful diagnos-
Transfusion-relatedacute lung injury: Past, pres- tic marker for transfusion-associated circulatory
ent, and future. Am J Clin Pathol 2008; 129: overload. Transfusion 2008;48: 1143-50.
287-97. 54. CliffordL, Iia Q, Yadav H, et al. Characterizing
44. Toy P, Gajic 0, Bacchetti P, et al. Transfusion- the epidemiology of perioperative transfusion-
related acute lung injury: Incidence and risk fac- associated circulatory overload. Anesthesiology
tors. Blood 2012;119:1757-67. 2015;122:21-8.
55. Narick C, Triulzi DJ, Yazer MH. Transfusion- 69. Malone DL, Hess JR, Fingerhut A. Massive
associated circulatory overload after plasma transfusion practices around the globe and a
transfusion. Transfusion 2012;52:160-5. suggestion for a common massive transfusion
56. RavalJS, Mazepa MA, Russell SL, et al. Passive protocol. J Trauma 2006;60:S91-6.
reporting greatly underestimates the rate of 70. Engstrom M, Schott U, Romner B, Reinstrup P.
transfusion-associated circulatory overload after Acidosis impairs the coagulation: A throm-
platelet transfusion. Vox Sang 2015; 108:387- boelastographic study. J Trauma 2006;61 :624-8.
92. 71. Counts RB,Haisch C, Simon IL, et al. Hemosta-
57. Pagano M, Ness P, Chajewski 0, et al. Hypoten- sis in massively transfused trauma patients. Ann
sive transfusion reactions in the era of prestor- Surg 1979; 190:91-9.
age Ieukoreduction, Transfusion 2015;55:1668- 72. Murray DJ, Pennell BJ,Weinstein SL,OlsonJD.
74. Packed red cells in acute blood loss: Dilutional
58. Li N, Williams L, Zhou Z, Wu Y. Incidence of coagulopathy as a cause of surgical bleeding.
acute transfusion reactions to platelets in hospi- Anesth Analg 1995;80:336-42.
talized pediatric patients based on the US 73. Harrigan C, Lucas CE, Ledgerwood AM, et al.
hemovigilance reporting system. Transfusion Serial changes in primary hemostasis after mas-
2014;54:1666-72. sive transfusion. Surgery 1985;98:836-44.
59. Cyr M, Hume H, Champagne M, et al. Anomaly 74. Martini WZ. Coagulopathy by hypothermia and
of the des-Arg9-bradykininmetabolism associat- acidosis: Mechanisms of thrombin generation
ed with severe hypotensive reactions during and fibrinogen availability. J Trauma 2009;67:
blood transfusions:A preliminary study. Transfu- 202-8; dtscussion, 8-9.
sion 1999;39: 1084-8. 75. Collins JA. Recent developments in the area of
60. Cugno M, Nussberger J, BiglioliP, et al. Increase
massive transfusion. World J Surg 1987; 11 :75-
of bradykinin in plasma of patients undergoing
81.
cardiopulmonary bypass: The importance of
76. Holcomb JB, Tilley BC, Baraniuk S, et al. Trans-
lung exclusion. Chest 2001; 120:1776-82.
fusion of plasma, platelets, and red blood cells in
61. Sihler KC, Napolitano LM. Complications of
a 1:1:1 vs a 1:1:2 ratio and mortality in patients
massive transfusion. Chest 2010; 137:209-20.
62. Dzik WH, Kirkiey SA. Citrate toxicity during with severe trauma: The PROPPR randomized
massive blood transfusion. Transfus Med Rev clinical trial. JAMA2015;313:471-82.
1988;2:76-94. 77. Sperry JL, Guyette FX, Brown JB, et al. Prehos-
63. Spinella PC, Holcomb JB. Resuscitation and pital plasma during air medical transport in trau-
transfusion principles for traumatic hemorrhagic ma patients at risk for hemorrhagic shock.
shock. Blood Rev 2009;23:231-40. N Eng!J Med 2018;379:315-26.
64. Wilson RF,BinkleyLE,Sabo FMJr, et al. Electro- 78. Roberts I, Shakur H, Ker K, et al, Antifibrinolytic
lyte and acid-base changes with massive blood drugs for acute traumatic injury. Cochrane Data-
transfusions. Am Surg 1992;58:535-44; discus- base Syst Rev 20l2;12:CD004896.
sion, 44-5. 79. O'Ouin RJ,Lakshminarayan S. Venous air embo-
65. Strauss RG. RBC storage and avoiding hyperka- lism. Arch Intern Med 1982; 142:2173-6.
lemia from transfusions to neonates and infants. 80. Mirski MA, Lele AV, Fitzsimmons L, Toung TJ.
Transfusion 2010;50: 1862-5. Diagnosis and treatment of vascular air embo-
66. Liu EA,Mannino FL, Lane TA. Prospective, ran- lism. Anesthesiology2007; 106: 164-77.
domized trial of the safety and efficacyof a limit- 81. Schonewille H, van de Watering LM, Loomans
ed donor exposure transfusion program for pre- DS, Brand A. Red blood cell alloantibodies after
mature neonates. J Pediatr 1994; 125:92-6. transfusion: Factors influencing incidence and
67. Bansal I, Calhoun BW, Joseph C, et al. A com- specificity.Transfusion 2006;46:250-6.
parative study of reducing the extracellular po- 82. Tormey CA, Stack G. The persistence and eva-
tassium concentration in red blood cells by nescence of blood group alloantibodies in men.
washing and by reduction of additive solution. Transfusion 2009;49:505-12.
Transfusion 2007;47:248-50. 83. Vamvakas EC, Pineda AA, Reisner R, et al. The
68. O'Leary MF, SzklarskiP, KleinTM, et al, Hemol- differentiation of delayed hemolytic and delayed
ysis of red blood cells after washing with differ- serologic transfusion reactions: Incidence and
ent automated technologies: Clinical implica- predictors of hemolysis. Transfusion 1995;35:
tions in a neonatal cardiac surgery population. 26-32.
84. Schonewille H, van de Watering LM, Brand A. 93. HayashiT, Hirayama F.Advances in alloimmune
Additional red blood cell alloantibodies after thrombocytopenia: Perspectives on current con-
blood transfusions in a nonhematologic alloim- cepts of human platelet antigens, antibody de-
munized patient cohort: Is it time to take pre- tection strategies, and genotyping. Blood Trans-
cautionary measures? Transfusion 2006;46:630- fus 2015;13:380-90.
5. 94. Ziman A, Klapper E, Pepkowitz S, et al. A sec-
85. Dumas G, Habibi A, Onimus T, et al. Eculizum- ond case of post-transfusion purpura caused by
ab salvage therapy for delayed hemolysis trans- HPA-5a antibodies: Successful treatment with
fusion reaction in sickle cell disease patients. intravenous immunoglobulin. Vox Sang 2002;
Blood 2016; 127:1062-4. 83:165-6.
86. Ruhl H, Bein G, Sachs UJ. Transfusion-associated 95. Loren AW, Abrams CS. Efficacy of HPA-1a
graft-versus-host disease. Transfus Med Rev (PIA1)-negative platelets in a patient with post-
2009;23:62-71. transfusion purpura. Am J Hematol 2004;76:
87. Klein HG. Transfusion-associated graft-versus- 258-62.
host disease: Less fresh blood and more gray 96. Vu K, Leavitt AD. Posttransfusion purpura with
(Gy) for an aging population. Transfusion 2006; antibodies agalnst human platelet antigen-4a fol-
46:878-80. lowing checkpoint inhibitor therapy: A case re-
88. KopolovicI, Ostro J, Tsubota H, et al. A system- port and review of the literature. Transfusion
atic review of transfusion-associated graft-
2018;58;2265-9.
versus-host disease. Blood 2015; 126:406-14.
97. Alessandrino EP, Della Porta MG, BacigalupoA,
89. Castro G, Merkel PA, Giclas HE, et al. Amoto-
et al. Prognostic impact of pre-transplantation
salen/UVA treatment inactivates T cells more
transfusion history and secondary iron overload
effectively than the recommended gamma dose
for prevention of transfusion-associated graft- in patients with myelodysplastic syndrome un-
versus-host disease. Transfusion 2018;58:1506- dergoing allogeneic stem cell transplantation: A
15. GITMO study. Haematologica 2010;95:476-84.
90. Williamson LM, Stalnsby D, Jones H, et al. The 98. Fung EB, Harmatz P, Milet M, et al. Morbidity
impact of universalleukodepletion of the blood and mortality in chronically transfused subjects
supply on hemovigilance reports of posttransfu- with thalassemia and sickle cell disease: A re-
slon purpura and transfusion-associated graft- port from the multi-center study of iron over-
versus-host disease. Transfusion 2007;47: 1455- load. AmJ HematoI2007;82:255-65.
67. 99. Food and Drug Administration. Notification pro-
91. Taaning E, Svejgaard A. Post-transfusion purpu- cess for transfusion related fatalities and dona-
ra: A survey of 12 Danish cases with special ref- tion related deaths. Silver Spring, MD: CBER
erence to immunoglobulin G subclasses of the Office of Communication, Outreach, and Devel-
platelet antibodies. Transfus Med 1994;4: 1-8. opment, 2019. [Availableat https://www.fda.
92. Shtalrid M, Shvidel L, Vorst E, et al. Post- gov/vaccines- blood-biologics/ report-problem-
transfusion purpura: A challenging diagnosis. Isr center-biologics-evaluation-research/transfu
Med AssocJ 2006;8:672-4. siondonation-fatalities.]
Perinatal.lssues in Transfusion Practice
Lani Lieberman MD; Gwen Clarke, MD; and Annika M. Svensson, MD, PhD
EMOLYTIC DISEASE OF THE FETUS cause early and/or severe hemolytic disease.l-'
and newborn (HDFN), fetal/neonatal al- The resulting increased hematopoietic drive
loimmune thrombocytopenia (FNAIT), causes a condition termed erythroblastosis feta-
and immune thrombocytopenia (ITP) affect lis, with liver and spleen enlargement secondary
pregnant women, their fetuses, and newborns. to extramedullary hematopoiesis, and portal hy-
The blood bank and transfusion service play crit- pertension. liver enlargement can lead to de-
ical roles in supporting the diagnosis and treat- creased production of albumin and associated
ment of these conditions, including the appro- decreased plasma oncotic pressure, with gener-
priate provision of Rh Immune Globulin (RhIG). alized edema, ascites, and effusions known as
hydrops fetalis. Severe HDFN can occur as early
as 18 to 20 weeks' gestation or earlier with anti-
TH bodies to Kl (of the KEL system), but may be
difficult to detect; severity usually increases in
subsequent pregnancies. Untreated, hydrops fe-
talis, with its associated high-output cardiovas-
HDFN refers to the destruction of fetal and new-
cular fallure secondary to anemia, can lead to fe-
born red cells by maternal red cell antibodies
tal death. The destruction of red cells also leads
specific for paternally inherited antigens on fetal
to elevated bilirubin levels. The maternal liver
red cells or erythroid precursors. HDFN can
clears the bilirubin, preventing the •accumula-
range from an isolated serologic finding with a
tion of bilirubin in the fetus. After birth, the in-
clinically unaffected newborn and a positive di-
fant's immature liver enzymatic pathways can-
rect antiglobulin test (DAT)result to severe ane-
not metabolize the unconjugated bilirubin,
mia and occasionally fetal demise.
which can increase to dangerous levels. The
hyperbilirubinemia can cause permanent braln
Pathophysiology
damage, known as kernicterus.' The maternal
Maternal immunoglobulin .G . (IgG) antibody antibody typically decreases in the neonate over
crosses the placenta Into the fetal circulation, 12 weeks, with a half-life of. about 25 days.
where it binds to fetal red cells or erythroid pre- Some antibodies. may cause more prolonged
cursors. The. immunoglobulin subclasses. IgG1 anemia or delayed-onset anemia. Al-Alalyan et al
and IgG3 are more likely than IgG2 or IgG4 to found that prolonged (mean postnatal age of
Lani Lieberman, MD, Assistant Professor, Department of Laboratory Medicine and Pathobiology, University of
Toronto, and Department of Clinical Pathology, University Health Network, Toronto, Ontario, Canada; Gwen
Clarke, MD, Clinical Professor, Department of Laboratory Medicine and Pathology, University of Alberta, and
Associate Medical Director, Clinical Services, Canadian Blood Services, Edmonton, Alberta, Canada; and Annika
M. Svensson, MD, PhD, Transfusion Medicine Consultant, Denver, Colorado
6S9
43.3 ± 15.7 days) hyporegenerative anemia est risk at delivery. After exposure, the maternal
(hemoglobin <8 gldL) is common among Rh- immune system must form the red cell antibody
isoimmunlzed infants, regardless of the use of and switch class from IgM to IgG before HDFN
intravascular intrauterine transfusion (lUT), in becomes a clinically significant risk. For all of
term and late-preterm infants." Other antibodies these reasons, the first pregnancy is rarely affect-
that cause anemia through erythropoietic sup- ed by HDFN.I2,13 Rick factors for FMH include
pression (anti-Kl, anti-Ir', anti-Ge, and rare cas- the following: abdominal trauma, placenta pre-
es of high-titer IgG anti-M antibodies) may also via, abruptio placentae, ectopic pregnancy,
contribute to prolonged anemia,"? threatened termination, or fetal death, as well as
Because of the widespread prophylactic use procedures such as amniocentesis, fetal blood
of RhIG in high-development-index nations, sampling (FBS), intrauterine manipulation, or
ABO incompatibility is currently the most com- abortion. 14
mon cause of HDFN. When present, ABO RhD is the most potent immunogenic red
HDFN is typically mild. 10 ABO HDFN is defined cell antigen. It remains the most important
by maternal/fetal ABO incompatibility, an ele- cause of HDFN in nations without robust access
vated bilirubin (corrected for gestational age), to prenatal care. 15,16 As little as 0.1 to 1 mL of D-
and a positive DAT result. If treatment is need- positive red cells can stimulate alloantibody pro-
ed, phototherapy is usually sufficient, and the duction." In ABO-incompatible, D-negative
need for neonatal exchange transfusion is un- mothers, the alloimmunization rate for the D
common. The incidence of ABO HDFN ranges antigen is markedly diminished, thus demon-
from 1% to 4%, depending on the ethnicity of strating a partially protective effect of ABO in-
the population. It develops when naturally oc- compatibility.P'" Anti-K1 is also an important
curring IgG anti-A and/or -B are transported cause of HDFN, and a potent immunogen. The
across the placenta and bind to fetal A and/or B level of hemolysis caused by the presence of
antigens on red cells. Group 0 mothers of Euro- antl-Kl is less than that caused by anti-D, be-
pean or Asian ancestry with group A infants are cause the K1 antigen is expressed on the early
most commonly affected; in populations of Afri- red cell precursors leading to reticulocytopenia
can ancestry, group B infants with group 0 and significant anemia." Other antibodies that
mothers are most likely to be affected. ABO have been less commonly reported to cause
HDFN rarely leads to severe anemia, because fe- moderate or severe disease include antibodies
tal ABO antigens are poorly developed and iso- against E, c, C, k, Kpa,Kpb,Ku, Jsa,Jsb,Jka, Fy",
hemagglutinins are neutralized by tissue and Fyb,S, s, and U.20,2I (See Table 13-5 in Chapter
soluble antigens. If a cord blood DAT result is 13.) The presence of multiple antibodies may
negative, clinically SignificantABO HDFN is un- lead to more severe HDFN.22
likely, and titration of these antibodies is not typ-
ically required. Diagnosis and Monitoring
The diagnosis of HDFN and ongoing laboratory
Maternal Alloimmunization
assessments involve the cooperation of the pa-
The biological characteristics that make certain tient, health-care provider, and blood bank/
individuals responsive to immunogenic stimuli transfusion laboratory personnel. The patient's
have not been clearly elucidated. Females can obstetric and transfusion history are important;
be alloimmunized to red cell antigens by preg- a previously affected pregnancy predicts the po-
nancy, blood transfusion, transplantation, or tential for future HDFN.23 During the first pre-
from unknown stimuli. 11 Minor fetomaternal natal visit, the maternal ABO and RhD type and
hemorrhage (FMH) occurs spontaneously in a an antibody screen to detect IgG-phase antibod-
high proportion of women during gestation. The ies [37 C with antihuman globulin (AHG)]
likelihood of FMH increases with gestational age should be performed. If a D-negative woman
(from 3% in trimester 1 to 12% and 45% in tri- does not have anti-D, she is a candidate for
mesters 2 and 3, respectively) and is at the high- RhIG administration to prevent RhD alloimmu-
C HAP T E R 2 3 Perinatal Issuesin TransfusionPractice 661
nization (see below). A positive antibody screen ter of 8 or lower may be accepted as the critical
requires antibody identification and titration. level, although some centers regard any K1 sen-
Antibodies against minor carbohydrate blood sitization as critical. The critical titer for other
group antigens such as antl-I, -PI, -Lea,and -Le\ antibody specificities is typically 16 to 32, and
whether IgM or IgG, do not cause HDFN and for all antibodies a "two tube" increase in titer
may be ignored because these antigens are poor- level is also considered important. 30-32 Once a
ly developed at birth. Treatment of the mother's critical titer is reached, Doppler fetal ultrasound
plasma with dithiothreitol (DTT) preferentially of the middle cerebral artery (MeA) is a nonin-
destroys IgM antibodies and can help distin- vasive way to assess fetal anemia.f Decisions
guish IgGfrom IgM antibodiesfor cases of anti-M about when to intervene with fetal transfusion
with increasing titer.24,25 After identifyinga clini- or delivery are based on the degree of fetal ane-
cally significant red cell antibody, the father mia and gestational age; an increase in the ve-
should be tested for the corresponding red cell locity of the end systolic MCA blood flow to
antigen, if possible, to stratify risk of fetal inheri- >1.5 multiples of the mean (MoM) on the ultra-
tance. Homozygousfathers have a 100% chance sound denotes moderate to severe anemia.
of offspringexpressing the red cell antigen; het-
erozygous fathers portend a 50% chance. For Treatment:
women sensitized to RhD, serologic methods
In cases of severe anemia, if it is too early in ges-
cannot readily determine paternal zygosity;
tation for delivery, fetal transfusion is indicated
when possible, a predicted genotype using pa-
to treat the anemia. The transfusion will sup-
ternal DNA testing is indicated." Alternatively,
press the fetal erythropoiesis and thus the pro-
fetal risk stratification through prediction of the
duction ofred cellswith the antigen correspond-
fetal red cell antigen type can be accomplished
ing to the maternal alloantibody. After birth,
by genotyping fetal DNA, either using amnio-
neonatal therapies such as phototherapy, simple
cytes or noninvasively with cell-free fetal DNA
transfusion, and exchange transfusion may be
(cffDNA)present in maternal plasma.27,28
indicated.
During an alloimmunized pregnancy, month-
ly to biweekly maternal antibody titers can be
Fetal Transfusion
used to deduce if the fetus is the source of ongo-
ing maternal immune stimulation and is at risk Red Blood Cell (RBC)units for IUT should be 1)
for clinically Significant HDFN. Once an anti- group 0 (in most cases) and.crossmatch compat-
body titer reaches a level above which clinically ible with maternal plasma, 2) irradiated to
significant HDFN could occur (the critical titer), prevent transfusion-associated graft-vs-host dis-
further titration may not be of benefit and non- ease (TA-GVHD),3) cytomegalovirus (CMV)
invasive clinical monitoring for fetal anemia reduced-risk (leukocytereduced or from a CMV-
should commence." The AABB-recommended seronegative donor), and 4) known to lack he-
method for titration is a salineAHG test incubat- mogiobin S to prevent sickling under low oxy-
ed for 60 minutes at 37 C ("tube" method; see gen tension. RBCunits collected within 5 to 7
Method 5-3). Other methods, such as albumin days are preferred, ff available, because of the
or gel, may result in higher titers than the rec- large transfusion volume and to prolong the cir-
ommended method and should be validated culation of transfused red cells. RBCunits may
with clinical findings and laboratory data to en- be washed or concentrated to a hematocrit of
sure appropriate interpretation. Because of the 70% to 85%. Most institutions will routinely use
potential for poor reproducibility of red cell anti- group 0, D-negative units for IUT, but type-
body titers, individual blood banks should vali- specific (D-positive)units may be used if anti-D
date their testing internally and keep previous is not causative and/or if the fetus is known to
specimens for subsequent comparison." The be D positive. If the maternal antibody is direct-
critical titer for anti-D is usually 8 to 32 in the ed at a high-prevalence antigen and no other
AHG phase. Because sensitization to the K1 an- compatible blood is available, the mother's
tigen may lead to hypoproliferativeanemia, a ti- washed, irradiated red cells or compatible, irra-
diated red cells from maternal siblings or rare ministration of intravenous immune globulin
donor registries can be used." (MG). Both have been used in small studies,
The volume of blood to be transfused can be early in gestation before IUT can be accom-
calculated by 1) multiplying the ultrasound- plished, or as alternatives to IUT, with the goal
estimated fetal weight in grams by the factor of blunting the effect of the maternal antibody.'
0.14 mUg to determine the fetal and placental In small case series, MG infusion has been
total blood volume, 2) multiplying this amount shown to stabilize anti-D titers, and results were
by the difference in posttransfusion (desired) best when the procedure was started before 28
and pretransfusion hematocrit, and 3) dividing weeks' gestation." Plasma exchange can tempo-
the resulting amount by the hematocrit of the rarily reduce antibody levels by as much as 75%
RBCunit to be transfused." For example: with and has been shown to reduce the risk of fetal
an estimated fetal weight of 1000 g, a desired death and/or morbidity in a mother with a pre-
posttransfusion hematocrit of 40% (0.4), a previous severe course of HDFN in one case se-
transfusion hematocrit of 15% (0.15), and a ries." The American Societyfor Apheresis classi-
measured hematocrit of the RBC unit of 85% fies HDFN as a Category III indication [basedon
(0.85), the volume to transfuse is 41.2 mL, as weak (grade 2) evidence] for the use of plasma
shown below. exchange before commencement of IUT.40
[1000 g x 0.14 mUg x (0.40 - 0.15)]/0.85 = Neonatal Treatment

41.2 mL
After delivery,hemoglobin and bilirubin must be
The volume and rate of the transfusion closelymonitored. The threat of kernicterus can
should be adjusted to accommodate the clinical be high in HDFN, especially in premature neo-
status of the fetus. Because these calculations nates." Phototherapy oxidizes elevated uncon-
are based on estimates and the fetal placental jugated bilirubin, allowing the oxidation prod-
volume can vary for any estimated fetal weight, ucts to be excreted in the urine. For severe
the hemogiobin or hematocrit should be jaundice unresponsive to phototherapy, most ex-
checked after transfusion to confirm that the de- perts recommend MG with the goal to avoid an
sired posttransfusion hemoglobin level has been exchange transfusion (AmericanAcademy of Pe-
achieved." When deliveryis not imminent, IUT diatrics), although other studies question its effi-
is repeated according to the severity of the dis- cacy, particularly given side effects such as he-
ease or based on an estimated decline in hema- molysls and necrotizing enterocohtis.P" In
tocrit of 1% per day. Generally, the strategy is to neonates who are unresponsive to phototherapy
and MG, a double-volume exchange transfu-
raise the hematocrit to >40% with transfusion
sion replaces approximately 85% to 90% of the
and to repeat transfusion when the calculated
blood volume and removes 50% of the bilirubin.
hematocrit has declined to 30% or below."
Exchange transfusion with reconstituted whole
Transfusion into the peritoneal cavity can be
blood (Chapter 24) is generally unnecessary if
used in cases where the umbilical cord artery
the infant received IUTs;however, small-volume
cannot be accessed, particularly when IUT is re-
"top-up" transfusions may be required until the
quired in early pregnancy. In the absence of hy-
neonate's erythropoiesis is able to support the
drops fetalis, the outcome of a successful IUT
hemoglobin requirement and residual maternal
treatment is generally good, with a low inci-
antibodies have disappeared.
dence of neurodevelopmental impairment and a
low risk of fetal loss.37
Prevention
Maternal Treatment Rh Immune Globulin
During pregnancy, alternative or adjunctive D-negative and some variant D-positivefemales
therapies to IUT have been recommended, in- are candidates for RhIGprophylaxis during preg-
cluding maternal plasma exchange and the ad- nancy to prevent RhD alloimmunization. RhIG
is made from pooled human plasma from indi- in populations of European ancestry, a majority
viduals either naturally or intentionally immu- of individuals with the serologicweak D pheno-
nized to the D antigen; recombinant products type may not benefit from the administration of
are in development. The product contains IgG- RhIG,because most have underlying RHD geno-
subtype anti-D. It is available in 300-, 120-, and types of weak D types 1, 2, or 3. Individuals
50-l-lgdoses. Appropriate ante- and postpartum with these genotypes have not been reported to
administration of RhIG reduces the risk of a form D alloantibody after exposure to conven-
D-negative mother becoming immunized by a tional D epitopes.52,53 Genotyping pregnant
D-positivefetus from about 16% to <0.1 %. The women with serologic weak D phenotype early
American Collegeof Obstetricians and Gynecol- in pregnancy to identify weak D genotypes 1, 2,
ogistsrecommends the first dose of RhIGbe giv- or 3 allows for selective administration of RhIG
en at 28 weeks' gestation because 92% of wom- to those who will benefit and avoids unneces-
en who develop anti-D during pregnancy do so sary treatment of women who will not. The
at or after 28 weeks.45-47 RhIG is indicated after practice of weak D genotypingin this population
any event that increases the risk for FMH. is becoming standard. If a woman is found to
D-negative females who have been previously have serologic weak D phenotype for the first
immunized to the D antigen, D-positivefemales, time at the time of delivery, RHD genotyping of-
and D-negative females whose infants are ten cannot practically be completed before the
known to be D negative are not candidates for 72-hour window for RhIG administration. In
RhIG.Women with apparent antibodies to both such cases, administration of RhIG is the pru-
D and C should be investigated for presence of dent choice, with subsequent RHD genotyping
anti-G before determining their candidacy for for weak D types to guide care for future preg-
RhIG. Unless specific reactivity to D is demon- nancies. Women proven to have weak D types
strated, a pregnant D-negativefemalewith anti-G 1, 2, or 3 can be managed as D-positivefor the
should receive RhIG. Outside of the United purpose of transfusion, thus conserving the sup-
States, RhIG prophylactic protocols may differ. ply of D-negative blood.52,54-56 Counseling fe-
Antenatal RhIG may be targeted to those wom- male patients found to have weak D genotypes
en who have a D-positive fetus determined by 1, 2, or 3 about their RHD status is important
cffDNAtesting. cffDNAobtained from maternal for their peace of mind and to minimize confu-
blood samples can identify pregnancies with sion. These females should be made aware that
D-negative fetuses and avoid RhIG administra- their serologic RhD status may be designated
tion.46,47 differently by different laboratories, and reas-
Administration of RhIG during pregnancy sured that they do not need RhIG in the future.
typically leads to a positive antibody screening At this time, there is not enough experience to
result in the mother, but the anti-D titer is low determine if other variant RHD genotypes can
and thus poses no risk to the fetus. Occasionally, form anti-D; thus, women with RHD genotypes
RhIG will cause a positive DAT result in the other than weak D types 1, 2, or 3 should be
newborn. The mechanism of action of RhIG has considered D negative for the purposes of trans-
not been completely elucidated. D-positive red fusion and RhIG administration.
cells are opsonized by RhIG and cleared by mac- After delivery of a D-positiveinfant, D-nega-
rophages in the spleen, which results in cyto- tive mothers without alloanti-D should receive
kine secretion and immunomodulation.w" Sim- RhIG. About 10% of the RhIG dose given at 28
ilar antigen-specificsuppression of the immune weeks' gestation will be present in the mother
response has been observed in a murine system when the infant is delivered at term. (The half-
where anti-K1 prevented immunization against life of IgG is about 25 days.) To.determine the
transfused K1-positivecellsbut not against other correct RhIG dose for postpartum administra-
antigens.P tion, a maternal blood sample is screened for
Historically,pregnant women with serologic FMH. Three tests are available to test for FMH:
weak D (see Chapter 11) were treated as D neg- the rosette screening test, the Kleihauer-Betke
ative and deemed eligible for RhIG.51 However, (KB)test, and the flow cytometry test.
The rosette test (Method 5-1) is a semiquan- assessment of the FMH by flow cytometry is
titative screen that is positive when more than used, a more precise dose of RhIGcan be calcu-
10 mL of D-positivefetal red cells are present in lated. The volume of fetal red cells Is estimated
the maternal circulation. The maternal speci- as follows:
men should be collected 1 to 2 hours after all of
the products of conception are delivered.57-59 (Fetalcells/total cells counted) x maternal
The test Is performed by incubating the D-nega- blood volume (mL) = FMH,whole blood (mL)
tive maternal blood sample with D antisera and
subsequently adding D-positive indicator red For example: For 6 fetal cells counted out of
cells, which form agglutinates (rosettes) with 2000 cells, and an estimated maternal blood
any fetal D-positive red cells in the maternal volume of 5000 mL, the FMH is calculated to
sample. The sample is read microscopicallyand be 15 mL. Femaleswith very high body mass in-
rosettes (agglutinates)are counted. If the screen- dices may have a total blood volume Significant-
ing test result is negative, a standard dose of ly higher than 5000 mL, which needs to be con-
300-llg RhIG is given. This is sufficient to pre- sidered in the calculation.62,63
vent immunization by 15 mL of fetal red cells or The estimated FMH volume is used to calcu-
30 mL of whole blood. A positive screening test late the dose of RhIG. A 300-11gdose of RhIG
result indicates that the FMHmay be > 15 mL of will suppress alloimmunization by 30 mL of fe-
fetal red cells (30 mL of whole blood) and re- tal whole blood. In the above example, the
quires further testing to quantify FMH and de- FMH is 15 mL, so the number of 300-pg RhIG
termine the appropriate dose of RHIG.Mothers vials to administer can be calculated as 15 mLi
with variant RHD genes may have false-positive 30 mL RhIG/vial = 0.5 vial. If the KB test is
rosette screening results. A positive DAT result used, due to the subjective nature of the test
may also reflect a false-positiveresult. To quanti- performance and interpretation, the RhIG dose
fy the volume of the FMH, a quantitative assay, is rounded up to the next whole number if the
such as the KB test or flow cytometric assess- number to the right of the decimal point is 2:5(if
ment, is used. The risk of FMH >30 mL at the <5, it is rounded down). An additional vial
time of deliveryis approximately 1 in 1250.12 should be added to all calculations. (See Table
The KB test capitalizes on the resistance of 23-1.) For example, the dose is calculated as fol-
fetal hemoglobin to acid treatment (Method 5- lows: 0.5 vial is rounded up to 1 vial, then 1 ex-
2). A thin smear of maternal blood is placed on a tra vial is added, resulting in a total of 2 vials to
slide, treated with acid, rinsed, counterstained, be administered.
and read microscopicallyby counting 1000 to
2000 cells to estimate the ratio of fetal to mater- 1.6 vials calculated from formula =
nal red cells. The maternal red cells appear as 2 (rounded up) + 1 (one vial added) =
pale "ghost" cells, and the fetal red cells are 3 vials
pink. KBresults can be confounded by the pres-
ence of fetal hemoglobin in the maternal cells 1.4 vials calculated from formula =
due to hereditary persistence of fetal hemoglo- 1 (rounded down) + 1 (one vial added) =
bin (HbF), elevated maternal HbF cells that in- 2 vials
crease during pregnancy, or hemoglobinopathies
such as sickle cell disease or thalassemia. In Postpartum RhIG should be given to the
such cases, flow cytometry can help to differen- mother within 72 hours of delivery. If prophy-
tiate between FMH and alternate causes. Row laxis is delayed, the American College of Obste-
cytometry measures fetal hemoglobin and/or D- tricians and Gynecologists recommends that
positive red cells. It has higher precision than RhIG treatment should still be provided." RhIG
the KBtest, although the requirement for rapid should also be administered if the RhD type of
turnaround time and lack of availabilityof flow the newborn is unknown or undetermined (eg,
cytometry or related expertise leads many labo- for a stillborn infant). RhIGis given by the intra-
ratories to continue using the KBmethod/":" If muscular (1M)or intravenous (IV) route. Some
C HAP T E R 2 3 PerinatalIssuesin TransfusionPractice 665
TABLE 23-1.Examples of RhlG Dose Selection Based on Differing Volumes of Fetomaternal

Hemorrhage in a 70-kg Female Using Vial Size of 300 ~g
0.9-1.4 3 4500
1.5-2.0 4 1200 6000
2.1-2.6 5 1500 7500
Notes:
1. Table reflects calculations basedon an example maternal blood volume of 5000 mL.
2. In the United States, 1 vial of 300 1-19(1500 IU) is neededfor each 15 mL of fetal red cells or 30 mL of fetal whole
blood. Other vial sizes are available.
preparations are for 1M administration only. decrease the risk for additional alloirrimuniza-
RhIG consists almost entirely of IgG, with only tion." Decreased alloimmunization in women
small amounts of other immunoglobulins. Ac- who received lUT strictly matched for Fy, Ik,
tive D immunization has a significant IgM com- and S antigens was shown in one study."
ponent; thus, new anti-Dproduced by the moth-
er is often detected in the saline phase and can
be completely or partially inactivated by 2-mer- p
captoethanol or DTT treatment, whereas reac-
tivity passively acquired from RhIG IgG re-
mains. Passivelyacquired antl-D rarely achieves Maternal thrombocytopenia (platelet counts
a titer >4. Failure of RhIGto prevent immuniza- <150,000/jlL) occurs inS% to .10%of pregnant
tion to D may result from a number of factors women." The causes of the thrombocytopenia
such as increased FMH volumes and high ma- include incidental thrombocytopenia of preg-
ternal body weight, leading to inadequate 1Min- nancy or gestational thrombocytopenia (74%),
jection/" However, in many cases, a cause for D
thrombocytopenia complicating hypertensive
immunization despite administration of RhIG
disorders in pregnancy (21%),and immunologic
cannot be determined."
disorders of pregnancy (4%).69Regardless of the
Red Cell Selection etiology, it is rare for maternal platelet counts to
decrease below 80,OOO/jlL,in any trimester."
Prevention of HDFN aims to reduce exposure of When the cause is immunologic, maternal IgG
females of childbearing ageto red cell alloanti-
antibodies to platelets can cross the placenta
gens. In life-threatening hemorrhage when un-
and cause fetal thrombocytopenia. Two catego-
crossmatched RBCsare needed, it is standard to
use Dmegattve RBCs initially for such females ries ofimrpune-mediated thrombocytopeni~ are
until. the blood type can be determined.v In recognized: cases. caused by alloantibodies
some countries, more extensive matching (K,C, (FNAIT) and those caused by autoantibodies,
c, E, or e) is undertaken for transfusion to these such as ITP and systemic lupus erythematosus
females." RBCunits that are antigenicallysimilar (SLE).The diagnostic distinction between these
to the mother may be selected in an attempt to conditions is important for therapy selection.
Fetal and Neonatal Alloimmune al (8-10 weeks), or fetal DNA present in a ma-
Thrombocytopenia ternal blood sample.81,82Amniocentesis has an
associated risk of pregnancy loss of 0.5% to
Pathophysiology
1.0%.83Chorionic villus sampling increases the
FNAIToccurs when the maternal immune sys- risk of alloimmunizationand is not recommend-
tem forms platelet-specificantibodies that cross ed. Noninvasive prenatal testing uses cffDNA
the placenta and destroy fetal platelets. Platelet techniques but is available only in some cen-
antigens represent specific polymorphisms in ters.81,82Genetic counseling of maternal siblings
platelet membrane glycoproteins. (See Chapter should be considered.
15.) In people of European ancestry, approxi-
mately 79% of FNAITcases are caused by anti- Treatment
bodies to human platelet antigen (HPA)-1a,
Antenatal intervention should be offered to all
which is present in about 98% of the US popula-
mothers of affectedfetuses who have had a pre-
tion. About 9% of cases are caused by anti-Hl'A.
vious pregnancy with FNAIT.The goal of treat-
5b, 4% by anti-Hl'A Ib, 2% by anti-HPA3a, and
ment is to avoid a fetal or neonatal ICH. Assess-
6% by other antibodies (including multiple anti-
ment of the fetus should begin at or before 20
bodies)." In people of Asian ancestry, HPA-4b
weeks' gestation. Mainstay antenatal treatment
and HPA5b are more often implicated than
includes close monitoring at a high-riskobstetri-
HPAla.71
cal center, with or without WIG, and with or
The reported incidence of affected pregnan-
without corticosteroids. The optimal dose,
cies ranges from 0.3 to 1 in 1000.72,73In 25% of
schedule, and gestational age to initiate MG is
FNAIT cases, the platelet antibody develops
not known. In most reports, MG is adminis-
during the first pregnancy and affects the first
tered at a dose of 1 g/kg per week; however, re-
pregnancy. In fact, the maternal antibody has
ported doses include 0.4 g/kg per day for 5
been detected as early as 17 weeks' gestation,
days, 0.5 g/kg per week, 0.8 g/kg per week,
and the fetus may develop thrombocytopenia as
1 g/kg every 2 weeks, and 2 g/kg per
early as 20 weeks' gestation. However, because week.76,84,85
FNAITis often not discovered until birth, when
There are no trials that have evaluated the
the newborn presents with petechiae, ecchy-
most appropriate mode of delivery; both routes,
moses, gastrointestinal bleeding, and/or intra-
cesarean section and vaginal delivery, are of-
cranial hemorrhage (ICH), the focus has been
fered.86,87The mode of delivery should be based
on prevention in subsequent pregnancies.
on obstetrician and patient preference. Further-
FNAIT-related ICH occurs in 0.02 to 0.1 in
more, procedures used to measure the infant's
1000 live births.74-76More than 50% of ICHs oc-
platelet count before delivery, such as FBSwith
cur in utero, often before 28 weeks' gesta-
tion.77,78Thirty-fivepercent of ICH events are fa- intrauterine platelet transfusions, are not recom-
mended, because the risk of morbidity and mor-
tal; nonfatal events are often associated with
tality from the procedure (11%) is greater than
significantneurologic consequences.":"
or equal to the risk of severe bleedingin utero or
at delivery." If FBSis performed, an irradiated,
Diagnosis
CMV-reduced-risk, antigen-negative platelet
A history of antenatal ICH or FNAITin a sibling transfusion should be administered at the same
is one of the most important predictors of time. Procedures during labor associated with
thrombocytopenia in a fetus." The mother and increased hemorrhagic risk to the fetus (eg, fetal
father should be typed for platelet antigens, and scalp electrodes, forceps)should be avoided.
the mother should be screened for alloantibod- After delivery, the greatest risk of infant
ies. DNAtesting of the father can determine the bleeding is present during the first few days of
zygosity of the antigen involved. Direct fetal life. The primary goal of management is to pre-
platelet genotypingcan be determined via amni- vent major hemorrhage such as ICH or death.
otic fluid (18-20 weeks), chorionic villus materi- Expert opinions propose a wide range of platelet
counts as a safe transfusion threshold. To date, thresholds that are safe in the ante- or peripar-
no randomized controlled studies have ad- tum period." Aninternational panel with exper-
dre~se.dthis question. For nonbl~~dmg, aSYIllP- tise in lTP nJ.iiIl~geIllentsuggested that,dur~ng
tomanc newborns, 30,000/IlLis suggested .as the first:tw9 trimesters, treatment shouldbe ini-
an appropriate threshold. For neonates with tiated when the patientis symptomatic, when
bleeding symptoms, such as rCB orgastrointes- platelet counts are <20,000 to 30,000/pL, or.to
tinal bleeding, platelets should be transfused to increase the platelet count before a procedure.
maintain their platelet ccuntssnitlally above Forwomen who require treatment, both IVIG and
100,000/IlL and then above SO,OOO/IlLfor at oral corticosteroids provide a good response."
least 7 days.85Hl'Aselected platelet transfusions Prospective studies report an Inctdence of se-
(maternal or donor), if available, are considered vereneonatal thrombocytopenia «SO,OOO/pL)
first-line therapy because they lead to higher
platelet increments and longer response durations
°
in 9% to 1S% and ICH in to 1.S% of infants."
Although maternal platelet counts do not pre-
when. compared to unselected platelet transfu- dict neonatal thrombocytopenia, mothers with a
sions. If HPA-selected platelets aren?t immedi- previous history of delivering a thrombocytope-
ately available, HPA-unselectedplatel~ts should nic infiiIlt are at a greater risk of delivering an-
be ordered.88Maternal platel~ts maysause a de- other thrombocytopenic infant.93,96The mode of
lay in providing til.etransfusion, as they are of- delivery should be based on obstetrician and pa-
ten logistically challenging to obtain in terms of tient preference, similar to FNAIT management
collection and component manipulation. The noted above.
typical platelet nadir occurs within 48 hours of In all infants born to mothers with a history
delivery,89,90and the majority of infants improve of ITP, an early postnatal platelet count should
within 1 to S weeks." In rare cases, thrombocy- be performed. 1M injections such as vitamin K
topenia may persist for up to 8 to 12 weeks." should be avoided until the platelet count is
known to be normal. A head ultrasound should
Immune Thrombocytopenia be performed to rule out ICH in thrombocytope-
Infants of mothers with thrombocytopenia from nic infants (platelet counts <SO,OOO/pL).
ITP, StE, or other autoi.ll1mup~(jiso,rdersmay Although treatment of the neonate is rarely
be thrombocytopenic because of placental trans- needed} most hemorrhagic events in neonates
fer of maternal autoantibodies. In general, occur 24 to 48 hours after delivery. Platelet
thrombocytopenia and sequelae in neonates counts should be repeated daily, as the nadir is
born to mothers with these conditions is less se- frequently seen 2 to S days after birth. In those
vere than in FNAIT. Although clinically signifi- with clinical hemorrhage or if the platelet
cant bleeding episodes are rare, newborns count remains <30,000/pL, NIG and/or
should be followed closely, as platelet counts platelet transfusion should be considered." Neo-
can decrease after birth. natal counts should be monitored until normal-
ITP is estimated to occur in 1 in 1000 to 1 in ized because thrombocytopenia can persist and
10,000 pregnant women." Fortunately, bleed- could represent an inherited form of thrombocy-
ing symptoms during pregnancy or at. the time topenia.
of delivery are uncommon." The management Literature and recommendations are avail-
of pregnant patients with ITP issimilarto that in able for ITp·management; but less so for other
nonpregnant women. Recommendations are autoimmune conditions. Maternal thrombocyto-
based on expert consensus. American Society of penia secondary to SLE is usually less severe
Hematology (ASH)guidelines state that there is than thrombocytopenia of ITP. Treatment of
no evidence to support obtaining intrapartum fe- thrombocytopenia related to these autoimmune
tal platelet counts routinely and that data are conditions is similar to the approach for an ITP
Itmited in support of speciflc platelet count pattern,"
KEY POINTS
1. HDFN is caused by maternal red cell antibodies that are specific to a paternally derived red cell
antigen. The maternal IgG antibody is transported across the placenta, where it destroys fetal
red cells, causing fetal anemia and hyperbilirubinemia.
2. The most common clinically significant antibodies that cause HDFN are anti-D and anti-KI;
anti-C, -c, and -E, along with others, are Significantbut less common. ABO HDFN is common
but usually causes only mild to moderate symptoms. Some antibodies, such as anti-I, -PI, -Le',
and -Leb, can be ignored during pregnancy.
3. Molecular typing of cffDNAcan be performed on maternal plasma to determine the fetal red
cell antigens. Paternal testing can also predict fetal inheritance. Molecular methods are re-
qulred to determine paternal RHD zygosity.
4. For lUT, the RBCs should be irradiated, CMV reduced-risk, hemoglobin S negative, group 0
(in most cases), and <7 days old.
S. The rosette test is a sensitive method for detecting FMH of approximately 10 mL or more.
The KB test is used to quantify the size of the FMH levels. Flow cytometry can more pre-
cisely measure fetal hemoglobin and/or D-positivered cells compared to KBtesting.
6. The calculated RhIG dose should be rounded up if the number to the right of the decimal point
is :?:0.5, or rounded down if the number is <0.5. In either case, an additional vial should be
added to the result.
7. In FNAIT,the maternal platelet antibody may develop as early as 17 weeks of gestation in the
first pregnancy, and fetal thrombocytopenia may develop as early as 20 weeks. Previous preg-
nancy outcomes predict future outcomes.
8. Irradiated, CMV-reduced-riskplatelets should be glven to treat neonatal thrombocytopenia and
avoid hemorrhage. Corresponding-HPA antigen-negative platelets (if available) will produce
greater posttransfusion platelet count increments.
9. HPA-selected platelet transfusions, if available, are considered first-line therapy for FNAIT. If
HPA-selected platelets are not immediately available, HPA-unselected platelets should be or-
dered.
10. Thrombocytopenia in neonates born to mothers with autoimmune conditions tends to be less
severe than thrombocytopenia in neonates born with FNAIT.
REFERENCES
1. Firan M, Bawdon R, Radu C, et al. The MHC antibodies in fetal alloimmune anemia. N EnglJ
class Irelated receptor, FeRn, plays an essential Med 1998;338:798-803.
role in the maternofetal transfer of gamma- 6. Endo Y, Ito S, Ogiyama Y. Suspected anemia
globulin in humans. Int Immunol2001; 13:993- caused by maternal anti Jraantibodies:A case re-
1002. port. BiomarkRes2015;3:23.
2. Pollock JM, Bowman JM. Anti-Rh(D)IgG sub- 7. Yasuda H, Ohto H, Nollet KE, et al. Hemolytic
classes and severity of Rh hemolytic disease of disease of the newborn with late onset anemia
the newborn. Vox Sang 1990;59:176-9. due to anti M: A case report and review of the
3. Dennery PA, Seidman DS, Stevenson DK. Neo-
Japanese literature. Transfus Med Rev 2014;28:
natal hyperbilirubinemia. N Engl J Med 2001;
344:581-90. 1-6.
4. Al-AlaiyanS, al Omran A. Late hyporegenerative 8. BlackallDP, Pesek GD, Montgomery MM, et al.
anemia in neonates with Rhesus hemolytic dis- Hemolytic disease of fetus and newborn due to
ease. J Perinat Med 1999;27: 112-15. anti-Ge3: Combined antibody-dependenthemol-
5. Vaughan II, Manning M, Warwick RM, et al. In- ysis and erythroid precursor cell growth inhibi-
hibition of erythroid progenitor cells byanti-Kell tion. AmJ Perinatol2008;25(9):541-5.
9. Arndt PA, Garratty G, Daniels G, et al. Late on- tus and newborn due to multiple maternal anti-
set neonatal anaemia due to maternal anti-Ge: bodies. AmJ Obstet GynecoI2015;213:68.e61-
Possible association with destruction of eryth- 5.
roid progenitors. TransfusMed 2005; 15(2):125- 23. Moise KJ Ir, Argoti PS. Management and pre-
32. vention of red cell alloimmunization in pregnan-
10. Ping L, Pang LH, Liang HF, et al. Maternal IgG cy: A systematic review. Obstet Gyneco12012;
anti-A and anti-B titer levels screening in pre- 120:1132-9.
dicting ABO hemolytic disease of the newborn: 24. De Young-Owens A, Kennedy M, Rose RL, et al.
A meta-analysis. Fetal Pediatr Pathol 2015; Anti-M isoimmunization: Management and out-
34:341-50. come at the Ohio State University from 1969 to
11. Delaney M, Wikman A, van de Watering L, et 1995. Obstet GynecoI1997;90:962-6.
al. Blood Group Antigen Matching Influence on 25. Stetson B, Scrape S, Markham KB. Anti-M allo-
Gestational Outcomes (AMIGO)study. Transfu- immunization: Management and outcome at a
sion 2017;57:525-32. single institution. AJPRep 2017;7:e205-1 O.
12. Bowman JM, Pollock JM, Penston LE. Fetoma- 26. Wagner FF, Flegel WA. RHD gene deletion oc-
ternal transplacental hemorrhage during preg- curred in the Rhesusbox. Blood 2000;95:3662.
nancy and after delivery. Vox Sang 1986;51: 27. Finning KM, Martin PG, Soothill PW, Avent
117-21. ND. Prediction of fetal D status from maternal
13. Sebring ES, Polesky HF. Fetomaternal hemor- plasma: Introduction of a new noninvasive fetal
rhage: Incidence, risk factors, time of occur- RHD genotyping service. Transfusion 2002;42:
rence, and clinical effects.Transfusion 1990;30: 1079.
344-57. 28. Van der Schoot E, De Haas M, Clausen FB. Ge-
14. Zipursky A, Pollock J, Chown B, Israels LG. notyping to prevent Rh disease: Has the time
Transplacental foetal hemorrhage after placental come? Curr Opin HematoI2017;24:544-50.
injury during delivery or amniocentesis. Lancet 29. Bachegowda LS, Cheng YH, Long T, Shaz BH.
1963;7:493-4. Impact of uniform methods on interlaboratory
15. Bhutani VK, Zipursky A, Blencowe H, et al.
antibody titration variability: Antibody titration
Neonatal hyperbilirubinemia and Rhesusdisease
and uniform methods. Arch Pathol Lab Med
of the newborn: Incidence and impairment esti-
2017; 141:131-8.
mates for 2010 at regional and global levels. Pe-
30. Zwingerman R, Jain V, Hannon J, et al. Alloim-
diatr Res 2013;74(Suppl1):86-100.
mune red blood cell antibodies: Prevalence and
16. Zipursky A, Paul VK. The global burden of Rh
pathogenicity in a Canadian prenatal popula-
disease. Arch Dis Child Fetal Neonatal Ed 2011;
96(2):F84-5. tion. J Obstet Gynaecol Can 2015;37(9):784-
17. Bowman JM. The prevention of Rh immuniza- 90.
tion. Transfus Med Rev 1988;2:129-50. 31. Judd WJ for the Scientific Section Coordinating
18. Bowman JM. Controversies in Rh prophylaxis. Committee. Guidelines for prenatal and perina-
Who needs Rh immune globulin and when tal immunohematology. Bethesda, MD: AABB,
should it be given? Am J Obstet Gynecol 2005.
1985;151 :289-94. 32. Slootweg YM, Lindenburg IT, KoelewijnJM, et
19. Ayache S, Herman JH. Prevention of D sensitiza- al. Predicting antt-Kell-mediated hemolytic dis-
tion after mismatched transfusion of blood com- ease of the fetus and newborn: Diagnostic accu-
ponents: Toward optimal use of RhIG. Transfu- racy of laboratory management. Am J Obstet
sion 2008;48:1990-9. GynecoI2018;219(4):393.el-8.
20. Reid ME, Lomas-Francis C, Olsson ML. The 33. Biale Y, Dvilansky A. Management of pregnan-
blood group antigen factsbook. 3rd ed. San Di- cies with rare blood types. Acta Obstet Gynecol
ego, CA:Academic Press, 2012. Scand 1982;61:219.
21. KoelewijnJM, Vrijkotte TG, van der Schoot CE, 34. Mandelbrot L, DaffosF, Forestier F. Assessment
et al. Effect of screening for red cell antibodies, of fetal blood volume for computer-assisted
other than anti-D, to detect hemolytic disease of management of in utero transfusion. Fetal Ther
the fetus and newborn: A population study in 1988;3:60-6.
the Netherlands. Transfusion2008;48:941. 35. Radunovic N, Lockwood CJ, Alvarez M, et al.
22. Markham KB, Rossi KQ, .Nagaraja HN, The severely anemic and hydropic isoimmune
O'Shaughnessy RW. Hemolyticdisease of the fe- fetus: Changes in fetal hematocrit associated
with intrauterine death. Obstet Gynecol 1992; 46. Daniels G, FinningK,Martin P, Massey E. Non-
79:390-3. invasive prenatal diagnosis of fetal blood group
36. Mari G, Norton ME, Stone J, et al. Society for phenotypes: Current practice and future pros-
Maternal-FetalMedicine (SMFM)clinicalguide- pects. Prenat Diagn 2009;29:101-7.
line #8: The fetus at risk for anemia-Diagnosis 47. Clausen FB. Integration of noninvasive prenatal
and management. Am J Obstet Gyneco12015; prediction of fetal blood group into clinical pre-
212(6):697. natal care. Prenat Diagn 2014;34:409-15.
37. Lindenburg IT, Smits-Wintjens VE, van Klink 48. Kumpel BM. On the immunologic basis of Rh
JM, et al. Long-term neurodevelopmenta1 out- immune globulin (anti-D)prophylaxis. Transfu-
come after intrauterine transfusion for hemolyt- sion 2006;46: 1652-6.
ic disease of the fetus/newborn: The LOTUS 49. Brinc D, Lazarus AH. Mechanisms of antl-D ac-
study. Am J Obstet Gynecol 2012;206: 141. tion in the prevention of hemolytic disease of
e141-8. the fetus and newborn. HematologyAm Soc He-
38. Margulies M, Voto LS,Mathet E, Marguiies M. matol Educ Program 2009: 185-91.
High-doseintravenous IgG for the treatment of 50. Stowell SR,Arthur CM, Girard-PierceKR,et al.
severe Rhesus alloimmunization. Vox Sang Anti-KEL sera prevents alloimmunization to
1991;61:181-9. transfused KELRBCsin a murine model. Hae-
39. Ruma MS, Moise KJIr, Kim E, et al. Combined matologica2015; 100:e394-7.
plasmapheresis and intravenous immune globu- 51. Sandler SG, RoseffSD, Domen RE,et al. Policies
lin for the treatment of severe maternal red cell and procedures related to testing for weak D
alloimmunlzation. Am J Obstet Gynecol 2007; phenotypes and administration of Rh immune
196:138.e131-6.
globulin: Results and recommendations related
40. Padmanabhan A, Connelly-SmithL, Aqui N, et
to supplemental questions in the Comprehen-
al. Guidelines on the use of therapeutic aphere-
sive TransfusionMedicine survey of the College
sisin clinicalpractice - Evidence-basedapproach
of American Pathologists, Arch Pathol Lab Med
from the Writing Committee of the American
2014;138:620-5.
Society for Apheresis: The eighth special issue.
52. Sandler SG, FlegelWA, Westhoff CM, et al. It's
J Clin Apher 2019;34(3):171-354.
41. American Academy of Pediatrics Subcommittee time to phase in RHD genotyping for patients
on Hyperbilirubinemia. Management of hyper- with a serologicweak D phenotype. Transfusion
bilirubinemia in the newborn infant 35 or more 2015;55:680-9.
weeks of gestation. Pediatrics 2004; 114:297- 53. Pham BN, Roussel M, Peyrard T, et al. Anti-D
316. investigations in individuals expressing weak D
42. Smits-WintjensVE, Walther FJ, Rath ME, et al. Type 1 or weak D Type 2: Allo-or autoantibod-
Intravenous immunoglobulin in neonates with ies? Transfusion2011 ;51:2679-85.
Rhesus hemolytic disease: A randomized con- 54. KackerS, VassalloR, KellerMA, et al. Financial
trolled trial. Pediatrics 2011; 127:680-6. implications of RHD genotyping of pregnant
43. Figueras-Aloy J, Rodriguez-Miguelez JM, women with a serologic weak D phenotype.
Iriondo-SanzM, et al. Intravenous immunoglob- Transfusion2015;55:2095-103.
ulin and necrotizing enterocolitis in newborns 55. Haspel RL,Westhoff CM. How do I manage Rh
with hemolytic disease. Pediatrics 2010; 125: typing in obstetric patients? Transfusion 2015;
139-44. 55:470-4.
44. Christensen RD, Ilstrup SJ, Baer VL, Lambert 56. Sandler SG, Chen LN, Flegel WA. Serological
DK. Increased hemolysis after administering in- weak D phenotypes: A review and guidance for
travenous immunoglobulin to a neonate with interpreting the RhD blood type using the RHD
erythroblastosis fetalis due to Rh hemolytic dis- genotype. BrJ Haematol20 17;179:10-19.
ease. Transfusion2015;55:1365-6. 57. Qureshi H, Massey E, Kirwan D, et al. BCSH
45. ACOG practice bulletin. Prevention of Rh D al- guideline for the use of antt-D immunoglobulin
loimmunization. Number 4, May 1999 (replac- for the prevention of haemolytic disease of the
es educational bulletin Number 147, October fetus and newborn. Transfus Med 2014;24:8-
1990). Clinical management guidelines for 20.
obstetrician-gynecologists.American College of 58. Judd WJ, Luban NLC, Ness PM, et al. Prenatal
Obstetrics and Gynecology. Int J Gynaecol Ob- and perinatal immunohematology: Recommen-
stet 1999;66:63-70. dations for serologic management of the fetus,
C HAP T E R 23 Perinatal Issuesin TransfusionPractice 671
newborn infant, and obstetric patient. Transfu- platelet alloantigens. Transfus Med 2004;14(6):
sion 1990;30: 175-83. 399-408.
59. Kelsey P, Reilly]1, Chapman JF, et al for the 72. Turner ML, BessosH, FaggeT, et al. Prospective
working party of the BCSH Blood Transfusion epidemiologic study of the outcome and cost-
and General HaematologyTask Forces. The esti- effectiveness of antenatal screening to detect
mation of fetomaternal haemorrhage. Transfus neonatal alloimmune thrombocytopenia due to
Med 1999;9:87-92. anti-HPA-la. Transfusion 2005;45: 1945-56.
60. Sandler SG, Delaney M, GottschallJL. Proficien- 73. Williamson LM, Hackett G, Rennie J, et al. The
cy tests reveal the need to improve laboratory natural history of fetomaternal alloimmuniza-
assays for fetomaternal hemorrhage for Rh im- tion to the platelet-specific antigen HPA-l a
munoprophylaxis. Transfusion 2013;53:2098- (PIAl, Zwa) as determined by antenatal screen-
102. ing. Blood 1998;92:2280-7.
61. Chen JC, Davis BH, Wood B, Warzynski MJ. 74. Kamphuis MM, Paridaans N, Porcelijn L, et al.
Multicenter clinical experience with flow cyto- Screening in pregnancy for fetal or neonatal allo-
metric method for fetomaternal hemorrhage de- immune thrombocytopenia: Systematic review.
tection. Cytometry 2002:50:285-90. BJOG2010;117(11): 1335-43.
62. Pham HP, Marques MB, Williams LA3rd. Rhe- 75. Kjeldsen-KraghJ, KillieMK, Tomter G, et al. A
sus Immune Globulin dosing in the obesity epi- screening and intervention program aimed to re-
demic era (letter). Arch Pathol Lab Med 2015; duce mortality and serious morbidity associated
139:1084. with severe neonatal allotmmune thrombocyto-
63. WOOEJ, Kaushal M. Rhesus Immunoglobulin penia. Blood2007;110(3):833-9.
dosage and administration in obese individuals 76. Winklehorst D, Murphy MF, Greinacher A, et
(comment). Arch Pathol LabMed 2017;141:17.
al. Antenatal management in fetal and neonatal
64. Koelewijn JM, de Haas M, Vrijkotte TG, et al.
alloimmune thrombocytopenia: A systematic re-
Risk factors for RhD immunisation despite ante-
view. Blood2017;129(11): 1538-47.
natal and postnatal anti-D prophylaxis. BJOG
77. Tiller H, Kamphuis MM, Flodmark 0, et al. Fe-
2009;116:1307-14.
tal intracranial haemorrhages caused by fetal
65. Callum JL, Waters JH, Shaz BH,et al. The MBB
and neonatal alloimmune thrombocytopenia:
recommendations for the Choosing Wisely cam-
An observational cohort study of 43 cases from
paign of the American Board of Internal Medi-
cine. Transfusion 2014;54:2344. an international multicentre registry. BMJOpen
66. Schonewille H, Klumper FJ, van de Watering 2013;3:3.
LM, et al. High additional maternal red cell allo- 78. BusselJB, Berkowitz RL,Hung C, et al. Intracra-
immunization after Rhesus- and K-matched in- nial hemorrhage in alloimmune thrombocytope-
trauterine intravascular transfusions for hemo- nia: Stratified management to prevent recur-
lytic disease of the fetus. Am J Obstet Gynecol rence in the subsequent affected fetus. Am J
2007;196: 143.e141. Obstet Gynecol201 0;203(2):135e13-14.
67. Schonewille H, Prinsen-Zander KJ, Reijnart M, 79. Winkelhorst D, Kamphuis MM, Steggerda SJ, et
et al. Extended matched intrauterine transfu- al. Perinatal outcome and long-term neurodevel-
sions reduce maternal Duffy, Kidd, and S anti- opment after intracranial haemorrhage due to
body formation. Transfusion 2015;55:2912-19. fetal and neonatal alloimmune thrombocytope-
68. Reese JA, Peck JD, Deschamps DR, et al. Plate- nia. Fetal Diagn Ther 2019;45(3): 184-91.
let counts during pregnancy. N Engl J Med 80. Bussel JB, Zabusky MR, Berkowitz RL, McFar-
2018;379(1):32-43. land JG. Fetal alloimmune thrombocytopenia. N
69. Kelton JG. Idiopathic thrombocytopenic purpu- EngiJ Med 1997;337:22-6.
ra complicating pregnancy. Blood Rev 2002; 81. Le Toriellec E, Chenet C, Kaplan C. Safe fetal
16:43-6. platelet genotyping: New developments. Trans-
70. Davoren A, Curtis BR,Aster RH,McFarland JG. fusion 2013;53: 1755-62.
Human platelet antigen-specific alloantibodies 82. SchefferPG, Ait Soussan A, Verhagen OJ, et al.
implicated in 1162 cases of neonatal alloim- Noninvasive fetal genotyping of human platelet
mune thrombocytopenia. Transfusion 2004;44: antigen-la. BJOG2011; 118:1392-5.
1220-5. 83. Wilson RD, Gagnon A, Audibert F, et al. Prena-
71. Ohto H, Miura S, Ariga H, et al. The natural his- tal diagnosis procedures and techniques to
tory of maternal immunization against foetal obtain a diagnostic fetal specimen or tissue:
Maternal and fetal risks and benefits. J Obstet mune thrombocytopenia (NAIT).Blood 2006;
Gynaecol Can 2015;37(7):656-68. 107(9):3761-3.
84. LakkarajaM, Berkowitz RL,Vinograd CA, et al. 91. Galea P, Patrick MJ, Goel KM. Isoimmune neo-
Omission of fetal sampling in treatment of sub- natal thrombocytopenic purpura. Arch Dis
sequent pregnancies in fetal-neonatal alloim- Child 1981;56(2):112-15.
mune thrombocytopenia. Am J Obstet Gynecol 92. Gill KK, Kelton JG. Management of idiopathic
2016;215(4):471.el-9. thrombocytopenicpurpura in pregnancy. Semin
85. Lieberman L, Greinacher A, Murphy MF, et al. Hemato12000;37:275-89.
Fetal and neonatal alloimmune thrombocytope- 93. Webert KE, Mittal R, Sigouin C, et al. A retro-
nia: Recommendations for evidence-basedprac- spective l l-year analysis of obstetric patients
tice, an international approach. Br J Haematol with idiopathic thrombocytopenic purpura.
2019;185(3):549-62. Blood2003; 102:4306-11.
86. van den Akker ES, Oepkes D, LoprioreE, et al. 94. Neunert C, Lim W, Crowther M, et al. The
Vaginaldeliveryfor fetusesat risk of alloimmune American Society of Hematology 2011
thrombocytopenia? BrJ Obstet Gynaecol 2006; evidence-based practice guideline for immune
113:781-3. thrombocytopenia.Blood2011;117:4190-207.
87. Ghevaert C, Campbell K, Walton J, et al. Man- 95. Provan D, Stasi R, Newland AC, et al. Interna-
agement and outcome of 200 cases of fetoma- tional consensus report on the investigationand
ternal alloimmune thrombocytopenia. Transfu- management of primary immune thrombocyto-
sion 2007;47:901-10. penia. Blood2010;115:68.
88. Chakravorty S, Roberts I. How I manage neona- 96. KoyamaS, Tomimatsu T, KanagawaT, et at. Re-
tal thrombocytopenia. BrJ Haemato12012;156: liable predictors of neonatal immune thrombo-
155-62. cytopenia in pregnant women with idiopathic
89. Allen D, Verjee S, ReesS, et al. Platelet transfu- thrombocytopenic purpura. Am J Hematol
sion in neonatal alloimmune thrombocytopenia. 2012;87:15-21.
Blood2007; 109(1):388-9. 97. Gernsheimer T, James A, Stasis R. How I treat
90. Kiefel V, Bassler D, Kroll H, et al. Antigen- thrombocytopenia in pregnancy. Blood 2013;
positive platelet transfusion in neonatal alloim- 121(1):38-47.
Neonatal and Pediatric
Transfusion Practice
Edward c.c. Wong, MD, and Rowena C. Punzalan, MD
RANSFUSION PRACTICE IN PEDIAT- tem function that necessitate special approaches

ric patients, particularly in neonates, dif- to component therapy. This is especially import-
fers from that in adults. 1 The differences ant for VLBW infants (<1500 g) and extremely
are related to physiologic changes occurring low-birthweight infants « 1000 g). The mean
during the transition from fetus to adolescent, cord blood hemoglobin. level of healthy neo-
blood volume, development of hematopoiesisand nates is 16.9 ± 1.6 g/dL, whereas that of
coagulation, immune system maturity, and physt- preterm neonates is 15.9 ± 2.4 g/dL. The he-
ologic response to hypovolemia and hypoxia that moglobin concentration normally declines
are variable in this heterogeneous population. All during the first few weeks of life, resulting in
of these factors contribute to the complexity and physiologicanemia of infancy, a condition that is
intricacies of pediatric transfusion practice. Ad- self-limited and usually tolerated without harm-
vances in neonatology now permit the .survival of ful effects in term infants but is potentially more
extremely premature infants, and most neonatal worrisome in preterm infants.3
transfusions are given to very low-birthweight The rate of decline in hemoglobin levels is a
(VLBW)infants.' This chapter discusses neonatal function of gestational age at birth. At 4 to 8
and pediatric transfusion practice during two dis- weeks after birth, hemoglobin decreases .to .as
tinct periods: the newborn period (birth to 4 low as 8.0 g/dL in preterm infants weighing
montl:ls),..and infancy..(after 4 months) through 1000 to 1500 g, and 7.0 g/dL in neonates
childhood. Transfusion practices relevant to spe- weighing <1000 gat birth." The physiologic de-
cial pediatric populations are also described. crease in hemoglobin concentration is caused by
several factors: 1) a decrease in erythropoietin
(EPO)that ismore prolonged in preterm infants,
resulting in diminished red cell production; 2)
N
decreased survival of fetal red cells; and 3) .in-
creasing blood volume due to rapid growth. Re-
duced EPO production results from increased
Neonates
oxygen delivery to tissues because of increased
Patients younger than 4 months have small pulmonary blood flow, elevated arterial pOz
blood/plasma volumes and immature organ sys- (partial pressure of oxygen) levels, increased red
Edward C.C. Wong, MD, Medical Director, Coagulation, Quest Diagnostics Nichols Institute, Chantilly, Virginia
and Adjunct Associate Professor, Pediatrics and Pathology, George Washington School of Medicine and Health
Sciences, and formerly Associate Director of Transfusion Medicine, Children's National Health System, Washing-
ton, District of Columbia; and Rowena C. Punzalan, MD, Medical Director, Transfusion Service, Versiti Blood
Center of Wisconsin and Children's Hospital of Wisconstn, and Associate Professor, Pediatrics (Hematology),
Medical College of Wisconsin, Milwaukee, Wisconsin
613
cel1 2,3-diphosphoglycerate (2,3-DPG), and in- interaction with vWF, and increased platelet gly-
creased hemoglobin A levels." coprotein Ib« (GPlba) expression compared to
Platelet counts in term and preterm neonates platelets from full-term infants. Despite this in-
are similar to those in children and adults. creased interaction, no significant differences
Thrombopoietin is found in fetal liver as early as were observed in the number of platelets that
6 weeks of gestation. Nevertheless, for various adhered in a stationary fashion to vWF, compar-
reasons, thrombocytopenia (platelet count ing platelets from these neonatal populations. 10
<150,000/j..lL) occurs in up to 25% of admis- Whether this reflects a compensatory mecha-
sions to neonatal intensive care units (ICUs).6 nism to a higher risk of bleeding in VLBW
In neonates, low levels of vitamin-Kdependent preterm infants or contributes to the higher risk
factors (Factors II, VII, IX, and X) and contact in VLBW preterm infants is unclear. A study of
factors (Factor XI, Factor XII, prekallikrein, and premature infants aged 33 weeks or less, having
high-molecular-weight kininogen) contribute to platelet counts < 100 ,000/].1L, showed that
altered coagulation test results." (See Table 24- those with high bleeding scores had higher ade-
1.) The naturally occurring anticoagulants (pro- nosine diphosphate (ADP) closure times using
teins C and S and antithrombin) are also de- the platelet function analyzer (PFA)-l00.11 How-
creased." In spite of these issues, the procoagu- ever, given the large amount of blood needed for
lant and anticoagulant systems are usually in testing, it remains to be seen whether this ap-
balance in healthy newborns, so spontaneous proach wil1 be acceptable to neonatology spe-
bleeding and thrombosis are rare." However, the cialists to guide platelet transfusions in patients
reserve capacity for both systems is limited. with high ADP closure times. Regardless of
Platelet function is likely different between these findings, serious bleeding and, less com-
VLBWpreterm and full-term infants. An in-vitro monly, thrombosis may occur in sick premature
arterial shear model using surface immobilized infants during the first week of life.
von Willebrand factor (vWF) found that plate- Preterm infants and neonates have underde-
lets from VLBW preterm infants had increased veloped immune function. It is not wel1 under-
TABLE 24-1. Screening Laboratory Tests for Hemostasis: Neonates vs Adults*
Preterm Neonates vs Neonates vs Older Approximate Age Adult

Full-Term Neonates Children/Adults Values Are Reachedt
aPTT Longer Longer 16 years
Prothrombin time Longer Same or longer 16 years
Thrombin time Longer Same or longer 5 years
Bleeding time Longer* Shorter 1 month
PFA-100 Longer* Shorter 1 month
ROTEM/TEG
Clotting time Same Shorter 3 months
Clot formation time Same Shorter 3 months
Maximal clot firmness Stronger Stronger 3 months
'Modified with permission from Revel-Vilk.9
tMaximum agereported.
tin samplesdrawn in the first 7 to 10 days of life.
apn = activated partial thromboplastin time; PFA = platelet function analyzer;ROTEM= rotational thromboelastometry;
TEG= thromboelastography(both ROTEMandTEGare viscoelastic methodsfor hemostasistesting in whole blood).
C HAP T E R 24 Neonatal and Pediatric TransfusionPractice 675
stood why unexpected red cell alloantibodies of waste, blood banks must be capable of providing
either immunoglobulinM (IgM)or IgG class are appropriately sized blood components." The
rarely produced during this time. Possibleexpla- vast majority of preterm infants weighing <1.0 kg
nations include deficient I'helper-cell function, or at <28 weeks' gestational age.need at least
enhanced I'suppressor-cell activity, and poor one transfusion."
antigen-presenting-cell function." Cellular im- Many factors, including iatrogenic blood loss
mune responses are also incompletely devel- from repeated phlebotomy, can lead to frequent
oped during this period and may make infants transfusions in sick newborns. Hypovolemia
susceptible to transfusion-associated graft-vs- (> I 0% blood volume loss) is not tolerated well
host disease (TA-GVHD). in newborns because of their limited ability to
Infants younger than 4 months of age have compensate with increased heart rate. In such
decreased ability for liver metabolism of the an- situations, the decreased cardiac output (along
ticoagulant citrate, making them vulnerable to with increased peripheral vascular resistance to
acidosis and/or hypocalcemia. Immature kid- maintain blood pressure) ultimately results in
neys also have lower glomerular filtration rates poor tissue perfusion and oxygenation, and met-
and concentrating ability than in older children, abolic acidosis.18As described later in more de-
leading to difficultiesin excreting excess potassi- tail, Red Blood Cell (RBC)transfusions are often
um, acid, and/or calcium. In addition, very administered to maintain a target hemoglobin
young infants <3 days old have blunted parathy- level in certain clinical situations and for symp-
roid hormone secretion in response to decreased tomatic anemia.2,19,20
ionized calcium, especiallyresulting from citrate
exposure during whole blood exchange." Erythropoietin Physiology and Therapy
In contrast to older children and adults, new-
Children Older Than 4 Months
borns have decreased EPO production in re-
Hemoglobin continues to increase after the neo- sponse to hypoxia, likely to prevent polycythe-
natal period until typical adult levels are mia in the hypoxic intrauterine environment.
achieved after adolescence. Levels of most pro- Most premature infants produce the smallest
coagulant and anticoagulant factors are not sig- amount of EPO expected for any degree of ane-
nificantly different from those in adults by 6 mia.21;22The·.reduction in EPO production is
months of age." greatest inpreterm infants because of persistent
liver-based EPO production, which is normally
regulated in the more hypoxic in-utero state.
R NIN Kidney-basedEPO production, which is regulat-
NE ed at higher p02 levels, does not normally occur
until after term delivery.'
This section reviews several general aspects of As an alternative to transfusion, the early ad-
transfusion practice to highlight situations that ministration of erythropoiesis-stimulatingagents
warrant special.attention in caring for the neo- (ESAs)such as recombinant human EPO or the
nate. In addition to the issues listed in the first longer-actingdarbepoeltln have been used to de-
section below, sections that follow address indi- crease transfusionsin at-riskneonates. However,
cations, thresholds, exchange transfusion, and a recent meta-analysis (3643 infants in 34 stud-
conditions specificto neonates. ies) showed that early administration of ESA(vs
no treatment or placebo). in preterm or low,
Considerations for Transfusion birthweight infants minimally reduced the total
amount of transfusion but did not reduce the
Body Sizeand Blood Volume
number of donor exposures. There was de-
Full-term newborns have a blood volume of ap- creased incidence of necrotizing enterocolitis
proximately 85 mLlkg; in contrast, preterm [NEC:relative risk, 0.62; 95% confidence inter-
newborns have about 100 mLlkg. IS To avoid val (CI), 0.48-0.801 and a trend toward
improved neurodevelopmental outcome in the can be used from a single RBCunit is increased,
ESAgroup." In contrast, earlier studies showed which may reduce the overall donor exposure
possible increased severity of retinopathy of pre- to a patient.
maturity" and increased incidence of infantile Luban and colleagues used theoretical calcu-
hemangiomas" with use of ESAsin preterm and lations in a variety of clinical settings to demon-
low-blrthwelght infants. As compliance with strate that red cells preserved in extended stor-
strict transfusion threshold criteria has image media present no major risk when used for
proved, clinicians have decreased phlebotomy small-volume transfusions.F Prospective ran-
rates and volumes and used point-of-care testing domized controlled trials to assess the outcomes
in VLBWinfants, resulting in a decrease in the of longer vs shorter storage times of RBCshave
rates of iatrogenic anemia and need for transfu- been performed in this population and found
sions.26,27Thus, in most cases, this approach that small-volume transfusion comparing older
combined with the use of aliquots from a single- AS-1 or AS-3 units to fresher CPDA units were
donor blood component unit for multiple trans- equivalent in terms of safety and efficacy.30,31
fusions achieves the same goal (ie, decreases (Seenext section.)
number of transfusions and donor exposures) Because it is unknown whether these theo-
without the need for EPO therapy. retical concerns for AS are significant for pa-
tients with renal or hepatic insufficiency, some
Cold Stress facilities may remove the AS from RBC units,
Hypothermia in the neonate can trigger or exag- particularly if multiple transfusions from the
gerate several responses, including 1) increased same unit are expected; however, this is techni-
metabolic rate, 2) hypoglycemia, 3) metabolic cally challenging and not possible in many facili-
acidosis, and 4) potential apneic events that may ties. The safety of AS-preserved RBCsin trauma-
lead to hypoxia, hypotension, and cardiac ar- related massive transfusions, extracorporeal
rest." In-line blood warmers are recommended membrane oxygenation (ECMO), cardiac sur-
for large-volume transfusions, including red cell gery, or exchange transfusions has not been
exchange transfusions, to combat the effects of studied. In a recent hospital survey in the Unit-
hypothermia. A radiant heater should never be ed States about large-volume transfusions in ne-
used to warm the blood being transfused be- onates, 43% of responders used RBC units
cause of the risk of hemolysis. Furthermore, to stored in AS-3, 29% used RBCs stored in AS-1,
prevent hemolysis in neonates undergoing and 28% used RBCsstored in CPD or CPDA.33
phototherapy, the blood-administration tubing AS-preserved RBC units should be used with
should be positioned to minimize exposure to caution in settings where large volumes are be-
phototherapy light." ing transfused.32,34,35
Ionized calcium and potassium levels should
RBCAdditive Solutions be monitored frequently during large-volume
transfusions. A blood warmer should be used to
Historically, RBCs transfused to children con- avoid hypothermia." These principles should be
tained only citrate-phosphate-dextrose-adenine applied to infants and small children as well.
(CPDA)-1anticoagulant-preservativesolution.30,31
However, as the use of additive solutions (AS) RBCAge and the Storage Lesion
containing adenine and mannitol evolved to ex-
tend the shelf life of RBCs,many experts began Small-volume, simple transfusions administered
to question their safety in neonates. One con- slowly have been shown to have little effect on
cern is the dose of adenine in AS and its relation serum potassium concentrations in infants
to renal toxicity. Mannitol is a potent diuretic younger than 4 months despite the high potassi-
with effects on fluid dynamics that can result in um levels in the supernatant of stored RBCs.In
fluctuations in the cerebral blood flow of calculating levels of infused potassium, Strauss
preterm infants. However, because the use ofAS mathematically determined that transfusion of
extends shelf life, the number of aliquots that an aliquot from an RBCunit that is sedimented
by gravity (80% hematocrit) and stored in an ex- qumng red cell exchange and large-volume
tended storage medium for 42 days would deliv- transfusion is that the RBC units selected should
er 2 mL of plasma contalning only 0.1 mmoliL be <14 days old, although this practice is varl-
of potassium when transfused at 10 mLlkg.37,38 able and dependent-on institutional policy.
This amount of potassium is much less than the The impact of RBC age and the. "storage .le-
dally requirement of 2 to 3 mmoliL for a patient sion" is seen in in-vitro and retrospective stud-
weighing 1 kg. It must be stressed that this ies, but clinical confirmation has not been found
calculation does not apply to the transfusion of in prospective studies. A randomized controlled
large volumes of RBCs (>20 mLlkg), such as trial conducted in Canada, Age of Red Blood
during surgery, exchange transfusion, or Cells in Premature Infants (ARIPI),randomly as-
ECMO.39,4o signed -Iow-birthweight infants to receive RBCs
The type of anticoagulant-preservative solu- that were 7 days old (mean = 5.1 days; n = 188)
tion used to store RBCsat collection determines or standard-issue RBCsdivided into aliquots and
the amount of potassium delivered.F'" In addl- stored for 2 to 42 days (mean = 14.6 days; n =
tion, special component processing, such as irra- 189).47 The primary composite endpoints in-
diation, can potentiate potassium leakage from cluded NEC, intraventricular hemorrhage
the red cell membrane. If irradiated components (IVH), and bronchopulmonary dysplasia. The
are stored for more than 24 hours, washing may study found no differences in the primary end-
be required to remove the excess potassium be- points between infants in the two arms, suggest-
fore transfusion to vulnerable patients." There ing that in the study population, the age of RBCs
are reports of severe adverse effects, including does not affect these common morbidities of
cardiac arrest and death, in infants who reo prematurity. ARIPI's generalizability has been
ceived either older RBC units or those that had questioned because of the study's liberal transfu-
been irradiated (>1 day before transfusion) via sion strategy (hemoglobin threshold of 10 gI dL),
central or intracardiac line.43,44Alternatively, use of SAGM (saline, adenine, glucose, and
and more practically, several institutions accept mannitol) units, and average duration of blood
AS units for low-volume neonatal transfusions storage. These practices do not reflect the trans-
without washing as long as these units do not fusion practices, storage solution, and age of
exceed a certaln storage age or time after lrradi- RBCs used in many centers in the United
ation.45. States."
Levels of2,3·DPG in red cells are known to
decline rapidly after 1 to 2 weeks of storage. Compatibility Testing
Older children and adult recipients are able to
replenish the missing 2,3·DPG in vivo and corn- MBB Standards jar Blood Banks anqJraRsfii·
pensate for hypoxia by increasing heart rate. In- sian Services (Standards) allows pretransfusion
fants younger than 4 months are not able to serologic testing for infants younger than 4
compensate as effectively because of low levels months to be limited.49(pp39-40)Initial patient test-
of intracellular 2,3·DPG that further decrease ing must include ABO and D typing of the pa-
with respiratory distress syndrome or septic tient's red cells and screening for unexpected
shock. Thus, if a large proportion of the neo- red cell antibodies using either plasma or serum
nate's blood volume is composed of transfused, from the infant or mother. During any hospital-
2,3·DPG·depleted blood, the resulting shift in ization, crossmatch·compatibility testing and reo
the hemoglobin oxygen dissociation curve fur- peat ABO and D typing need not be conducted
ther increases oxygen affinity for hemoglobin as long as all of the following criteria are met: 1)
and reduces oxygen availability to the tissues." the antibody screening result is negative; 2)
However, this shift in the hemoglobin dissocia- transfused RBCs are group 0, ABO identical, or
tion curve might be minimized by opposing ABO compatible; and 3) transfused cells are ei-
shifts to the curve from decreased pH and in- ther D negative or the same D type as the pa-
creased pCOz that is associated with hypoxia. tient. Testing the infant's reverse type for anti-A
The usual recommendation for newborns reo and/or anti-B is not necessary. However, before
non-group-O RBCs can be issued, testing of the nates when approximately 10% of blood volume
infant's plasma or serum is required to detect has been lost or they have symptomatic anemia.
passively acquired maternal anti-A or antl-B and Several guidelines have been published over
should include an antiglobulin phase. If ABO an- the past 15 years regarding the indications for
tibody is present, maternally ABO-compatible RBCtransfusions in neonates. 19,20,36,53,54 Most of
RBCs are transfused until the acquired antibody the recommendations are based on experience
is no longer detected. acquired in clinical practice rather than pub-
If an unexpected non-ABO alloantibody is de- lished evidence. To this end, a critical need ex-
tected in the infant's or mother's specimen, RBC ists for clinical studies in this area. Table 24-2
units provided for the infant must lack the corre- lists the indications from one of these guide-
sponding antigen(s) or be compatible by anti- lines. 19
globulin crossmatch. This regimen should con- When 10 mLlkg of RBCswith a hematocrit
tinue until the maternal antibody is no longer of >80% (achieved by allowing an RBC unit to
detected in the infant's plasma or serum. sediment by gravity undisturbed with aliquot
The policy of the hospital transfusion service de- ports on the bottom) is transfused, the expected
termines the frequency for reevaluating the pa- increase in hemoglobin concentration in a neo-
tient's antibodies. Once a negative antibody nate is approximately 3 gI dL. A similar volume
screening result is obtained, crossmatches and of RBCs with AS usually has a hematocrit of
use of antigen-negative blood are no longer re- 65%, and its transfusion results in a projected
quired in infants younger than 4 months be- posttransfusion hemoglobin increase of 2 g! dL.
cause of their immature immunologic status. (SeeTable 24-3 for blood component dosing rec-
Multiple observational studies have shown ommendations and the expected results. 55)
that alloimmunization to red cell antigens is rare Two randomized controlled trials have com-
during the neonatal period.12,50,51 For this rea- pared the outcomes of restrictive (hemoglobin =
son, repeated typing and screening, which is re- 7 gldL) vs liberal (hemoglobin = 10 g/dl.) RBC
quired for adults and children older than 4 transfusion thresholds in VLBW infants. The
months and contributes to significant iatrogenic Iowa trial revealed a lower rate of transfusion
blood loss, is unnecessary in younger infants. Al- events (3.3 vs 5.2; P = 0.025) with a restrictive
so, the transfusion service should avoid transfus- compared to a liberal strategy.56 However, rates
ing any components that may passively transfer of periventricular leukomalacia and death were
high-titer alloantibodies to recipients.F higher in the restrictive arm. The Premature In-
fants in Need of Transfusion (PINT) study from
Indications and Thresholds for Canada found no significant difference between
Transfusion the two arms, which had similar thresholds as
the Iowa study, in the composite endpoint of
Sick neonates are more likely to receive RBC death or bronchopulmonary dysplasia, retinopa-
transfusions than any other patient age group, thy of prematurity (stage >3), or brain injury
and RBCs are the component most often trans- (periventricular leukomalacia, intracranial hem-
fused during the neonatal period.' Symptoms of orrhage Grade 4, or ventrlculomegalv)." The
anemia in neonates include tachycardia and/or follow-up of the PINT study revealed that at 18
tachypnea, increased episodes of bradycardia to 24 months after birth, infants in the restric-
and/or apnea, poor weight gain despite ade- tive arm had more cognitive delay than those in
quate feeding, increased oxygen requirement, the liberal arm." On the other hand, follow-up
and increased blood lactate. Because of the poor studies on the patients in the Iowa trial showed
specificity of the signs and symptoms of anemia, decreased brain structure and function at ages 7
prophylaxis for anemia is a major reason for sim- to 10 years in the liberal arm, suggesting an op-
ple transfusion. Specifically, a venous hemoglo- posite effect in long-term neurocognitive out-
bin < 13 g! dL in the first 24 hours of life necessi- comes.59,60 However, these follow-up studies
tates consideration of RBC transfusion.19,20,36 were post-hoc analyses and not powered to de-
Red cell replacement is considered for sick neo- tect differences.
C HAP T E R 2 4 Neonatal and Pediatric TransfusionPractice 679
TABLE24·2. Transfusion Guidelines for RBCsin Infants Younger than 4 Months 19

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.~
1. Hematocrit <20% with low reticulocyte count and symptomatic anemia (tachycardia,tachypnea,poor
feeding).
2. Hematocrit <30% and any of the following:
a. On <35% oxygen hood.
b. Onoxygen by nasalcannula.
c. On continuous positive airway pressureand/or .intermittent mandatory ventilation or mechanical
ventilation with meanairway pressure<6 cm of water.
d. With significant tachycardia or tachypnea (heart rate> 180 beats/minutefor 24 hours, respiratory
rate>80 beats/minutefor 24 hours).
e. With significant apneaor bradycardia(>6 episodes in 12 hours or 2 episodes in 24 hours requiring
bagand mask ventilation while receivingtherapeutic doses of methylxanthines).
f. With low weight gain «10 g/day observedover 4 dayswhile receiving<:100kcal/kg/day).
3. Hematocrit <35% and either of the following:
a. On>35% oxygen hood.
b. Oncontinuous positive airway pressure/intermittent mandatoryventilation with mean airway pres-
sure <:6-8 cm of water.
4. Hematocrit <45% and either of the following:
a. On extracorporeal membraneoxygenation.
b. With congenital cyanotic heart disease.
A recent meta-analysisof 16 randomized and among neonates with preoperative hematocrit

45 nonrandomized studies in neonates showed <40% was 7.5%, compared to 1.4% among those
no significant difference between liberal and re- with hematocrit :::::40%.62
strictive transfusion strategies in terms. of mor- There are two ongoing studies looking at
tality, NEC, retinopathy of prematurity, chronic transfusion thresholds in premature and low-
lung disease, and intraventricular. hemorrhage, birthweight infants. The Transfusion of Prema-
although many studies had a high risk of bias." ture Infants Study (TOPS), currently being
Therefore, there is still no definitive answer to conducted in the United States, is examining
the question of appropriate threshold for pro- whether death and neurodevelopmental out-
phylactic RBCtransfusion in neonates. come at 22 to 26 months corrected age is less
Despite the lack of differencebetween liberal common among premature infants who received
and restrictive transfusion thresholds in out- transfusion at high hemoglobin thresholds.f In
comes for neonates in general,neonates undergo- Europe, the ETTNO trial will be randomly as-
ing surgery may have different risk profiles. In a signing 920 VLBWinfants to receive RBCtrans-
recent analysis of the pediatric database of the fusions based on· a restrictive or liberal
American College of Surgeons national quality threshold, and evaluating death rates and neuro-
program, the prevalence of in-hospital mortality developmental outcome at 24 months."
TABLE 24-3. Blood Components and Dosing of Small Volumes in Neonataland Pediatric Patients"
Red Blood Cells 10-15 mUkg Hemoglobin increase:2-3 g/dL *

FreshFrozenPlasma 10-15 mL/kg 15%-20% rise in factor levels (assuming 100%
recovery)
Platelets[whole-blood-derived 5-10 mUkg or 50,000/1-lLrise in platelet count (assuming 100%
(WBO)or apheresis] 1 WBO unitl10 kg recovery)!
(patients ~10 kg)
CryoprecipitatedAHF 1-2 units/10 kg 60-100 mg/dL rise in fibrinogen (assuming 100%
recovery)
*Oependenton anticoagulant-preservativesolution, with 3 g/dL incrementfor CPOand CPOA-1,and 2 g/dL for AS-1, AS-3,
AS-5, AS-?, and SAGM.
tAssumes ~5.5 x 1Q10 platelets in 50 mL of plasma (WBO) and 3.0 x 1011 platelets in 250 to 300 mL plasma(apheresis).
AHF = antihemophilic factor; CPO = citrate-phosphate-dextrose; CPOA-1 = cltrate-phosphate-dextrose-adenlne-t: AS =
additive solution; SAGM = saline-adenine-glucose-mannitol.
Neonatal Exchange Transfusion Two critical objectives of exchange transfu-

sion are the removal of unconjugated bilirubin
The most common indication for exchange
and maximization of albumin-binding of residu-
transfusion in neonates is hyperbilirubinemia
al bilirubin. In addition, in antibody-mediated
caused by hemolytic disease of the fetus and
hemolytic processes, exchange transfusion re-
newborn, discussed in detail in Chapter 23. Oc- moves both free antibody and antibody-coated
casionally, it is used to eliminate toxins, drugs, red cells, replacing these with antigen-negative
or chemicals administered to the mother near red cells.
the time of delivery. Exchange transfusion is also Exchange transfusion needs to be performed
used when toxic doses of drugs have been ad- before the development of kernicterus. In full-
ministered to the infant or accumulate at high term infants, kernicterus rarely develops at
levels as a result of prematurity and/or an in- bilirubin levels <25 mg/dl., However, in sick
born error of metabolism.v'" VLBWinfants, kernicterus can occur at bilirubin
levels as low as 8 to 12 mgldL.68
Physiology A double-volume exchange transfusion (two
In neonates, excessively high levels of unconju- 85-mLlkg transfusions for full-term infants and
gated bilirubin may cross the blood-brainbarrier, two 100-mLlkg transfusions for VLBWinfants)
concentrate in the basal ganglia and cerebellum removes approximately 70% to 90% of circulat-
and cause irreversible damage, known as kernic- ing red cells and approximately 50% of total bill-
terus. Infants are susceptible to hyperbilirubin- rubin." After the first exchange transfusion, bili-
emia because their immature liver conjugates rubin may reequilibrate between extravascular
bilirubin poorly and because their incompletely tissue and plasma, which may necessitate anoth-
developed blood-brain barrier allows bilirubin er exchange transfusion.
transit. Phototherapy (blue/green light in the
Component Choice
range of 460-490 nm, which converts unconju-
gated bilirubin into a water-soluble isomer, aid- Typically, RBCs are resuspended in ABO-
ing excretion) is the current treatment of choice compatible, thawed Fresh Frozen Plasma (FFP)
for hyperbilirubinemia; exchange transfusion is for an exchange transfusion. No single method
reserved for patients who fail phototherapy." of combining components has been shown to be
superior to another. RBCs<5 to 7 days old and fants just after birth. If umbilical venous cathe-
stored in CPDA-1have been used to avoid high ters are not available, small saphenous catheters
levels of potassium and to maximize red cell sur- may be used. Two exchange transfusion tech-
vtval." When using AS-RBCunits, some blood niques are commonly employed: isovolumetric
banks elect to remove the additive-containing and manual push-pull. In isovolumetric ex-
plasma to reduce the volume transfused. change transfusion, two catheters of identical
Most transfusion services provide RBCunits size provide.vascular access. The catheters al-
that are hemoglobin S negative, cytomegalovi- low simultaneous withdrawal and infusion of
rus (CMV) reduced-risk (leukocyte reduced or blood and are regulated by a single peristaltic
CMV seronegative), and irradiated. Irradiation pump. The umbilicalvein is typicallyused for in-
should be performed just before the exchange to fusion, and the umbilical artery is used for with-
prevent potentiation of the potassium storage le- drawal. The manual push-pull technique uses a
sion. If units are not irradiated immediately be- single vascular access portal witha three-way
fore use, some experts recommend removing stopcock attached to the unit of blood, the pa-
the supernatant of red cells or washing the unit tient, and a graduated discard container. A stan-
to avoid the complications of hyperkalemic car- dard filter and in-line blood warmer are recom-
diac arrhythmias." mended.
The glucose load administered during ex- With both techniques, the absolute maxi-
change transfusion can be high in some cases, mum volume of each withdrawal and infusion
which stimulates the infant's pancreas to release depends on the infant's body weight and hemo-
insulin and results in rebound hypoglycemia." dynamic status. Usually, no more than 5 mL/kg
Therefore, infant plasma glucose levels should body weight or 5% of the infant's blood volume
be monitored during the first few hours follow- is removed and replaced during a 3- to 5-minute
ing exchange transfusion. Because resuspended cycle." The exchange transfusion should not be
RBCsdo not include platelets, the platelet count performed rapidly because sudden hemodynam-
should be assessed after completion of the ex- ic changes may affect cerebral blood flow and
change transfusion. shift intracranial pressure, contributing to IVH.73
A double-volume exchange transfusion in ne- A total double-volume exchange transfusion typ-
onates rarely necessitates the infusion of >1 icallytakes 90 to 120 minutes."
RBC unit. The unit's hematocrit should be ap-
proximately 45% to 60%, and the unit should Conditions Specific::to Neonates
have sufficient plasma (based on estimated Hemolytic disease of the fetus and newborn and
blood volume) to provide clotting factors." If neonatal alloimmune thrombocytopenia are dis-
the neonate's condition requires a higher postex- cussed in Chapter 23.
change transfusion hematocrit, a small-volume
RBC transfusion may be given or a unit with a Polycythemia
higher hematocrit can be used for the initial ex-
change transfusion. The reconstituted blood Neonatal polycythemia is defined as venous he-
should be well mixed to sustain the intended matocrit >65% or hemoglobin >22 g/dL at any
hematocrit throughout the exchange. time during the first week after birth. Approxi-
Postexchange transfusion hematocrit, plate- mately 5% of all newborns develop polycythe-
let count, and bilirubin should be measured. mia for a variety of reasons, most commonly in-
This can be performed by using the last portion trauterine growth restriction and gestational
of blood removed for the exchange. Complica- diabetes. Once the hematocrit rises above 65%,
tions have been reviewed in detail elsewhere.70 viscosity increases and oxygen transport de-
creases. However, in neonates, the exponential
rise in viscosity can occur at a hematocrit as low
Vascular Access and Technique
as 40%.74Congestive heart failure can result be-
Umbilical venous catheters are used for ex- cause infants have limited capability to increase
change transfusions in preterm and full-term in- their cardiac output to compensate for hypervis-
cosity. Central nervous system abnormalities, caused by decreased red cell mass, typically be-
pulmonary and renal failure, and NEC can occur cause of surgery, anemia from underlying chron-
from the resultant decreased blood flow. ic disease, or hematologic malignancies. Chron-
A partial exchange is used to normalize the ic RBC transfusions are administered to combat
hematocrit to between 55% and 60% and im- tissue hypoxia and suppress endogenous hemo-
prove tissue perfusion while maintaining blood globin production in children with hemoglob-
volume. This exchange is accomplished by re- inopathies. Table 24-4 can help guide transfu-
moving whole blood and replacing it with nor- sion decisions in patients older than 4 months.
mal saline or other crystalloid solutions. Plasma Before receiving any RBCtransfusion, all pe-
is not used to replace whole blood because NEC diatric patients older than 4 months require
has been reported as a complication of plasma ABO and Rh testing and screening for the pres-
transfusion. 75 ence of clinically Significant antibodies in the
The formula below can be used to approxi- same frequency as adult patients. Compatibility
mate the volume of replacement fluid required testing should be performed according to MBB
and the volume of whole blood that must be Standards_49(40)
withdrawn for the partial exchange:
Special Populations
Volume of replacement fluid =
Sickle Cell Disease
Blood volume x (Observed Hct - Desired Hct)
Observed Hct Chronic transfusion in patients with sickle cell
disease (SCD) decreases the proportion of red
Necrotizing Enterocolitis cells containing hemoglobin S, reduces sickling,
and prevents blood viscosity increase. The main
NEC, a serious condition in neonates, is charac- indication for chronic transfusion is primary and
terized by ischemic necrosis of the intestinal secondary prevention of stroke. It has been
mucosa, associated with inflammation, invasion shown that the risk of recurrent stroke is re-
of enteric-gas-forming organisms, and dissection duced to < 10% if hemoglobin levels are main-
of gas into the abdominal cavity. Small studies tained between 8 and 9 gldL and the percent-
have suggested the possibility of RBC transfu- age of hemoglobin S stays <30%.78,79 In children
sion as an independent risk factor for develop- with SCD and silent cerebral infarcts (SCIs),
ment of NEC in neonates, and a previous meta- transfusion therapy is more effective at prevent-
analysis showed an increased risk of NEC (odds ing stroke and SCI recurrence than hydroxyurea
ratio = 2) in neonatal transfusion recipients." and phlebotomy" or observation.t'r"
However, a recent, prospective, multicenter, ob- Simple or partial manual exchange transfu-
servational cohort study of VLBWinfants found sions can be administered every 3 to 4 weeks.
that severe anemia itself, rather than transfu- Automated red cell exchange has also been used
sion, was independently associated with NEC.77 to prevent iron overload in patients with SCD
who require chronic transfusion.f See Table 24-
4 for a list of other indications for simple or
RBC TRANSFUSION fiN chronic RBC transfusion in SCD. RBCs for pa-
INfANTS LDER THAN 4 tients with SCD should be screened for hemo-
MONTH N CHILDREN globin S and, ideally, leukocyte reduced to pre-
vent HLA alloimmunization and platelet
The most significant differences between RBC refractoriness in preparation for possible hema-
transfusions in infants older than 4 months and topoietic progenitor cell transplantation.
adults are 1) blood volume, 2) the ability to tol- Red Cell Alloimmunization in SCD. Al-
erate blood loss, and 3) age-appropriate hemo- though the benefits of chronic transfusion thera-
globin and hematocrit levels. In this age group, py in patients with SCD have been demonstrat-
the most common indication for RBC transfu- ed, it is not without risks. Patients with SeD have
sion is treatment or prevention of tissue hypoxia the highest rates of red cell alloimmunization of
TABLE 24·4. Transfusion Guidelines for RBCsin Patients Older than 4 Months 19,20
1. Emergencysurgical procedure in patient with significant postoperativeanemia.
2. Preoperativeanemiawhen other corrective therapy is not available.
3. Intraoperativeblood loss> 15% total blood volume.
4. Hematocrit <24% and:
a. In perioperativeperiod, with signs and symptoms of anemia.
b. While on chemotherapy/radiotherapy.
c. Chronic congenital or acquired symptomatic anemia.
5. Acute blood loss with hypovolemia not responsiveto other therapy.
6. Hematocrit <40% and:
a. With severepulmonary disease.
b. On extracorporealmembraneoxygenation.
7. Sickle cell diseaseand:
a. Cerebrovascularaccident.
b. Acute chest syndrome.
c. Splenic sequestration.
d. Aplastic crisis.
e. Recurrentpriapism.
f. Preoperativelywhen generalanesthesiais planned(target hemoglobin = 10 g/dL).
8. Chronictransfusion programs for disorders of red cell production (eg, p-thalassemia major and Dia-
mond-Blackfansyndrome unresponsiveto therapy).
any patient groUp.84,85 Antibodies are produced units are considerably more costly" and may be
against common RH, KEL,FY,and JKsystem an- difficultto obtain."
tigens. Many SCD treatment centers perform an In academic institutions in the United States
extended red cell phenotype or genotype analy- and Canada, the most common protocol for pre-
sis of a patient's red cells before beginning trans- vention of red cell alloimmunization in nonallo-
fusion therapy, and preferentially transfuse red immunized patients with SCD is transfusion of
cell antigen-matched (RH and KEL)units to re- ABO/Rh-compatible units that are additionally
duce the rate of red cell alloimmiinization.w" matched for C, E, and K antigens." Once pa-
tients have developed a red cell antibody, exten-
Red cell genotyping can identify extended phe-
sion of matching to additional red cell antigens
notype, provide information about variant Rh al- (FtIFyh, Jk"IJlch, S) is often used to prevent fur-
leles that may appear phenotypically negative ther allolrnmunlzatlon."
but form antibodies against Rh antigens, and Other Complications of RBC Transfu-
provide results about common .silencing muta- sions in SCD. SCD transfusion recipients also
ttons." However, particularly for patients who face risks such as iron overload (which can
are not yet alloimmunized, this process of cause hepatic and cardiac dysfunctionj'" and the
matching for red cell antigens is not uniformly risks of increased donor exposure during red cell
practiced because phenotypically compatible exchange. In addition, patients with SCD may
be at risk of life-threatening, delayed hemolytic expert panel made a recommendation to pro-

transfusion reactions. If a patient's hemoglobin vide RBCsmatched for C, c, E, e, and K to thal-
level decreases after transfusion, the patient assemia patients without alloantibodies, and
might have developed "hyperhemolytic" syn- RBCsmatched for C, c, E, e, K, Ft, Fyb,]ka, ]kb,
drome. This poorly understood phenomenon is S, and s to those with one or more alloantibod-
characterized by destruction of the patient's ies; however, evidence to support their recom-
own red cells along with transfused cells and mendation was weak. 105
mayor may not display allo- or autoantibodies.
If hyperhemolytic syndrome is suspected, case AgeofRBCs
reports suggest that stopping and withholding
transfusion and administering corticosteroids in To avoid hyperkalemia in patients with renal
combination with intravenous immune globu- failure or anuria or patients who receive RBCs
lin may be beneficial.94,95 These patients should rapidly (ie, for ECMO), RBCs <7 days old may
also be monitored closely for the formation of be provided. The practice of providing fresh
RBCs for all pediatric patients is based on less
autoantibodies.F'"
Alternatives to Transfusion in SeD. Early evidence than for adults, in whom the evidence
hydroxyurea therapy has been shown to from randomized controlled trials is weak.
In the TOTALrandomized noninferiority trial
decrease transfusions, hospitalizations, vaso-
of 290 children (aged 6-60 months), most with
occlusive pain crises, and acute chest syndrome.
malaria or SCD, with hemoglobin :::;5g/dL and
This is now standard-of-carein children with he-
moglobin SSor S~o-thalassemia.91,96 The optimal lactate :::::5mmollL, older RBCs (25 to 35 days)
did not result in statisticallydifferent worsening
clinical circumstances for hematopoietic progen-
of lactate levels 24 hours after transfusion com-
itor cell transplantation in SCD have not been
pared to fresher RBCs (1 to 10 days). There
determined."
were also no differences in clinical assessment,
Thalassemia cerebral oxygen saturation, electrolyte abnor-
malities, adverse events, survival, and 30-day re-
Thalassemia with severe anemia is treated with covery between the groups.!" Although data
transfusion to improve tissue oxygenation and
are being analyzed from a multicenter trial (ABC
suppress extramedullary erythropoiesis in the
PICU study) comparing clinical outcomes in
liver, spleen, and marrow, decreasing long-term
children in the ICU who receive RBCs stored
associated complications. Most transfusion pro- <7 days or those of standard age,107there is cur-
tocols aim for target hemoglobin between 8 to rently no evidence to support the transfusion of
10 g/dL to allow normal growth and develop- fresh RBCsin the pediatric population.
ment. Iron overload can occur due to the pa-
tient's underlying physiology even without
Whole Blood
transfusion. If this occurs, it is treated with che-
lation therapy beglnning early in childhood." There are limited indications for transfusion of
Besides iron overload, alloimmunization can whole blood in children. Since the early 1990s,
cause significant problems in patients receiving after Manno et al108studied the hemostatic ef-
chronic transfusions for thalassemia. The inci- fects of fresh whole blood vs reconstituted
dence of alloantibody formation in this popula- whole blood following open-heart surgery in
tion ranges from 5.6% to 24.7% worldwide.?"?' children for 24 hours after surgery, fresh whole
and is 19% in the United States.!" However, blood has been advocated in order to improve
among those who received RBCs matched for hemostasis and decrease systemic inflammation.
C, c, E, e, and K, alloimmunization rates were In that study, children <2 years undergoing car-
reported as 0 to 3.5%,100103 with another study diopulmonary bypass (CPB) surgery who re-
noting there was not a statistically significant ceived whole blood <48 hours old had signifi-
trend toward decreased red cell alloimmuniza- cantly decreased mean 24-hour postoperative
tion when patients received RBCs matched for blood loss compared to children who received
C, c, E, e, and K1.104Recently, an international reconstituted whole blood. Of note, children >2
years of age undergoing simpler defect repairs alloimmunization to paternally inherited platelet
did not have a greater benefit from fresh whole antigens. (See Chapter 23.)
blood <48 hours old over reconstituted whole
blood. Improved hemostasis was attributed. to Indicati()ns
better platelet function in fresh whole blood. In
Most platelet transfusions in preterm and full-
2004, Mou et allo9 examined fresh whole blood
term infants are performed for platelet counts
«48 hours old) vs reconstituted blood (RBCs
<50,000/I-lL to treat active bleeding or prevent
and FFP)for priming of the CPB circuit in chilo occurrence or extension of NH. 114, 115Prophylac-
dren <1 year old. Neither postoperative bleed-
tic platelet transfusions in this population are
ing nor inflammatory markers were significantly controversial. 116,117 (See Table 24-5 for transfu-
different between groups. Interestingly, circuit sion indications and thresholds.) Unlike aduit
priming with whole blood was associated with a patients who rarely have severe bleeding com-
longer ICU stay and increased fluid overload. plications until platelet counts decline to
Thus, the use of fresh whole blood after CPB <10,000/1lL, preterm infants with other com-
may be warranted if it can be obtained from the plicating illnesses may bleed at higher platelet
local blood center or hospital blood center to counts.l'" This increased risk may be attribut-
use in the postoperative setting, but it may not able to I) lower concentrations of plasma coagu-
be as useful for the CPBprime. lation factors, 2) circulation of an anticoagulant
Recently, a large US pediatric hospital report- that potentiates thrombin inhibition, 3) intrinsic
ed on the experience of using fresh whole blood or extrinsic platelet dysfunction/hyperreactivity,
in both pediatric cardiac and pediatric craniofa- or 4) increased vascular fragility,especiallyintra-
cial reconstructive surgeries.l'v"! In both stud- cranial vessels.
ies, donor exposure, compared to either pub- A severe complication of prematurity is NH,
lished studies (cardiac surgery) or a historical which occurs in approximately 40% of preterm
control group before the implementation of neonates in the first 72 hours after birth. Al-
whole blood (reconstructive surgery), was sig- though prophylactic platelet transfusions may
nificantly decreased; other outcomes were not increase platelet counts and shorten bleeding
assessed. However, logisticalissues in obtaining times, this approach has not been shown to re-
whole blood vs the process of providing recon- duce the incidence of NH, and the severity of
stituted whole blood need to be considered in thrombocytopenia appears to be independent of
the decision to use this product. the risk of NH Grade >2.118Hence, the use of
platelets in this situation and the appropriate
platelet dose remain controversial.!'?
In the PiaNeT2 study, a European multi-
center randomized controlled trial in 660
preterm neonates, there was a higher rate of
Mild-to-moderate thrombocytopenia (platelet new major bleeding or death within 28 days in
count <150,000/I-lL) is the most common he- those who received transfusion for a platelet
mostatic abnormality in ill preterm and full-term count threshold of 50,000/1-lL than in those
infants, affecting approximately 20% to 25% of treated using a platelet count threshold of
infants in neonatal ICUS.6,112The most common 25,OOO/I-lL, with 88% of transfusions apparent-
cause is increased destruction of platelets that is ly occurring within protocol. Although the rea-
generally associated with a variety of self-limited son for improved survival in the lower-threshold
conditions.!" Other causes of thrombocytope- group is unclear, this study indicates that plate-
nia include impaired platelet production, abnor- let transfusion for a threshold of 25,000/1-lL is at
mal platelet distribution, andlor platelet dilu- least as safe as for a threshold of 50,000/1-lL.120
tion secondary to massive transfusion. In rare In older infants and children, platelets are
instances, thrombocytopenia may be the result most often prophylactically administered for
of in-utero destruction associated with maternal chemotherapy-induced thrombocytopenia. The
TABLE 24-5. Transfusion Guidelines for Platelets in Neonates and Older Children 19,20
With Thrombocytopenia
1. Plateletcount <1O,OOO/~L
with failure of plateletproduction.
2. Plateletcount <30,OOO/~L
in neonatewith failure of plateletproduction.
3. Plateletcount <50,OOO/~L
in stableprematureinfant:
a. With active bleeding,or
b. Beforean invasiveprocedure,with failure of plateletproduction.
4. Plateletcount <1OO,OOO/~L
in sick prematureinfant:
a. With active bleeding,or
b. Beforean invasiveprocedurein patientwith DIC.
Without Thrombocytopenia
1. Active bleedingin associationwith qualitativeplateletdefect.
2. Unexplainedexcessivebleedingin a patient undergoingcardiopulmonarybypass.
3. PatientundergoingECMOwith:
a. A plateletcount of <1OO,OOO/~L,
or
b. Higherplateletcounts and bleeding.
DIG = disseminated intravascular coagulation; = extracorporeal membrane oxygenation.
transfusion threshold for these patients is usual- Components and Dose

ly a platelet count between 10,000 and
The use of whole-blood-derivedplatelets at dos-
20,0001IlL, although the platelet count should
es of 5 to 10 mLlkg body weight have been
not be the sole determinant for transfusion. The demonstrated to raise the platelet count of an
PLADO study, which focused on prophylactic average full-term newborn by 50,000 to
platelet dose, showed that in children and adults 100,0001IlL, depending on the concentration
with hypoproliferative thrombocytopenia, proof the platelet component used.38,112 A similar
phylacticplatelet transfusions for platelet counts dosing regimen is typically used for apheresis
s:lO,OOOIIlL,in concert with different doses of platelets and older children. (SeeTable 24-3 for
platelets, had no effect on the incidence of mod- platelet dosing for desired increment, and Table
erate to severe bleeding.!" However, subgroup 24-5 for indications for platelet transfusion.) Al-
analysis of the 198 children in this study though 15-to 60-minute posttransfusionplatelet
showed that children were at a higher risk of counts can help evaluate platelet survival, these
bleeding over a wider range of platelet counts counts are not necessarilygood predictors of he-
compared to adults. In the pediatric patients, mostatic efficacy.
the rate of Grade 2 or greater bleedingwas high- When possible,the platelet component should
est in the youngest patient cohorts-86% in pa- be ABO group specific or compatible. Transfu-
tients aged 0 to 5 years, 88% in 6- to 12-year- sion of ABO-incompatible plasma should be
olds, and 77% in 13- to 18-year-olds,compared avoided in children, and especially in infants,
to 67% in adults-suggesting other measures because of their small blood and plasma vol-
may be needed to prevent thrombocytopenic umes." If it becomes necessary to administer
bleeding in pediatric cancer patients. 122 ABO-incompatibleplatelets to an infant, plasma
TABLE 24-6. Transfusion Guidelines for Plasma Components in Neonatesand OlderChildren19,20

Fresh Frozen Plasma (FFP)
1. Support during treatment of disseminatedintravascular dissemination.
2. Replacementtherapy.
a. When specific factor concentratesare not available,including, but not limited to, antithrombin; pro-
tein Cor S; and Factor II, FactorV, FactorX, and FactorXI.
b. DUringtherapeutic plasma exchangewhen FFPis indicated (cryopoor plasma, plasmafrom which
the cryoprecipitate has been removed).
3. Reversalof warfarin in an emergencysituation may also be an alternative,such as before an invasive
procedure with active bleeding.
Note: FFPis not indicated for volume expansion or enhancementof wound healing.
Cryoprecipitated AHF
1. Hypofibrinogenemiaor dysfibrinogenemiawith active bleeding.
2. Hypofibrinogenemiaor dysfibrinogenemiawhile undergoing an invasiveprocedure,
3. FactorXIII deficiency with active bleeding or while undergoing an invasive procedure in the absenceof
FactorXIII concentrate.
4. Limited directed-donor cryoprecipitatefor bleedingepisodes in small children with hemophiliaA (when
recombinant and plasma-derivedFactorVIII products are not available).
5. In the preparation of fibrin sealant.
6. Von Willebrand diseasewith active bleeding, but only when both of the following are true:
a. Deamino-O-argininevasopressin (DOAVP)is contraindicated, not available,or does not elicit
response.
b. Virus-inactivated plasma-derivedFactorVIII concentrate (which contains von Willebrand factor) or
von Willebrand recombinant concentrateis not available.
may be removed either by volume reduction or

washing with saline resuspension. (Refer to lo-
cal protocols.) However, routine, centrifugation N ES
to remove plasma from the platelets should be
AND (HU.DR
avoided because it.has been shown.to decrease
the posttransfusion platelet count increment.F' Plasma "
In addition, when platelets are stored in a sy- ,:".
ringe, the pH has been shown to decrease rapid- Plasma is frequently used to replace coagulation
ly, a potential problem for an already ill and aci- factors in preterm and full-term infants with
dotic recipient.124,125 Therefore, when volume hemorrhage or before surgery, particularly if
reduction in an open system is used and the multiple factor deficiencies are present, such as
component is placed in a syringe, the processing in hemorrhagic disease of the newborn due to
should be performed as close to the time of issu- vitamin K deficiency. It is occasionally used to
ance as possible, and the component should be replace anticoagulant factors when anticoagula-
infused within 4 hours from the start of process- tion is contraindicated in infants with life-threat-
ing. ening thrombosis. Table 24-6 provides the indi-
cations for the transfusion of plasma, which are tion's pediatric critical care unit. 128Plasmatrans-
similar in older infants, children, and adults. fusions should be considered with caution in
FFPis plasma that was frozen within 8 hours this population.
of collection and is used within 24 hours of
thawing. Twenty-four hours after being thawed, Cryoprecipitated Antihemophilic
the plasma can be relabeled as Thawed Plasma, factor
which is considered an acceptable substitute
that can be used over the next 4 days. (See Cryoprecipitate transfusions are primarily used
Chapter 6.) Another form ofplasma used in chil- to treat conditions resulting from 1) decreased
dren is Plasma Frozen Within 24 Hours After or dysfunctional fibrinogen (congenital or ac-
Phlebotomy (PF24), used within 24 hours of quired) or 2) Factor XIIIdeficiency, if Factor XIII
thawing. The usual dose of plasma is 10 to 15 concentrate leg, Corifact (CSL Behring)]is not
mLlkg, which is expected to increase all factor available. Cryoprecipitate is also glven in con-
activity levels by 15% to 20% unless there is a junction with platelets and FFP to treat dissemi-
marked consumptive coagulopathy.!" The evi- nated intravascular coagulation (DIC).Typically,
dence for use of plasma to correct coagulopathy 1 unit is sufficient to achieve hemostatic levels
without bleeding is weak. 117,126 in an infant.
To limit donor exposure for each recipient ABO-compatible cryoprecipitate is preferred
while minimizing plasma waste, blood can be in neonates because transfusion of a large vol-
collected into a system with multiple, integrally ume of ABO-incompatible cryoprecipitate may
attached bags that create ready-to-freeze ali- result in a positive direct antiglobulin test result
quots." Once thawed, the aliquots can be subdi- and, in very rare cases, mild hemolysis.129,130
vided further for several patients if they can be Cryoprecipitate transfusion is not recom-
used within a 24-hour period. A common prac- mended for patients with Factor VIII deficiency
tice at some institutions is to use group ABplas- because standard therapy for this condition is
ma because a single unit can provide multiple infusion of recombinant or virus-inactivated,
small-volume doses for several neonates. How- monoclonal-antibody-purified, plasma-derived
ever this can be difficultto maintain, as ABplas- Factor VIII products.!" In the United States,
ma is usually not in plentiful supply. A plasma cryoprecipitate should be used to treat von Will-
product that is being increasinglyused is solvent! ebrand disease only if plasma-derived, virus-
detergent-treated plasma (SD plasma). SD plas- inactivated concentrates or recombinant prod-
ma is plasma pooled from many donors that has ucts containing vWF are not available." Recom-
undergone processing to remove lipid-enveloped binant vWF [Vonvendi(BaxaltaInc)]was recent-
viruses. A recent study showed decreased mor- ly approved in the United States for treatment
tality in children in the ICU who received SD and prevention of bleeding in adults with von
plasma compared to those who received FFP or Willebrand disease; there is currently an open
PF24.127In addition, the risk of allerglcreactions study in children aged 6 years and above. See
and transfusion-related acute lung injury Tables 24-3 and 24-6 for dosing and guidelines
(TRAU) appears reduced due to the process of regarding plasma and cryoprecipitate use.117, 131
filitration to remove cellular debris, as well as
the dilution of individual donor plasma proteins.
Use of this type of plasma will depend on the
risk/benefit analysis performed by transfusion N S
services.
A recent prospective observational study
showed that plasma transfusions were inde- The role of granulocyte transfusion for sepsis in
pendently associated with an increased risk of neonates is unclear, and this treatment is rarely
new or progressive multiple-organ dysfunction, used, although dosage appears to correlate with
nosocomial infections, and prolonged length of efficacy. For patients of any age, it is important
stay in 831 children admitted to one institu- to establish the following factors before the
C HAP T ER 24 Neonatal and Pediatric TransfusionPractice 689
transfusion of granulocytes: 1) evidence of bac- mains no definitive evidence for the benefit (or
terial or fungal septicemia; 2) absolute neutro- lack thereof) of granulocyte transfusions.
phil count <SOO/llL, chronic granulomatous
disease, or leukocyte adhesion deficiency; and
3) diminishing storage pool (such that 7%of nu-
cleated cells in the marrow are granulocytes
that are metamyelocytes or more mature).132,133
Granulocyte concentrates are produced by
standard apheresis techniques or by pooling
buffy coats from whole blood. A typical dose for Vascular Access
infants is 10 to 15 mLlkg body weight, which is
approximately 1 x 109to 2 x 109 polymorpho- Vascular access is the most difficult aspect
nuclear cells/kg.19,132For larger children and of transfusion administration in patients young-
adults, a minimum dose of 1 x 1010is suggest- er than 4 months old, particularly preterm in-
ed.19,131,132
If a patient weighs over 20 kg, either fants who require long-term or continuous intra-
one-halfof a unit or the whole unit can be trans- venous infusions. The umbilical vein is most
fused, depending on its volume and the volume frequently cannulated after birth to facilitate ad-
ministration of fluids and transfusions and to
status of the patient.
monitor central venous pressure. Vascular cath-
Treatment is usually administered dailyuntil
eters (24-gauge) and small needles (2S-gauge)
an adequate neutrophil count is maintained
generally can be safely used for RBC transfu-
and/or the patient shows clinicalimprovement.
sionswithout causing hemolysis if constant flow
Granulocytes must be irradiated, given that
rates are applied. The outcomes of transfusions
recipients are either neonates with immature using smaller-gauge needles and catheters or
'l-cell function or older immunosuppressed intra-osseous catheters in neonates have not
individuals, both groups that are at in- been evaluated.
creased risk for TA-GVHD. Because granulo-
cytes cannot be leukocyte reduced, if the re- Filters and Transfusion Sets
cipient is CMV seronegative, the component
should be obtained from CMV-seronegative The plastic tubingin the administration sets can
donors to prevent virus transmtsston.' Granulo- add a significant amount of dead-space volume
cytes should be ABO compatible with the red to the transfusion and should be accounted for
cells of the recipient because of the significant when preparing a transfusion dose. Pediatric in-
red cell content in these components.49(P35) fusion sets created for platelets and other small-
Many institutions also .provide D-compatible volume components have less dead space than
components to decrease Rh alloimmunization. standard sets. When using normal saline to flush
To obtain higher doses of granulocytes for blood components into the patient, ceasing the
larger patients, donors can receive mobilizers flush before the normal saline reaches the pa-
such as steroids, granulocyte colony-stimulating tient may decrease the risk of dilution. For stan-
factor (G-CSF),or a combination. A multicenter dard filterrequirements, see Chapter 18.
study that randomly assignedneutropenic adults
and children with systemic bacterial or fungal Rate of Administration
infection to receive standard antimicrobialswith The rate of blood component administration is
or without granulocytes (from donors stimu- dictated by the clinical needs of the patient,
lated with G-CSFand steroid) showed no bene- whether neonate or older child. Therefore, ad-
fit in mortality or microbialresponse in the gran- ministering a simple transfusion over 2 to 4
ulocyte group. However, the study closed early hours is adequate in nonemergent situations.
because of poor accrual and was insuffidently However, in states of shock or severe bleeding, a
powered to adequately address the efficacy of rapid infusion is often required. Rapid .infusion
granulocyte transrustons.F' Therefore, there re- devices should be used with caution; in neo-
nates, syringe push can be used instead. In small of the component or filtration at the bedside.
infants, despite concerns from neonatologists Thisprocess eliminates the need for the nurse to
that rapid blood infusion rates may adversely af- transfer blood from the pack to a syringe at the
fect intravascular volume and electrolyte levels, bedside for delivery by a syringe pump. Remov-
an increased risk of NH in these small and frag- ing this additional step reduces the risk of con-
ile patients has not been clearly demonstrated tamination, mislabeling, or damage to the unit
and remains a theoretical risk.!" that results in blood loss or spillage."
Reducing donor exposures is readily accom-
Aliquoting for Small-Volume plished by aliquoting, which enables a recipient
Transfusion to receive multiple small-volume transfusions
from a single unit until it reaches its expiration
The purpose of creating small-volumealiquots is date.31,140 Some hospital transfusion services as-
to limit donor exposures, prevent circulatory sign a single unit of RBCsto one or more infants
overload, and decrease blood waste.30,135-138 Sev- based on their weight. 137-139
eral technical approaches can be used." Once an aliquot is produced at either the
Small-volume RBC transfusion aliquots are blood center or hospital blood bank, it must be
commonly prepared with a multiple-pack sys- labeled with the expiration date and the origin
tem.54,139 Quad packs, employed mostly by
and disposition of each smaller unit, and often
blood centers, are produced from a single unit of the parent unit must be recorded. Aliquot expi-
whole blood that is diverted into a primary bag ration dates vary from institution to institution
with three integrally attached smaller bags. The and depending on the oversight authority; thus,
plasma is then separated and diverted into one the local standard and standard operating proce-
bag during component preparation. The remain- dures should always be followed.
ing red cells are drawn into the smaller bags as
needed for transfusion. Each of the smaller units
Extracorporeal Membrane
has the same expiration date as the original unit
Oxygenation
because the system's original seal has remained
intact and a "closed system" is maintained. A ECMO is a prolonged treatment whereby blood
hospital transfusion service can then remove (ei- is removed from the patient's venous circula-
ther by heat sealer or metal clips)each aliquot as tion, circulated through a machine to remove
needed. For hospital transfusion services that do CO2 and replenish O2, and then returned to the
not have a sterile connection device (Fig24-1), patient. In neonates and children, ECMO has
this method provides three aliquots from a sin- become a lifesavingtreatment for meconium as-
gle untt." However, some blood components piration syndrome, persistent pulmonary hyper-
may still be wasted when used in aliquots that tension of the newborn, congenital diaphrag-
are larger than the dose selected for each patient matic hernia, and respiratory failure due to
based on body weight. sepsis. It is also used for postoperative support
Hospital transfusion services that have a ster- following cardiac surgery. Because standardized
ile connection device have multiple options to guidelines for transfusion practice in ECMO
produce aliquots, such as transfer packs leg, have not been established, centers typically es-
T3000 (Charter Medical), Fig 24-2], small- tablish their own criteria. Table 24-7 provides
volume bags, or tubing that has integrally at- some guidelines for ECMO.141
tached syringes. Syringesets (Fig24-3) offer the Bleeding complications are frequent during
greatest accuracy for obtaining the desired vol- ECMO treatment and are usually multifactorial.
ume to be transfused based on vo1ume-per- Causes include 1) systemic heparinization, 2)
weight calculations. Some syringe sets have an platelet dysfunction, 3) thrombocytopenia, 4)
attached ISO-micronin-line filter for use during other coagulation defects, and/or 5) the nonen-
the aliquoting process so that, when issued by dothelial ECMO circuit. Thrombotic complica-
the blood bank, the cells are ready to be placed tions are also common. Hospital blood banks
on a syringe pump without further manipulation and transfusion services must be in close com-
FIGURE 24-1. A sterile tubing welder. (Courtesy of R. Punzalan, Versiti Blood Center of Wisconsin and
Children's Hospital of Wisconsin.)
FIGURE24-2. T3000 neonatal/pediatric150-mL aliquot system. (Courtesyof P. Niston,CharterMedical, Ltd.)

FIGURE 24·3. Neonatal syringe set. (Reproduced with permission from Charter Medical, Ltd.)
munication with the ECMO staff and observe to manage resistance to heparin, without strong
local protocols to ensure safe, efficient, and con- data to support the practtce.i" A retrospective
sistent care. cohort study showed tighter control of heparin
There has been increased use of antithrom- and decreased incidence of thrombosis without
bin infusion during ECMO, prophylacticallyor increased bleeding when antithrombin was glv-
TABLE 24·7. Blood Component Preparation Protocols for ECM0141
Cardiac arrest 5-10 min 2 units RBCs O-neg RBCs <14 days, AS
ECMOcircuit disruption 5-10 min 2 units RBCs O-neg RBCs <14 days, AS
Progressive septic shock 30 min 2 units RBCs O-neg RBCsor <10 days,
(nonneonate) type specific any preservative
Neonatetransferred for 1-2 hours 2 units RBCs O-neg RBCs <10 days,
ECMO 1 unit FFP AB plasma CPOor CPOA
1 unit platelets
Cardiac ICU 30-60 min 2 units RBCs Type specific <7 days, AS
Gradual respiratory or Hours to days 2 units RBCs Type specific <10 days, CPO
cardiac failure on conven-
tional support
- -
ECMO = extracorporeal membrane oxygenation; RBCs = Red Blood Cells; AS = additive solution; FFP = Fresh Frozen
Plasma; CPO= citrate-phosphate-dextrose; CPDA= citrate-phosphate-dextrose-adenine;ICU= intensive care unit.
en prophylactically to neonates on ECMO.143 transfusion medicine specialist and the coagula-

However, a retrospective review of the US Pedi- tion laboratory should be aware of unusual he-
atric Health Information Systems database mostatic monitoring requirements of these de-
showed that childrenwho received at least 1 dose vices.146
of antithrombin during ECMO had more bleed-
ing and thrombotic events than those who did Massive Transfusion
not.!" Therefore, studies on the role of anti-
Trauma is the leading cause of death in infants,
thrombin replacementduringECMOare needed.
Children, and young adults aged 1 to 21 years.
ECMOinitiation typicallyrequires 1 to 2 units
Although trauma rarely leads to hemorrhagic
of ABO and Rh group-compatible and cross-
shock and massive transfusion, resuscitation af-
match-compatible RBCs for blood priming. In
ter trauma can be challenging.
emergencies,uncrossmatchedblood may be used.
The evidence to support a pediatric massive
In addition, a single unit of group-compatible
transfusion protocol (MTP)is limited, but sever-
FFP should be allocated to the ECMO patient.
al pediatric institutions use MTPs to improve
RBC units are usually negative for hemoglobin
the outcomes of patients who have massive
S, relatively fresh «5 to 7 days old), irradiated
bleeding with trauma.!" Although studies indi-
(many patients requiring ECMO are neonates
cate that an MTP in the pediatric setting is feasi-
and/or potential transplant recipients), and
ble not only for providing rapid and balanced
CMV seronegative or leukocyte reduced." Be-
blood component support but also for decreasing
cause the ECMO circuit consumes platelets,
the risk of thromboembolic events, the role of
higher platelet counts are often maintained.
the MTP and the optimal ratio of components
However, there are no evidence-based guide- has not been studied in the pediatricsetting_14S-150
lines for transfusion in patients on ECMO_
A large pediatric trauma trial has not been un-
dertaken and, given the rarity of patients, would
Ventricular Assist Devices be difficult to accomplish.148-152 In a recently
The use of ventricular assist devices (VADs) in published multicenter trial, adults with severe
children is increasing, especially with the US trauma and major bleeding who received early
Food and Drug Administration (FDA)approval administration of plasma, platelets, and RBCsin
of the first long-term VAD (Berlin Heart EXCOR a 1:1:1 ratio compared with a 1:1:2 ratio did not
VAD)in children, in 2011, as a bridge to cardiac have significant difference in mortality at 24
transplantation or as a bridge from ECMO for hours or at 30 days. However, there were fewer
cardiac support!" In children, the most con- deaths from bleeding at 24 hours in the 1:1:1
cerning complicationsinclude stroke and device- group, so this ratio has been adopted by many
related thromboembolism. These occur in large trauma centers, including for older children.t"
part due to activation of the endothelium, he- Tranexamic acid has been used for massive
mostatic proteins, and the fibrinolyticsystem, as trauma-associated bleeding in adults. In the
well as platelets and leukocytes, resulting in CRASH-2study, a multicenter randomized con-
increased thrombin generation necessitating trolled trial in >7500 massively bleeding pa-
thromboprophytaxts.I" Therefore, 8 to 24 hours tients, early (within 1-3 hours) treatment with
after implantation of a VAD,aggressiveanticoag- tranexamic acid was associated with decreased
ulation and antiplatelet treatment are initiated, bleeding-associated mortality (5.3% vs 7.7% in
with close monitoring of anticoagulant levels, the placebo group; p <0.0001). However, late
including use of viscoelastic testing. Oral antico- (>3 hours) treatment with tranexamic acid was
agulant drugs (eg, warfarin) are considered in associated with increased bleeding-associated
children older than 12 months when oral or en- mortality (4.4% vs 3.1% in the placebo group;
teral feeding is optimized. Because of the very p = 0.004). No patients <16 years old were
small extracorporeal size of the device, priming enrolled in that study.!" The use of tranexamic
is typicallynot required. However, the pediatric acid in pediatric trauma has not been studied.
TABLE 24-8. Irradiation Guidelinesfor Children Who Require Cellular Blood Components19.20.131
1. Prematureneonatesweighing <1200g at birth.
2. Any patientwith:
a. Knownor suspectedcellularimmunedeficiency.
b. Significant immunosuppressionrelatedto chemotherapyor radiationtreatment.
3. Any patient receiving:
a. Componentsfrom blood relatives.
b. HLA-matchedor crossmatchedplateletcomponents.
c. Granulocytetransfusion.
T randomized study to completely prevent CMV

transmissionby transfusionin VLBWneonates. 158
However, in the absence of superior benefit for
Acute Transfusion Reactions using both approaches, providing leukocyte-
reduced components is generally felt to be ade-
In general, acute transfusion reactions are more quate to prevent transfusion-transmitted CMV.160
commonly reported in children than in adults. 155
In particular, there is increased incidence of fe- Leukocyte Reduction
brile nonhemolytic, allergic, and hypotensive re-
actions in pediatric patients.155,156 The benefits of transfusing leukocyte-reduced
components include reducing the risk of
CMV Prevention transfusion-transmitted CMV, preventing febrile
nonhemolytic transfusion reactions, and de-
The manifestation of CMV infection in suscepti- creasing the risk of HLAalloimmunization. 161
ble individuals (eg, neonates, immunocompro-
mised patients) is variable, ranging from Irradiation
asymptomatic seroconversion to multiorgan in-
volvement, viremia, and death. CMV may be TA-GVHDhas been reported most frequently in
transmitted transplacentally, during the birth newborns with confirmed or suspected congeni-
process, during breastfeeding, as a result of per- tal immunodeficiency.The majority of TA-GVHD
sonal contact with infected persons, or from cases reported in nonimmunocompromised in-
transfusion. With current technologies, the risk fants have occurred after intrauterine transfu-
of acquiring CMV from transfusion is likely low- sion and subsequent postnatal exchange transfu-
er than the historically reported 1% to 3%.157 In sion, and were almost always associated with
a recent study, it was found that breast milk was transfusion from maternal directed donation.162,163
the main source of CMV transmission in neo- There have also been rare cases of TA-GVHDin
nates. 158 The risk of transfusion-transmitted CMV extremely premature infants, infants with neo-
infection may be higher in low-birthweight in- natal alloimmune thrombocytopenia, or infants
fants (<1200 g) who were born to seronegative on ECMO.42,162 Mortality from TA-GVHD is
mothers and have received multiple transfu- >90%.
sions.37,159 Cellular blood components are irradiated to
Reduction of CMV transmission from trans- prevent TA-GVHDin immunocompromised re-
fusion is achieved by leukocyte reduction or pro- cipients and for situations when the patient may
vision of blood from CMV-seronegativedonors, share an HLA haplotype with the blood donor.
as evidenced by similar efficacyin adults.'? Us- (See Table 24-8.) Expert opinion and practices
ing both approaches has been shown in a non- differ on this topic. Therefore, protocols should
be 'based .on the patient populations served, Pathogen Inactivation

equipment available, and procedures that have
the highest likelihood of ens\.)1'ingthat a patient One part()ge~ inactivation technique '.r:or plate-
lets at1dplasma has been ,approvedby tn~FDA
who needs irradiated blood receives it-The pro-
[INTERCEPT (Cerus Corp)]. Other 'systems
cesses of irradiation, irradiator quality control,
[Mirasol(Terumo BCT)and Theraflex (Iv1~coPh-
and quality assurance are addressed in Chapters
arma)] have been used in several European
1,2,6, and 17.
countries for many years, includirg .~.~.chil-
dren.167,168 In a recent meta-analysisof random-
Volume Reduction and Washing
ized controlled trials (only some of which in-
Theplasma volume of. the component (usually clude children), the use .of pathogen-reduced
platelets, because RBCs have little plasma) is platelets was associated with increased risk of
usually reduced in transfusions to patients who platelet refractoriness and platelet. transfusion
cannot tolerate increased intravascular volume requirement, but was not associated with in-
(eg, patients with renal ischemia or compro- creased mortality, clinicallysignificantbleeding,
mised cardiac function). In 1993, the AABB or other adverse.events. 169 There are very few
Committee on, Pediatric •Hemotherapy stated studies describing the use of pathogen-reduced
that volume reduction of platelet concentrates platelets in pediatric patients. Seven years of Eu-
should be reserved .for infants who have total ropean.hemovtgilance .:data, .including in chil-
body fluid restrictions.!" Methods for platelet dren, showed minimal adverse events, .usually
volume reduction haVe.been published.165 How- mild, and no increase in TRALI, TA-GVHD,
ever, there is no optimal centrifugation rate and transfusion-transmitted infection, or death after
preparation method. As with any platelet modi- transfusion of pathogen-reduced platelets. 170
fication, the total number of platelets typically In an Austrian study that examined RBCand
decreases, and platelet activation may occur plasma utilization (as a marker of significant
with these procedures. 166 bleeding before and after implementation of
Saline-washed RBCsand platelets are admin- pathogen-reduced platelets), there was nostatis-
istered to reduce the risk of recurrence of severe tically significant difference in RBC and FFP
allergic reactions from plasma. Other uses of transfusion or transfusion-related adverse events
washing may include removal of anticoagulant- among children and neonates.!" In a retrospec-
preservative solutions and removal of high levels tive review, posttransfusion platelet count incre-
of potassium in RBCs, although centrifugation ments in children with cancer who received
and volume reduction alone may be adequate. pathogen-reduced platelets using riboflavin
Maternal cells should be washed to remove ma- were not as increased as in those who received
ternal antibodies and irradiated if these are be- untreated. platelets, but there was no difference
ing used for fetal or neonatal transfusion. in bleeding outcomes.I?
ble;.and transfUsedcellsat'~eitherpnegative or.the.s~m.~PitVB~~sthepatient..Testingtliei~:

...
·f~t'sreyers~ •.typ~forany-J:.~~<pr ang-Bis not necessary: < > i·•••·...•··.·••.•
···i'
~•••~Wa11-VolllIne. (1O-15 ....In~/l{?1.~~rppletr~sfUsions of RB~s (reg8l'dlessof storage solu~orih
.'.'.••.
xmen .adIninisteredslgVltyth~V~~~enshown to·havelittl~ ~ffeston serum potassiumconce~·.....
. tt~tionsjn infants <4 mOllth?pfage despite elevatedpotassiumlevels.in the plasmaofstored
RBCs. .
······6.p~ingc()rrigol1eBt.prep~~tibnritaliquotsaremadewithasteri1econnection.devfce,.they~e
·..........•
cOflsidered.to have been preParedWithin.a "closed'systemt and the originalunit's'expiration
...::.,"~(~~~ cari~el1sedf()iitllene\y ~!~\J.c)t.i .•• ·••...••
·
.:'-,,.,
•...•.. ,i',:':iy;,., ..........•.•..........•...................
'•...•.•......••• '<.'.'\c":'
' '....................•.........
'.'"1.'." T,t~sfusionofJ:.B.s-incOIIlBa~bl~Blastlla/~hould. bE),~yOid,~d,Iti .iBr~ts~d clii1drenbec~V?E)i Of.,
·.'...•~eir smallblood and plas~~v9lymes.,IfABOout'of-groyp:pi&telettransfusionbeco~es n~ce~i
,s~,.plasInarpay.be r~~9y~qbyVolumereduction.•••• '..........ii i ' > i.\ •••••.•.•••.•••
\ ••· >i· ···· >i
'8-. q~r0nic ¥BGtr~sfUsion m~rapx9f indefiniteduration.to.~~nt~ he~o~obin.S.leyelsb~19Vl
~g1is the..therapy of choiS~~Pt~duceme risk of strokerecurrel1c~in patientsvvithSCpo seq
B~Hentshave the hi~estr~~~~.... of.ali0immunizatignto req c~lltllinor antigensof~YP~ti~l1t ....
~pyp ..The.most cpmrpOnlyf?r~~d ~tibodies arE).to g~, ~ph~, ~d J~ syst~m ~~g~l1~'
1V1~ySCp ••. treatfilent c~nt~rsttyi~o.prevent red cell •.all?itr1~.l1nizatio~ ••bytllatching c?~p?,
•.·ri~Pts•~qrecipi~nts·.foKpli~notyJ)i~ally.sirnilar.
. ~tigen'prgfi1esi!h/.ssttategyis ~gt tlle.s~~~ .at
.'."y:. ·all··centers andis controversia~' because"obtaining"enough pneftotypiCallysimilaruiIitsmay'b~',
..•...'.' HifffC111fClfId.c9stly
·I); t.p>(jecreasethe··
.••••.•.....••.•••
·•.•••••.....••...•
,..•"' •.•.............••
.: ...........•.........••....•.•••..•••..
risk.·.of·.trarls~sio~.tr~s~ittedcJ:v1V•.infe~~?l1>il1
:-:,(":' \ .••·.·.•. •..•••
· ·.•.•••..•.•... i··... .: .
·•·•.•.•.•
suscep~lepopulati0n~.•sUf'~·
as!pw.-birthw.eightneonates.•.... ~d.irpmY11ocompromised. children,•.these patients.sli0uldr~C~iV~
blood componentsthat either have been leukocytereduced or are from CMV-seronegativedo·
REFERENCES
1. Hillyer CD, Mondoro TH, Josephson CD, et al. young infant. Am J Pediatr Hemato1 1990; 12:
Pediatric transfusion medicine: Development of 95-104.
a critical mass. Transfusion 2009;49:596-601. 8. Monagle P, Barnes C, IgnjatovicV, et al. Devel-
2. Hume H, Bard H. Small volume red blood cell opmental haemostasis: Impact for clinical hae-
transfusions for neonatal patients. Transfus Med mostasis laboratories. Thromb Haemost 2006;
Rev 1995;9:187-99. 95:362-72.
3. Blanchette V, Doyle J, Schmidt B, et al. Hema- 9. Revel-VilkS. The conundrum ofneonatal coagu-
tology. In: AvecyG, Fletcher M, MacDonald M, lopathy. HematologyAm SocHematol Educ Pro-
eds. Neonatology:Pathophysiologyand manage- gram 2012;2012:450-4.
ment of the newborn. 4th ed. Philadelphia: JB 10. Cowman J, Quinn N, Geoghegan S, et al. Dy-
Lippincott, 1994:952-99. namic platelet function on von Willebrand fac-
4. Brugnara C, Platt as. The neonatal erythrocyte tor is differentin preterm neonates and full-term
and its disorders. In: Nathan DG, Orkin SH, eds. neonates: Changes in neonatal platelet function.
Nathan and Oski's hematology of infancy and J Thromb Haemost 2016;14:2027-35.
childhood. 7th ed. Philadelphia: WB Saunders, 11. Deschrnann E, Saxonhouse MA, Feldman HA,
2009:21-66. et al. Associationbetween in vitro bleeding time
5. Dame C, Fahnenstich H, Freitag P, et al. Eryth- and bleeding in preterm infants with thrombo-
ropoietin mRNA expression in human fetal and cytopenia.JAMAPediatr 2019;173(4):393-4.
neonatal tissue. Blood 1998;92:3218-25. 12. DePalma L. Review: Red cell alloantibody for-
6. Roberts I, Stanworth S, Murray NA. Thrombo- mation in the neonate and infant: Consider-
cytopenia in the neonate. Blood Rev 2008;22: ations for current immunohematologic practice.
173-86. Immunohematology 1992;8:33-7.
7. Andrew M, Paes B, Johnston M. Development 13. Dincsoy MY, Tsang RC, Laskarzewski P, et al.
of the hemostatic system in the neonate and The role of postnatal age and magnesium on
parathyroid hormone responses during "ex- 27. Carroll PD, Widness ]A. Nonpharmacological,
change" blood transfusion in the newborn peri- blood conservation techniques for preventing
od. ] Pediatr 1982; 100:277-83. neonatal anemia-Effective and promising strat-
14. Monagle P, Barnes C, IgnjatovicV, et al. Devel- egies for reducing transfusion. Semin Perinatol
opmental haemostasis. Impact for clinical hae- 2012;36:232-43.
mostasis laboratories. Thromb Haemost 2006; 28. Barcelona SL, Cote C]. Pediatric resuscitation in
95:362-72. the operating room. Anesthesiol Clin North Am
15. Sisson TR, Whalen LE, Telek A. The blood vol- 2001;19:339-65.
ume of infants. II. The premature infant during 29. Luban NL, Mikesell G, Sacher RA. Techniques
the firstyear of life.] Pediatr 1959;55:430-46. for warming red blood cells packaged in differ-
16. Fabres], Wehrli G, Marques MB, et al. Estimat- ent containers for neonatal use. Clin Pediatr
ing blood needs for very-low-birth-weight in- 1985;24:642-4.
fants.Transfusion 2006;46: 1915-20. 30. StraussRG, BurmeisterLF,Johnson K, et al. AS-I
17. Lin]C, Strauss R, Kulhavy]C, et al. Phlebotomy red cells for neonatal transfusions: A random-
overdraw in the neonatal intensive care nursery. ized trial assessing donor exposure and safety.
Pedatrics 2000;1 06:EI9. Transfusion 1996;36:873-8.
18. Wallgren G, Hanson ]S, Lind ]. Quantitative 31. StraussRG,BUrmeisterLF,Johnson K, et al. Fea-
studies of the human neonatal circulation. 3. sibilityand safety of AS-3red blood cellsfor neo-
Observations on the newborn infants central cir- natal transfusions.J Pediatr 2000; 136:215-19.
culatory responses to moderate hypovolemia. 32. Luban NL, Strauss RG, Hume HA. Commentary
Acta Paediatr Scand 1967;(SuppI179):45. on the safety of red cells preserved in extended-
19. RoseffSD, Luban NL, Manno CS. Guidelines for storagemedia for neonatal transfusions. Transfu-
assessing appropriateness of pediatric transfusion 1991;31:229-35.
sion. Transfusion 2002;42: 1398-413. 33. Pyles R, Lowery ], Delaney M. The use of red
20. Saifee NH, Lau W, Kelr A. Intrauterine, neona- cell units containing additives in large volume
tal, and pediatric transfusion. In: Marques MB, neonatal transfusion in neonatology units in the
USA (letter). ISBTScience Series 2017;12:322-
Schwartz ], Wu Y, eds. Transfusion therapy:
3.
Clinical principles and practice. 4th ed. Bethes-
34. Rock G, Poon A, Haddad S, et al. Nutricel as an
da, MD: AABBPress, 2019:291-341.
additivesolution for neonatal transfusion. Trans-
21. Ohls RK.Evaluation and treatment of anemia in
fus Sci 1999;20:29-36.
the neonate. In: Christensen RD, ed. Hemato-
35. Tuchschmid P, Mieth D, Burger R, et al. Poten-
logic problems of the neonate. Philadelphia:WB
tial hazard of hypoalbuminemia in newborn ba-
Saunders, 2000:137-69, bies after exchange transfusions with ADSOL
22. Halvorsen S, Bechensteen AG. Physiology of
red blood cell concentrates. Pediatrics 1990;85:
erythropoietin during mammalian development. 234-5.
Acta Paediatr SuppI2002;91: 17-26. 36. New HV,Berryman], Bolton-MaggsPHB, et al.
23. Ohlsson A, Aher SM. Earlyerythropoiesis-stimu- Guidelines on transfusion for fetuses, neonates
lating agents in preterm or low birth weight in- and older children. Br ] Haematol 2016; 175:
fants. Cochrane Database Syst Rev 2017; 11: 784-828.
CD004863. 37. StraussRG. Data-driven blood banking practices
24. KandasamyY, Kumar P, Hartley L. The effect of for neonatal RBC transfusions. Transfusion
erythropoietin on the severity of retinopathy of 2000;40: 1528-40.
prematurity. Eye 2014;28;814-18. 38. Strauss RG. Transfusion therapy in neonates.
25. Doege C, Pritsch M, Fruhwald MC, et al. An as- Am] DisChild 1991;145:904-11.
sociation between infantile haemangiomas and 39. Strauss RG. Neonatal transfusion. In: Anderson
erythropoietin treatment in preterm infants. KC, Ness PN, eds. Scientificbasis of transfusion
Arch Dis Child Fetal Neonatal Ed 2012;97:F45- medicine: Implications for clinical practice. 2nd
9. ed. Philadelphia:WB Saunders, 2000:321-6.
26. Henry E, Christensen RD, Sheffield M], et al. 40. Strauss RG. Routinely washing irradiated red
Why do four NICUs using identical RBCtrans- cells before transfusion seems unwarranted.
fusion guidelines have different gestational age- Transfusion 1990;30:675-7.
adjusted RBC transfusion rates? ] Perinatol 41. McDonald TB, Berkowitz RA. Massive transfu-
2015;35:132-6. sion in children. In: Jefferies LC, Brecher ME,
eds. Massive transfusion. Bethesda, MD: AABB, 56. Bell EF, Strauss RG, Widness JA, et al. Random-
1994:97-119. ized trial of liberal versus restrictive guidelines
42. Ohto H, Anderson KC. Posttransfusion graft- for red blood cell transfusion in preterm infants.
versus-host disease in Japanese newborns. Pediatrics 2005;115:1685-91.
Transfusion 1996;36: 117-23. 57. KirpalaniH, Whyte RK,Andersen C, et al. The
43. Lee AC, Reduque LL, Luban NLC, et al. premature infants in need of transfusion (PINT)
Transfusion-associatedhyperkalemic cardiac ar- study: A randomized, controlled trial of a re-
rest in pediatric patients receivingmassive trans- strictive (low) versus liberal (high) transfusion
fusion. Transfusion 2014;54:244-54. threshold for extremely low birth weight in-
44. Hall TL, Barnes A, Miller JR, et al. Neonatal fants.J Pediatr 2006; 149:301-7.
mortality following transfusion of red cells with 58. Whyte RK,KirpalaniH, Asztalos EV,et al. Neu-
high plasmapotassium levels. Transfusion 1993; rodevelopmental outcome of extremely low
33:606-9. birth weight infants randomly assigned to re-
45. Fung MK, RoseffSD, Vermoch KL.Blood com- strictive or liberal hemoglobin thresholds for
ponent preferences of transfusion services blood transfusion. Pediatrics2009; 123:207-13.
supporting infant transfusions: A University 59. McCoy TE, Conrad AL,Richman LC, et al. Neu-
HealthSystem Consortium benchmarking study. rocognitive profiles of preterm infants randomly
Transfusion2010;50: 1921-5. assigned to lower or higher hematocrit thresh-
46. Wong EC, Luban NL. Hematology and oncolo- olds for transfusion. Child Neuropsychol 2011;
gy. In: Slonim AD, Pollack MM, eds. Pediatric 17:347-67.
critical care medicine. Philadelphia: Lippincott, 60. Nopoulos PC, Conrad AL, Bell EF, et al. Long-
Williamsand Wilkins, 2006:157-95. term outcome of brain structure in premature
47. Fergusson DA, Hebert P, Hogan DL, et al. Effect infants: Effects of liberal vs restricted red blood
of fresh red blood cell transfusions on clinical cell transfusions. Arch Pediatr Adolesc Med
outcomes in premature, very low-birth-weight 2011;165:443-50.
infants: The ARIPI randomized trial. JAMA 61. KeirA, Pal S, TrivellaM, et al. Adverseeffects of
2012;308: 1443-51. red blood cell transfusions in neonates: A sys-
48. Patel RM, Josephson CD. Storage age of red tematic review and meta-analysis. Transfusion
blood cells for transfusion of premature infants. 2016;56:2773-80.
JAMA2013;309:544-5. 62. Goobie SM, Faraoni D, Zurakowski D, et al. As-
49. Gammon R, ed. Standards for blood banks and sociation of preoperative anemia with postoper-
transfusion services. 32nd ed. Bethesda, MD: ative mortality in neonates. JAMAPediatr 2016;
AABB,2020. 170:855-62.
50. Strauss RG. Selection of white cell-reduced 63. Kirpalani H, Bell E, D'Angio C, et al. Transfu-
blood components for transfusions during early sion ofprematures (TOP)trial: Doesa liberal red
infancy. Transfusion 1993;33:352-7. blood cell transfusion strategy improve neuro-
51. Turkmen T, Oiu D, Cooper N, et al. Red blood logically-intact survival of extremely-low-birth-
cell alloimmunization in neonates and children weight infants as compared to a restrictive strat-
up to 3 years of age. Transfusion2017;57:2720- egy? Version 1.0 (October 8,2012). Bethesda,
6. MD: National Institute of Child Health and Hu-
52. Josephson CD, Castillejo M, Grima K, et al. man Development, 2012. [Avallableat https://
ABO-mismatched platelet transfusions: Strate- www.nichd.nih.gov/sites/defaultlfiles/aboutl
gies to mitigate patient exposure to naturally oc- Documents/TOP_Protoco1.pdf.]
curring hemolytic antibodies. Transfus Apher 64. ETTNO Investigators. The 'Effects of Transfu-
Sci 2010;42:83-8. sion Thresholds on Neurocognitive Outcome of
53. Girelli G, Antoncecchi S, Casadei AM, et al, Extremely Low Birth-Weight Infants (ETINO)'
Recommendations for transfusion therapy in Study: Background, aims, and study protocol.
neonatology. BloodTransfus 2015;13:484-97. Neonatology2012;101 :301-5.
54. RoseffSD. Pediatric blood collection and trans- 65. BallardRA,Vinocur B, ReynoldsJW, et al. Tran-
fusion technology. In: Herman JK, Manno CS, sient hyperammonemia of the preterm infant.
eds. Pediatric transfusion therapy. Bethesda, N Eng]J Med 1978;299:920-5.
MD: AABBPress, 2002:217-47. 66. Leonard N. The early detection and manage-
55. Wong E, RoseffSD, eds. Pediatric hemotherapy ment of inborn errors presenting acutely in the
data card. Bethesda, MD: AABB,2015. neonatal period. Eur J Pediatr 1985;143:253-7.
67. American Academy of Pediatrics Subcommittee 81. DeBaun MR, Gordon M, McKinstry RC, et al.
on Hyperbilirubinemia. Management of hyper- Controlled trial of transfusionsfor silent cerebral
bilirubinemia in the newborn infant 35 or more infarcts in sickle cell anemia. N Engl J Med
weeks of gestation. Pediatrics 2004; 114:297- 2014;371 :699-71O.
316. 82. Adams RJ,McKie VC, Hsu L, et al. Prevention
68. Kliegman RM, Stanton BMD, St. Geme J, Schor of a first stroke by transfusions in children with
NF, eds. Nelson's textbook of pediatrics. 20th sicklecell anemia and abnormal results on tran-
ed. Philadelphia: WB Saunders, 2016. scranial doppler Ultrasonography.N EnglJ Med
69. Valaes T. Bilirubin distribution and dynamics of 1998;339:5-11.
bilirubin removal by exchange transfusion. Acta 83. KellyS, 000010 K, Marsh A, et al. Erythrocyta-
Paediatr Scand 1963;52:604. pheresisfor chronic transfusiontherapy in sickle
70. Wong EC,PisciottoPT.Technicalconsiderations/ cell disease: Survey of current practices and re-
mechanical devices. In: HillyerCD, Strauss RG, view of the literature, Transfusion 2016;56:
Luban NLC, eds. Handbook of pediatric transfu- 2877-88.
sion medicine. London: Elsevier Academic 84. RosseWF, Gallagher D, KinneyTR, et al. Trans-
Press, 2004: 121-8. fusion and alloimmunization in sickle.cell dis-
71. Luban NL. Massive transfusion in the neonate. ease: The cooperative study of Sickle cell dis-
TransfusMed Rev 1995;9:200-14. ease. Blood 1990;76:1431:7.
72. Weisz B, BelsonA, Milbauer B, et al. [Complica- 85. RosseWF, Telen M, Ware RE. Transfusion sup-
tions of exchange transfusion in term and port for patients with sickle cell disease. Bethes-
preterm newborns]. Harefuah 1996;130:170-3. da, MD: AABBPress, 1998.
73. Bada HS, Chua C, SalmonJH, et al. Changes in 86. Vichinsky EP, Luban NL, Wright E, et al. Pro-
intracranial pressure during exchange transfu- spective RBC phenotype matching in a stroke-
sion. J Pediatr 1979;94: 129-32. prevention trial in sickle cell anemia: A multi-
74. Maheshwari A, Carlo WA. Plethora in the new- center transfusion trial. Transfusion 2001 ;41 :
born infant (polycythemia). In: Kliegman RM, 1086-92.
87. Yazdanbakhsh K, Ware RE, Nofzat-Pirenne F.
Stanton BF, Schor NF, et al, eds. Nelson text-
Red blood cell alloimmunization in sickle cell
book of pediatrics. 20th ed. Philadelphia: Blsevi-
disease:Pathophysiology,risk factors, and trans-
er,2015:887-8.
fusionmanagement. Blood2012; 120:528-37.
75. Black VD, Rumack CM, Lubchenko LD, Koops
88. Chou ST, Jackson T, Vege S, et al. High preva-
BL. Gastrointestinal injury in polycythemic in-
lence of red blood cell allotmmuntzatton in sick-
fants. Pediatrics 1985;76:225-31. le cell disease despite transfusion from Rh-
76. Mohamed A, Shah PS. Transfusion associated
matched minority donors. Blood 2013; 122:
necrotizing enterocolitis: A meta-analysis of ob- 1062-71.
servational data. Pediatrics 2012;129:529-40. 89. Kacker S, Ness PM, Savage WJ, et al. Cost-
77. Patel RM, Knezevic A, Shenvi N, et al. Associa- effectiveness of prospective red blood cell anti-
tion of red blood cell transfusion, anemia, and gen matching to prevent alloimmunization
necrotizing enterocolitis in very low-birth- among sickle cell patients. Transfusion 2014;
weight infants. JAMA2016;315:889-97. 54:86-97.
78. Ware RE, Helms RW, SWiTCH Inyestigators. 90. HillyerKL,Hare VW, Josephson CD, et al. Part-
Stroke with transfusions changing to .hydroxy- ners for life: The transfusion program for pa-
urea (SWiTCH).Blood 2012;119:3925-32. tients with sickle cell disease offered at the
79. Estcourt LJ, Fortin PM, Hopewell S, et al. Blood American Red Cross Blood Services, Southern
transfusion for preventing primary and second- Region,Atlanta, Georgia. Immunohematology
ary stroke in people with sickle cell disease. Co- 2006;22:I08-11.
chrane Database Syst Rev2017;1 :003146. 91. Yawn BP, Buchanan GR, Afenyi-AnnanAN, et
80. Ware RE, Davis BR, Schultz WH, et al. Hy- al. Management of sicklecell disease: Summary
droxycarbamide versus chronic transfusion for of the 2014 evidence-based report by expert
maintenance of transcranial doppler flow veloci- panel members. JAMA2014;312: 1033-48.
ties in.children with sickle cell anaemia- TCD 92. Tahhan HR, Holbrook CT, Braddy LR, et al.
with transfusions changing to hydroxyurea Antigen-matcheddonor blood in the transfusion
(TWiTCH):A multicentre, open-label, phase 3, management of patients with sickle cell disease.
non-inferiority trial. Lancet 2016;387:661-70. Transfusion1994;34:562-9.
93. Wood JC, Cohen AR, Pressel SL, et al. Organ globinopathies:A systematic review and guide-
iron accumulation in chronicallytransfusedchil- line. Transfusion2018;58:1555-66.
dren with sickle cell anaemia: Baseline results 106. Dhabangi A, AinomugishaB, Cserti-Gazdewich
from the TWiTCH trial. Br J Haematol 2016; C, et al. Effect of transfusion of red blood cells
172:122-30. with longer vs shorter storage duration on ele-
94. Petz LD, Calhoun L, Shulman IA, et al. The vated blood lactate levels in children with se-
sickle cell hemolytic transfusion reaction syn- vere anemia: The TOTALrandomized clinical
drome. Transfusion 1997;37:382-92. trial. JAMA2015;314:2514-23.
95. Win N, Doughty H, Telfer P, et al. Hyperhemo- 107. Age of blood in children in pediatric intensive
lytic transfusion reaction in sickle cell disease. care units (ABCPICU).St.louis, MO: Washing-
Transfusion2001;41:323-8. ton UniversitySchoolof Medicine, 2015. [Avall-
96. Wong TE, Brandow AM, Llm W, et al. Update able at https://clinicaltrials.gov/ct2ishow/
on the use of hydroxyurea therapy in sickle cell NCTOI977547.]
disease. Blood2014;124:3850-7. 108. Manno CS, Hedberg KW, Kim HC, et al. Com-
97. Arnold SD, Bhatia M, Horan J, et al. Haemato- parison of the hemostatic effects of fresh whole
poietic stem cell transplantation for sickle cell blood, stored whole blood, and components af-
disease-Current practice and new approaches. ter open heart surgery in children. Blood 1991;
BrJ HaematoI2016;174:515-25. 77:930-6.
98. Olivieri NF, Brittenham GM. Iron-chelating 109. Mou SS, Giroir BP, Molitor-KirschEA, et al.
therapy and the treatment of thalassemia.Blood Fresh whole blood versus reconstituted blood
1997;89:739-61. for pump priming in heart surgery in infants.
99. Dhawan HK,KumawatV, Marwaha N, et al. AI- N EnglJ Med 2004;351: 1635-44.
lolmmunization and autoimmunization in trans- 110. Jobes DR, Sesok-Pizzini D, Friedman D. Re-
fusion dependent thalassemia major patients: duced transfusionrequirement with use of fresh
Study on 319 patients. Asian J Transfus Sci whole blood in pediatric cardiac surgicalproce-
2014;8:84-8. dures. Ann Thorac Surg2015;99:1706-11.
100. RomphrukAV,SimtongP, ButryojanthoC, et al. 111. Thottathil P, Sesok-PizziniD, Taylor JA, et al.
The prevalence, alloimmunization risk factors, Whole blood in pediatric craniofacial recon-
antigenic exposure, and evaluation of antigen- struction surgery. J Craniofac Surg 2017;28:
1175-8.
matched red blood cells for thalassemia transfu-
112. BlanchetteVS,Kuhne T, Hume H, et al. Platelet
sions: A 10-year experience at a tertiary care
transfusiontherapy in newborn infants.Transfus
hospital. Transfusion2019;59: 177-84.
Med Rev 1995;9:215-30.
101. Davoudi-KiakalayehA, Mohammadi R, Pour-
113. Castle V, Andrew M, KeltonJ, et al. Frequency
fathollah AA, et al. Alloimmunization in thalas-
and mechanism of neonatal thrombocytopenia.
semia patients: New insight for healthcare. Int J
J Pediatr 1986;108:749-55.
Prev Med 2017;8: I01. 114. Andrew M, Vegh P, Caco C, et al. A random-
102. VichinskyE, Neumayr L,Trimble S, et al. Trans- ized, controlled trial of platelet transfusions in
fusion complications in thalassemia patients: A thrombocytopenic premature infants. J Pediatr
report from the Centers for DiseaseControl and 1993;123:285-91.
Prevention. Transfusion2014;54:972-81. 115. Honohan A, van't Ende E, Hulzebos C, et al.
103. Pujani M, Pahuja S, Dhingra B, et al. Alloimmu- Posttransfusion platelet increments after differ-
nisation in thalassaemics: A comparison be- ent platelet products in neonates: A retrospec-
tween recipients of usual matched and partial tive cohort study. Transfusion2013;53:3100-9.
better matched blood. An evaluation at a tertia- 116. New HV,Stanworth SJ,EngelfrietCP,et al. Neo-
ry care centre in India. Blood Transfus 2014; natal transfusions.Vox Sang2009;96:62-85.
12(Suppll):sI00-4. 117. Poterjoy BS, Josephson CD. Platelets, frozen
104. Michall-Merianou V, Pamphili-Panousopoulou plasma, and cryoprecipitate:What is the clinical
L, Piperi-Lowes L, et al. Alloimmunization to evidence for their use in the neonatal intensive
red cell antigens in thalassemia: Comparative care unit? Sellin PerinatoI2009;33:66-74.
study of usual versus better-match transfusion 118. von LindernJS, van den Bruele T, LoprioreE, et
programmes.Vox Sang 1987;52:95-8. al. Thrombocytopenia in neonates and the risk
105. Compernolle V, Chou ST, Tanael S, et al. Red of intraventricular hemorrhage: A retrospective
blood cell specificationsfor patients with hemo- cohort study. BMCPediatrics 2011; 11:16.
119. Josephson CD, Su LL, Christensen RD, et al. 132. Price TH, Boeckh M, Harrison RW, et al. Ef-
Platelet transfusion practices among neonatolo- ficacy of transfusion with granulocytes from
gistsin the United States and Canada: Results of G-CSF/dexamethasone-treated donors in neu-
a survey. Pediatrics 2009;123:278-85. tropenic patients with infection. Blood 2015;
120. Curley A, Stanworth SJ, Willoughby K, et. al. 126:2153-61.
Randomized Trial of Platelet-Transfusion 133. Marfin AA, Price TH. Granulocyte transfusion
Thresholds in Neonates. N Engl J Med 2019; therapy. J Intensive Care Med 2015;30:79-88.
380(3):242-51. 134. Ballabh P. Intraventricular hemorrhage in pre-
121. Slichter SJ, Kaufman RM, Assmann SF, et al. mature infants: Mechanism of disease. Pediatr
Dose of prophylactic platelet transfusions and Res 2010;67(1):1-8.
prevention of hemorrhage. N Eng!J Med 20 I0; 135. Wang-Rodriguez J, Mannino FL, Liu E, et al. A
362:600-13. novel strategy to limit blood donor exposure
122. Josephson CD, Granger S, Assmann SF, et al. and blood waste in multiply transfused prema-
Bleeding risks are higher in children versus ture infants. Transfusion 1996;36:64-70.
adults given prophylactic platelet transfusions 136. Liu EA, Mannino FL, Lane TA. Prospective, ran-
for treatment-induced hypoproliferative throm- domized trial of the safety and efficacy of a limit-
bocytopenia. Blood2012;120:748-60. ed donor exposure transfusion program for pre-
123. Honohan A, Tomson B, van der BornJ, et al. A mature neonates. J Pediatr 1994;125:92-6.
comparison of volume-reduced versus standard 137. Bednarek FJ, Weisberger S, Richardson DK, et
HLA/HPA-matched apheresls platelets in allo- al. Variations in blood transfusions among new-
immunized adult patients. Transfusion 2012;52: born intensive care units. SNAP II study group.
742-51. J Pediatr 1998;133:601-7.
124. PisciottoPT, Snyder EL,SnyderJA,et al. In-vitro 138. Maier RF, Sonntag}, Walka MM, et al. Chang-
characteristics of white cell-reduced single-unit ing practices of red blood cell transfusions in in-
platelet concentrates stored in syringes.Transfu- fants with birth weights less than 1000 g.J Pedi-
sion 1994;34:407-11. atr 2000; 136:220-4.
125. Diab Y, Wong E, Criss VR, et al. Storage of ali- 139. Levy GJ, Strauss RG, Hume H, et al. National
quots of apheresis platelets for neonatal use in survey of neonatal transfusion practices: 1. Red
syringeswith and without agitation. Transfusion blood cell therapy. Pediatrics 1993;91 :523-9.
2011;51:2642-6. 140. Goodstein MH, Herman JH, Smith JF, et al. Met-
126. Motta M, Del Vecchio A, RadicioniM; Clinical abolic consequences in very low birth weight in-
use of fresh-frozen plasma and cryoprecipitate in fants transfused with older AS-1 preserved
neonatal intensive care unit. J Matern Fetal erythrocytes. Pediatr Pathol Lab Med 1999; 18:
Neonatal Med 2011 ;24: 129-31. 173-85.
127. Camazine MN, Karam 0, Colvin R, et al. Out- 141. Friedman DF, Montenegro LM. Extracorporeal
comes related to the use of frozen plasma or membrane oxygenation and cardiopulmonary
pooled solvent/detergent treated plasma in criti- bypass. In: Hillyer CD, Strauss RG, Luban NLC,
cally ill children. Pediatr Crit Care Med 2017; eds. Handbook of pediatric transfusion medi-
18:e215-23. cine. London: Elsevier Academic Press, 2004:
128. Karam 0, LacroixJ, RobitailleN, et al. Associa- 181-9.
tion between plasma transfusions and clinical 142. Wong TE, HuangYS, Weiser J, et al. Antithrom-
outcome in critically ill children: A prospective bin concentrate use in children: Amulticenter
observational study. Vox Sang2013; 104:342-9. cohort study. J Pediatr 2013; 163:1329-34.
129. AABB,America's Blood Centers, American Red 143. Stansfield BK,Wise L, Ham PB 3rd, et al. Out-
Cross, Armed Services Blood Program. Circular comes following routine antithrombin III re-
of information for the use of human blood and placement during neonatal extracorporealmem-
blood components. Bethesda, MD: AABB, brane oxygenation. J Pediatr Surg 2017;52:609-
2017. 13.
130. Cushing M, Bandarenko N, eds. Blood transfu- 144. Wong TE, Nguyen T, Shah SS,et al. Antithrom-
sion therapy: A handbook. 13th ed. Bethesda, bin concentrate use in pediatric extracorporeal
MD: AABB,2020 (in press). membrane oxygenation: A multicenter cohort
131. Wong ECC, RoseffSD, Bandarenko N, eds. Pe- study. Pediatr Crit Care Med 2016;17:1170-8.
diatric transfusion: A handbook. 5th ed. Bethes- 145. Eghtesady P, Almond CS, Tjossem C, et al; Ber-
da, MD: AABB,2020 (in press). lin Heart Investigators. Post-transplant out-
comes of children bridged to transplant with the megalovirus (CMV)seronegativeblood products

Berlin Heart EXCORPediatric ventricular assist for the prevention of transfusion-associated
device. Circulation 2013;128(11 Supp11):S24- CMV infection after marrow transplant. Blood
31. 1995;86:3598-603.
146. Massicotte MP, Bauman ME, Murray J, Almond 158. Josephson CD, Caliendo AM, Easley KA, et al.
CS.Antithrombotic therapy for ventricular assist Blood transfusion and breast milk transmission
devices in children: Do we really know what to of cytomegalovirus in very low-birth-weight in-
do? J Thromb Haemost 2015;13(Suppl 1):S343- fants: A prospective cohort study. JAMAPediatr
50. 2014; 168:1054-62.
147. Horst J, Leonard JC, VogelA, et al. A survey of 159. Brady MT, Milam JD, Anderson DC, et al. Use
US and Canadian hospitals' paediatric massive of deglycerolizedred blood cells to prevent post-
transfusion protocol policies. Transfus Med transfusion infection with cytomegalovirus in
2016;26:49-56. neonates. J Infect Dis 1984;150:334-9.
148. Hendrickson JE, Shaz BH, Pereira G, et al. Im- 160. Heddle NM, Boeckh M, Grossman B, et al for
plementation of a pediatric trauma massive the Clinical Transfusion Medicine Committee.
transfusion protocol: One institution's experi- AABBcommittee report: Reducing transfusion-
ence. Transfusion2012;52: 1228-36. transmitted cytomegalovirusinfections. Transfu-
149. Nosanov L, Inaba K, Okoye 0, et al. The impact sion 2016;56: 1581-7.
of blood product ratios in massively transfused 161. Trial to Reduce Alloimmunization to Platelets
pediatric trauma patients. Am J Surg 2013; study group. Leukocyte reduction and ultravio-
206:655-60. let B irradiation of platelets to prevent allolm-
150. Hwu RS, Spinella PC, Keller MS, et al. The ef- munizatlon and refractoriness to platelet trans-
fect of massive transfusionprotocol implementa- fusions. N EnglJ Med 1997;337:1861-9.
tion on pediatric trauma care. Transfusion2016; 162. Sanders MR, Graeber JE. Posttransfusion graft-
56:2712-19. versus-host disease in infancy. J Pediatr 1990;
151. Chidester SJ,WilliamsN, WangW, et al. Apedi- 117:159-63.
atric massive transfusion protocol. J Trauma 163. Rtihl H, Bein G, Sachs UJH. Transfusion-associ-
Acute Care Surg 2012;73:1273-7. ated graft-versus-hostdisease. TransfusMed Rev
152. Dehmer JJ, Adamson WT. Massive transfusion 2009;23:62-71.
and blood product use in the pediatric trauma 164. Strauss RG, Levy GJ, Sotelo-AvilaC, et al. Na-
patient. Semin Pediatr Surg 2010; 19:286-91. tional survey of neonatal transfusion practices:
153. HolcombJB, TilleyBC, Baraniuk S, et al. Trans- II. Blood component therapy. Pediatrics 1993;
fusion ofplasma, platelets, and red blood cells in 91(3):530-6.
a 1:1:1 vs a 1:1:2 ratio and mortality in patients 165. MoroffG, FriedmanA, Robkm-KlineL, et al. Re-
with severe trauma: The PROPPRrandomized duction of the volume of stored platelet concen-
clinicaltrial. JAMA2015;313:471-82. trates for use in neonatal patients. Transfusion
154. The CRASH-2collaborators. The importance of 1984;24: 144-6.
early treatment with tranexamic acid in bleed- 166. Schoenfeld H, Muhrn M, Doepfmer UR, et al.
ing trauma patients: An exploratory analysis of The functional integrity of platelets in volume-
the CRASH-2randomised controlled trial. Lan- reduced platelet concentrates. Anesth Analg
cet2011;377(9771):1096-101,1101.el-2. 2005; 100:78-81.
155. Oakley FD, Woods M, Arnold S, Young PP. 167. Knutson F, OsselaerJ, PterelliL,et al. A prospec-
Transfusion reactions in pediatric compared tive, active haemovigilance study with com-
with adult patients: A look at rate, reaction type, bined cohort analysis of 19,175 transfusions of
and associated products. Transfusion 2015;55: platelet components prepared with amotosalen-
563-70. UVAphotochemical treatment. Vox Sang 2015;
156. Bolton-MaggsPHB,ed, et al on behalf of the Se- 109:343·52.
rious Hazards of Transfusion (SHOT) Steering 168. McCullough J, Vesole DH, Benjamin RJ, et al.
Group. The 2017 annual SHOT report. Man- Therapeutic efficacyand safety of platelets treat-
chester, UK: SHOT Office, 2018. [Availableat ed with a photochemical process for pathogen
h ttps:/ /www.shotuk.org/shot-reports/ (ac- inactivation: The SPRINT trial. Blood 2004;
cessed December 5, 2019).J 104:1534-41.
157. Bowden RA,Slichter SJ,SayersM, et al. A com- 169. Estcourt LJ, Malouf R, Hopewell S, et al.
parison of filtered leukocyte-reduced and cyto- Pathogen-reduced platelets for the prevention of
bleeding. Cochrane Database Syst Rev 2017;7: nent utilization and patient safety in a large Aus-
CD009072. trian RegionalMedical Centre. Vox Sang 2017;
170. Knutson F, Osselaer J, PierelliL, et al. A prospec- 112:47-55.
tive, active haemovigilance study with com- 172. Trakhtman P, Karpova0, BalashovD, et al. Effi-
bined cohort analysis of 19,175 transfusions of cacy and safety of pathogen-reduced platelet
platelet components prepared with amotosalen- concentrates in children with cancer: A retro-
UVA photochemical treatment. Vox Sang 2015; spective cohort study. Transfusion 2016;56(Sup-
109:343-52. pl 1):S24-8.
171. Amato M, Schennach H, Astl M, et al. Impact of
platelet pathogen inactivation on blood compo-
Therapeutic Apheresis
Jennifer Webb, MD, MSCE, and Chester Andrzejewski lr, PhD, MD
HERAPEUTICAPHERESIS(TA) IS AN N R RI
extracorporeal therapy used in the treat-
ment and management of various diseas- The major goals of TA are 1) to remove a patho-
es and is accomplished either through the re- logic cellular.,and/or humoral element(s) from a
moval and discard of selected blood constituents patient's blood; 2) to replace a deficient sub-
or via the collection of selected blood elements stance(s), typicallyin the context of the removal
and subsequent ex-vivo manipulation and re- of a pathologic constituent; 3) to modulate cellu-
turn of those elements to the patient It is relat- lar functionality (eg, via further extracorporeal
ed to but distinct from "preparative apheresis," manipulations and the return of those cellular
in which blood components are collected from elements, such as performed with the use of ul-
blood donors for therapeutic uses (covered in travioletlight exposure in photopheresis); and/
Chapter 6). Standards and guidelinesfor TA per- or 4) to collect various autologous nonpatholog-
formance, health-care provider education, quali- ic cellular populations for further manipulation
fications and clinical privileging,and procedural and therapeutic uses (eg, cellular therapies in-
documentation have been promulgated by volving stem cells, dendritic cells, and chimeric
AABB, the American Society for Apheresis antigen receptor Tcell production).
(ASFA), and the College of American Patholo- The various types of apheresis procedures are
gists (CAP).I-7In 2012, the National Heart, listed in Table 25-1. In the most commonly per-
Lung, and Blood Institute (NHLBI)sponsored a formed clinical apheresis procedure, therapeu-
National Institutes of Health (NIH) State of the tic plasma exchange (TPE), 1.0 to 1.5 plasma
Science Symposium in Therapeutic Apheresis volumes are typically removed in a single ses-
exploring various aspects related to TA, includ- sion. A series of treatments may be necessary to
ing scientificopportunities in TA,what challeng- achieve the desired clinical effect Targeting of
es/barriers exist to increasing TA research, and larger volumes for removal increases the risk of
prioritization of TAresearch activities+" coagulopathy, citrate toxicity, or electrolyte
Jennifer Webb, MD, MSCE, Assistant Professor of Pediatrics, Division of Pediatric Hematology, George Washing-
ton University, School of Medicine and Health Sciences, and Director of the Therapeutic Apheresis Program,
Children's National Hospital, Washington, DC; and Chester Andrzejewski Ir, PhD, MD, Medical Director, System
Blood Banking/Transfusion and Apheresis Medicine Services, Baystate Health, Assistant Professor, Department of
Pathology, University of Massachusetts Medical School-Baystate, Springfield, Massachusetts, and Adjunct Assis-
tant Professor, Department of Pathology, Tufts University School of Medicine, Boston, Massachusetts
J. Webb has disclosed no conflicts of interest. C. Andrzejewski has disclosed financial relationships with Amgen
and Haemonetics.
705
TABLE 25-1. Therapeutic Apheresis Modalities
Blood
Procedure Removed
Therapeutic plasma Plasma Reduction of an abnormal plasma Albumin, crystal-
exchange protein (eg, autoantibody) loid, or plasma
Redcell exchange Redcells Sickle-cell-disease-related Red Blood Cells
complications
Leukocytapheresis Buffy coat Leukemiawith leukostasis As needed
Thrombocytapheresis Buffy coat Thrombocytosis As needed
Erythrocytapheresis Redcells Erythrocytosis As needed
Extracorporeal Buffy coat (reinfused) Graft-vs-host disease, None
photopheresis CutaneousT-cell lymphoma
Selectiveadsorption Specific plasma protein, Hypercholesterolemia None
antibody, or lipid
Adsorptive cytapheresis Activated monocytes or Inflammatory bowel disease None
granulocytes
Rheopheresis High-molecular-weight Age-related macular None
plasma proteins degeneration
Double-filtration Pathogenicsubstances Atopic dermatitis None
plasmapheresis of a particular size or
molecular weight
imbalances, depending on the replacement flu- space, or recirculation at the openings of a dou-
ids used." In plasmapheresis, used generally in ble-lumen catheter will result in a less-than-
the donor setting although the term is often predicted decrease. In a I.O-plasma-volumeex-
used interchangeably with TPE, lesser volumes change, approximately two-thirds of a targeted
of plasma (::s: I L) are removed, resulting in little "idealized substance,,,12such as immunoglobu-
to no need for fluidreplacement. lin M (IgM)or fibrinogen, is typicallyremoved,
The effectiveness of apheresis in removing provided that the constituent does not diffuse
pathologic substances depends on the concen- Significantlyfrom the extravascular to the intra-
tration of the substance in the blood, its volume vascular compartments.
of distribution into the extravascular space, the
degree of protein-binding of the targeted sub-
stance, the volume of blood processed and plas- VI LITI
ma removed, and the equilibrium between the
substance's blood and extravascular concentra- Several technological platforms for performing
tions. Most efficient at the procedure's begin- apheresis exist. The most commonly available
ning stages, component removal decreases ones are described below.
exponentially and asymptoticallywith time be- Continuous-flow centrifugation apheresis de-
cause of the required addition of replacement vices have a rotating channel designed to intro-
fluidsneeded to maintain physiologicstabilityin duce whole blood at one site. The blood ele-
the patient. Continued production of the sub- ments will subsequently separate largely by
stance, mobilization from the extravascular density as blood flows through the channel. The
C HAP T E R 2 5 TherapeuticApheresis 707
resulting layers of plasma, platelets, leukocytes, are available. In adsorptive cytapheresis, also
and red cells can be selectively removed. The known as granulocyte-monocyte apheresis (GMA),
targeted constituent is diverted into a collection monocyt~s and granulocytes are selec~vely re-
bag, while the remaining blood components are moved from whole blood through adhesion or
returned to the patient with appropriate replace- filtration. Inactivated leukocytes and the re-
ment fluids. To achieve optimal separation and maining blood components are returned to the
treatment blood volume, modern apheresis de- patient. Thi~ technique has been used to treat
vices calculate and control blood-withdrawal autoimmune, inflammatory conditions such as
flow rates, amounts of anticoagulant solution ulcerative colitis and psoriasis.
and replacement fluids needed, as well as centri-
fuge speed.
Intermittent-flow centrifugation devices pro- N 0
cess smaller volumes of blood in cycles, where a
cycle consists of whole blood being drawn and
separated, and then components are reinfused All patients should be evaluated by a physician
or diverted. Multiple cycles are often required to familiar with apheresis before beginning any
achieve the goals of treatment. The volume of course of TA. General considerations in formu-
whole blood drawn. in a cycle is targeted to lating an overall therapy plan are identified in
achieve a specific red cell threshold. This leads Table 25-2. A more comprehensive evaluation
to larger extracorporeal volumes achieved than of the patient is typically performed in this set-
in continuous-flow devices. ting, with the clinicalevaluatton focusing on se-
Filtration-based apheresis devices also oper- lect elements at subsequent TPE encounters in
ate by continuous flow. Anticoagulated whole
stable patients receiving an extended series of
blood is passed through a microporous filter that
treatments, unless changes in the patient's clini-
allows plasma to pass through while retaining
cal status dictate otherwise.
the blood cells. The separated plasma can then
The indication, type of procedure, choice of
be diverted into a waste bag or, as with selective
replacement fluid (Table 25-3), vascular access,
adsorption, further processed and returned to
frequency and number of treatments, and the
the patient. This type of device is not suitable for
treatment goal or endpoint should be document-
the removal of select cellular components in cir-
culation (cytapheresis). Variants of this tech-
ed in the patient's record. During the initial eval-
nique are rheopheresis or double-filtration plas-
uation, the nature of the procedure and its ex-
mapheresis (DFPP), which further processes the pected benefits, possible risks, and available
separated plasma through a secondary filter that alternatrves must be explained to the patient,
removes high-molecular-weight molecules, and informed consent must be documented. De-
thereby reducing plasma viscosity, or targets a pending on institution-specific policies, these in-
specific component. Secondary filters of differ- formed consent discussions should be repeated
ent pore sizes remove different pathogenic sub- periodiSally (eg, annually) and documented in
stances, such as immune complexes, auto anti- the medical record. Procedures should be per-
bo~i~s, or lipoproteins. . formed only in settings where there is ready ac-
·In selective adsorption, blood or plasma ..is cess to care fo!. untoward reactions, including
passed over a column or matrix that has a high equipment, medtcatlons, and personnel trained
affinity for a specific component, such as immu- in managing serious adverse events, such as ana-
noglobulin G (IgG) in immunoadsorption (IA) or phylaxls, metabolic alkalosis, air embolus, and
lipoproteins in lipoprotein apheresis (LA). The hypotensive reactions. Comorbidities that may
effluent is returned to the patient after being re- affect the patient's ability to tolerate apheresis
united with cellular components. This approach and concurrent medication reconciliation
has the advantage of highly specific removal of should be noted. Other specificpoints to consid-
the target of interest. However, it is restricted to er when evaluating a patient include the follow-
the few conditions for which affinity adsorbents ing:
TABLE 25-2. Elements* in a Comprehensive Evaluation and Treatment Plan for Patients Being
Considered for TA Interventions
- - "~ =
Clinical diagnosis and differential diagnosis / chief complaint / clinical objective / reason for referral
History of the pertinent present illness
Patient's past / family medical histories, including prior TA and hemotherapies (eg, blood transfusions and
IVIG infusions) and history of reactions, comorbidities
Pertinent review of systems and medication reconciliation
Pertinent physical examination, including vascular access assessment
TA indication / rationale / outcome objectives
Selection of apheresis modality and equipment
Volume / duration / frequency targets / fluid balance
Per procedure
For entire series
Vascular access device considerations as needed
Choice of replacement fluids
$ Blood prime, if required
$ Specially prepared units (eg, antigen negative,phenotyped), if applicable
Physiologic / adverse reaction / clinical outcomes monitoring
Coordination of pre- / intra- / post- / and interprocedurallaboratory testing, drug / blood component adminis-
tration, and other procedures (eg, dialysis)
Use of ancillary equipment prn
Blood warmers
Fetalmonitors
Ventilatory support devices
$ SeparateIV accessfor calcium infusion
Patient educational materials and discharge instructions
Needfor initial series extension / concurrent immunosuppressive therapies
Coagulopathies/ anticoagulation medication when relevant
*Identified elements are typically considered in the clinical evaluation at the time of the initial consultation of all patients
being assessedfor TA. In stable individuals undergoing an extendedseries of TA, apheresis medicine specialists may elect
to exclude select aspects from the patient's evaluation at each subsequent apheresis session (if such elements were
obtained at time of initial consultation) unless changesin the clinical status of the patient warrant otherwise.
TA = therapeutic apheresis; IVIG = intravenous immune globulin; prn = as needed.
$ Transfusion/apheresis history: Notation Neurologic status: Mental status and the

of prior transfusion/apheresis interventions ability to consent and cooperate.
and reactions, outcomes from such therapies, Cardiorespiratory status: Adequate venti-
and speciaiblood component requirements. lation and oxygenation capabilities,presence
TABLE25-3. Comparison of Replacement Fluids
Crystalloids Low cost Two- to threefold more volume

Nonallergenic required
No viral risk Hypo-oncotic
Lacking coagulation factors and
immunoglobulins
Albumin Iso-oncotic Higher cost
Low risk of reactions Lacking coagulation factors and
immunoglobulins
Plasma Iso-oncotlc Viral transmission risk
Normal levels of coagulationfactors, Increased citrate load
immunoglobulins, and other plasma ABO compatibility required
proteins Higher risk of allergic reactions
Cryoprecipitate-poor lso-oncotlc Sameas plasma
plasma Reduced high-molecular-weight von
Willebrand factor and fibrinogen
Normal levels of most other plasma
proteins
of hyper- or hypovolemia, and any cardiac ar- fectious diseases and pertinent specific disease
rhythmias. markers, should be collected before the first
Renal and metabolic status: Fluid balance, treatment. Coagulation monitoring may be ap-
alkalosls, and electrolyte abnormalities, in- propriate when albumin is used as the primary
cluding hypocalcemia, hypokalemia, and hy- replacement fluid during frequently repeated
pomagnesemia. procedures. The collection of an "archive speci-
Hematologic status: Presence of clinically men" before the initial apheresis session should
significant anemia, thrombocytopenia, coag- also be considered, because concentrations of
ulopathy, bleeding, or thrombosis. plasma constituents are altered by TA. Similarly,
Medications: Recently administered intra- apheresis medicine practitioners should make
venous immunoglobulin (IVIG)and antibody colleagues and patients aware that diagnostic
biologics, angiotensin-converting enzyme testing after apheresis can be differentiallyaffect-
(ACE)inhibitors, drugs with high albumin- ed depending on the analyte of interest, the re-
binding properties, and anticoagulants. placement fluids used, and the frequency of TPE
sessions.
Appropriate laboratory monitoring is guided
by the TA indication, type and frequency of pro-
R
cedures, duration of therapeutic course, and
concomitant medical conditions. In general, it is TA requires excellent vascular access to achieve
wise to obtain test results, including a complete adequate flow rates." Peripheral access is pre-
blood cell count, blood type and antibody detec- ferred and generally requires at least a 17-gauge
tion test (antibody screen), coagulation profile, needle for blood withdrawal and at least an 18-
and electrolytes, before starting treatment. Oth- gauge catheter for return. Patients without ade-
er diagnostic study results, such as tests for in- quate peripheral veins, those who require multi-
ple frequent procedures, and those unable to access an AVfistula.AVfistulasincrease demand

provide hand compressions to maintain vascular on cardiac output and should therefore be per-
flow may require placement of central venous formed only in selected patients with sufficient
catheters (CVCs).CVCsfor apheresis must have cardiovascularreserve.":"
rigid walls to accommodate the negative pres- The choice of placement site for a CVCis in-
sure generated in the withdrawal line. Dual- fluenced by the anticipated duration of treat-
lumen catheters similar to the types used in he- ment. Subclavian or internal jugular access is
modialysis are preferred, although single-lumen generally preferable for TPE courses lasting up
catheters can be used for intermittent-flow pro- to several weeks. Femoralvein access should be
cedures. For a short series of procedures, a non- used only temporarily because of the higher risk
tunneled catheter may be adequate; however, of infection. Patients requiring extended treat-
for long-term access, tunneled and cuffed CVCs ment courses usually have tunneled CVCplace-
allow for stable, safe, durable points of apheresis ment. With proper care, tunneled catheters can
access.lv" Compared to nontunneled catheters, be used for prolonged periods. Confirmation of
tunneled CVCs have lower rates of infection CVCplacement, usuallyperformed radiographic-
and may be suitable for outpatient procedures ally, should be obtained before initiation of the
over months to years. first apheresis session and whenever a CVC re-
Peripherally inserted central catheters (ie, placement occurs.
PICC lines) typically do not support the high Good catheter care is very important for pa-
flow rates used in adult apheresis; however, for tient safety and to maintain CVC patency. Re-
single points of access or pediatric patients who sponsibility for CVC maintenance needs to be
typicallyuse slower flow rates, PICC lines have clearly defined, and apheresis medicine services
been successfullyused as an alternative CVCop- need to be actively involved in either directly
tion. One device that has been successfullyused providing this care or in its coordination with
in pediatric patients receiving extracorporeal other parties involved in treating the patient.
photopheresis (ECP) is the 5-fr, single-lumen, Catheters need to be flushed regularly. Heparin
Turbo-flo(Cook Medical) PICC.!6It offers a sin- (10-1000 U/mL) or 4% trisodium citrate is usu-
gle point of access of moderate duration (6-12 ally placed in each lumen after each use to pre-
months). Like other dual-lumen CVCs, there is vent occlusion by clots. If a port becomes clot-
no needlestick when connecting to the aphere- ted, instillation of a fibrinolytic agent, such as
sis equipment because the hub is external; how- urokinase or recombinant tissue plasminogen
ever, there are additional home-care needs with activator, may restore patency. Routine dresstng
a PICC line, as many require daily flushing and care is essential to prevent insertion-site infec-
frequent cap changes. tions.
Implanted subcutaneous ports, also known Venous access devices may cause thrombo-
as totally implantable venous access devices sis. Infrequently, their placement may result in
(TIVADs),may provide an option for some pa- severe complications, such as pneumothorax,
tients requiring long-term treatment, such as arrhythmia, or perforation of the heart or great
chronic red cell exchange for complications of vessels, occurring with up to 1% of proce-
sickle cell anemia. Procedures using double- dures." Other complications such as arterial
lumen ports tend to be longer, often last for puncture, deep hematomas, and AV formation
years, and can provide stable access with few are more frequent, occurring with 1% to 3% of
lifestyle limitations; however, they have more procedures." Bacterial colonization may lead to
procedural complications than procedures us- catheter-associated bloodstream infection, espe-
ing other CVCS.!7An arteriovenous (AV)fistula cially in immunosuppressed patients. Inadver-
may also be used for apheresis, but personnel tent disconnection of catheters may produce
should be suitably trained before attempting to hemorrhage or an air embolism.
ANTI ON TABLE25-4. Reported Frequency of Adverse

Reactions to Apheresis*
Acid-citrate-dextrose solution A (ACD-A)is the
most commonly used anticoagulant, although
heparin combined with ACD-Ais also used, par-
ticularly in the setting of large-volume leukocy-
tapheresis for hematopoietic progenitor cell or Access/device 1.56
mononuclear cell collection in small patients Hypotension 0.77
who may not be able to metabolize large vol- Urticaria 0.63
umes of citrate.22,23 Heparin anticoagulation is
necessary for LAand may be desirable for select- Nausea/vomiting 0.23
ed patients undergoing TPEwho are particularly Shivering 0.11
susceptible to hypocalcemia, such as small chil- Flushing 0.09
dren, or in the setting of severe .metabolicalka-
losis, liver failure, or renal failure, Heparin is Arrhythmia 0.03
also the standard anticoagulant for ECP.With ci- Anaphylaxis 0.022
trate anticoagulation, coagulation monitoring is Vertigo 0.01
generally not necessary, although knowledge
and monitoring of ionized calcium levels may be Abdominal
:. "'.. pain
' .
0.008
helpfulin selected patients. In patients with nor- Other (includes seizures, 0.39
mal hepatic function, the infused citrate is rapid- back pain, hypertension,
ly metabolized and rarely causes systemic anti- etc)
coagulation. Total 5.8
·Adapted from Mortzell-Henrikssonet al.27
A s
Although TA is very safe, complications do oc-
cur; adverse events, mostly mild, occur in ap- tinuous intravenous (IV) calcium supplementa-
proximately.4% to 6% of procedures (Table 25- tton." Metabolism of citrate to bicarbonate
4).24,27 Other than adverse events related to vas- leads to a mild metabolic alkalosis, which can
cular access, symptomatic hypocalcemia due to exacerbate hypocalcemia and may cause hypo-
citrate anticoagulation is the most common ad- kalemia."
verse effect, typically characterized by perioral Allergic reactions are most common with
and digital paresthesiae. Nausea or other gastro- plasma replacement, although they may also oc-
intestinal symptoms may also occur, although cur with albumin." Most reactions are mild,
tetany is rare. Cardiac.arrhythmia is very rare, characterized by urticaria or cutaneous flushing.
Severe allergicreactions can involve the respira-
but patients with preexisting hypocalcemia or
tory tract with dyspnea, wheezing, and (rarely)
significant prolongation of the. QT interval stridor. Most allergic reactions respond quickly
should be monitored carefully. Calcium supple- to IVdiphenhydramine. Anaphylaxisis very rare
mentation may alleviate symptoms of citrate but can occur. Patients receiving large volumes
toxicity; a typical dose is 1 g of calcium gluco- of plasma, such as those with thrombotic throm-
nate per liter of albumin infused. Citrate also bocytopenic purpura (TTP),may be most at risk
chelates ionized magnesium, so. it is possible for allergic reactions." Premedication with an
that hypomagnesemia contributes to the symp- antihistamine, or possiblysteroids, is not neces-
toms of citrate toxicity; however, one random- sary for routine apheresis but may be indicated
ized clinical trial showed no benefit of adding for patients with repeated or previous severe
magnesium during leukocytapheresis with con- reactions. Solvent/detergent-treated, pooled
donor plasma has also been used as replacement When plasma is exposed to foreign surfaces
fluid successfullyin patients with a history of se- such as plastic tubing or filtration devices, the
vere allergic reactions to conventional plasma, kinin system can be activated, resulting in pro-
though experience is limited.32,33 duction of bradykinin. Infusion of plasma con-
Respiratory difficulty during or immediately taining bradykinin can cause abrupt hypoten-
following apheresis can have many causes, such sion. Patients taking ACE inhibitors are more
as pulmonary edema, pulmonary embolism, air susceptible to these hypotensive reactions be-
embolism, or leukostasis. If using plasma as the cause these drugs block the enzymatic degrada-
replacement fluid, transfusion reactions such as tion of bradykintn." Hypotensive reactions are
transfusion-related acute lung injury (TRALI), more likely during selective adsorption proce-
anaphylaxis, or transfusion-associated ctrculato- dures because these devices expose plasma to a
ryoverload (TACO)are possible." Hemothorax very large surface area. Holding ACE inhibitors
or hemopericardium resulting from vascular ero- for 24 to 48 hours ahead of procedures or transi-
sion from a CVC is rare but may be fatal." Pul- tioning the patient to an angiotensin-receptor
monary edema due to volume overload or cardi- blocker may decrease the frequency of reaction;
ac failure is usually associated with dyspnea, an however, because some ACE inhibitors have a
increase in diastolic blood pressure, and charac- long duration of action, medications should be
teristic chest radiograph findings. Predominantly reviewed to ensure they are stopped with
ocular reactions (periorbital edema, conjunctival enough time to avoid unnecessary adverse reac-
swelling, and tearing) have occurred in individu- tions."
als sensitized to the ethylene-oxide gas used to Intensive TPE without plasma replacement
sterilize disposableplastic apheresis kits.36,37 depletes coagulation factors. A l.Oplasma-
Hypotension during apheresis can be' a sign volume exchange typically reduces coagulation
of citrate toxicity, hypovolemia, or a vasovagal, factor levels by 25% to 50%, although most fac-
allergic, drug, or transfusion reaction. Hypovole- tor levels recover quickly", levels of fibrinogen,
mia can occur early in treatments of small pa- a large intravascular molecule, are reduced by
tients when the return fluid consists of the sa- -66%. If the patient has normal hepatic synthet-
line used to prime the apheresis circuit. ic function, coagulation factor levels typicallyre-
Vasovagalreactions are characterized by brady- turn to near normal within I to 2 days." Thus,
cardia and hypotension. Such reactions usually many patients can tolerate TPE every other day
respond well to a fluid bolus and placing the pa- for 1 to 2 weeks without developing significant
tient in the Trendelenburg position. coagulopathies requiring plasma replacement.
When hypotension occurs during plasma or Bleeding due to coagulation factor depletion
red cell exchanges,potential transfusionreactions is rare. For patients at risk, plasma may be used
such as acute hemolysis, bacterial contamina- for replacement at the end of the procedure.
tion, or anaphylaxisshould be considered. Hypo- Apheresis can also cause thrombocytopenia. In-
tension is more frequent in children, the elderly, tensive TPE can cause hypogammaglobulin-
neurology patients, anemic patients, and those emia, which may affect the accuracy of subse-
treated with intermittent-flow devices that have quent serologic testing. Serum levels of IgG and
large extracorporeal volumes. Continuous-flow IgM recover to 40% to 50% of the preapheresis
devices typicallydo not have large extracorpore- level by 48 hours." In addition, WIG or subcu-
al volumes but can produce hypovolemia if re- taneous IgG is often used to treat conditions
turn flow is inadvertently diverted to a waste that respond to TPE and is often administered
collection bag, either through operator oversight following completion of a series of TPE proce-
or device malfunction. Hypovolemia may also dures. The absolute immunoglobulin level at
be secondary to inadequate volume or protein which a patient becomes at risk for infection has
replacement. During all procedures, it is essen- not been established.
tial to carefully document and maintain records Albumin-bound drugs are removed by TPE.
of the volumes processed, removed, and re- This can produce subtherapeutic levels of medi-
turned. cations unless dosages are adjusted. Biologic
C HAP T E R 25 TherapeuticApheresis 713
therapeutics, such as IVIG, antithymocyte glob- Few randomized studies have been performed
ulin, and monoclonal antibodies, have a long in- in pediatric patients receiving therapeutic apher-
travascular half-life and are readily removed by esis, and much of the evidence is extrapolated
apheresis. Drugs should be administered follow- from adult data, single-centerexperiences, anec-
ing a TPE procedure to avoid impairing their ef- dotal reports, or expert opinion, thus highlight-
fectiveness. ing the need for future research in this fleld."
Collapsed or kinked tubing, malfunctioning
pinch valves, or improper threading of tubing
may damage red cells in the extracorporeal cir- THERAP,Eunc APHERESiS
cuit. Instrument-related hemolysis occurs in INDICATiONS
0.06% of TA procedures." Hemolysis can also
occur with the use of hypotonic replacement Although there are many case reports of suc-
fluids or ABO-incompatible plasma. The opera- cessful treatment of various diseases and condi-
tor should carefully observe plasma collection tions by apheresis, there are few high-quality,
lines for pink discoloration suggestive of hemol- prospective, clinical trials. ASFAhas published
ysis. Other types of equipment failure, such as evidence-based clinical guidelines categorizing
problems with the rotating seal, leaks in plas- treatment lndications.' The classifications used
ticware, and roller pump failure, are rare. in the guidelines are defined as follows:
Fatalities during apheresis are rare and are re-
portable to the Food and Drug Administration • Category I: Disorders for which apheresis is
(FDA).42 Venous thromboembolism (VTE), in- accepted as first-linetherapy, either as a pri-
cluding pulmonary emboli (PE), shortly after or mary stand-alone treatment or in conjunc-
during ECP has been reported to the FDA, and tion with other modes of treatment.
an investigation is ongoing." Historically, death Category II: Disorders for which apheresis
has been reported in 0.006% to 0.09% of thera- is accepted as second-linetherapy, either as a
peutic procedures, with most fatalities attribut- stand-alone treatment or in conjunction with
able to underlying medical conditions.2s,44,4s other modes of treatment.
Category III: Disorders in which the opti-
mal role of apheresis therapy is not estab-
PEDIATRiC APHERES6S Iished. Decision-making for patients should
be individ11alized.
Although apheresis is safe and generally well Category IV: Disorders in which published
tolerated in pediatric patients, special attention evidence demonstrates or suggests that
should be paid to this population. Given the apheresis is ineffective or harmful. Institu-
relatively small total blood volumes (TBVs)of tional review board approval is desirable if
children, priming of the machine with blood or apheresis treatment is undertaken in these
albumin may be necessary to avoid excessive, circumstances.
unsafe extracorporeal volume. In addition, pedi-
atric patients may be sensitive to rapid fluid Fuller discussions of the conditions treated
shifts, so significantlyslower flow rates are used. with Tkmay be found by consulting more com-
Vascular access devices are often necessary due prehensive medical texts on the respective dis-
to inadequate peripheral access or inability to orders. An overview of some of the more com-
tolerate the duration of the procedure peripher- monly encountered diseases treated by TA is
ally because of the requirement to stay still. Be- provided below.
cause of the slow flow rates, devices and config-
urations may differ from cve use in adults,
Theriillpil!~tic Plasma Exchange
although generally they must be apheresis or
hemodialysis capable. Often, empiric calcium In both alloimmune and autoimmune disorders,
and/or magnesium supplementation is used to pathogenic factors, that are circulating in the
avoid toxicity from citrate antlcoagulation.w" blood are targets for removal by TPE. Autoanti-
body-mediated diseases include acute and not been established.f TPEresults in greatly im-
chronic inflammatory demyelinating polyradicu- proved survival in TTP patients; however, treat-
loneuropathies (AIDP and CIDP), antiglomeru- ment failures do occur.56,57The use of cryopre-
lar basement membrane disease, and myasthe- cipitate-poor plasma, which is deficient in vWF,
nia gravis. Examples of conditions where the for replacement fluid has not been shown to im-
goal is to remove problematic alloantibodies in- prove clinical response or outcomes." although
clude renal transplantation with presensitization it is often tried in refractory TTP cases. Adjunc-
and antibody-mediated organ transplant rejec- tive therapies including rituximab and immuno-
tion. Diseasesin which immune complexes may suppression are often used to augment treat-
be pathogenic and can be removed by apheresis ment and prevent relapse. In a Phase III trial,
include rapidly progressive glomerulonephritis, caplacizumab, an anti-vWF immunoglobulin
cryoglobulinemia, and vasculitis. Other indica- fragment (nanobody) in conjunction with TPE
tions include conditions treated by removing increased the rate of recovery, decreased recur-
protein-bound drugs, toxins, or high concentra- rence rates, and decreased mortality in patients
tions of lipoproteins. In addition, there is emerg- with acquired TTP when compared to placebo,
ing evidence on the immunomodulatory effects leading to its approvalby the FDA.59
of TPE, which may be partially responsible for Hemolytic uremic syndrome (HUS), also
the clinical effects seen in autoimmune dlsor- called infection-associated thrombotic microan-
ders." IndicationsforTPEare listedin Table25-5. giopathy (TMA), is a similar condition that oc-
In TTP, an absolute or functional deficiency curs more commonly in children than in adults
of the von Willebrand factor (vWF)-cleaving and is often confused for TTP. HUS may follow
metalloprotease ADAMTS13 results in the accu- diarrheal infections with verotoxin-secreting
mulation ofhigh-molecular-weightvWF multim- strains of Escherichia coli (straln 0157:H7) or
ers, with subsequent intravascular platelet Shigella.Compared to patients with classicTTP,
activation and platelet-rich thrombi formation in those with HUS have more renal dysfunction
the microvasculature." In many cases, an inhib- and less prominent neurologic and hematologic
itor of ADAMTS13can be demonstrated. TPEis findings. Although diarrhea-associated HUS
first-line treatment for TTP, with the goal of re- rarely responds to TPE, atypical HUS (aHUS)
moving both the inhibitor and large vWF multi- may respond (also called complement-mediated
mers while simultaneously replacing the defi- TMA, caused by complement factor deficien-
cient enzyme through replacement with Fresh cies or autoantibodies to Factor H); however, pa-
Frozen Plasma (FFP). Early identification and tients with aHUS caused by membrane cofactor
treatment of patients with severe ADAMTS13 protein (MCP or CD46) mutations may not re-
deficiency is critical in this illness, which has a spond to TPE. Eculizumab, a monoclonal anti-
90% mortality rate when left untreated. Early body directed against C5a, thereby inhibiting
initiation of TPE reduces mortality to <15%.50 terminal complement activation, has been
The PLASMICscore, a validated tool based on shown to be effective in treating individuals
readily available clinical parameters, has been with aHUS.60Treatment with eculizumab has
shown to discriminate between TTP and other been shown to prevent progression to renal
thrombotic microangiopathies with high sensi- transplantation and dialysis dependence in pa-
tivity and reliability, and may serve as a useful tients with aHUSand is the preferred therapy in-
tool in clinical practice.T" TPE is typicallyper- stead of prolonged TPEwithout signsof renal re-
formed dally until the platelet count and lactate covery.6!,62
dehydrogenase level normalize, often with the Secondary forms of microangiopathic hemo-
former being >150,000/pL for 48 hours, but lytic anemia (MAHA),associated with systemic
treatment duration should be guided by the in- lupus erythematosus, cancer, hematopoietic pro-
dividual patient's course. After response has genitor/stem cell transplantation, chemothera-
been achieved, intermittent apheresis or plasma py, or immunosuppressive medications, may be
infusion tapers may be instituted, but the effica- clinicallyindistinguishable from classicalTTP.In
cy of these approaches in preventing relapse has many of these cases, however, ADAMTS13
TABLE 25-5. Indications for Therapeutic Plasma Exchange5
Acute disseminated encephalomyelitis Steroid refractory II QOD (3-6)

Acute inflammatory demyelinating Primary treatment QOD (5-6)
polyradiculoneuropathy (Guillain-Barre
syndrome)
Acute liver failure High-volume TPE QD (3)
Routine TPE III QD (variable)
Amyloidosis, systemic IV
Antiglomerular basement membrane DAH I QD or QOD
disease (Goodpasture syndrome) (variable)
Dialysis independence
Dialysis dependence, no DAH III
Atopic (neuro-) dermatitis (atopic eczema), III Weekly
recalcitrant (variable)
Autoimmune hemolytic anemia, severe Severe cold agglutinin disease II QD or QOD
(variable)
Severe WAIHA III
Burn shock resuscitation III 1-3 within 24-36
hours
Cardiac neonatal lupus III 3/week to
monthly
Catastrophic antiphospholipid syndrome QD or QOD
(variable)
Chronic focal encephalitis (Rasmussen III QOD (3-6)
encephalitis)
Chronic inflammatory demyelinating 2-3/week
polyradiculoneuropathy
Coagulation factor inhibitors III QD (variable)
Complex regional pain syndrome Chronic III QOD (5-7)
Cryoglobulinemia Symptomatic/severe II QOD (3-8)
Dilated cardiomyopathy, idiopathic NYHA II-IV III QOD (5)
Erythropoietic protoporphyria, liver disease III QOD (variable)
Familial hypercholesterolemia Homozygotes/heterozygotes II Every 1-2 weeks
(Continued)
TABLE 25·5. Indications for Therapeutic Plasma Exchanqe" (Continued)
Focalsegmentalglomerulosclerosis Recurrent in transplanted co or aOD

kidney (variable)
Steroid resistant in native III
kidney
HELLPsyndrome Postpartum III aD (variable)
Antepartum IV
HemophagocyticIymphohistiocytosis; III co (variable)
hemophagocytic syndrome; macrophage-
activating syndrome
Heparin-inducedthrombocytopenia and Beforecardiopulmonary III oo or aOD
thrombosis (HIT/HITT) bypass (variable)
Thrombosis III
Hypertriglyceridemic pancreatitis Severe III aD (1-3)
Preventionof relapse III
Hyperviscosity in Symptomatic co (1-3)
hypergammaglobulinemia Prophylaxisfor rituximab co (1-2)
Immune thrombocytopenia Refractory III aOD (6)
IgA nephropathy(Berger disease) Crescentic III aOD (6-9)
Chronic progressive III
Lambert-Eatonmyasthenic syndrome II co or aOD
(variable)
Multiple sclerosis Acute attack/relapse II aOD (5-7)
Chronic III Variable
Myastheniagravis Acute, short-term treatment I on or aOD
Long-term treatment (variable)
II
Myeloma cast nephropathy II aOD (10-12)
Nephrogenicsystemic fibrosis III oc or aOD
(5-14)
Neuromyelitis optica spectrum disorders Acute attack/relapse II aOD (5-10)
Maintenance III
N-methyl-D-aspartatereceptor antibody I aOD (5-6)
encephalitis
TABLE25·5. Indications for Therapeutic Plasma Exchange" (Continued)
Overdose,envenomation,and poisoning Mushroom poisoning II 00 (variable)

Envenomation III
Drug overdose/poisoning III
Paraneoplasticneurologic syndromes III 00 or 000 (5-6)
Paraproteinemic demyelinating IgG/lgA/lgM 000 (5-6)
neuropathies/chronic acquired
Multiple myeloma III
demyelinating polyneuropathies
Anti-MAG neuropathy III
Multifocal motor neuropathy IV
Pediatric autoimmune neuropsychiatric PANDASexacerbation II 00 or 000 (3-6)
disorders associatedwith streptococcal
Sydenhamchorea, severe III
infections (PANDAS);Sydenhamchorea
Pemphigus vulgaris Severe III 00 or 000
(variable)
Phytanic acid storage disease II 00 (variable)
(Refsum disease)
Posttransfusion purpura III 00 (variable)
Progressive multifocalleukoencephalopa- III 000 (variable)
thy associatedwith natalizumab
Pruritus due to hepatobiliary diseases Treatment resistant III Weekly (3)
(variable)
Psoriasis Disseminated pustular IV
Redcell alloimmunization Pregnancy,GA <20 weeks III 1-3/week
Scleroderma (systemic sclerosis) III 1-3/week
Sepsis with multiorgan failure III 00 (variable)
Steroid-responsive encephalopathy II 00 or 000 (3-9)
associated with autoimmune thyroiditis
(Hashimoto encephalopathy)
Stiff-person syndrome III 000 (3-5)
Sudden sensorineural hearing loss III 00 or 000 (1-3)
Systemic lupus erythematosus Severecomplications II 00 or 000 (3-6)
Thrombotic microangiopathy, coagulation THBD, DGKEand PLG III 00 or 000
mediated mutations (variable)
(Continued)
TABLE 25-5. Indications for Therapeutic Plasma Exchanqe' (Continued)
Thrombotic microangiopathy, complement Factor H autoantibodies 00 (variable)

mediated Complementfactor gene III
mutations
Thrombotic microangiopathy, drug Ticlopidine 00 or 000
associated (variable)
Clopidogrel III
Gemcitabine/quinine IV
Thrombotic microangiopathy, infection STEC-HUS,severe III 00

associated pHUS (variable)
III
Thrombotic microangiopathy, thrombotic I 00

thrombocytopenic purpura (variable)
Thrombotic microangiopathy, III 00 (variable)
transplantation associated
Thyroid storm II 00 or 000
(variable)
Toxic epidermal necrolysis Refractory III 00 or 000
(variable)
Transplantation,cardiac Desensitization II 00 or 000
Antibody-mediated rejection (variable)
III
Transplantation, hematopoietic stem cell, Major-HPC(M), HPC(A) II 00 (variable)

ABOi Major/Minor ABOiwith pure III
red cell aplasia
Transplantation, hematopoietic stem cell, III 000 (4-5)
HLA sensitization
Transplantation, liver Desensitization,ABOi LD 00 or 000
(variable)
Desensitization,ABOi DOor III
antibody-mediatedrejection
Transplantation, lung Antibody-mediated rejection III 000 (variable)
or desensitization
Transplantation, renal, ABOcompatible Antibody-mediated 00 or 000
rejection (variable)
Desensitization,LD
Desensitization,DO III
Transplantation, renal, ABOincompatible Desensitization,LD 00 or 000

(variable)
Antibody-mediated rejection II
C HAP T E R 2 5 Therapeutic Apheresis 719
TABLE25·5. Indications for Therapeutic Plasma Exchanqe' (Continued)
RPGN, Cr ::;5.7 mg/dL III

EGPA III
Vasculitis, IgA (Henoch-Sch6nlein Crescentic RPGN III 00 or 000
purpura) ----------- (4-11)
Severe extrarenal III
disease
Vasculitis, other HBV-PAN II 000 (9-12)
sehcet disease III
Idiopathic PAN IV
Voltage-gated potassium channel II 000 (5-7)
antibody related diseases
Wilson disease Fulminant 00 or 000
(variable)
000 = every other day; TPE = therapeutic plasma exchange;00 = daily; DAH = diffuse alveolar hemorrhage;WAIHA = warm
autoimmune hemolytic anemia; NYHA = New York Heart Association (class); HELLP = hemolysis, elevated liver enzymes,
low platelet count (syndrome); IgA = immunoglobulin A; MAG = myelin-associated glycoprotein; GA = gestational age;
THBD = thrombomodulin; DGKE = diacylglycerol kinase epsilon; PLG = plasminogen; STEC-HUS = Shiga-like toxin
hemolytic uremic syndrome; pHUS = S. pneumoniaehemolytic uremic syndrome; LD = living donor; DO = deceaseddonor;
ABOi = ABOincompatible; HPC(M) = hematopoietiCprogenitor cells collected from marrow; HPC(A)= HPCsfrom apheresis;
ANCA = antineutrophil cytoplasmic antibodies; RPGN = rapidly progressive glomerulonephritis; EGPA = eosinophilic
granulomatosis with polyangiitis; HBV = hepatitis B virus; PAN = polyarteritis nodosa; Cr = creatinine.
activity is normal or only moderately reduced, rna viscosity measurements may not be useful in
and the response to TPE is typically poor. Trans- guiding therapy in some patients because they
plant-associated microangiopathic hemolysis re- may not correlate with symptoms.Normal plasma
sponds to eculizumab, but rarely apheresis, due viscosity is 1.4 to 1.8 centipoise (cP). Because
to the different pathophysiologic mechanism of most patients are asymptomatic until plasma vis-
endothelial injury and complement activa- cosity is >6.0 to 7.0 cP, those with mild eleva-
tion.63,64HELLPsyndrome (hemolysis, elevated
tions may not require treatment. In general, hy-
liver enzymes, low platelet count), a pregnancy-
perviscosity becomes a concern when M protein
associated TMA, responds to TPE in the postpar-
tum period, but TPE has not been as effective concentrations reach 3 g/dL for IgM, 4 g/dL for
antenatally.65 IgG, and 6 g/dL for IgA.66Nonetheless, symp-
In multiple myeloma, where patients may ex- tomatic patients, regardless of plasma viscosity
hibit hyperviscosity-related adverse sequelae, levels, especially those with visual and neurolog-
the goal is to remove excessive amounts of the ic symptoms, are candidates for urgent apheresis
tumor-associated paraprotein (M protein). Plas- treatments.
Conflicting data exist regarding the efficacy IPE may be beneficial." Early initiation of IPE
of IPE for treating acute renal failure in myelo- may facilitate response, and some clinical re-
ma. A randomized controlled trial of IPE vs sponses may not manifest until later in follow-
conventional care showed no difference in mor- Up.75IPE may also be effectivein neuromyelitis
tality or renal function at 6 months"; however, optica (NMO), also called Devic disease, even in
among dialysis-dependent patients, 43% in the the absence of NMO antlbodies."
IPE group and none in the control group recov- IPE to treat select peripheral nervous system
ered renal function. Another randomized clini- diseases leg, AIDPs such as Guillain-Barresyn-
cal trial showed no benefit of IPE in a com- drome (GBS)]is well established, with random-
posite outcome measure of death, dialysis ized controlled clinical trials substantiating its
dependence, and glomerular filtration rate68; efficacy in GBS patients not exhibiting sponta-
however, biopsy confirmation of the renal diag- neous recovery.' IYlGinfusions alone or follow-
nosis was not required in this trial. Similarly, a ing a course of IPE appear equally effective."
retrospective cohort study showed no benefit of No biomarkers are currently known to predict
IPE in either reducing mortality or preserving response to upfront therapeutic strategy.As a re-
renal function." If IPE is to be undertaken, bi- sult of concomitant autonomic nervous system
opsy confirmation of cast nephropathy may be involvement, patients often exhibit blood pres-
advisable.With the advent of new drugs to treat sure and pulse variability that can complicate
multiple-myeloma patients, including proteo- initial sessions of IPE but typicallybecomes less
some inhibitors and monoclonal antibodies, the pronounced during later IPE sessions. Given
role for IPE in this setting continues to evolve, equivalent efficacy and similar severity and fre-
as effects are often transient and overall survival quencies of adverse events with either IPE or
clearly depends on patient response to chemo- MG, IPE appears to be a less expensive first-
therapy." line therapy option for treating patients with
Patients receiving rituximab (anti-CD20) for GBS.78
IgM Waldenstrom macroglobulinemia may ex- In focalsegmental glomerulosclerosis(FSGS),
perience a transient increase in M protein levels a circulating factor that increases glomerular
after drug initiation, and a short IPE series be- permeability with resultant proteinuria has been
fore starting this medication may be indicated in hypothesized, as FSGSfrequently recurs after re-
select patients. Patients with pretreatment IgM nal transplantation and can produce allograft
levels >4 g/dL may benefit from IPE to avoid failure.79,80IPE may effectivelyremove the per-
developing symptomatic hyperviscoslty.Y' meability factor and induce remission in recur-
IPE has been evaluated as a therapeutic rent FSGSfollowingrenal transplantation. How-
treatment for acute liver failure and shown to be ever, response to IPE in primary FSGShas not
efficaciousin treating liver failure due to Wilson been well studied.
disease.' High-volume plasma filtration was su- IPE may provide adjunctive immunosup-
perior to standard medical therapy at improving pression in treating or preventing antibody-
transplant-free survival in patients with liver fail- mediated rejection (AMR) of solid-organ trans-
ure.? Molecular Adsorbent Recirculating Sys- plants. AMR presenting in the early posttrans-
tem (MARS)therapy (beyond the scope of this plantation period may respond better to IPE
chapter) is another extracorporeal technique than later AMR.81IPE before transplantation of
that may bridge patients with liver failure to an ABO-incompatible kidney may prevent
transplantation." hyperacute rejection, and IPE followingimplan-
IPE may be used to treat central nervous sys- tation is often used to treat AMR in this
tem (CNS) acute disseminated encephalomyeli- setting_82,83 IPE in conjunction with immuno-
tis. Experience in treating chronic progressive modulatory therapies, such as IYlG, rituximab,
multiple sclerosiswith IPE has been discourag- or bortezomib, before transplantation can re-
ing. However, for acute CNS inflammatory deduce the risk of rejection in HLA-alloimmunized
myelinating diseases unresponsive to steroids, patients.84,85
Cytapheresis from extravascular sites. Myelogenous leuke-

mia cells commonly have a higher density than
The goal of cytapheresis is to remove excessive
lymphocytic cells and can be difficult to separate
or pathogenic leukocytes, platelets, or red cells.
from red cells by centrifugation. Use of hy-
Indications for cytapheresis are listed in Table
25-6. droxyethyl starch enhances red cell sedimenta-
In leukemia, high cell counts (typically tion by rouleaux formation and can improve cy-
> 1OO,OOO/pL) can produce microvascular stasis tapheresis efficiency in acute myelogenous
with headache, mental status changes, visual leukemia. In a typical leukocyte-reduction pro-
disturbances, or dyspnea. The leukocyte count cedure, 2.0 blood volumes are processed and
at which a patient becomes symptomatic is vari- the collection rate is determined by the white
able but occurs more often in myelomonocytic cell count and inlet flow rate, removing up to
and monocytic leukemias. Typically, patients 20% of TBV into the collection bag. Many pa-
with acute or chronic lymphocytic leukemia are tients with hyperleukocytosis are receiving
asymptomatic at higher cell counts, and cyta- hyperhydration with crystalloid during their
pheresis may not be required. Cytapheresis apheresis procedures to mitigate tumor lysis
commonly results in a less-than-predicted reduc- complications; however, pediatric patients may
tionin leukocyte count, despite excellent collec- benefit from FFP or albumin replacement due to
tion, because of mobilization and reequilibration the risk of hypovolemia caused by the proce-
of marginalized intravascular cells and cells dure. Advances in the available automated
TABLE 25·6. Indications for Therapeutic Cytapheresls"
Every 2-3 weeks

(variable)
Hyperleukocytosis Symptomatic Leukocytapheresis II QD (variable)
Prophylactic or III
secondary
Inflammatory bowel Ulcerative colitis/ Adsorptive III Weekly (5-10)
disease Grohn disease cytapheresis
Polycythemia vera; Polycythemia vera Erythrocytapheresis Variable
eryth rocytosis
Secondary III
erythrocytosis
Psoriasis Disseminated Adsorptive III Weekly (5)
pustular cytapheresis
Thrombocytosis Symptomatic Thrombocytapheresis II QD (variable)
Prophylactic or
III
secondary
Vasculltis sehcet disease Adsorptive Weekly (5)
II
cytapheresis
QD = daily.
apheresis platforms may further improve fluid Red Cell Exchange

management in these critically ill patients."
Indications for red cell exchange are listed in Ta-
Massive thrombocytosis, typically> 1,000,0001
ul., can occur in essential thrombocythemia, in ble 25-7. Red cell exchange is most commonly
polycythemia vera, or as a reactive phenome- performed in sickle cell disease (SeD). The goal
non. Such patients may be at risk of both throm- is to reduce the burden of hemoglobin-S-
bosis and hemorrhage. Hemorrhage risk is partly containing red cells, provide donor red cells
due to acquired von Willebrand disease second- containing hemoglobin A, and restore oxygen-
ary to the very high platelet count. Reduction in carrying capacity in the setting of severe ane-
platelet count is commonly less than predicted mia. A common goal is to reduce hemoglobin S
because of mobilization of platelets to the pe- to <30%, with a final hematocrit not to exceed
ripheral blood, primarily from the spleen. -30% to avoid hyperviscosity in patients with
TABLE 25·7. Indications for Therapeutic Red Cell Exchanqe"
Babesiosis Severe Red cell exchange II Once

Erythropoieticpro- Red cell exchange III 3/week
toporphyria, liver
disease
Hematopoieticstem Minor-HPC(A) Red cell exchange III Once
celltransplantation,
ABOi
Malaria Severe Red cell exchange III 1-2
treatments
Red cell alloirnmunl- Exposureto RhD+red cells Red cell exchange III Once
zation, preventionand
treatment
Sicklecell disease, Acutestroke Red cell exchange Once
acute Acutechest syndrome, severe II
Othercomplications III
Sicklecell disease, Stroke prophylaxis Redcell exchange I As needed to
nonacute maintaintar-
get HbS
Recurrent vaso-occlusive pain II
crisis
Pregnancy II
Preoperative management III Once
~--~~~~- --~" ~~ ..--------===.='-

HPC(A) = hematopoietic progenitor cells collected from apheresis; HbS = hemoglobin S; ABOi = ABO incompatible.
chronic anemia. Prevention or treatment of they have similar hematocrits. Chronic red/cell
acute stroke is an indication for red cell ex- exchange may carry a lower risk of alloimmuni-
change in patients with SC;D;Retrospective data zation than simple transfusion in SCD patients,
has shown that patients with SCDwho received despite the increased donor exposures associat-
red cell exchange as their treatment for acute ed with the procedure."
stroke had fewer episodes of recurrent stroke. Isovolemichemodilution, an optional modifi-
For patients with elevated cerebral blood flow cation to automated red cell exchange, involves
velocity, determined by transcranial Dopplerim- initial depletion of the patient's red cells to a pre-
aging, transfusion reduces the risk of stroke.87,88 determined hematocrit, and replacement. with
Chronic redcell exchange, typicallyevery 4 to 6 saline or albumin. This can reduce donor expo-
weeks, can normalize cerebral blood flow while sures by reducing the volume of RBCs needed
minimizing iron overload. Acute chest syn- for exchange." Although it is generally well tol-
drome, another serious complication of SCD, erated, patient volume status, preprocedure he-
presents as dyspnea, chest pain, and cough, of- matocrit, and cerebral vascular autoregulation
ten accompanied by fever, leukocytosis,decreas- should be considered before undertalcing this
ing hematocrit, hypoxia, and pulmonary infil- modification. In general, isovolemic hemodilu-
trates. Respiratory failure can develop, and tion may be performed in patients who are he-
death occurs in -3% of cases." Red cell ex- modynamlcally stable and who maintain a pre-
change is indicated for progressive infiltrates procedure hematocrit >23% to 26%. Studies
and hypoxemia refractory to conventional thera- have shown that decreasing the patient's hema-
py and simple transfusion, and there is some evi- tocrit by 6% to 8% may be safe and well tolerat-
dence supporting early intervention with ex- ed; however, this should be modified based on
change in pediatric patients with acute chest the individual patient's needs and tolerance of
syndrome.P?' Red cell exchange may have a procedure.92,96
role in other sickle cell syndromes, including In addition, red cell exchange may be used in
multiorgan failure, hepatic/splenic sequestra- severely ill patients to decrease parasite burden
tion, priapism, and intrahepatic cholestasis. In in bloodborne infections, such as malaria and
addition, it may be used for preventing iron babesiosis. Red cell exchange may also be used
overload and for preventing or managing vaso- followinglarge-volumetransfusion of Rh-positive
occlusive pain crises. AABBand the American RBCs to an Rh-negative female of childbearing
Society of Hematology (ASH)have developed potential, to limit Rh sensitization.
consensus statements on transfusion for sickle
cell disease that include guidance on the utility Extracorporeal Photopheresis
and goals of red cell exchange in this popula-
tion, both for acute and chronic complications." In ECP, the buffy-coatlayer is collected from pe-
Red Blood Cells (RBCs)for replacement must ripheral blood, treated with 8-methoxypsoralen
be ABO compatible and negative for antigen(s) and ultraviolet A light, and reinfused into the
to known, clinically significant alloantibodies patient. The treatment causes crosslinking of
present in the recipient. For patients with SCD, leukocyte DNA, which prevents replication and
RBCs,should also be phenotype-matched for C, induces .apoptosis. ECP was developed to treat
E, and K, if possible." Units for these patients cutaneous .Tcell lymphoma, although itts in-
should also be tested for hemoglobin S and creasingly used for other indications (Table 25-
found to be negative before including them as 8). ECP has complex immunomodulatory ef-
part of the RBC replacement, Itis desirable to fects, including induction of monocyte differen-
use relatively fresh units to maximize posttrans- tiation into dendritic cells, alteration of f-cell
fusion red cell survival. RBCunits containing ei- subsets, and changes in cytokine .profiles.97,98
ther citrate-phosphate-dextrose-adenine(CPDA)- ECP has been shown to be effective in acute
1 or additive solutions (AS)may be used. It is and chronic skin-associated graft-vs-hostdisease
desirable that all units used in a given procedure (GVHD), although response rates are lower in
contain the same anticoagulant solution so that non-skin-associated GVHD. ECP often offers a
TABLE 25·8. Indications for Photopheresls"

Indication Modifying Conditions Category Typical·Course*·(duration)
Atopic (neuro-) dermatitis III Every2 weeks (12 weeks)
(atopic eczema), recalcitrant
CutaneousT-cell lymphoma; Erythrodermic Every2-4 weeks (variable)
Mycosis fungo ides; Sazary Nonerythrodermic III
syndrome
Graft-vs-host disease Acute II Weekly, tapering to every 2 weeks
Chronic II Weekly (4), then every 2 weeks
(8-12 weeks)
Inflammatory bowel disease Crohn disease III Weekly (4), every 2 weeks (8)
Nephrogenic systemic fibrosis III Every2-4 weeks (5)
Pemphigus vulgaris Severe III Every2-4 weeks (variable)
Psoriasis Disseminated pustular III Weekly (4 months)
Scleroderma (systemic III Every4-6 weeks
sclerosis) (6-12 months)
Transplantation, cardiac Cellular/recurrent II Weekly or every other week
rejection (several months)
Rejection prophylaxis II
Transplantation, liver Desensitization,ABOi III Weekly or every 2-8 weeks
(several months)
Acute rejection/immune III
suppression withdrawal
Transplantation, lung Bronchiolitis obliterans II Weekly (5), every 2 weeks (4),
syndrome every month (3)
'One cycletypicallyconsistsof extracorporealphotopheresison 2 consecutivedays.
ABOi= ABOincompatible.
steroid-sparing approach to management of crease rejection severity and allow for reduced
GVHD.99,100 immunosuppressive dosages, even in the set-
ECP for solid-organ transplant rejection has ting of hemodynamic compromtse.Pv'?' ECP
been carefullystudied in cardiac and lung trans- may have a role in stabilizing lung function in
plantation. In a randomized clinical trial, bronchiolitis obliterans syndrome after lung
prophylactic ECP, in conjunction with previous- transplantation, 105,106
generation immunosuppression (ie, not includ-
ing calcineurin inhibitors or mycophenolate), Selective Adsorption
resulted in fewer rejection episodes, decreased
HLAantibodies, and reduced coronary artery in- The currently established indications for selec-
timal thickness; but no difference was found in tive adsorption of plasma proteins are listed in
time to first rejection episode, incidence of he- Table 25·9, although in general these are not
modynamic compromise, or survival at 6 or 12 widely used, as they require specialized equip-
months.101,102 In cardiac rejection, ECP may de- ment.
TABLE 25-9. Indications for SelectiveAdsorption"
Acute inflammatory dernyellnat- Primary treat- IA on or aOD (5-6)

ing polyradiculoneuropathy ment
(Gulllaln-Barre syndrome)
Age-related macular degenera- High-risk Rheopheresis II 8-10 (2/week) over
tion, dry 8-21 weeks
Amyloidosis, systemic Dialysis-related P2-microglobulin II 3/week with dialysis
column
Atopic (neuro-) dermatitis IA III 10-12 treatments over
(atopic eczema), recalcitrant 4-6 weeks
DFPP III Weekly
chrq?iGinflammatory demye- IA 2-3/week, then tapered
linatiog polyradiculoneuropathy
Coagulation factor inhibitors IA III aD
Cryoglobulinemia Symptomatic/ IA II Every 1-3 days (3-8)
severe
Dilated cardiomyopathy, NYHAII-IV IA II oc or 000 (5)
idiopathic
Familial hypercholesterolemia Homozygotes LA Once every 1-2 weeks
Heterozygotes II (indefinite)
, ..
Focal segmental Recurrent in kid- IA aD or 000 at initia-
glomerulosclerosis ney transplant tion, wean based on
response
Recurrent in kid- LA II Twice per week (3),
neytransplant! once per week (6)
steroid-resistant
in native kidney
Hypertriglyceridemic Severe LA III on (1-3)
pancreatitis Prevention of III
relapse
Immune thrombocytopenia Refractory IA III 1-3/week (variable)
Lipoprotein (a) Progressive ath- LA II Once every 1-2 weeks
hyperlipoproteinemia erosclerotic car- (indefinite)
diovascular
disease
Multiple sclerosis Acute attack! IA II aOD (5-7)
relapse
Chronic III Variable
(Continued)
TABLE 25-9. Indications for Selective Adsorption' (Continued)
Myasthenia gravis Acute, short- IA co or aOD (3-6)

term treatment
Long-term treat- II
Every 1-2 weeks
ment
Neuromyelitis optica spectrum Acute attack! IA II aD or aOD (variable)
disorders relapse
N-methyl-D-aspartate receptor IA co or aOD (5-12)
antibody encephalitis
Paraneoplasticneurologic IA III 2/week (6)
syndromes
Pemphigus vulgaris Severe IA III oo (3), then weekly
and tapering (variable)
Peripheralvascular disease LA II 1-2/week
(variable)
Phytanic acid storage disease LA II aD (variable)
(Refsum disease)
Sudden sensorineural hearing Rheopheresis/LA III en (1-2)
loss
Thrombotic microangiopathy, STEC-HUS, IA III on (variable)
infection associated severe
Transplantation, renal, ABO Antibody- IA co or aOD
compatible mediated relec- (5-6)
tion
Desensitization,
LD
Desensitization, III
DD
Renaltransplantation, ABO Desensitization, IA oo or aOD (variable)
incompatible LD
Antibody- II
mediated rejec-
tion
Voltage-gated potassium chan- IA II oo or aOD (5-10)
nel antibody related diseases
- ~
IA = immunoadsorption; QD = daily; QOD = every other day; DFPP= double-filtration plasmapheresis;NYHA = New York
Heart Association (class); LA = lipoprotein apheresis;STEC-HUS= Shiga-like toxin hemolytic uremic syndrome; LD = living
donor; DD = deceaseddonor.
Selective removal of low-density lipoprotein communications. Some of these items related to

(LDL) can be accomplished by passing heparin- the dir~ctcliIfical care of.the patient have been
ized plasma over a dextran-sulfate column or highlighted in earlier sections of this chapter
beads coated with anionic polyacrylate ligands, (see "Patient Evaluation and Management").
or by precipitation of heparin-LDL complexes in Other information regarding specific aspects of
acidified plasma. LA treatments need to be re- me apheresis intervention itself should be docu-
peated indefinitely, typically at 2-week inter- mented in nursing procedure records. Lot num-
vals. There is evidence that LA reduces the inci- b~rs of disposables, identification numbers of
dence of major coronary events and stroke. 107In apheresis devices and related equipment (eg,
addition, atherosclerotic plaque regression may blood warmers, medications, and blood compo-
occur in some patients." Potentially beneficial, nents given during the procedure), and patient
secondary effects of LA include reduction in lev- educational materials are some items that
els of C-reactive protein, fibrinogen, tissue fac- shOuld be noted. Checklist approaches to such
tor, and soluble adhesion molecules.108,109LA is
do.cllmentation are useful (Fig 25-1). Using an
also used to treat primary or recurrent FSGS, al-
electronic medical record to document all fluids
though the mechanism of action is not well de:
arid medications administered during an aphere-
fined."? LA may also be performed with DFPP
sis intervention can enhance patient care by
techniques, as described earlier in this chapter.
IgG can be selectively removed by passing contributing more accurate data for the calcula-
plasma over a column of staphylococcal protein tion ofa pa.tleIlt,:stotal fluid status, thus poten-
A bound to silica. The putative mechanism of tially averting fluid-related adverse consequenc-
action is the removal of pathogenic autoantibod- es (Fig 25-2). It also facilitates investigation of
ies or immune complexes, although an alterna- any adverse events. Additionally, electronic
tive mechanism has been proposed in immune medical alerts that highlight a patient's recent
thrombocytopenia (ITP).111Staphylococcal pro- apheresis procedure may provide useful guid-
tein A adsorption treatment can be performed ance at the time of test ordering and test inter-
manually or in conjunction with automated pretatiqpJor clintctans Who Il1aybe less familiar
TPE. Affinity columns containing IgG antibod- wtthhow therapeutic apheresis procedures may
ies, IgE monoclonal antibodies, or ABO blood impact results from laboratory testing performed
group carbohydrate ligands.have been tested in in theperiaphereslsinterval (Fig 25-3).112
clinical trials but are not currently approved for Navigating payment systems for apheresis
use in the United States. services in today's diverse health-care insurer
environm~ntSqnbe challenging. Readers are re-
ferred to an annually updated reimbursement
A HE guicie.from<.ASFAdetaileci information in this
c area. 113
DaCU Nff PAYME Apheresis medicine is a professionallydiverse
AND P IDER discipline involving specialists across a broad
clinical spectrum. Hospital clinical privileging of
CREDENTIAUNG
both physicians and nurses involved in the deliv-
Pertinent clinical documentation is essential for ery of apheresis medicine services is a topic of
ensuring patient safety, health-care team com- interest and discussion within professional soci-
munications, adherence to regulatory and ac- eties and health-care institutions. Although no
creditation agency requirements, monetary re- uniform approach has been mandated regarding
imbursement of hospitals and providers, and practitioners' professional requirements, institu-
medicolegal protections. Although no specific tions interested in more formally addressing this
formats are mandated, certain items should be subject may wish to.review an ASFAcommen-
included in practitioners' written/electronic tary on the topic.'
(A)
(8)
FIGURE 25-1. Computer-screen displays of select sections of an apheresis procedural note documented in a
patient's electronic medical record used at 8aystate Medical Center, Springfield, MA. Note checklist format and
time-out review features incorporating various patient safety and regulatory/accreditation agencies' required
documentation items (A) and information regarding disposables used in the procedure (8).
FIGURE 25·2. Computer-screendisplay of a select section of an apheresis procedural note detailing volumes
and identities of fluids processed, removed, and replacedduring a therapeutic plasma exchangefor a patient
with thrombotic thrombocytopenic purpura; these details are documented in the patient's electronic medical
record usedat BaystateMedicalCenter,Springfield, MA.
FIGURE 25·3. Electronicmedicalalert usedat BaystateMedical Center,Springfield, MA.

REFERENCES
1. Gammon R, ed. Standards for blood banks and peutic apheresis: Knowledge gaps and research
transfusion services. 32nd ed. Bethesda, MD: opportunities in the area of hematology-
AABB,2020. oncology.J ClinApher 2015;31(I ):38-47.
2. College of American Pathologists Commission 10. Winters JL, Cooper LT, RatcliffeNR, et al. Na-
on LaboratoryAccreditation. Transfusion medi- tional Heart, Lung, and Blood Institute state of
cine checklist. Northfield,IL:CAP,2019. the science symposiumin therapeutic apheresis-
3. Andrzejewski C, Linz W, Hofmann J, et al. Therapeutic apheresis in cardiovasculardisease.
American Society for Apheresis white paper: J ClinApher 2015;30(3):183-7.
Considerationsfor medical staff apheresis medi- 11. WintersJL. Plasmaexchange: Concepts, mecha-
cine physician credentialing and privileging. nisms, and an overview of the American Society
J ClinApher 2012;27(6):330-5. for Apheresis guidelines. Hematology Am Soc
4. Marshall CS, Andrzejewski C, Carey PM, et al. Hemato1Educ Program2012;2012:7-12.
Milestones for apheresis education.J ClinApher 12. Chopek M, McCullough J. Protein and bio-
2012;27(5):242-6. chemical changes during plasma exchange. In:
5. Padmanabhan A, Connelly-SmithL, Aqui N, et Ulmas J Berkman E, eds. Therapeutic herna-
al. Guidelines on the use of therapeutic aphere- pheresis: A technical workshop. Bethesda, MD:
sisin clinicalpractice - Evidence-basedapproach AABB,1980:13-24.
from the Writing Committee of the American 13. Kaufman RM, ed. Special issue: Strategies for
SOCietyfor Apheresis: The eighth special issue. long-term intravenous access for apheresis pro-
J ClinApher 2019;34(3):171-354. cedures: History, applications and challenges.
6. American Society for Apheresis. Guidelines for Transfusion2018; 58{Sl):545-613.
therapeutic apheresis clinical privileges. J Clin 14. Adamski J. Vascular access considerations for
Apher 2007;22(3):181-2. extracorporeal photopheresis. Transfusion
7. American Society for Apheresis. Guidelines for 2018;58{Suppl1):590-7.
documentation of therapeutic apheresis proce- 15. Kalantari K. The choice of vascular access for
dures in the medical record by apheresis physi- therapeutic apheresis.J Clin Apher 2012;27(3):
cians.J ClinApher 2007;22(3):183. 153-9.
8. National Institutes of Health. Therapeutic 16. Kapadia E, Wong E, Perez-Albuerne E, Jacob-
apheresis: Recap from an NHLBI working sohn D. Extracorporeal photopheresis per-
group, November 28-29, 2012. Bethesda, MD: formed on the CELLEX(R)compared with the
National Heart, Lung, and Blood Institute, UVAR-XTS(R) instrument is more efficient and
2012. [Availableathttps:llwww.nhlbLnih.gov/ better tolerated in children with steroid-
events/20 12/therapeutic-apheresis (accessed refractory graft-versus-host disease. Pediatr
January 27,2020).] BloodCancer 2015;62(8):1485-8.
9. KarafinMS, SachaisBS,Connelly-SmithL, et al. 17. ShresthaA, Iawa Z, KochKL,et al. Use of a dual
NHLBIstate of the science symposiumin thera- lumen port for automated red cell exchange in
adults with sickle cell disease. J Clin Apher plasma exchange. J ClinApher 2001; 16(3):120-
2015;30(6):353·8. 4.
18. Agarwal AK. Systemic effects of hemodialysis 30. Wang KY,Friedman DF,DaVeigaSP. Immediate
access. Adv Chronic Kidney Dis 2015;22(6): hypersensitivity reaction to human serum albu-
459-65. min in a child undergoing plasmapheresis.
19. Rao NN, Dundon BK, Worthley MI, Faull RJ. Transfusion.2019;59(6):1921-3.
The impact of arteriovenous fistulae for hemodi- 31. Reutter JC, Sanders KF,Brecher ME, et al. Inci-
alysis on the cardiovascular system. Semin Dial dence of allergic reactions with fresh frozen
2016;29(3):214-21. plasma or cryo-supernatant plasma in the treat-
20. Dariushnia SR, Wallace MJ, Siddiqi NH, et al. ment of thrombotic thrombocytopenic purpura.
Quality improvement guidelines for central ve- J Clin Apher 2001; 16(3):134-8.
nous access. J Vasc Interv Radiol201 0;21 (7): 32. Lee LJ, Roland KJ, Sreenivasan GM, et al.
976-81. Solvent-detergent plasma for the treatment of
21. Lee KA, Ramaswamy RS. Intravascular access thrombotic microangiopathies: A Canadian ter-
devices from an interventionalradiology per- tiary care centre experience. Transfus Apher Sci
spective: Indications, implantation techniques, 2018;57(2):233-5.
and optimizing patency. Transfusion 2018; 33. Sidhu D, SnyderEL,Tormey CA.Two approach-
58(SI ):549-57. es to the clinical dilemma of treating TTP with
22. Ceppi F, Rivers J, Annesley C, etal. Lymphocyte therapeutic plasma exchange in patients with a
apheresis for chimeric antigen receptor f-cell history of anaphylactic reactions to plasma.
manufacturing in children and young adults J ClinApher20 17;32(3):158-62.
with leukemia and neuroblastoma. Transfusion 34. AskariS, Nollet K,Debol SM, et al. Transfusion-
20 18;58(6):1414-20. related acute lung injury during plasma ex-
23. MandaI S, Baron BW, Mischeaux M, et al. Col- change: Suspecting the unsuspected. J Clin
lection of peripheral blood stem cells from a 7 Apher 2002; 17(2):93-6.
35. Duntley P, SieverJ, KorwesML, et al. Vascular
month-old girl weighing 7 kg with the use of
erosion by central venous catheters. Chest
combined heparin and citrate anticoagulation.
1992;101(6):1633-8.
J ClinApher 2013;28(4):309-10.
36. Leitman SF,BoltanskyH, Alter HI, et al. Allergic
24. Matsuzaki M, Hiramori K,Imaizumt T, et al. In-
reactions in healthy platelet-pheresis donors
travascular ultrasound evaluation of coronary
caused by sensitization to ethylene oxide gas.
plaque regression by low density lipoprotein-
N Eng!J Med 1986;315(19):1192-6.
apheresisin familialhypercholesterolemia.J Am 37. Purello D'Ambrosio F, SavicaV, Gangemi S, et
Coli CardioI2002;40(2):220-7. al. Ethylene oxide allergy in dialysis patients.
25. McLeod BC, Sniecinski I, Ciavarella D, et al. Nephrol DialTransplant 1997;12(7):1461-3.
Frequency ofimmediate adverse effectsassociat- 38. Owen HG, Brecher ME. Atypicalreactions asso-
ed with therapeutic apheresis. Transfusion ciated with use of angiotensin-converting
1999;39(3):282-8. enzyme inhibitors and apheresis. Transfusion
26. Norda R, Berseus 0, Stegmayr B. Adverse 2003;34(10):891-4.
events and problems in therapeutic hemaphere- 39. Perkins KA. Contraindication of angiotensin-
sis.A report from the Swedish registry. Transfus converting enzyme (ACE)inhibitors for patients
Apher Sci2001 ;25(1):33-41. receiving therapeutic plasma exchanges.
27. Mortzell Henriksson M, Newman E, Witt V, et Nephrol Nurs J 2008;35(6):571-4.
al. Adverse events in apheresis: An update of 40. Orlin JB, Berkman EM. Partial plasma exchange
the WAAregistrydata. TransfusApher SCi2016; using albumin replacement: Removal and recov-
54(1):2-15. ery of normal plasma constituents. Blood 1980;
28. HaddadS, LeitmanSF,WesleyRA,et al. Placebo- 56(6):1055-9.
controlled study of intravenous magnesium sup- 41. Flaum MA, Cuneo RA, Appelbaum FR, et al.
plementation during large-volumeleukapheresis The hemostatic imbalance of plasma-exchange
in healthy allogeneic donors. Transfusion 2005; transfusion. Blood 1979;54(3):694-702.
45(6):934-44. 42. Food and Drug Administration. Transfusion/do-
29. Marques MB, Huang ST. Patients with throm- nation fatalities.SilverSpring,MD: CBEROffice
botic thrombocytopenic purpura commonly de- of Communication, Outreach, and Develop-
velop metabolic alkalosis during therapeutic ment, 2019. [Available at https://www.fda.
gov/vaccines- blood-biologics/ report -problem- Medical Center. Transfus Apher Sci 2017;
center-biologics-evaluation-research/transfu 56(4):591-4.
siondonatlon-fatalitles (accessed January 27, 53. Li A, Khalighi PR, Wu 0, Garcia DA. External
2020).] validation of the PLASMICscore: A clinicalpre-
43. Food and Drug Administration. Death and pul- diction tool for thrombotic thrombocytopenic
monary embolism related to extracorporeal pho- purpura diagnosisand treatment. J Thromb Hae-
topheresis (ecp)treatment - Letter to health care most 2018;16(1):164-9.
providers. (February 5, 2018) Silver Spring, 54. Tiscia GL, Ostuni A, CascavillaN, et al. Valida-
MD: Divisionof Industry and Consumer Educa- tion of PLASMICscore and follow-up data in a
tion, 2018. [Availableat https://www.fda.gov/ cohort of patients with suspected rnicroangiopa-
medical-devices/letters-health-care-providers/ thies from Southern Italy.J Thromb Thromboly-
death-and-pulmonary-embolism-related-extra sis 2018;46(2): 174-9.
corporeal-photopheresis-ecp-treatment-letter- 55. Bandarenko N, Brecher ME. United States
health-care (accessedJanuary 27,2020).] thrombotic thrombocytopenic purpura aphere-
44. KiprovDD, Golden P, Rohe R, et al. Adverse resis study group (USTIP ASG):Multicenter sur-
actions associated with mobile therapeutic vey and retrospective analysisof current efficacy
apheresis: Analysisof 17,940 procedures. J Clin of therapeutic plasma exchange. J Clin Apher
Apher 2001;16(3):130-3. 1998;13(3):133-41.
45. Food and Drug Administration. FDAexecutive 56. Howard MA, Williams LA, Terrell DR, et al.
summary (prepared for the February 26, 2016 Complications of plasma exchange in patients
meeting of the Gastroenterology-UrologyDevic- treated for clinically suspected thrombotic
es Panel): Classificationof centrifuge-typethera- thrombocytopenic purpura-hemolytic uremic
peutic apheresis devices. Silver Spring, MD: syndrome. Transfusion2006;46( 1):154-6.
FDA,2016. [AVailableat https://www.fda.gov/ 57. Page EE, Kremer HovingaJA, Terrell DR, et al.
media/96 112/ download (accessedJanuary 27, Thrombotic thrombocytopenic purpura: Diag-
2020).] nostic criteria, clinical features, and long-term
46. Kim HC. Therapeutic pediatric apheresis. J Clin outcomes from 1995 through 2015. BloodAdv
Apher 2000; 15(1-2):129-57.
2017;1 (10):590-600.
47. KimYA,Sloan SR.Pediatric therapeutic aphere-
58. Zeigler ZR, Shadduck RK, Gryn JF, et al. Cryo-
sis: Rationale and indications for plasmaphere-
precipitate poor plasma does not improve early
sis, cytapheresis, extracorporeal photopheresis,
response in primary adult thrombotic thrombo-
and LDL apheresis. Pediatr Clin North Am
cytopenic purpura (TTP). J CUn Apher 2001;
2013;60(6): 1569-80.
48. Meyer EK, Wong EC. Pediatric therapeutic 16(1):19-22.
apheresis:A critical appraisalof evidence. Trans- 59. ScullyM, Cataland SR,Peyvandi F, et al. Capla-
Ius Med Rev 2016;30(4):217-22. cizumab treatment for acquired thrombotic
49. Terrell DR, Williams LA,Vesely SK, et al. The thrombocytopenic purpura. N EnglJ Med 2019;
incidence of thrombotic thrombocytopenic 380(4):335-46.
purpura-hemolytic uremic syndrome: All pa- 60. Legendre CM, Licht C, Muus P, et al. Terminal
tients, idiopathic patients, and patients with se- complement inhibitor eculizumab in atypical
vere ADAMTS-13 deficiency. J Thromb Hae- hemolytic-uremic syndrome. N Engl J Med
most 2005;3(7):1432-6. 2013;368(23):2169-81.
50. LaraPN Ir, Coe TL, Zhou H, et al. Improvedsur- 61. Fakhouri F, Hourmant M, Campistol JM, et al.
vival with plasma exchange in patients with Terminal complement inhibitor eculizumab in
thrombotic thrombocytopenic purpura- adult patients with atypical hemolytic uremic
hemolytic uremic syndrome. Am J Med 1999; syndrome: A single-arm, open-label trial. Am J
107(6):573-9. KidneyDis 2016;68(1):84-93.
51. Bendapudi PK, Hurwitz S, Fry A, et al. Deriva- 62. Legendre CM, Licht C, Loirat C. Eculizumabin
tion and external validation of the PLASMIC atypical hemolytic-uremic syndrome. N Eng! J
score for rapid assessment of adults with throm- Med 2013;369(14):1379-80.
botic microanglopathies:A cohort study. Lancet 63. Iodele S, Dandoy CE, Myers KC, et al. Newap-
HaematoI2017;4(4):eI57-e64. proaches in the diagnosis, pathophysiology,and
52. JajoskyR, FloydM, Thompson T, ShikleJ. Vali- treatment of pediatric hematopoietic stem cell
dation of the PLASMIC score at a University transplantation-associated thrombotic microan-
giopathy. Transfus Apher Sci 2016;54(2):181- 77. Randomised trial of plasma exchange, intrave-
90. nous immunoglobulin, and combined treat-
64. Jodele S, Laskin BL, Dandoy CE, et al. A new ments in Gulllain-Barre syndrome. Plasma Ex-
paradigm: Diagnosis and management of HSCT- change/Sandoglobulin Gulllain-BarreSyndrome
associated thrombotic microangiopathy as multi- Trial Group. Lancet 1997;349(9047):225-30.
system endothelial injury. Blood Rev 2015; 78. Winters JL, Brown D, Hazard E, et al. Cost-
29(3):191-204. minimization analysis of the direct costs of TPE
65. Simetka 0, Klat J, Gumulec J, et al. Early identi- and WIg in the treatment of Guillain-Barresyn-
fication of women with HELLP syndrome who drome. BMC Health ServRes 2011; 11:10 1.
need plasma exchange after delivery. Transfus 79. De VrieseAS, Sethi S, Nath KA,et al, Differenti-
Apher Sci 2015;52(1):54-9. ating primary, genetic, and secondary FSGSin
66. Somer T. Rheology of paraproteinaemias and the adults: A clinicopathologic approach. J Am Soc
plasma hyperviscosity syndrome. Baillieres Clin NephroI2018;29(3):759-74.
Haematol 1987; 1(3):695- 723. 80. Wada T, Nangaku M. A circulating permeability
67. Johnson WJ. Treatment of renal failure associat- factor in focal segmental glomerulosclerosis:
ed with multiple myeloma. Arch Intern Med The hunt continues. Clin Kidney J 2015;8(6):
1990;150(4):863. 708'15.
68. Clark WF, Stewart AK, Rock GA, et al. Plasma 81. Al-BadrW, KallogjeriD, Madaraty K,et al. A ret-
exchange when myeloma presents as acute re- rospective review of the outcome of plasma ex-
nal fallure. Ann Intern Med 2005;143(11):777. change and aggressive medical therapy in anti-
69. Movilli E, Guido J, Silvia T, et al. Plasma ex- body mediated rejection of renal allografts: A
change in the treatment of acute renal fallure of single center experience. J Clin Apher 2008;
myeloma. Nephrol Dial Transplant 2007; 23(6):178-82.
22(4):1270-1. 82. Sivakumaran P, VoAA,VillicanaR, et aI. Thera-
70. Abramson HN. The multiple myeloma drug peutic plasma exchange for desensitization prior
pipeline-2018: A review of small molecules and to transplantation in ABO-incompatiblerenal al-
their therapeutic targets. Clin Lymphoma My- lografts.J Clin Apher 2009;24: 155-60.
eloma Leuk 2018;18(9):611-27. 83. Tobian AAR,Shirey RS,Montgomery RA, et al.
71. Treon SP, Branagan AR, Hunter Z, et al. Para- Therapeutic plasma exchange reduces ABO ti-
doxical increases in serum IgM and viscosity ters to permit ABO-incompatible renal trans-
levels following rituximab in Waldenstrom's plantation. Transfusion2009;49(6):1248-54.
macroglobulinemia. Ann Oncol 2004; 15( 10): 84. Padmanabhan A, Ratner LE, Jhang JS, et al.
1481-3. Comparative outcome analysis of ABO-
72. Larsen FS, Schmidt LE, Bernsmeier C, et al. incompatible and positive crossmatch renal
High-volume plasma exchange in patients with transplantation: A Single-center experience.
acute liver failure: An open randomised con- Transplantation 2009;87(12):1889-96.
trolled trial. J HepatoI2016;64(1 ):69-78. 85. Jordan SC, Choi J, KahwajiJ, Vo A. Progress in
73. Gerth HU, Pohlen M, Tholking G, et al. Molec- desensitization of the highly HLAsensitized pa-
ular adsorbent recirculating system can reduce tient. Transplant Proc 2016;48(3):802-5.
short-term mortality among patients with acute- 86. Schulz M, Bug G, BialleckH, et al. Leucodeple-
on-chronic liver failure-a retrospective analysis. tion for hyperleucocytosis-First report on a
Crit Care Med 2017;45(10):1616-24. novel technologyfeaturing electronic interphase
74. Gwathmey K, Balogun RA, Burns T. Neurologic management. Vox Sang2013;105(1):47-53.
indications for therapeutic plasma exchange: 87. Adams RJ,McKie VC, Hsu L, et aI. Prevention
2013 update. J Clin Apher 2014;29(4):211-19. ofa first stroke by transfusions in children with
75. Llufriu S, Castillo J, Blanco Y, et al. Plasma ex- sickle cell anemia and abnormal results on tran-
change for acute attacks of CNS demyelination: scranial doppler ultrasonography. N Eng!J Med
Predictors of improvement at 6 months. Neurol- 1998;339(1):5-11.
ogy 2009;73(12):949-53. 88. Lee MT. Stroke Prevention Trial in. Sickle Cell
76. Bonnan M, Valentino R, Olindo S, et al. Plasma Anemia (STOP): Extended follow-up and final
exchange in severe spinal attacks associated results. Blood2006; 108(3):847-52.
with neuromyelitis optica spectrum disorder. 89. Vichinsky EP, Neumayr LD, Earles AN, et al.
Mult Scler 2009;15(4):487-92. Causes and outcomes of the acute chest syn-
drome in sicklecell disease. N EnglI Med 2000; for ECP treatment in patients with steroid-
342(25):1855-65. refractory acute GVHD. Patient Prefer Adher-
90. Saylors RL, Watkins B, Saccente S, Tang X. ence 2015;9: 105-11.
Comparison of automated red cell exchange 101. Barr ML, Baker CI, Schenkel FA, et al. Prophy-
transfusion and simple transfusion for the treat- lactic photopheresis and chronic rejection: Ef-
ment of children with sickle cell disease acute fects on graft intimal hyperplasia in cardiac
chest syndrome. Pediatr Blood Cancer 2013; transplantation. Clin Transplant 2000; 14(2):
60(12):1952-6. 162-6.
91. Turner 1M, KaplanIB, Cohen HW, Billett HH. 102. BarrML, Meiser BM, Eisen HI, et al. Photopher-
Exchange versus simple transfusion for acute esis for the prevention of rejection in cardiac
chest syndrome in sickle cell anemia adults. transplantation. N Engl I Med 1998;339(24):
Transfusion2009;49(5):863-8. 1744-51.
92. Biller E, Zhao Y, Berg M, et al. Red blood cell 103. Dall'AmicoR, Montini G, Murer L, et al. Extra-
exchange in patients with sickle cell disease- corporeal photochemotherapy after cardiac
Indications and management: A review and transplantation: A new therapeutic approach to
consensus report by the therapeutic apheresis allograft rejection. Int I Artif Org 2000;23( 1):
subsection of the AABB. Transfusion 2018; 49-54.
58(8):1965-72. 104. ColvinMM, Cook Il., Chang P, et al. Antibody-
93. VichinskyEP, Luban NLC,Wright E, et al, Pro- mediated rejection in cardiac transplantation:
spective RBCphenotype matching in a stroke- Emerging knowledge in diagnosis and manage-
prevention trial in sickle cell anemia: A multi- ment: A scientificstatement from the American
center transfusiontrial. Transfusion2001;41(9): Heart Association. Circulation 2015; 131(18):
1086-92. 1608-39.
94. Wahl SK, GarciaA, Hagar W, et al. Lower allo- 105. Hachem R, CorrisP. Extracorporealphotophere-
immunization rates in pediatric sickle cell pasis for bronchiolitis obliterans syndrome after
tients on chronic erythrocytapheresis compared lung transplantation. Transplantation 2018;
to chronic simple transfusions. Transfusion
102(7):1059-65.
2012;52(12):2671-6. 106. Iaksch P, Scheed A, KeplingerM, et al. A pro-
95. Sarode R, Matevosyan K, Rogers ZR, et al. Ad-
spective interventional study on the use of ex-
vantages of isovolemichemodilution-redcell ex-
tracorporeal photopheresis in patients with
change therapy to prevent recurrent stroke in
bronchiolitis obliterans syndrome after lung
sickle cell anemia patients. I Clin Apher 2011;
transplantation. J Heart Lung Transplant 2012;
26(4):200-7.
96. KimI, Ioseph R, Matevosyan K, Sarode R. Com- 31(9):950-7.
parison of Spectra Optia and COBE Spectra 107. Masakl N, Tatami R, Kumamoto T, et al. Ten-
apheresis systems' performances for red blood year follow-up of familial hypercholesterolemia
cell exchange procedures. Transfus Apher Sci patients after intensive cholesterol-lowering
2016;55(3):368-70. therapy. Int Heart I 2005;46(5):833-43.
97. Bladon1, TaylorPC. Extracorporealphotophere- 108. Kobayashi S, Oka M, Moriya H, et al. LDL-
sis: A focus on apoptosis and cytoklnes. I Der- apheresis reduces P-selectin, CRP and fibrino-
matol Sci2006;43(2):85-94. gen-Possible important implications for im-
98. Morelli AE, Larregina AT. Concise review: proving atherosclerosis. Ther Apher Dial 2006;
Mechanisms behind apoptotic cell-based thera- 10(3):219-23.
pies againsttransplant rejection and graftversus 109. WangY.Effectsof heparin-mediatedextracorpo-
host disease.Stem Cells2016;34(5):1142-50. real low-density lipoprotein precipitation be-
99. Del Fante C, Scudeller L, Viarengo G, et al. Re- yond lowering proatherogenic lipoproteins-
sponse and survival of patients with chronic Reduction of circulating proinflammatory and
graft-versus-hostdisease treated by extracorpo- procoagulatory markers. Atherosclerosis 2004;
real photochemotherapy: A retrospective study 175(1):145-50.
according to classicaland National Institutes of 110. Muso E, Mune M, YoriokaN, et al. Beneficial
Health classifications.Transfusion 2012;52(9): effectof low-densitylipoprotein apheresis (LDL-
2007-15. A) on refractory nephrotic syndrome (NS)due
100. Zhang H, Chen R, Cheng I, et al. Systematicre- to focal giomerulosclerois (FGS).Clin Nephrol
view and meta-analysis of prospective studies 2007;67(06):341-5.
111. Silverman GJ, Goodyear CS, Siegel DL. On the 113. Berman K. Therapeutic apheresis: A guide to
mechanism of staphylococcal protein A immu- billing and securing appropriate reimbursement.
nomodulation. Transfusion 2005;45(2):274-80. 2019 ed. Vancouver, Canada: American Society
112. Levy R, Pantanowitz L, Cloutier D, et al. Devel- for Apheresis, 2019. [Availableat https://www.
opment of electronic medical record charting for apheresis.org/page/ ApheresisReimbursem (ac-
hospital-based transfusion and apheresis medi- cessed December 6, 2019).]
cine services: Early adoption perspectives.
J Pathol Inform 2010; 1:8.
The Collection and Processing of
Hematopoietic Progenitor Cells
Laura S. Connelly-Smith, MBBCh, OM, and Michael L. Linenberger, MD, FACP
LURIPOTENT HEMATOPOIETIC STEM stimulating factor (GM-CSF),and stem cell fac-

cells, capable of self-renewal and differenti- tor (SCF)_Plerixafor,a C-X-Cchemokine receptor
ation, and committed lineage-restricted pro- type 4 (CXCR4)receptor antagonist, allows for
genitor cells are collectivelyreferred to as hema- rapid mobilization of HPCs into the blood." UCB
topoietic progenitor cells (HPCs). These cells are contains HPCs that may be sufficient to reconsti-
capable of reconstituting marrow function when tute hematopoiesis.6
transplanted into conditioned recipients. HPCs CD34 is a cell-surface antigen that is ex-
for clinical use can be collected either from mar- pressed on progenitor cells. Even though its ex-
row [HPC(M)],from peripheral blood after mobi- act function is undefined and is not specific to
lization [HPC(A)],or from umbilical cord blood HPCs, CD34 is used for identifying and quanti-
(UCB)collected at birth [HPC(CB)].1,2 HPCs tend fying HPCs and can be used to select and enrich
to reside with mesenchymal elements within the HPCs for transplantation."
marrow microenvironment, their interaction
generating a niche that supports and regulates
hematopoiesis.' Cytoadhesive interactions be- I Nt IL
tween membrane receptors and ligands ex-
pressed on the microenvironment stromal cells Hematopoietic progenitor/stem cell transplanta-
and within the extracellular matrix ensure that tion (HSCT)is an established procedure to treat
HPCs mostly situate within the marrow." Disrup- many acquired and congenital disorders of the
tlon of these cellular and microenvironmental in- hematopoietic system ranging from hematologic
teractions allows mobilization of the HPCs into malignancies to nonneoplastic immune disorders
the peripheral blood from where they can be col- and hemoglobinopathies to disorders of enzyme
lected. Mobilization can be accelerated with an metabolism. Several nonhematologic malignant
increase in progenitor cells into the bloodstream and nonmalignant disorders are also selected in-
following treatment with chemotherapy and by dications for HSCT (Table 26-1). Certain indica-
the administration of certain exogenous cyto- tions for HSCT occur more frequently and are
kines such as granulocyte colony-stimulating dependent on patient age. For instance, immuno-
factor (G-CSF),granulocyte-macrophage colony- deficiency and inborn errors of metabolism are
Laura S. Connelly-Smith, MBBCh, DM, Assistant Medical Director, Apheresis and Cellular Therapy, Seattle Can-
cer Care Alliance, Associate Professor, Division of Hematology, Department of Medicine, University of Washing-
ton, and Assistant Member, Clinical Research Division, Fred Hutchinson Cancer Research Center; and Michael L.
Linenberger, MD, FACP, Medical Director, Apheresis and Cellular Therapy, Seattle Cancer Care Alliance, Profes-
sor, Division of Hematology, Department of Medicine, University of Washington, and Member, Clinical Research
Division, Fred Hutchinson Cancer Research Center, Seattle, Washington
137
TABLE 26-1. Diseases Treated with Autologous or Allogeneic HSCT

Hematologic Malignancies
Leukemia
Multiple myeloma
Myelodysplasia
Non-Hodgkin lymphoma
Hodgkin disease
Myeloproliferative neoplasms (MPN)
Other Clonal Disorders of Marrow/Marrow Failure Syndromes
Severeaplastic anemia
Paroxysmal nocturnal hemoglobinuria
Fanconianemia
Dlamond-Blacktananemia
Pure red cell aplasia
Amegakaryocyticthrombocytopenia
Inherited Metabolic Disorders
Mucopolysaccharidoses
Leukodystrophies
Gaucherdisease
Lysosomal storage disorders
Congenital Immunodeficiency
Severecombined immunodeficiency (SCID)
Wiskott-Aldrich syndrome
Omenn syndrome
X-linked Iymphoproliferative syndrome
Chronic granulomatous disease
Leukocyteadhesion deficiency
DiGeorgesyndrome
Hemoglobinopathies
Sickle cell disease
Thalassemia
Other Nonhematologic Malignancies
Germ cell tumors
Neuroblastoma
C HAP T E R 2 6 The Collection and Processing of HPCs 739
TABLE 26·1. Diseases Treated with Autologous or Allogeneic HSCT (Continued)
Ewing sarcoma
Wilms tumor
Autoimmune Diseases
Systemic sclerosis
Multiple sclerosis
more common in children, whereas adults more fore receiving the stored HPC(A)product. Nota-
commonly have clonal disordersof their marrow bly, some resource-limited transplantation
or hematologic malignancies. The decision to centers have successfully used short-term stor-
perform any HSCTrequires careful consideration age of HPC(A) products without cryopreserva-
with the complex integration of numerous vari- tion for autologous transplantation."
ables framinga treatmentplan.These include pa- Autologous patients must be medically well
tient-specific goals, age, prognosis, diseasepro- enough to undergo mobilization and HPC pro-
gression, previous therapy, comorbidlties, curement, and should be assessed for their suit-
availabilityof a suitable HPC source, and the type ability. Significant levels of prior chemotherapy
of transplant being considered (eg, autologous vs or radiation or ongoing marrow disease involve-
allogeneic, myeloablative vs nonmyeloablative ment may make HPC mobilization and collec-
therapy). tion difficultbecause of reduced HPC quality or
number. Patients must undergo a full medtcal
evaluation With history and physical evaluation
Autologous Transplantation
per institutional policies and procedures, accred-
Autologous HSCTrefers to the donor and recipi- itation standards, and regulatory require-
ent being the same person, In autologous trans- ments.'?" Unlike allogeneic transplantation, eli-
plantation, the reinfusion of the patient's stem gibility requirements are not mandated py the
cells allows for the recovery of the marrow fol- Food and Dr,ugAdministration(FDi\) for autolo-
lowing the provision of high-dose therapy. The gous HSCT, so screening questionnaires to iden-
antitumor effect comes from the chemotherapy tify relevant Infectious disease risk factors are
and radiotherapy used during the conditioning not required. Similarlylaboratory testing for hu-
phase of transplantation. Chemotherapy re- man immunodeficiency virus types I and 2
ceived by patients can be associated with signifi- (HW1I2J, hepatitis B virus, hepatitis C virus,
cant toxicity but requires less immune suppres- syphilis, and human Tcell lymphotropic virus
sion than allogeneic transplantation, Autologous types I and II (HTLVVII) is not mandated but
transplantation is most successful for patients should be assessed, because autologous prod-
whose marrow is not actively affected with the ucts are cryopreserved and stored with other
disease. For patients with marrow involvement products; thus, the presence of these viruses is a
in their malignant disease, there is always con- cross-contamination risk. I I (For discussion of in-
cern over the collection and reinfusion of cancer fectious disease testing, see Chapter n Autolo-
cells. Peripheral blood has, for the most part, re- gous patients can still proceed to transplantation
placed marrow as a graft source in the majority if tested positive for one of the relevant infec-
of the autologous HSCTsperformed in the Unit- tious diseases, but the apheresis product should
ed States." For autologous transplantation, the be quarantined and stored in such a way as to
collected HPC product is cryopreserved until minimize contamination risks (see "Cryopreser-
the patient has been deemed ready for.precondi- vation"). All autologous HPC products must,
tioning with high-dose preparative therapy be- however, be labeled to indicate that they are for
autologous use only and that they have not been various accrediting bodies, such as AABBand
evaluated for infectious substances, unless all the Foundation for the Accreditation of Cellular
otherwise-applicable screening and testing has Therapy (FACT),and by the National Marrow
been performed as for allogeneic products." Donor Program (NMDP). For HPC(CB) trans-
plantation, the screening and testing process in-
Allogeneic Transplantation volves the mother and her samples. In the Unit-
The aim of allogeneic transplantation is to re- ed States, there are currently no donor tests that
place a diseased or nonfunctioning hematopoiet- have been approved for UCBsamples; therefore,
ic and/or immune system with normal HPCs infectious disease testing is performed on a ma-
from a healthy related or unrelated donor. When ternal sample. The maternal blood samples
allogeneic transplantation is used to treat malig- serve as a surrogate for the UCB unit, and test-
nant conditions, patients may receive aggressive ing should reflect the health of the mother at
conditioning with radiotherapy and/or chemo- the time the unit is collected, ideally having
therapy regimens. Myeloablative regimens in- been taken within 7 days of UCBcollection.
duce cytotoxicity and prevent rejection of the Relevantinfectious disease agents that can be
new graft. Posttransplantation immunosuppres- transmitted by HPCs to the transplant recipient,
sion is required to prevent graft-vs-hostdisease and for which there is an FDA-licensedscreen-
(GVHD). In allogeneic transplantation, the ing test, include HIV,hepatitis B and C viruses,
transplanted cells are also seen as therapeutic in Treponema pallidum, HTLV-I1II,Trypanosoma
view of the anticipated graft-vs-neoplasm(GVN) cruzi, West Nile virus, and cytomegalovirus
effect. For patients with inborn errors of metab- (CMV). Appropriate screening measures have
olism, congenital immunodeficiency, or other been developed for Zikavirus (ZIKV),such as re-
diseases and conditions in which germline mu- view of medical and travel history. Two FDA-
tations are present in the patient's cells, the goal licensed ZIKV assays are available to screen
of allogeneic HSCT is to reconstitute a healthy whole blood and blood components for transfu-
lymphohematopoietic system while avoiding sion. One is also FDA-approvedfor use in testing
GVHD. plasma or serum specimens from HPC donors.
In the allogeneic setting, the healthy donor However, the FDA does not consider this test
undergoes the HPC collection. Screening and appropriate for preventing transmission of ZIKV
testing for infectious diseases are mandated in through HCT/Ps, because ZIKVis readily de-
the United States to determine whether trans- tected in HCT/Ps after viral RNA is no longer
plantation of mobilized peripheral blood or UCB detectable in plasma.!2
poses a risk for transmitting a relevant infectious If donor screening or testing detects a risk of
disease to the recipient. US federal regulations transmitting a relevant infectious disease, the
are laid out in Title 21 of the Code of Federal potential HPC donor is considered ineligible.
Regulations (CFR),Part 1271 [coveringhuman The donor, recipient, and their physicians are in-
cells, tissues, and cellular and tissue-based prod- formed of the donor's ineligible status, and a
ucts (HCT/Ps)].!! Screening and testing include risk/benefit analysis is performed to determine
a standardized health questionnaire, a physical whether the donor's HPCs might still be consid-
examination, a review of the relevant medical ered safe to use. If a decision is made to proceed
records, and applicabletesting for relevant infec- with transplantation using the ineligible donor's
tious diseases. Although marrow products are HPCs, justification must be documented under
administered under Sections 375 and 379 of the "urgent medical need" criteria, as defined by
Public Health Service Act and are not subject to the FDA. Finally, in addition to screening and
CFR Part 1271, marrow donors are screened testing for infectious diseases, the donor's medi-
and tested similarly to mobilized peripheral cal fitness is evaluated to determine whether the
blood and UCB donors, because all three prod- donor is healthy enough to safelyundergo HPC
ucts are treated similarly under the standards of mobilization and collection.
a lower survival with DOEI mismatching, and

in particular if this mismatch is added to a mis-
match at HLA-A, -B, -C, and -DRBI. As a result,
HLA-DOEl, ,DPBl, and -DRB3!4IS may be
Histocompatibility added to prioritize donors with minimal or per-
missiblemismatching at these loci, especiallyif a
One of the major barriers influencing the clinical 10/10 or 12/12 match is being considered."
success of allogeneic HSCT is compatibility High-resolution 8/8- and 1011O-matched trans-
across the highly polymorphic classicalHLAsys- plants of HPCs·from unrelated donors are, at
tern. The transplantation physician preferably least in part, responsible for recently improved
selects an HLA-compatibledonor-in order tore- survival outcomes of matched unrelated-donor
duce the risk of graft failure, GVHD,and mortal- transplantations for several diseases.18·25The
ity.13.1SThe HLA alleles that are important for great diversity of HLA alleles and haplotypes
HSCT compatibility are the Class I genes HLA- makes identifying an unrelated donor matched
A, -B, and -C, and the ClassII genes HLA-DRBl, at allelic resolutions a major challenge for most
-DOEl, and -DPBl. Other HLA genes are less patients.26,27 The likelihood of finding an opti-
important either because they have low levels of mally HLA-matched unrelated donor (MUD)
polymorphism (because they are pseudogenes) ranges from 75% for patients of European ances-
or because they are minimally or not expressed. try to 16%for South or Central Americans ot.Af-
The ..human major histocompatibility complex rican ancestry." In addition, the time needed to
(MHC), where HLA genes are found, displays procure a graft from an unrelated donor may be
"linkage disequilibrium," where certain alleles up to 8 weeks." With these ongoing limitations
are inherited together more frequently than to donor availability, alternate sources of stem
would occur by chance. This inheritance pat- cells have been identified and include UCB, as
tern is not random and explains why an HLA- well as haploidentical and mismatched related
identical sibling is more likely to be an exact donors or mismatched unrelated donors.
match and thus preferred as the optimal donor. Studies show that survival after UCB trans-
Unfortunately, -70% of patients will not have a plantation is similar to other graft sources, and
suitable siblingmatch. emerging data demonstrate acceptable out-
Molecular techniques (high-resolution DNA- comes with haploidentical donor transplanta-
based tissue typing) have supplanted serologic tion.3037 Due to the immaturity of UCB T cells,
methods for HLAtyping. This has led to a signif- the HLA matching requirements for UCB are
icant improvement in resolution and matching less stringent than those for marrow and mobi-
at the allele level. Allogeneictransplantation us- lized peripheral blood HPCs, but HLAmatching
ing either HPC(A) or HPC(M) that are mis- is still an important factor for engraftment. Con-
matched at HLA-A, -B, -C, and -DRBlloci (by ventionally, HLA matching of UCB for HSCT
high-resolution typing) are associatedwith a 5% has used low-/intermediate-resolution typing
to 10% decrease in survival with each mis- for HLA-Aand HLA-B(antigeniclevel) and high-
match." Thus, patients should be typed using resolution typing for HLA-DRBI (allelic .level).
high-resolutton techniques at the RLA -A, -B, -C, HLAmatching for UCB units is generally based
and -DRBlloci to facilitate finding an optimal on three loci, and a maximum of 216 HLAmis-
donor. The matching process considers bothof matches is considered acceptable, with a very
the inherited alleles for each of these four loci, high transplant-related mortality (TRM)associat-
for a total of eight possible matches; alternative- ed with greater rnismatch" Provided that suffi-
ly, ifDOEtand DPBl are also considered, 10 or cient cell dose is achieved, the outcomes of 4/6-
12 alleles are examined. Because two large stud- matched UCB transplants are comparable to
ies failed to show an individual impact of DOEI that of HLA-matched unrelated donors, albeit
on survival, it is common practice in US centers with an increased risk of nonrelapse mortality
to consider an 8/8-matched donor as "fully (NRM).L In a study analyzing the role of HLA-C
matched." Other studies, however, have shown on UCB transplantation, Eurocord and NMDPI
CIBMTR (Center for International Blood and when an alternative donor is lacking, reduction
Marrow Transplant Research) reported higher of strong-reactingDSAsmust be attempted."
TRM in patients receiving a UCB unit with an
HLA-C mismatch; also, contemporary mis- Other Considerations for Donor
matching at HLA-Cand HLA-DRBlwas associ- Selection
ated with the highest risk of mortality." When
using a single unit of HPC(CB), HLA-Cantigen Donor selection for allogeneic transplantation is
mismatching was shown to increase TRM, par- determined largely by histocompatibility. How-
ticularly if combined with HLA-DRB1 mis- ever, several other variables are taken into con-
matching. When using double HPC(CB) units sideration, especially when more than one
(as in most adult patients), there are no guide- equivalent HLA-matched unrelated donor is
lines for HLA matching between the two units available and eligible. These include the under-
other than the minimum requirement, for each lying disease and the disease stage, the clinical
unit, of 4/6 HLA matches with the patient's stability of the patient, CMV status of the pa-
HLA specificities. Nevertheless, some centers tient and donor, ABO blood group matching
prefer to use at least a 4/6 match between the with the patient, donor gender and age, weight
two units. With UCB, it is also possible to select discrepancy between donor and patient, and, in
permissible HLAmismatches with reduced im- some institutions, killer cell immunoglobulin-
munogenicity, such as the noninherited mater- like receptor (KIR)status of the donor.' Male
nal antigens (NIMA).NIMA matching is not es- gender, younger age, nulliparity, matching of
sential, although in patients with hematologic ABO and CMV status, and greater size of the
malignancies, NIMA-matched grafts have been donor relative to the recipient may have positive
associated with a lower TRM and improved en- effects. For example, survival is greater for recip-
graftment, leading to reduced overall mortality ients of grafts from younger, unrelated donors
compared with HLA-mismatched, non-NIMA- aged 18-32 years compared to grafts from older
matched grafts." donors, after controlling for HLAcompatibility."
Haploidentical transplantation, using an HPC Other variables specificto HPC(CB)donation in-
donor who is a parent, a child, or a sibling clude maternal history as well as UCB collec-
matched at only half of the HLAloci with the re- tion, processing, and unit storage conditions."
cipient, is becoming more common as a result of A scoring system, the "cord blood apgar," has
advances in GVHDprevention. A major compo- been derived to determine the utility of the
nent of success with haploidentical transplanta- HPC(CB)unit by considering the total nucleated
tion is the use of posttransplantation cyclophos- cell (TNC) count, CD34+ cell count, colony-
phamide." Advantages of haploidentical HPC forming-unit (CFU) count, mononuclear cell
transplants include faster procurement and near- (MNC) content, and product volume."
universal donor availabilitybecause the majority
of patients needing an allogeneic HPC trans- Graft Source
plant have access to one first-degreerelative.29,42
Haploidentical HSCT presents unique challeng- Graft sources include HPC(A), HPC(M) and
es, including overcoming immunologic barriers HPC(CB). The choice of graft source is deter-
between donor and recipient with chemothera- mined mostly by the transplantation center's
py or HPC product manipulation. HSCT in re- preference and experience. Donor preference
cipients with antibodies against donor HLA de- should also be considered. The three sources of
terminants have a higher risk of primary graft grafts have biologic differences related to their
failure and associated adverse outcomes." It is different cellular composition. HPC(A)products,
essential to test the recipient for preformed the preferred source for autologous and alloge-
donor-specific antibodies (DSAs)for HLA. The neic transplantation in adults, engraft faster, re-
antibody level at which DSAshave a significant constitute immunity more quickly, and may ex-
impact is not clear, nor is the appropriate course ert a greater GVN effect when compared to
of action when DSAs are detected; however, HPC(M) and HPC(CB). These features make
C HAP T E R 26 The Collection and Processing of HPCs 743
HPC(A) attractive for patients undergoing non- ly posttransplantation period. HPC( CB) grafts
myeloablative (NMA) and reduced-intensity have more immature T cells and, thus, are less
conditioning (RIC) regimens and may offer an immunologically reactive. Consequently, they
advantage in overall and dtsease-free survival for are associated with higher risk of engraftment
selected patients with late-stage hematologic failure (particularly with NMA regimens) and
rnalignancies." Chronic GVHD (cGVHD) con- delayed immune reconstitution. By cornpari-
tinues .to be a major long-term complication of son, the potential for neoplastic disease relapse
HPC(A)grafts. may be lower for patients receiving HPC(CB)
HPC(M)was the primary source of HPCs be- grafts for malignancy and minimal residual dis-
fore the availability of mobilized peripheral ease at the time oftransplantation." The risk of
blood grafts collected by Ieukocytapheresis, and GVHD with UCB depends on the degree of HLA
allogeneic HPC(M)from pediatric siblingdonors disparity with the recipient.
continues to be used predominantly in chilo
dren. HPC(M) products have sign~ficantlyfew-
er T cells than HPC(A)products, and as a result lL
have a higher risk of engraftment failure and de-
layed immune reconstitution, as well as a poten- Patients undergoing HPC collection for a future
tial risk of disease relapse (ie, less GVN effect). autologous transplantation are evaluated for
However, HPC(M) grafts are a,ssqciated with their medical fitness and abilityto tolerate either
lower risk of cGVHD. Inrecipients ofmYeloab· marrow harvest or mobilization treatment with
lative transplants from unrelate4. donors for he- G·CSF and leukocytapheresis. Allogeneic do-
matologic malignancies, a randomized con- nors, once selected, also require evaluations to
trolled trial indicated that HPC(M) and ensure the safety of the donation process (ie, do-
mobilized HPC(A)graftsyield.equivalent effects nor suitability) for them, in addition to assess-
on survival. However, marrow gra,ftswere asso- ments to minimize the risk of transmission of In-
ciated with less cGVHD but more graft fail- fectious disease agents (te, donor eligibility).
ure." In children, the risk 0fengraftment failure Allogenetc donor eligibility is determined
is lower with HPC(ML as they usually receive through compliance with the screening require-
adequate CD34+ cells dueto their smaller body ments mandated by FDA regulations (21 CFR
weight. Children may also better tolerate infec- 1271). Suitabilitydetermination and risk assess-
tious complications when Irnmune reconstitu- ment tor relilt~dand unrelated donors are based
tion is delayed, compared to adults, who also of on accreditation ..standards and published con-
ten have more serious medical comorbidities. sensus guldelines and criteria.50·53 After screen-
The risk of relapse of neoplastic diseases with ing results are available, the donor is document-
HPC(M) (by virtue of lesser GVN effect) may be ed to be either: eligible and suitable; acceptable
mitigated by myeloablative regimens, which as a donor despite ineligibility (and justified
children can tolerate better than adults. through urgent medical need criteria); or de-
HPC(CB)units are cryopreservedand usually clared ineligible or medically unsuitable and de-
readily available from cord blood banks after ac- ferred from donation.
ceptable matched units are identified. They are Informed consent must be obtained from all
typicallyof smallvolume with l-log-fewerTNCs autologous patients and allogeneic donors or
and CD34+ cells/recipient weight, compared to their designated representatives before any col-
HPC(A)and HPC(M) grafts. For most adults, 2 lection procedure. The consent process .in-
units [double cord transplantation or double eludes providing the donor with information re-
HPC(CB) transplants] are required in order to garding the risks and benefits of the selected
constitute an adequate dose for a successful procedure, any tests required to protect the reo
transplantation. When double cord units are cipient, alternative collection methods, and pro-
used, eventually only one HPC(CB) unit en- tection of donor health information. In the case
grafts and the other one disappears after contrlb- of allogeneic transplantation, the donor should
uting to cellular immune support during the ear- be made aware of any potential consequences of
not proceeding with the donation once the re- A high recipient-to-donor weight ratio may
cipient has started conditioning therapy. The do" require a relatively high volume of marrow to be
nor is always given the opportunity to ask collected from a small donor who has limited to"
questions and to refuse donation. Consent must tal blood volume. In such cases, preoperative
be obtained from the maternal donor for UCB autologous blood storage can be considered to
collection, processing, testing, storage, and med- avoid the potential need for allogeneic red cell
ical use." A medical order is required for pro" transfusion. The NMDP guidelines suggest re-
curement and must include collectiongoals.50(P59) moving no more than 20 mL of marrow per ki-
Facilities that collect HPCs are required to logram of donor weight. 57 The volume of the
ensure donor access to medical care based on harvest is dictated by the recipient's weight, the
the risks and clinical situation associated with underlying diagnosis,and by treatment protocol.
each type of donation. Licensed providers, nurs- Usually a minimum of 2 to 3 x 108 nucleated
es, and allied health professionals must be cells/kg recipient weight are needed to facilitate
trained and experienced in HPC procurement efficient engraftment. Checking the nucleated
and product handling. They must be competent cell concentration of an aliquot from the mar-
to manage any potential adverse events (AEs)in" row product midway through the harvest can
curred by donors and patients and any technl- help estimate the total volume of marrow need-
cal variances that might affect the integrity or ed to reach the TNC goal and will prevent un-
quality of the HPC product. necessary blood loss. Determination of CD34+
cell concentration midway through the proce-
Marrow HPCCollection dure is not practical, given the slow turnaround
time for this assay and the high priority to mini-
Marrow harvest is an invasive operative proce- mize the procedure and anesthesia time.
dure performed under sterile conditions and Marrow harvest techniques vary consider"
typically under general anesthesia. In addition ably, depending on institutional practice; how"
to undergoing relevant donor screening, infec- ever, a standard approach by experienced opera-
tious disease testing, and HLAcompatibilitytest" tors optimizes donor safety and product
ing, marrow donors must also be physicallysuit" quality, 57 At least two individuals are required to
able for donation and be able to tolerate the type perform the procedure, one of whom is the se-
of anesthesia required to perform the harvest nior surgeon. After the induction of anesthesia,
successfully. The donor's preoperative physical the donor is placed in the prone position and the
status can be assessed using the American Sod- posterior iliac crests are asepticallyprepared and
ety of AnesthesiologistsPhysical Status (ASAPS) draped with sterile barriers. An 11- to l-t-gauge
classification system." Donors with preexisting needle with attached syringe flushed with anti"
ischemic heart disease, cardiac failure, cerebro- coagulant is inserted into each posterior iliac
vascular disease, insulin-dependent diabetes crest, and approximately 5 mL of marrow is as"
mellitus, or liver or renal dysfunction are at pirated. The needle and syringe are then rotated
higher risk of AEswith general anesthesia." Do- up to 180 degrees, and the aspiration is repeat"
nors with serious oropharyngeal, neck, back, ed. Large-volume aspiration is avoided to pre-
spine, or hip conditions; abnormal platelet func- vent significant peripheral blood contamination
tion or coagulopathy; or history of malignant hy- of the product. The aspirated marrow is then
perthermia should be precluded due to the risks collected into a large collection bag containing
of anesthesia. Donors should be counseled heparin anticoagulant mixed with acid-citrate-
about potential bone, nerve, or vessel damage dextrose formula A (ACD"A)anticoagulant, tis"
and bleeding during or after the marrow har- sue culture medium, or physiologic buffer. Ap-
vest. Autologous donors may have had previous proved collection systems incorporate In-line fll-
radiation therapy to the pelvis, which will limit ters to remove bone spicules, aggregates, and
the amount of marrow nucleated cells available debris. The process Is repeated at different bone
in the posterior iliac crests. sites (aiming to minimize the number of skin
puncture sites) until the target TNC count or monly placed and used for both HPC(A) collec-
maximum safe donor-volume limit is reached. tion and longer-term intravenous (IV) fluid/
Marrow donation is generally considered a medication administration during the transplan-
safe procedure in healthy donors, and serious tation process. Peripheral venous access is usual-
complications of marrow harvest are rare. Mi- ly sufficient for adult allogeneic donors who can
nor complications, such as pain at the site of tolerate one or two procedures using two large-
harvest, fatigue, insomnia, nausea, dizziness, bore needles (16/17 gauge). Nevertheless, 7%
and anorexia, occur frequently but resolve in to 10% of healthy volunteer donors overall and
most donors by 1 month after the procedure." approximately 20% of female donors require a
A large study of unrelated marrow donors ob- temporary central venous catheter. 59 CVCs are
served a 3% prevalence of persistent discomfort usually placed in the internal jugular veins, and
compared to baseline at 6 months." Not surpris- less commonly in the subclavian, avoiding femo-
ingly, marrow donors suffer more immediate ral access because of the higher risk of infection
postdonation physical side effects compared to and discomfort. CVCs are associated with addi-
donors of HPC(A) products; however, their tional risks such as bleeding, thrombosis, and
overall experience over the long term is equlva- pneumothorax. Correct placement of the line
lent.60,61 Marrow donors often have significant needs to be confirmed before apheresis for
decreases in hemoglobin concentration after the HPC(A)collection.50(p58)
procedure. Depending on the marrow volume For patients undergoing autologous trans-
to be collected, marrow donors may be advised plantation, HPCs are mobilized into the periph-
to store an autologous unit of Red Blood Cells eral circulation using hematopoietic growth
(RBCs)before the procedure. Historically,up to factors, most commonly G-CSF(filgrastim)with
70% to 76% of marrow donors received at least or without mobilization chemotherapeutic
1 autologous RBC unit during or shortly after agents.64.68 More recently, HPC mobilization in
marrow harvest.58,59,62 More recently, the utility patients has been augmented with adjunctive
of autologous blood storage and transfusion in use of plerixafor, the reversible CXCR4receptor
this setting has been questioned.f If the donor antagonist.69·72 Long-acting, pegylated G-CSF
requires an allogeneic RBC or platelet transfu- (pegfilgrastim) and biosimilar forms of G-CSF
sion before or during the procedure, the units are also now available as alternative mobilizing
should be irradiated to prevent contamination of agents, with efficacyequivalent to filgrastim.66,73
the graft by viable leukocytes from these blood Other hematopoietic growth factors, such as
components. Marrow donors should be made GM-CSF,are rarely used for HPC mobilization.
aware during the informed consent process that Adult allogeneic donors undergo HPC mobiliza-
they might need a transfusion. tion with G-CSFalone. Chemotherapy is never
used to mobilize stem cells in healthy donors,
Peripheral Blood HPCCollection and plerixafor is still an experimental agent for
this indication." Mobilized HPC(A) collections
Over the last few decades, HPC(A) products using G-CSFhave been safely performed in sib-
containing mobilized peripheral blood HPCs ling pediatric donors, but this is not a uniform
have been the most frequently used graft source practice in all pediatric transplantation centers."
of CD34+ cells for autologous and allogeneic Myeloid hematopoietic growth factors, such
transplantation. Collection of HPC(A) is per- as G-CSF, cause expansion of the granulocytic
formed mostly as an outpatient procedure be- population in the marrow and the release of pro-
cause of the ease of the procedure and limited teases that disrupt several cytoadhesive. "an-
side effects. Adequate vascular access is re- chors" that retain HPCs within the hematopoi-
quired to accommodate the high flow rates nec- etic niche. This includes the binding of CXCR4
essary for efficient blood processing. In patients on the HPC membrane with stromal-cell-
undergoing HPC(A) collection for autologous derived factor 1 (SDF1)in the marrow stroma.
transplantation, a semipermanent (tunneled) When used alone, G-CSFis administered daily
apheresis central venous catheter (CVC)is com- over at least 4 to 5 days to achieve a major
mobilization effect and significant increase in number of CD34+ cells for efficient engraft-
circulating CD34+ stem/progenitor cells. As ment. The minimum number of cells needed for
monotherapy or after mobilization chemothera- transplantation is commonly cited as 2 x 106
py, G·CSFis typically dosed at 10 ug/kg/day; CD34+ cells/kg recipient weight, although
however, higher daily doses (up to 16 ug/kg/ 5 x 106 CD34+ cells/kg is more destrable.":"
day in two divided doses) have been grv- Patients with poor mobilization may benefit
en.65,66,68 For practical purposes, G-CSF doses from higher-dose G-CSFand/or the addition of
are often rounded to the nearest vial size, and plerixafor in combination with G-CSF.Various
effective dosing may be based on adjusted ideal clinical studies have demonstrated that plerixa-
body weight in obese patients.66,68 When mobi- for in combination with G-CSFSignificantlyin-
lizing HPCs from "steady state" (ie, without pri- creases HPC collection yields compared to G-
or chemotherapy), apheresis is usually initiated CSF alone.69-72 Several clinical scenarios where
96-120 hours after the start of G-CSFadminis- plerixaforwas used as a salvage agent have been
tration, when the peripheral CD34+ cell con- described.66,72 Randomized clinical trials in pa-
centration is peaking. Leukocytapheresisis opti- tients with multiple myeloma or lymphoma
mized for collection of MNCs and will yield a have demonstrated that the addition of plerixa-
product rich in CD34+ cells. Daily G-CSF ad- for to G-CSFcan double the circulating CD34+
ministration and leukocytapheresis continues cell HPCs, allowing fewer apheresis procedures
until the desired number of CD34+ cells are ob- to collect a sufficientquantity of HPCS.70,71,80
tained. Procurement of HPC(A)from a patient or do-
The time course of HPC mobilization and nor by leukocytapheresis typically involves pro-
apheresis after chemotherapy plus G-CSFvaries cessing two to three times the total blood vol-
based on the treatment regimen and the pa- ume (TBV).This translates to processing 10-12 L
tient's baseline clinical and hematologic sta- for an adult in accordance with the instrument
tuS.66,68 Certain chemotherapy agents at specific hardware, software programming, and manufac-
doses will cause significant but transient mar- turer's instructions. Large-volume leukocyta-
row suppression. During the recovery from mar- pheresis (LVL),which involves processing be-
row suppression, there is a substantial increase tween three and six times the TBV,is frequently
in the number of circulating CD34+ cells, and performed to collect higher numbers of CD34+
this effect can be magnifiedby the dailyadminis- cells in children and adults undergoing autolo-
tration of G-CSF. This cherno-mobllizatlon ap- gous HSCT.81
proach leads to a sharp increase in circulating Anticoagulation used for HPC(A) collection
CD34+ cells, typically at 9 to 11 days from the varies according to local institutional practices
administration of chemotherapy, at which time and instrument requirements. The standard an-
daily leukocytapheresis is initiated.65,66,68 The ticoagulant is ACD-A,either alone or in combi-
threshold for starting leukocytapheresis is typi- nation with heparin." Toxicities and side effects
callywhen the CD34+ cell countis at lO/I-lLor of ACD-A are related to hypocalcemia, with
greater. Side effectsof G-CSFare common, mild, symptoms of paresthesias, muscle irritability,
and transient. These include bone pain, myalgia, and, less commonly, muscle spasm and cardiac
headache, insomnia, and flu-likesymptoms; less arrhythmia. For LVL procedures, the require-
frequently, sweating, anorexia, fever, chills, and ment for high volumes of ACD-A over the
nausea occur.58-62,68,75 Potentially serious compli- course of treatment can lead to a large net-
cations, such as splenic rupture, severe throm- positive fluid balance and potential fluid over-
bocytopenia, and acute lung injury are rare. load complications. Up to 20% of donors experi-
For some autologous patients and, rarely, alence minor apheresis/ collection-related AEs,
logeneic donors, mobilization of HPCs can be such as citrate toxicity, nausea, fatigue, chills,
challenging and result in inadequate collections. hypertension, hypotension, allergic reactions, or
Cases of poor mobilization may require addi- syncope.59,62,75 Use of heparin results in modest
tional pharmacotherapy or repeat attempts at systemic anticoagulation for a short time after
mobilization in order to achieve an adequate completion of the procedure, with mild bleeding
C HAP T E R 2 6 The Collection and Processingof HPCs 747
risks in patients with severe thrombocytopenia. define the potential financial and survival bene-
Adjunctive use of heparin also carries a small fits.
risk of heparin-induced thrombocytopenia
(HIT)PThe benefit of heparin-based anticoagu- Umbilical Cord Blood B-IPCCollection
lation during HPC(A)collection is that the total
VCB Can be collected either before delivery of
volume of anticoagulant can be two to three
the placenta (in utero) .or after delivery of the
times smaller, thereby decreasing risk of volume
placenta (ex utero) in either a vaginal or cesare-
overload and citratetoxicity.
an birth. Several reports have suggested higher
Once HPC collection starts, daily apheresis
collection volumes with cesarean deliveries and
continues until the target CD34+ cell goal is when VCBis collected in utero.89,90Both in- and
reached. Prediction algorithms can be used to
ex-utero collection methods continue to be rou-
estimate the required processed blood volume tinely used, but in-utero collections are more
to achieve a desired CD34+ cellyield. The algo- common." VCB is collected by cannulating the
rithms use a precollection CD34+ count along umbilical vein and allowing the placental blood
with the expected collection efficiency for the to be removed by gravityinto collection contain-
standard or large-volume leukocytapheresis pro- ers to which anticoagulation, most commonly
cedure.83,84Institutions vary in their practice in citrate-phosphate-dextrose (CPD),has been add-
terms of when to commence HPC(A) collec- ed. Closed systems using collection bags have
tions, with some centers measuring CD34+ cell reduced the incidence of bacterial contamina-
levels before collection in each patient, and oth- non." The collection bag is weighed to estimate
ers using this approach only in patients with the collection volume. The volume collected
known risk factors of mobilization failure. For varies but usually ranges from 50 to 200 mL.
most allogeneic donors, sufficiel1tnumbers of There is a close correlation between the weight
HPCs can be obtained in one to two collections of the bag and the TNC count, and often cord
commencing on day 4 or 5 ofG·CSF administra- blood banks will have a rninlmum volume
tion. A complete blood.count, including platelet threshold for the VCB to be shipped to the pro:
count, is required within 24 hours before leuko- cessinglaboratory. Some cord blood banks estab-
cytapheresis. Thisis especiallyimportant for au- lish thresholds for banking based on TNC and
tologous patients recovering from mobilization will measure the TNC at the collection site. Fac-
chemotherapy, because they may still be throm- tors associated with higher VCB volumes and
bocytopenic at the time of collection, and the greater yield of nucleated cells include birth-
apheresis procedure itself reduces the blood weight, placental weight, gestational age, induc-
platelet count by 30% to 50%.51,59,75,81 tion of labor, prolonged labor, cesarean section,
Some studies have demonstrated that the ab- early cord clamping, firstborn infants, European
solute lymphocyte count on day 15 after trans- ancestry, and female infant gender." The train'
plantation (ALC-I5)is an independent prognos- ing and experience of the VCB collector will of-
tic factor for improved survival after autologous ten alsohave an effecton the volumes obtained.
HSCTfor certain hematologic malignancies." In All studies focused on HPC(CB)have demon-
addition, increasing the number of apheresis col- strated the .impact that infused TNCs, CFVs,
lections to achieve an absolute count of >0.5 x and CD34+ cells have on engraftment, trans-
109 lymphocytes/kg has been associated with plant-related events, and survival.93,94Some-
early ALCrecovery and improved outcomes.v'" times, and particularly in adults, there is an in-
Although it may be possibleto perform addition- sufficient cell dose in a single HPC(CB) for
allymphocyte collections without mobilization, transplantation. This :cell dose limitation has
the costs of additional days of collection and pro- been overcome by the development of double
cessing are not inconsequential. More data, in- HPC(CB) transplantations." A large CIBMTR
cluding randomized clinical trials, need to be study demonstrated that double HPC(CB)trans-
obtained and analyzed to confirm the applicabil- plants had Similaroutcomes to single HPC(CB)
ity of these observations in other populations to transplants in patients with acute myelogenous
leukemia." This has been confirmed by other processing is commonly used for plasma reduc-
studies in patients with hematologic malignan- tion, red cell reduction, and buffy-coatprepara-
cies.97,98 According to several different reports, a tion before cryopreservation and storage. For
minimum TNC count of 3 x 107/kg at cryopres- clinical indications, plasma or volume reduction
ervation needs to be obtained for engraft- may be required for a minor-ABO-mismatched
ment.94,95 For double HPC(CB) transplants, allograft (marrow or peripheral blood) with high
some centers require that each HPC(CB) unit titers of anti-A and/or antl-B to reduce the
has a minimum TNC count of 1.5 x 107/kg at amount of donor isoagglutinins and risks of re-
cryopreservation." A cryopreserved TNC dose cipient hemolysis.101 Plasma volume reduction
of 2.5 x 107 cells/kg, however, has typically might also be required to prevent fluid overload
been accepted as the minimum cell dose for suc- in recipients with small body weight or those
cessful engraftment in single HPC(CB) trans- with renal disease or cardiac failure. Beforestor-
plants." In double HPC(CB)transplants, an ar- age, product volume reduction is often neces-
bitrary precryopreservation CD34+ cell dose of sary to concentrate the cells and reduce the vol-
:2: 1 x 105/kg has been used by several transplan- ume before addition of cryopreservative and
tation centers as the minimum accepted cell freezing. The smaller volumes followingplasma
doses for each unit." reduction allow for a smaller number of bags
that require storage, thawing, and infusing.
Smaller volume translates into reduced risk of
PROCESSING HUMAN dimethylsulfoxide (DMSO) toxicity at the time
PROGENI EllS of infusion.
After collection, HPC products are transferred Red Cell and Plasma Reduction
from the operating room or apheresis center to
the processing facility either at room tempera- Red cell reduction may be required to prevent
ture or at 2 to 8 C, depending on the type of hemolysis of donor cells in major-ABO-
product and anticipated duration of transporta- mismatched allograftsthat contain an unaccept-
tion and storage. The HPC product may subse- able volume ofred cellsfor a recipientwith anti-A
quently undergo quality control (OC) testing and/or anti-B isoagglutinin titers >16.101 The
that may include cell enumeration, flow cyto- amount of acceptable incompatible red cells in
metric immunophenotypic studies, sterility test- the final product is defined by institutional poli-
ing, and viabilitystudies. Some products will re- cy but is usually in the range of 20 to 30 mL or
quire further processing, cryopreservation, and 0.2 to 0.4 mLlkg (for pediatric recipients). 101
storage. Cryopreserved HPC products that are Most HPC(A) products have low hematocrits
anticipated to be stored for many years should and smaller product volumes such that the risk
be maintained under conditions that retain cell of a hemolytic transfusion reaction is minimal.
viability and function as demonstrated by stabili- Bycomparison, HPC(M)products have high he-
ty studies. Processingrefers to all aspects of ma- matocrits and larger total volume and red cell
nipulation, cryopreservation, packaging, and la- volume such that red cell reduction is frequently
beling of cellular therapy products. During required. Cord blood banks often reduce red
processing, steps are taken to guarantee the po- cells from products before cryopreservatton.!"
tency and purity of the product for transplanta- This minimizes unwanted red cells that lyse on
tion and/or storage. thawing and reduces the final volume to opti-
HPC processing methods are often divided mize storage space and limit costs.
into basic/routine (minimal manipulation) pro- Classically, HPC product red cell reduction
cedures that are commonly used in clinical cell employs sedimentation agents such as hy-
processing laboratories, and more specialized droxyethyl starch (HES) and centrifugation to
manipulations that involve complex technolo- pellet the red cells, or gravity, where hanging
gies and more than minimal manipulation of the the bag allows for sedtmentation."? If both ma-
product (see next section). 100 Centrifuged-based jor and minor ABOmismatches exist (termed bl-
directional ABO incompatibility), red cells and solution (eg, 10% dextran followed by 5% albu-
plasma may need to be reduced from the prod- min), transfer into an appropriately sized bag for
uct. For larger marrow volumes, buffy-coatcon- centrifugation, and resuspension of the cell pel-
centration of marrow may be achieved by cen- letts) before delivery to the patient-care unit for
trifugation and harvesting using an apheresis or infusion. Some laboratories perform two centrif-
cell-washing device.100,103Newer apheresis in- ugation steps-removing the supernatant from
struments can efftciently reduce red cells from the first spin and centrifuging that portion a sec-
larger-volume marrow products with excellent ond time before combining the two cell pellets;
CD34+ cell recovery.'?' Regardlessof the meth- this optimizes cell recovery. lOB In order to mini-
od used, both volume and red cell reduction mize cell loss, several processing laboratories are
procedures must be weighed against the poten- now thawing and diluting the HPC(CB)product
tial loss of HPC numbers and viability and in- without washing the cells.102,108-111 With these
creased risk of contamination during the addi- no-wash methods, studies have demonstrated a
tional manipulations. high rate of engraftment with a low incidence of
serious adverse reactions following HPC(CB)in-
Storage Preparation fusion.109,110
Most allogeneic HPC(M) and HPC(A)products In selected circumstances, such as for pa-
are Infused fresh within 48 to 72 hours after col- tients with a documented severe DMSO allergy,
lection. If the fresh products are not intended to washing thawed HPC products may be the safer
be infused within that time, they are cryopre- choice. Washing to remove DMSO and other
served. lOS HPC(A)that are transported or stored additives raises concerns over loss of HPCs due
for more than a few hours should be maintained to additional manipulation and increased expo-
at 2 to 8 C and, ideally, diluted to a cell concen- sure time of cells to DMSO. Various procedures
tration that minimizes metabolic stress. Bycom- have been developed to remove DMSO through
parison, HPC(M) products do notrequire dilu- manual or automated methods, and some of
tion and may be stored and transported at room these have been associatedwith reduced infusion-
temperature for up to 48 to 72..hours without related AEs.112-114Importantly, the cellular com-
Significantloss of CD34+ .cell viability or num- ponent of thawed HPC(A)products, particular-
bers.l'" Autologous HPC(A) products are con- ly granulocytes, may also contribute to toxicities
centrated and cryopreserved until the patient during or shortly after reinfusion.us Separating
has undergone preconditioning therapy and is out these cellular components and debris after
ready for transplantation. thawing to mitigate toxicity is technically chal-
lenging; however, limiting the number of
Wilshing infused nucleated cells can reduce severe
infusion-related complications.!" In most set-
Washing the thawed HPC product after cryo- tings, it is safe to infuse the cryopreserved
preservation removes lysed red cells, hemoglo- HPC(A) product directly into the patient after
bin, and the cryoprotectant (le, DMSO). Be- the thawing process is completed at the bedside.
cause some HPCs may be lost with postthaw
manipulation and washing of cryopreserved Thawing
HPC(M) and HPC(A) products, this processing
step is not routinely performed. Bycomparison, Cryopreserved HPC(A)or HPC(M) products are
cryopreserved HPC(CB) products are routinely usually thawed at the time and place?f th~
processed after thawing even if they have under- planned infusion-typically at the bedside. This
gone red cell reduction before storage.102Histor- is done to minimize the amount of time cells are
ically, most institutions based their HPC(CB) exposed to DMSO in the liquid state after thaw-
processing methods, including the thawing! ing and before infusion. Rapidthawing in a 37 C
washing process, on the procedure adopted by waterbath followed by infusion as quickly and
the New York Placental BloodProgram.!" This safely as. possible is recommended to avoid
involves slow, sequential addition of a wash DMSO-mediated cytotoxicity.!" Caution must
be used, however, because frozen plastic con- separates the cell populations of a cell product
tainers are prone to breakage.!" The cryopre- based on density alone. However, as the liquids
served product is usually brought to the bedside are passed through the chamber housing the
in a liquid-nitrogen dry shipper. After removal, cells in the direction opposite (counterflow) to
the product must be handled with care while it the centrifugal force, adjustment of flow rate
is verified to determine the product's identity and/or centrifugation speed also allows separa-
and ensure the integrity of the bag. The product tion of cell populations based on size.119,120
is then placed into a clean or sterile plastic over- Through this process, cellswith specificsize and
wrap bag and submerged in a 37 C waterbath. density profiles can be separated from other
While the bags are in the waterbath, the prod- cells. This method has been successfully used
uct is gently massaged to achieve an icy slush to isolate monocytes from apheresis products
consistency. If the product bag breaks, the cells for vaccine applications and lymphocytes for
may be recovered from the outer bag; however, adoptive immunotherapy. This method was
a risk/benefit discussion should occur regarding also historically used for T-cell depletion of
the need for infusing potentially contaminated HPC grafts.
cells. The same risk applies when the washing
of a cellular component is required. If bag leak- Cell-Selection Systems
age occurs, a hemostat should be used to pre-
In positive cell selection, strategies are em-
vent loss of the product, and the contents
ployed to mark and isolate a target cell popula-
should be aseptically diverted into a transfer
tion of interest that becomes the desired HPC
bag.!" A sample should also be sent for culture.
product. Immunomagnetic cell-selection sys-
Local hospital and laboratory policies must be
tems such as the CliniMACSsystem (Miltenyi
followedfor recipients with multiple products to
Biotec Bergisch) incorporate monoclonal
be infused at a given time. Many facilitiesthaw
antibody-based technologies to target cell-
and infuse products sequentially to ensure that
surface antigens. For positive cell selection of
the previous unit was infused safelyand without
CD34+ cells or CD34+ enrichment, a magneti-
major AEsbefore thawing and infusing the next
cally labeled CD34 monoclonal antibody re-
bag.
agent is mixed with the HPC product to first tag
the cells.Magneticallylabeled target cells are re-
tained as the cell suspension passes through a
column in which a magnetic field is generated.
Unlabeled cells pass through the column and
are collected in a negative-fraction bag. Once
Specializedprocessing of HPC products can op-
timize product purity and potency beyond levels the column has been washed to ensure that only
obtained by routine methods. More-specialized the CD34+ cells are retained, bound cells are
then released from the column by removal of
manipulations require unique reagents and in-
the magnetic field. The CD34+ cells pass into a
strumentation, and these are discussed else-
separate collection bag as the HPC product. Tar-
where. Therefore, the descriptions of methods
get cell recovery has been reported to be 50% to
in this chapter are brief and focus on their appli-
100% when successfully performed.121,122 This
cation to HPCs.
strategy is very effective at enriching the prod-
uct for CD34+ cells, with purities of 90% to
Elutriation
99% commonly obtained. Other cellular constit-
Counterflow centrifugal elutriation separates uents of the original product are removed from
cell populations based on two physical charac- the HPC product, and thereby the procedure
teristics-size and density (sedimentation coeffi- also becomes a method for passive T-cellreduc-
cient)-and uses a stream of fluid/media flow- tion or depletion.
ing in a direction usually opposite to the In negative-selection procedures, the unde-
direction of gravity sedimentation. A centrifuge sirable target cell population(s) is actively re-
C HAP T E R 26 TheCollection and Processingof HPCs 751
moved from the HPC product. 122 Negative- for leukemia or other disorders associated with
selection methods have been used to remove an expected prolonged period of neutropenia. 128
CD3+/CD 19+ cells as the target in HPC prod- Most expansion cultures contain a cytokine
ucts in an effort to reduce the risk of severe cocktail that includes SCF, Pms-llketyrosine ki-
GVHD. However, in order to reduce the inci- nase 3 (FLT-3) ligand, and thrombopoietin,
dence of graft rejection, some investigators sup- along with novel and/or proprietary ingredi-
plement the HPC graft by adding back a smaller ents. The media, culture vessels, and culture du-
fixed dose of CD3+/CD 19+ cells.123 One major ration vary from protocol to protocol.
benefit to negative selection is the retention of
other cell populations such as natural killer cells
in the HPC graft, which could aid in providing c PRESE N
an antitumor effect. Newer targets for negative
"specific" selection include T-cellreceptor (TcR) Cryopreservation allows for long-term storage of
a/~, and CD45RA+ cells.124,125 Historically, HPC products and is used predominantly for
negative selection was achieved through physi- HPC(CB)and autologous HPC(A)products that
cal methods such as soybeanlectin agglutination have been collected for future hematopoietic
followed by sheep erythrocyte-rosette depletion rescue following high-dose therapy. Allogeneic
(E-rosetting), or counterflow elutriation. More products may be cryopreservedif the transplan-
recently, immunologic approaches to T-cell de- tation is unexpectedly delayed (eg, due to the
pletion have used monoclonal antibodies, such recipient's medical condition), if the donor will
as anti-CD3 monoclonal antibodyconjugated im- not be availableat the time of transplantation, or
munomagnetic beads. in situations where more donor. cells are ob-
tained than needed (and excess cells are stored
Cell Expansion for future use). During HPC product cryopreser-
vation, a cryoprotectant agent is required to pre-
Because the infused dose of nucleated, CD34+, vent ice crystal formation within and outside of
and colony-formingcells in an HPC product is cells. The freezing process must also.be slow,
positively correlated with the speed of engraft- ideally using an automated step-wise method
ment and patient outcomes, much effort has (controlled rate). Ice crystal formation can result
been focused on the abilityto expand HPCs and in cell injury and death, resulting in reduced via-
other progenitors ex vivo. Ex-vivo expansion ble cell recovery at thawing and potential slow
has the potential to increase the number of lym- engraftment after transplantation. The concen-
phocytes, committed progenitors, and long-term tration ·of cryoprotectant and the steps in the
repopulating hematopoietic stem cells. In recent cryopreservation protocol must be validated by
years, HPC(CB)products have been the focus of each HPC processing facility for each product
expansion trials because UCB HPCs possess type. Standard protocols are used, but consen-
higher proliferative and self-renewal capacity, sus is lacking on a universally accepted method.
and the use of these products is limited by the Similarly, the maximum acceptable duration of
relatively small quantity of nucleated and cryopreserved storage for HPCproducts remains
CD34+ cells.126,127 Ex-vivo expansion could in- undefined; 129-!31
crease the hematopoietic stem cell content and DMSO is the most frequently used cryopro-
improve the availabilityof adequately sized and tectant for HPC product cryopreservation. This
matched DCB units for transplantation. It may highly polar solvent penetrates the cell mem-
also eliminate the need for double HPC(CB) brane, reduces intracellular ice formation, and
transplants for those without an adequately prevents cell damage due to dehydration during
sized single HPC(CB) graft.126,m Ex-vivo ex- freezing.132 The most common additives for
panded HPC(CB)products that lack significant cryopreservation of HPC(A)or HPC(M) are plas-
numbers of T cells and contain more committed ma in saline, human albumin in saline, dextran,
progenitors might also provide a bridge to early and HES. HES and dextran are the preferred
granulocyte recovery after aggressive.treatment and optimal additives for cryopreserving
HPC(CB) products.F"!" Because of concern desired 1 to 2 C/minute. The freezing rate can
about adverse reactions to DMSO (see "Patient be monitored by placing the probe of an elec-
Care"), newer cryoprotectants, such as treha- tronic temperature monitor inside the cassette,
lose, are currently being assessed as alternate against the cryopreservation bag. Cell Viability,
cryoprotectants for HPCS.136Laboratories vary recovery, and engraftment are generally compa-
in their final concentration of DMSO. A final rable to controlled-rate freezing.141Followlng
concentration of 10% (volume/volume) is most both controlled-rate and noncontrolled-rate
frequent, but lower concentrations (to <5%) freezing, the HPC product is transferred to a
have been reported as equally effective.137-139 Us- storage freezer. Most clinical facilities store HPC
ing low DMSO concentrations could decrease products in the liquid or vapor phase of liquid
cell toxicity, as well as reduce the risk of AEs at nitrogen (LN2;-195 Cor -150 to -125 C, re-
the time of infusion, but might also affect cell respectively). Some processing laboratories may
covery due to poor cryopreservation. Each pro- also place cassettes directly into LN2vapor stor-
cessing facility should validate the freezing solu- age without slow cooling to -80 C first. It has
tion and the DMSO concentration for optimal not been determined if viability and engraftment
viable cell recovery of the target population. potential are increased by storage in the liquid
Some laboratories use HES as a way to decrease phase vs the vapor phase; however, cryopre-
the concentration of DMSO (eg, 5% DMSO plus served HPC(A) and HPC(CB) have been evalu-
6% HES).140HESis a nonpenetrating (extracellu- ated in both phases, and they remain viable and
lar), macromolecular cryoprotectant; it likely function for at least 15 and 23.5 years, respec-
functions by forming a glassy shell, or mem- tively.142,143
In theory, storage in the liquid phase
brane, around the cells, retarding the extrusion would avoid the risk of transient warming
of water out of the cell and into the extracellular events when the freezer is entered. HPC cryo-
ice crystals. preserved products requiring quarantine can be
After addition of cryoprotectant, the HPC stored in the vapor phase of LN2in an attempt
product must be cooled slowly to preserve post- to minimize contamination risk. Other methods
thaw viability and function. Automated may include overwrap of collection bags or
controlled-rate freezing, using an instrument physical segregation during storage.
with computer programming capability, decreas-
es HPC product temperature incrementally and
in a closely monitored fashion. The HPC prod- UAUTY CONTR
uct is cooled at a rate of 1 C/minute until the
transition temperature from liquid to solid. At HPC product collection, transport, receipt, pro-
this point, when the solution starts to freeze, cessing, storage, and release are significant pro-
there is a release of latent heat of fusion. The cesses for QC monitoring in any quality manage-
freezer program then triggers a period of super- ment program involving HPCs. Policies and
cooling to counteract this heat release. Follow- procedures related to these issues include the
ing solidification of the HPC product, cooling re- operational techniques and activities that deter-
sumes at the rate of 1 C/minute until the mine the accuracy and reliability of the institu-
product has reached -60 C. From this point on, tion's personnel, equipment, reagents, and
the product is cooled at a controlled rate deter- operations, along with the handling and manip-
mined by facilitypolicy until it reaches -100 C. ulation of HPC products. Quality standards used
HPC products may also be frozen at a manu- in QC address testing and characterization of
al "noncontrolled rate." For this approach, the the product to ensure its safety, purity, and po-
HPC product with additives and cryoprotectant tency, as well as the requirements for release
is transferred into metal freezing cassettes and and distribution of the product for infusion. The
placed horizontally on shelves in a -80 C me- extent of QC testing primarily depends on the
chanical freezer. The metal cassette can be complexity of the manufacturing process and
wrapped in disposable absorbent pads or styro- the nature of the planned treatment, and wheth-
foam insulation to adjust the cooling rate to the er the HPC product is within the scope of stan-
dard practice or in the context of a clinical trial. ered the in-vitro "gold standard" for measuring
Testing requirements for the release of cellular progenitor potency. The practical application of
therapy products from the processing facility the CFU assay is limited by the 2-week culture
must be clearly defined and must comply with period and poor interlaboratory standardiza-
local, state, and federal regulations. Voluntary tion. The results of this assay correlate with the
standards are provided by AABB,FACT,the Col- speed and likelihood of engraftment of HPCs
lege of American Pathologists (CAP), The Joint from marrow, peripheral blood, or UCB.145·149 A
Commission, and the International Organiza- reasonable correlation between engraftment
tion for Standardization (ISO). These standards speed and the more rapidly available CD34+
define the requirements for QC, and accredita- cell count (flow cytometry) has made this assay
tion mandates compliance. the accepted surrogate QC test for graft potency.
Commonly performed QC tests for HPCs in- The clonogenic assay is still useful, despite diffi-
clude TNC count and differential, cell viability, culties in standardization, for HPC(CB)products
CD34+ cell count and viability, sterility testing, that are stored for long periods.
and, for allogeneicproducts, T-cellcontent. CFU Sterility testing of the product assesses for
assays are performed in some centers; however, aerobic and anaerobic bacteria and fungi. In US
these are not uniformly validated for standard laboratories, culture-based methods are the
clinical practice or release criteria. Several of most common tests, and these must be validat-
these tests.may be performed during in-process ed by each processing laboratory for their prod-
manipulations,such as red cellreduction, plasmal ucts and reagents. Other rapid methods are
volume reduction, MNC andlor cell subset en- needed for products that undergo more exten-
richment, cell depletion, and selection of target sive manipulation, and these may include Gram
cells. Cell count (or TNC count) and differential staining, endotoxin measurement, and myco-
are commonly performed on a validated hema- plasma testing.
tology analyzer. CD34+ cell enumeration is per- Additional considerations for release testing
formed by flow cytometry. Most CD34+ cell- include labeling and assessment of product ap-
enumeration strategies are based on guidelines pearance (eg, color, turbidity, and container in-
of the International Society for Cellular Therapy tegrity). Cell composition, storage conditions,
(ISCT).144TNC and CD34+ counts are general product expiration, patient identification, prod-
measures of product quantity but do not provide uct identification, processing laboratory name
information about viability. Cell viability may be and address, warnings, and precautions are
determined using various methods, including common labeling elements required for HPC
trypan blue, acridine orange, and 7-aminoacti- product release. The implementation of labeling
nomycin D (7-AAD).The use of 7-AAD,a fluo- standards by the International Society of Blood
rescent chemical compound with DNA affinity, Transfusion (ISBT)has helped to move standard-
in flow-cytometric analyses offers advantages ization forward in this matter. 150ISBT 128 is a
over traditional trypan blue staining. These in- codingand labeling system that was developed
clude decreased subjectivity, increased accuracy for blood and blood components to improve
(particularly with thawed HPC products), and quality, safety, and traceability in blood bank-
the ability to combine with CD34+ assessment. ing. The goal of ISBT 128 for HPC products is to
Because turnaround times for flow-based assays globally standardize terminology, coding, label-
may limit the laboratory's ability to release a ing, and identification of medical products of.hu-
product for fresh infusion, microscope-based man origin. Today the standard is managed by
methods using vital dyes or fluorescent stains the International Council for Commonality in
may be particularly useful for quick assessment Blood Banking Automation (ICCBBA).Full im-
of overall nucleated cell viability. plementation of ISBT 128 improves traceability,
The colony-formationassay, or clonogenic as- transparency, vigilance and surveillance, and in-
say with the counting of CFUs,has been consid- teroperability.
SH!PPIN durable material that is able to withstand

TRAN HPC shocks, pressure changes, and extremes of tem-
perature. Unrelated allogeneic fresh products
eEL U TS
that are shipped directly to a transplantation
Shipping and transportation refer to the physical center are typicallyintended for immediate infu-
act of transferring a product within or between sion into a patient who has already undergone
facilities. Transplantations requiring unrelated- preparative chemoradiotherapy. These products
donor allogeneicHPCproducts, and some related- generally travel in the custody of a qualifiedrep-
donor allogeneic transplantations that occur resentative or properly trained courier who has
when the donor and recipient are not located to- direct control of the product at all times. The
gether, require the cellular therapy products to NMDP standards and guidelines provided by
be shipped and transported between different cellular therapy accreditation programs recom-
geographic locations. Less frequently, cryopre- mend that HPC(A) products be infused within
served autologous HPC products need to be 48 hours of collection. Several studies have
transported to a different location from the origi- shown that maintaining a product temperature
between 2 and 8 C throughout the shipment re-
nal collection site. Standards for sale and effec-
tains optimal product integrity and potency at
tive shipping and transportation are defined by
the time of infusion.I05,I06,153 This temperature
accreditation organizations.50(PP42.44),150 Shipping
can maintain CD34+ cell viability more effec-
involves the HPC product leaving the control of
tively than shipment at room temperature, par-
trained personnel in the facility that collected,
ticularly for HPC(A)and shipping times of 24 to
received, stored, andlor distributed the product
72 hours. Shipment at cold temperature is par-
after collection. Transportingrefers to the trans-
ticularly important for products with higher
fer of a product between or within facilities
nucleated-cell concentrations.
when the product remains under the control of
Cryopreserved products are shipped in dry
a facility'strained personnel. shippers charged with LN2• These containers
The necessary conditions for shipping and
maintain temperatures below -150 C for up to
transport vary depending on the type of product, 2 weeks, and their temperatures are continuous-
its state (fresh or cryopreserved), and the dis- ly monitored with electronic data loggers.!"
tance involved.151 HPC products may be
The HPC product should not be exposed to
shipped on public roads andlor on aircraft. All gamma irradiation or x-ray devices; instead, it
procedures involved in the transportation or should be manually inspected, if necessary. Ap-
shipment should be compliant with applicable propriate records must accompany the product,
regulations and standards. Shipping and trans- and the chain of custody of the HPC product
porting HPC products is tightly regulated by the must be clearly documented as the product is
FDA, Department of Transportation, Interna- transferred from the transfer facility to the re-
tional Air Transport Association, International ceiving facility via the courier. Upon receipt of
Civil Aviation Organization, AABB,and FACT. the product, trained personnel at the receiving
The requirements for continuous temperature facility should promptly follow the instructions
monitoring and the procedures for packaging, for opening the container and inspecting the
labeling, and documentation are all designed .to HPC product. At that point, a decision is made
maintain the integrity of the HPC product while to accept, reject, or quarantine the HPC prod-
protecting the health and salety of personnel in- uct. 50(pp44·45)
volved in the transfer process.
During shipment, products must be placed in
secondary containers that are sealed to prevent
leakage and that have been validated to main-
tain the temperature range required for the in- For fresh HPC products, once the physician car-
tegrity of the product under the shipping condi- ing for the patient has ordered and approved the
tions.50(p43),152 Containers should be made of HPC product for infusion, the product should be
delivered to the patient-care area without delay. are infused slowly to start and with sufficient ob-
The cell product is usually handled by the clini- servation to detect symptoms. After ensuring
cal staff caring directly for the patient and in- that there is no immediate AE with the initial
fused by personnel who are trained in monitor- volume, the product can be infused more quick-
ing and management of infusion-related events. ly, based on patient tolerability.P' The number
The procedure for infusion is similar to that used of bags of cryopreserved product to be infused is
for most blood components.v-!" After patient determined by the number of cells collected and
and product identification procedures are per- the maximum dose of DMSO that can be given
formed in accordance with local institutional in one day (1 g/kg/ day or 1 mLlkg/ day of HPC
guidelines and policies, the product is usually in- product cryopreserved in final 10% DMSO con-
fused by N gravity drip directly through a CVC, centration).1l3,15~,156 Bags with the highest
and typically without a pump, although some CD34+ cell content are often infused first, fol-
centers may use pumps. HPC products should lowed by bags with a lower number of CD34+
never be leukocyte reduced or irradiated. Even cells. Some centers limit the total number of
though the practice is hot universal, some insti- TNCs to be infused per day due to AEs attribut-
tutions use a standard blood filter at the bedside. ed to a high number of granulocytes in the HPC
One group found no clear disadvantage to filtra- product.11S,116,IS7,IS8
tion as they demonstrated no difference in any Infusion-related AEs associated with HPC
markers of viability or potency for products after products may be similar to those that occur with
routine filtration.1SSBecause HPC(M) products transfusion of blood components. These include
are routinely filtered in the operating room, fil- allergic,hemolytic, and febrile reactions, as well
tration at the bedside (eg,using a standard blood as fluid overload. Certain AEs have been at-
filter that excludes> 170 microns) is often not tributed specificallyto DMSO, including nausea,
performed; however, including this practice is at vomiting, headache, changes in blood pressure
the discretion of institutional policy. The pore and pulse, and cough. The incidence of AEs fol-
size of the standard blood filter is many times lowing infusion of cryopreserved HPC products
larger than the typical white cell or red cell (5- ranges from 6% up to 70%, with variable de-
10 microns), and filtration should remove un- grees of severity.1l3,IS9The variability probably
wanted clotted cells or macroaggregates with- reflects differences in processing, premedica-
out removal of the intended transfused product. tion, patient monitoring, and the classificationof
If a standard blood filter is used, the laboratory AEs. Additional AEs include facial flushing, skin
should validate this process. Normal saline is rashes, nausea, vomiting, fever, chills, abdomi-
usually the only fluid administered concomitant- nal pain, hypotension, hypertension, bradycar-
ly with the HPC product, and it may be used to dia, and tachycardia.113,160In one study of 1191
flush the product bag and IVtubing after the bag aduit patients, hypoxia requiring oxygen was
empties to maximize the cell dose infused. It noted in 29% of infusions, chest tightness in
may also be added directly to the bag if the flow 16.7%, and shortness of breath in 8.3%.160Neu-
rate becomes too slow. rologic toxicities have also been reported, in-
At ._the time of infusion of cryopreserved cluding amnesia, encephalopathy, seizures, and
HPC(A) products, they are thawed in a 37 C stroke.115,161-163
Fortunately, severe infusion-
waterbath at the patient's bedside. Thawing of related AEsare uncommon.
the product is usually completed by trained cel- Many patients may receive premedication in
lular therapy technicians before immediately an attempt to prevent allergic, DMSO-related,
handing over to the clinical staff for infusion. and febrile nonhemolytic reactions, along with a
The preparation of any HPC product that recombination of hydration, antihistamines, anti-
quires washing and/or dilution before infusion, pyretics, and anti-inflammatoryagents.1S4,159 In-
including HPC(CB) products after thawing, is travenous hydration (eg, for 2-6 hours before
usually completed in. the processing laboratory and up to 6 hours after infusion) along with di-
before the final product is delivered to the pa- uretics may be needed with DMSO-containing
tient's bedside for infusion. Thawed products HPCs. Patients' vital signs and oxygen saturation
should be closely monitored during and up to federal levels to ensure the safety, purity, and po-
several hours after infusion of cryopreserved tency of cellular therapy products. The FDAand
HPC products. After infusion of HPC(CB),moni- the Centers for Medicare and Medicaid Services
toring for AEs is required for 24 hours. All moni- (CMS) are the main governing bodies that pro-
toring information should be captured on an ac- vide federal oversight. State health departments
companying infusion form. When completed, may also enforce local regulations for collection
this form is returned to the laboratory. centers and processing laboratories/manufactur-
The transplantation physicianand the medical ing facilities.Individuals and organizations or in-
director of the cell therapy laboratory should be stitutions involved in cellular therapy processing
notified immediately of any unexpected or mod- must be familiarwith the requirements of these
erate-to-severeinfusion-related AE. An investiga- agencies as regulated by force of law. The FDA
tion into whether the AEis HPC product related was granted authority to establish regulations
and whether it might represent transmission of for all HCT/Ps, including hematopoietic stem/
infectious disease from the product should be progenitor cells derived from peripheral blood
carried out and guided by the signs or symptoms and UCB, by Section 361 of the Public Health
of the reaction. Important data include laboratory Service (PHS)Act. FDArequirements are aimed
test results (eg, direct antiglobulin test, antibody at protecting the public health by preventing the
titer, Gram stain, or blood cultures), imaging (eg, introduction, transmission, and spread of infec-
chest x-ray), and echocardiogram results. Prod- tious disease. The general and permanent rules
uct and infusion-related AEs may be reduced by published in the Federal Register by the depart-
certain cell-processing techniques, such as red ments and agencies of the federal government
cell or plasma reduction, postthaw washing, or are codifiedin the CFR.The FDA's Center for Bi-
dilution. Some transplantation centers routinely ologicsEvaluationand Research (CBER)regulates
wash cryopreserved products after thawing and HCT/Ps under Title 21 CFR 1270 and 1271.165
before infusion in order to reduce the DMSO HPCs that are minimally manipulated and
content.l'v'" DMSO depletion has been shown collected for transplantation in an autologous
to reduce the frequency of AEs but could result fashion or transplanted into a first- or second-
in loss of CD34+ cells and introduction of bacte- degree relative are regulated solely under Sec-
rial contaminatton.v':'?' For these reasons, most tion 361 of the PHS Act and are subject to the
transplantation centers avoid postthaw washing jurisdiction of CBERunder 21 CFR 1271.166 If,
or other manipulations. in the manufacturing process, the relevant bio-
Clinical outcome data, including delayed or logic or functional characteristics of an HPC
failed engraftment and infusion-related AEs, product are altered (eg, genetically modified, ex-
should be reviewed regularly and discussed with panded ex vivo, or combined with a drug, de-
the institutional quality management group. The vice, or biologic), or the cells are intended for
medical director's review should include assess- transplantation into a non-first-or-second-degree
ment of the HPC product's quality indicators (eg, relative (unrelated donors), the HPC product is
dose, viability, and, if available, CFUs),any asso- subject to regulation under Section 351 and 361
ciated deviations, and the presence ofinfusion re- of the PHSAct and 21 CFR 1271 as a drug and/
actions. Particular attention is paid to any poten- or biologic product. These products require li-
tiallaboratory-related issues that could contribute censing or an exemption from licensing from
to less-than-optimaloutcomes. the FDAas part of an IND application. Issuance
of HPCs from unrelated donors facilitated
through the NMDP/Be The Match Registry
may be administered under the registry's IND or
an institutionally held IND protocol. Similarly,
minimally manipulated, unrelated UCB intend-
ed for hematopoietic or immunologic reconstitu-
The collection and provision of HPCs are regu- tion in patients with disorders affecting the he-
lated in the United States both at the state and matopoietic system must be FDA licensed or
used under an IND. Minimallymanipulated mar- laboratory inspections. Having sufficient person-
row that is not combined with another.regulated nel that are.highly trained and competent, ade-
article (with some. exceptions) andislntended quate facilltlesand equipment, and a solid quality-
for homologous use is not considered anH CTIF. management plan are essential for a successful
Manufacturers of HCT/Ps are required by program,
FDAregulations to use a tracking and labeling
system that enables each productto be tracked
from the donor to the recipient and from there- NCLJJSION
cipient back to the donor. The FDArequires in-
stitutions that manufacture HCT/Ps to register HSCT is an established lifesaving treatment, es-
with the agency and list theirHCT/Ps.167 The pecially for patients with hematologlc.mallgnan-
FDA also inspects laboratories that manufacture cies. Use ofHSCT has more recently expanded
or process FDA-regulated products, such as to nonmalignant disorders, such as thalassemia,
HCTIP processing laboratories, to verify that sickle cell disease, and autoimmune conditions.
they comply with relevant regulattons.!" The As HPC biology becomes better understood and
CMS regulates all laboratory testing (except re- the ability to engineer HPC grafts evolves, clini-
search) performed on humans in the United cal applicationswill continue to grow. New pro-
States through the Clinical Laboratory Improve- cessing techniques, technology, 'mobilization
ment, Amendments (CLIA).16,9Regulations re-
quire that laboratories be certified under CLIA
methods, and ex-vivo manipulations add this to
complex, developing field. To maintain a high-
as both a general requirement and a prerequisite qualityproduct that ensures safe engraftment and
for receiving Medicare and Medicaid reimburse-
optimal patient outcomes, each step of the pro-
ment. They provide minimal standards for facili-
cedure from donor qualificationthrough product
ties, equipment, and personnel."? CMS can re-
voke certification or impose fines on laboratories infusion must be performed appropriately by col-
that failto comply with its regulations. lection facilities, processing laboratories, and
In the United States, cellular therapy collec- transplantation centers. Transplantation logistics
tion centers and processing facilitiesmay also be and care are overseen by accreditation organiza-
accredited by organizations such as FACT,171 tions and regulatory agencies, which require up-
CAP/72 or AABB.173 Accreditation through these dates and modifications to their applicable rules,
organizations is voluntary; however, they allow regulations, and standards to maintain high-
institutions and facilities to be officiallyrecog- quality care.
nized as having operations of high quality that Additional focused descriptions on HPC pro-
meet federal and localstate regulations. curement and cellular processing, including
Maintaining standards and compliance with UCB banking and UCB use, maybe found in
regulations is verified through institutional and other AABBpublications.
~. >PJI~gellE~ic HPG products. ~ayeti1e.abilitY .to replace defectiye Hetrtatopoiesis.··atid· cellullJl"a~d•

h~lnoral. immunologic
•....> •.•.•••...• fUnctionsv.thereby.providinga gr~ft-vs-l1eoplasm•.effect..••·•..•.•••••••.•••.•••••• \.)••i ·•·•·•••···
·•i. <....
...6. Sc~eeningand. testing forrelevatl~infectious diseasesinano~el1~icHPG d?nors is mandat~d?Y'
tl1eFDA. The donor is decl(l1'edineligibleif risk for such diseaSei~discovered. The donor lTI.a y....•.•
7. ~rG
still donate if there is .anurgentrnedicaLneed •and j~stific~ti?ncriteria are met. < •••.••.•.••..•••
processing techniques that reduce product voluIIle, plaslTIa,red cells, and (postti1aw)
".'•••••.•••...••..••.••.••.•.•••.••..•••
cryoprotectant ..areuse~?a~e~o~.W'aft.specificatio~a~.d.ti1~.~~ciRi~nt'sclini.cal~eeds.Sp~?i~I~
. . .··ized..processing.or -.more-than-ininimal.manipuIationinvolvesspecial.ins~U.lTIel1tsandre~gents
fbrcen·.sele:tio?,de?letio?i·~Xpan.~ion,.0~.fUrtherJll?~ifi?ati6ri)?S. {.. i ••.••
.....••..•.. \ •..••.....••.••••..•....••..
, .···?;c';.<
:~8•.:.~!~Sfor a~t?logous.tr~splantatio~ ~e.cryopreserv~d
g
~tfi!
bM~o..Cryoprese~ationfti1~~~.·
••••...
·in.·.g
'. ...•..•.
infUsion,.and infUsion-rel.a
..
·.t
e d...•....
·
. •.
adverse-eventmonito.r
operating procedures that maximize patient safety and minimize i.n
...•.....
.......•.
·.. e..•.....
qtoxicity.
lii re pers.onn
...••... ....••
·.t.·.·.· eI.•.••
~d> •. s.tan
>......
.••••••• '.
a.·.·
...•..•..•
.•....
...•.......•........•.......•.••..
d....•....•.•.•.•..
ar
.............•...•.
•.·•..
.•••.
9~ ClGis an important component of any quality managem~nt?r0W'~~d.inc1udes testin~ ~~~
c~aracterization of the prod~ct to ensure. a •safe.and effe:tiy~ P~?d~ct for infusion. COIIllTI?~ .
.................
i Cl~testing inc1udesTNGco~t ~d viability, CD34+ C~llC?~t~d.viability,micrObi?lo~q ....
...••.••.•.•.•. <.·.. i i.
t~sting, ~dT-celicontentf?ra110geneic. products.·•..•.••••.•••. i:. < < .: ••••.·•••·•·.••.• ··.i•.\ <
•··••••••
to. Regulatory and.accreditation standards ••.outline·.·requirelTIe?tsf?r·.product·•procurementt·h~' • ••
•••..••dlirlg, and·.processing; labeling;\ storage,..transport~tio~,•••• and··s~ipping;. recipie~tout:omes;
"'.••" ,traSking, and reporting; d()c1.l.Illehtand record control;and'fadlltysafetyand operational' c6riY:C
• trols; . '" . . . ..' . . ..
REFERENCES
1. Saad A, Marques MB, Mineishi S. Donor and marker for diverse progenitors. Stem Cells
graft selection strategy. In: Abutalib S, Hari P, 2014;32: 1380-9.
eds. Clinicalmanual of blood and bone marrow 8. D'Souza A, Fretham C. Current uses and out-
transplantation. Hoboken, NJ: John Wiley & comes of hematopoietic cell transplantation
Sons Ltd, 2017:1-8. (HCT): CIBMTR summary Slides, 2018. Mil-
2. Panch SR,SzymanskiI, Savani BN, et al. Sourc- waukee, WI: Center for International Bloodand
es of hematopoietic stem and progenitor cells Marrow Transplant Research, 2018. [Available
and methods to optimize yields for clinical cell at https://www.cibmtr.orgiReferenceCenter/
therapy. Biol Blood Marrow Transplant 2017; SlidesReports/SummarySlides/Pages/index.
23:1241-9. aspx (accessedDecember 8,2019).]
3. Bianco P. Bone and the hematopoietic niche: A 9. Kardduss-UruetaA, Gale RP, Gutierrez-Aguirre
tale of two stem cells.Blood2011; 117:5281-8. CH, et al. Freezingthe graftis not necessary for
4. Yu VW, Scadden DT. Hematopoietic stem cell auto transplants for plasma cell myeloma and
and its bone marrow niche. Curr Top Dev Biol lymphomas. BoneMarrow Transplant 2018;53:
2016;118:21-44. 457-60.
5. Devine SM, Flomenberg N, Vesole DH, et a1. 10. Donor history questionnaire-HPC, Apheresis
Rapid mobilization of CD34+ cells following and HPC, Marrow. Version2.0. Bethesda, MD:
administration of the CXCR4 antagonist AABB,2019. [Availableat http://www.aabb.
AMD3100 to patients with multiple myeloma org/trn/ questionnaires/Pages/ dhqhpc.aspx (ac-
and non-Hodgkin's lymphoma. J Clin Oncol cessed December 12, 2019).]
2004;22: 1095-102. 11. Food and Drug Administration. Human cells,
6. Broxmeyer HE, Hangoc G, Cooper S, et al. tissues, and cellular and tissue-based products;
Growth characteristicsand expansion of human donor screening and testing, and related label-
umbilicalcord blood and estimation of its poten- ing. Fed Regist2005;70( 100):29949-52.
tial for transplantation in adults. Proc Nat! Acad 12. Food and Drug Administration.Guidance for in-
SciU SA 1992;89:4109-13. dustry: Donor screening recommendations to
7. Sidney LE, Branch MJ, Dunphy SE, et al. Con- reduce the risk of transmission of Zika virus by
cise review: Evidence for CD34 as a common human cells, tissues, and cellular and tissue-
based products. (May 2018) Silver Spring, MD: 22. Schetelig J, Bornhauser M, Schmid C, et aI.
CBEROffice of Communication, Outreach, and Matched unrelated or matched sibling donors
Development, 2018. [Available at https:11 result in comparable survival after allogeneic
www.fda.gov/downloads/BiologicsBloodVac stem-cell transplantation in elderly patients with
cines/GuidanceComplianceRegulatoryInforma acute myeloid leukemia: A report from the co-
tionl Guidances/Tissue/UCM488582. pdf.] operative German Transplant Study Group.
13. Lee SJ, Klein J, Haagenson M, et aI. High- J Clin OncoI2008;26:5183-91.
resolution donor-recipient HLAmatching con- 23. Gupta V, Tallman MS, He W, et aI. Comparable
tributes to the success of unrelated donor mar- survival after HLA-well-matched unrelated or
row transplantation. Blood2007; 110:4576-83. matched sibling donor transplantation for acute
14. LoiseauP, Busson M, Balere ML, et aI. HLAas- myeloid leukemia in first remission with unfa-
sociation with hematopoietic stem cell trans- vorable cytogenetics at diagnosis. Blood 2010;
plantation outcome: The number of mismatches 116: 1839-48.
at HLA-A,-B, -C, -DRBl, or -Dilll is strongly 24. Walter RB, Pagel JM, Gooley TA, et aI. Compari-
associated with overall survival. BiolBloodMar- son of matched unrelated and matched related
row Transplant 2007;13:965-74. donor myeloablative hematopoietic cell trans-
15. Spellman S, Setterholm M, Maiers M, et aI. Ad- plantation for adults with acute myeloid leuke-
vances in the selection of HLA-compatible do- mia in first remission. Leukemia 2010;24:1276-
nors: Refinements in HLAtyping and matching 82.
over the first 20 years of the National Marrow 25. Peters C, Schrappe M, von Stackelberg A, et aI.
Donor Program Registry. BioI Blood Marrow Stem-cell transplantation in children with acute
Transplant 2008;14(9 Suppl):37-44. lymphoblastic leukemia: A prospective interna-
16. Spellman SR, Eapen M, Logan BR, et al. A per- tional multicenter trial comparing sibling donors
spective on the selection of unrelated donors with matched unrelated donors-The ALL-SCT-
and cord blood units for transplantation. Blood BFM-2003 trial. J Clin OncoI2015;33: 1265-74.
26. Marsh SGE, Albert ED, Bodmer WF, et aI. No-
2012;120:256-65.
17. Shaw BS,Spellman SR.HLAtyping and implica- menclature for factors of the HLA system, 2004.
Tissue Antigens 2005;65:301-69.
tions. In: Abutallb S, Hari P, eds. Clinicalmanu-
27. Robinson J, Halliwell JA, McWilliam H, et aI.
al of blood and bone marrow transplantation.
The IMGT/HLA database. Nucleic Acids Res
Hoboken, NJ: John Wiley &Sons Ltd, 2017:9-
2013;41: 1013-17.
18.
28. Gragert L, Eapen M, Williams E, et aI. HLA
18. National Marrow Donor Program/Be The
match likelihoods for hematopoietic stem-cell
Match. HLAtyping and matching. Minneapolis,
grafts in the U.S. registry. N Eng! J Med 2014;
MN: NMDP, 2018. [AVailableat https:llbethe 371 :339-48.
matchclinical. org/Transplant -Therapy-and- 29. Barker IN, KrepskiTP, DeFor TE, et.al, Search-
Donor-Matching/HLA-Typing-and-Matchingl ing for unrelated donor hematopoietic stem
(accessed December 8, 2019).] cells: Availability and speed of umbilical cord
19. Zhang MJ, Davies SM, Camitta BM, et aI. Com- blood versus bone marrow. BioIBlood Marrow
parison of outcomes after HLA-matched sibling Transplant 2002;8:257-60.
and unrelated donor transplantation for children 30. Weisdorf D, Eapen M~RuggeriA, et aI. Alterna-
with high-risk acute lymphoblastic leukemia. tive donor transplantation for older patients
BioI Blood Marrow Transplant 2012;18:1204- with acute myeloid leukemia in first complete
10. remission: A Center for International Blood and
20. SaberW, Opie S, RizzoJD, et aI. Outcomes after Marrow Transplant Research-Eurocord analysis.
matched unrelated donor versus identical sib- BioiBloodMarrow Transplant 2014 ;20:816-22.
ling hematopoietic cell transplantation in adults 31. Marks DI, Woo KA, Zhong X, et aI. Unrelated
with acute myelogenous leukemia. Blood umbilical cord blood transplant for adult acute
2012; 119:3908-16. lymphoblasticleukemia in first and second com-
21. Saber W, Cutler CS, Nakamura R, et aI. Impact plete remission: A comparison with allografts
of donor source on hematopoietic cell transplan- from adult unrelated donors. Haematologica
tation outcomes for patients with myelodysplas- 2014;99:322-8.
tic syndromes (MDS). Blood 2013;122:1974- 32. Eapen M, Rubinstein P, Zhang MJ, et aI. Out-
82. comes of transplantation of unrelated donor um-
bilical cord blood and bone marrow in children 42. Ciurea SO, BayraktarUD. "No donor"? Consid-
with acute leukaemia: A comparison study. Lan- er a haploidentical transplant. Blood Rev 2015;
cet 2007;369: 1947-54. 29:63-70.
33. LaughlinMJ, Eapen M, Rubinstein P, et al. Out- 43. SpellmanS, BrayR, Rosen-BronsonS, et al. The
comes after transplantation of cord blood or detection of donor-directed, HLA-specificalloan-
bone marrow from unrelated donors in adults tlbodies In recipients of unrelated hematopoietic
with leukemia. N Engl J Med 2004;351 :2265- cell transplantation is predictive of graft failure.
75. Blood2010;115:2704-8.
34. Milano F, Gooley T, Wood B, et al. Cord-blood 44. Brand A, DoxladlsIN, Roelen DL. On the role of
transplantation in patients with minimal residu- HLA antibodies in hematopoietic stem cell
al disease. N Engl] Med 2016;375:944-53. transplantation. TissueAntigens 2013;81 :1-11.
35. Luo Y, Xiao H, Lai X, et al. Tcell-replete 45. KollmanC, SpellmanSR,Zhang M, et al. The ef-
fect of donor characteristics on survivalafter un-
haploidentical HSCT with low-dose antH-
related donor transplantation for hematologic
lymphocyte globulin compared with matched
malignancy.Blood2016;127:260-7.
siblingHSCTand unrelated HSCT.Blood 2014; 46. McCulloughJ, McKenna D, KadidloD, et al. Is-
124: 2735-43. sues in the quality of umbilical cord blood stem
36. Wang Y, Liu OF, Xu LP, et al. Haploidenticalvs cells for transplantation. Transfusion 2005;45:
identical-sibling transplant for AML in remis- 832-41.
sion: A multicenter, prospective study. Blood 47. Page KM,Zhang L, MedizabalA, et al. The cord
2015;125:3956-62. blood apgar: A novel scoring system to optimize
37. Solomon SR, Sizemore CA, Sanacore M, et al. the selection of banked cord blood grafts for
Total body irradiation-based myeloablative hap- transplantation. Transfusion2012;52:272-83.
loidentical stem cell transplantation is a safe and 48. Stem Cell Trialists' CollaborativeGroup. Alloge-
effective alternative to unrelated donor trans- neic peripheral blood stem-cell compared with
plantation in patients without matched sibling bone marrow transplantation for hematologic
donors. BioiBloodMarrow Transplant 2015;21: malignancies: An individual patient data meta-
1299-307. analysis of nine randomized trials. J Clin Oncol
38. Barker IN, Scaradavou A, Stevens GE. Com- 2005;23:5074-87.
bined effectof total nucleated cell dose and HLA 49. Anasetti C, Logan BR, Lee SJ, et al. Peripheral-
match on transplantation outcome in 1061 cord blood stem cellsversus bone marrow from unre-
blood recipientswith hematologic malignancies. lated donors. N Eng!J Med 2012;367: 1487-96.
Blood2010; 115:1843-9. 50. HaspelRL,ed. Standards for cellular therapy ser-
39. Eapen M, Klein JP, Sanz GF, et al. Effect of vices. 9th ed. Bethesda, MD: AABB,2019.
donor-recipient HLAmatching at HLAA, B, C, 51. Lown RN, PhilippeJ, Navarro W, et al. Unrelat-
and DRB1 on outcomes after umbilical-cord ed adult stem cell donor medical suitability:Rec-
blood transplantation for leukaemia and myelo- ommendations from the World Marrow Donor
dysplasticsyndrome: A retrospective analysisby Association Clinical Working Group Commit-
tee. BoneMarrow Transplant 2014;49(7):880-6.
Eurocord-European Group for Blood and Mar-
52. Worel N, Buser A, Greinix HT, et al. Suitability
row Transplantation; Netcord; Center for Inter-
criteria for adult related donors: A consensus
national Blood and Marrow Transplant Re- statement from the Worldwide Network for
search. Lancet Onco12011; 13:1214-21. Blood and Marrow Transplantation Standing
40. Van Rood H, Stevens CE, SmitsJ, et al. Reexpo- Committee on Donor Issues. BioiBloodMarrow
sure of cord blood to noninherited maternal Transplant 2015;21 :2052-60.
HLA antigens Improves transplant outcome in 53. Bitan M, van Walraven SM, Worel N, et al. De-
hematological malignancies. Proc Nat! Acad Sci termination of eligibilityin related pediatric he-
USA 2009; 106:19952-7. matopoietic cell donors: Ethical and clinicalcon-
41. BasheyA, Zhang MJ, McCurdy SR, et al. mobi- siderations. Recommendations from a Working
lized peripheral blood stem cells versus unstim- Group of the Worldwide Network for Bloodand
ulated bone marrow as a graft source for Tcell- Marrow Transplantation Association.BioiBlood
replete haploidentical donor transplantation Marrow Transplant 2016;22:96-103.
using post-transplant cyclophosphamide. J Clin 54. Babic AM, Regan DM. Umbllical cord blood
OncoI2017;35:3002-9. banking. In: Fung MK, Grossman BJ, Hillyer
C HAP T E R 2 6 TheCollection and Processingof HPCs 761
CD, Westhoff CM, eds. Technical manual. 18th 64. Narayanasami U, Kanteti R, Morelli], et al. Ran-
ed. Bethesda, MD: AABB,2014:729-52. domized trial of G-CSF versus chemotherapy
55. Hackett N], De Oliveira GS, ]aln UK, Kim ]Y. and G-CSFmobilization of hematopoietic pro-
ASAclass is a reliable independent predictor of genitor cells for rescue In autologous transplan-
medical complications and mortality following tation. Blood 2001;98:2059-64.
surgery. Int] Surg 2015;18:184-90. 65. Kroger N, Zeller W, Fehse N, et al. Mobilizing
56. Fleisher LA, Fleischmann KE, Auerbach AD, et peripheral blood stem cells with high-dose
al. 2014 ACC/AHA guideline on perioperative G-CSFalone is as effective as Dexa-BEAMplus
cardiovascular evaluation and management of G-CSF in lymphoma patients. Br ] Haematol
patients undergoing noncardiac surgery: A re- 1998;102:1101-6.
port of the American College of Cardiology/ 66. Yuan S, Wang S. How do we mobilize and col-
American Heart AssociationTask Force on prac- lect autologous peripheral blood stem cells?
tice guidelines. ] Am Coli CardioI2014;64:e77- Transfusion 2017;57: 13-23.
137. 67. Sung AD, Grima DT, Bernard LM, et al. Out-
57. Spitzer TR. Bone marrow collection. In: Loper comes and costs of autologous stem cell mobili-
K, Areman EM, eds. Cellular therapy: Princi- zation with chemotherapy plus filgrastim vs fil-
ples, methods, and regulations. 2nd ed. Bethes- grastim alone. Bone Marrow Transplant 2013;
da, MD: AABB,2016:294,305. 48:1444-9.
58. Miller ]P, Perry EH, Price TH, et al. Recovery 68. Gertz MA. Review: Current status of stem cell
and safety profile of marrow and PBSC·donors: mobilization. Br] Haematol201 0;150:647-62.
Experience of the National Marrow Donor Pro- 69. DiPersio ]F, Stadtmauer EA, Nademanee A, et
gram. BiolBlood Marrow Transplant 2008;14: al, for the 3102 Investigators. Plerixafor and fil-
29-36. grastim versus placebo and filgrastim to mobi-
59. Pulsipher MA, Chitphakdithal P, Logan BR, et lize hematopoietic stem cells for autologous
al. Acute toxicities of unrelated bone marrow stem cell transplantation in patients with multi-
ple myeloma. Blood2009; 113:5720-6.
versus peripheral blood stem cell donation: Re-
70. DiPersio]F, Micallef IN, StiffP], et al. Phase III
sults of a prospective trial from the National
prospective randomized double-blind placebo-
Marrow Donor Program. Blood 2013;121(1):
controlled trial of plerixafor plus granulocyte
197-206.
colony-stimulating factor compared with place-
60. Switzer GE, Bruce ]G, Harrington D, et al.
bo plus granulocyte colony-stimulatingfactor for
Health-related quality of life of bone marrow
autologous stem-cellmobilization and transplan-
versus peripheral blood stem cell donors: A pre-
tation for patients with non-Hodgkin's lympho-
specified subgroup analysis from a Phase III rna.] Clln OncoI2009;27:4767-73.
RCT-BMTCTNprotocol 0201.. Bioi Blood Mar- 71. Smith YR, Popat U, Ciurea S, et al Just-in-time
row Transplant 2014;20(1): 118-27. rescue plerixafor in combination with chemo-
61. Burns LJ, Logan BR, Chitphakdithal P, et al. Re- therapy and granulocyte-colony stimulating fac-
covery of unrelated donors of peripheral blood tor for peripheral blood progenitor cell mobiliza-
stem cells versus recovery of unrelated donors tion. AmJ HematoI2013;88:754-7.
of bone marrow: A prespecified analysis from 72. Costa LJ, Abbas], Hogan KR,et al. Growth fac-
the Phase III Blood and Marrow Transplant tor plus preemptive ('just-in-time') plerixafor
Clinical Trials Network Protocol 0201. Biol successfully mobilizes hematopoietic stem cells
BloodMarrow Transplant 2016;22(6): 1108-16. in multiple myeloma patients despite prior
62. Pulsipher MA, Chitphakdithal P, Miller ]P, et al. lenalidomide exposure. Bone Marrow Trans-
Adverse .events among 2408 unrelated donors plant 2012;47:1403-8.
of peripheral blood stem cells: Results of a pro- 73. Becker P, Schwebig A, Brauninger S, et al.
spective trial from the National Marrow Donor Healthy donor hematopoietic stem cell mobili-
Program. Blood2009; 113:3604-11. zation with biosimilar granulocyte-colony-
63. Spitzer TR, Sugrue MW, Gonzalez C, et al. stimulating factor: safety, efficacy,and graft per-
Transfusion practices for bone marrow harvests: formance. Transfusion 2016;56:3055-64.
A survey analysis from the AABBBone Marrow 74. Schroeder MA, RettigMP, Lopez S, et al. Mobi-
Quality Improvement Initiative Working Group lization of allogeneic peripheral blood stem cell
(letter). Bone Marrow Transplant 2017;52: donors with intravenous plerixafor mobilizes a
1199-200. unique graft. Blood2017; 129:2680-92.
75. Pulsipher MA, LevineJE, HayashiRJ,et al. Safe- after autologous stem cell transplantation in
ty and efficacyof allogeneic PBSCcollection in non-Hodgkin Lymphoma: A prospective study.
normal pediatric donors: The pediatric blood Biol BloodMarrow Transplant 2008: 14;807-16.
and marrow transplant consortium experience 86. Porrata LF, Litzow MR, Inwards DJ, et al, In-
(PBMTC)1996-2003. Bone Marrow Transplant fused peripheral blood autograft absolute lym-
2005;35:361-7. phocyte count correlates with day 15 absolute
76. Weaver CH, Hazelton B, Birch R, et al, An anal- lymphocyte count and clinicaloutcome after au-
ysis of engraftment kinetics as a function of the tologous peripheral hematopoietic stem cell
CD34 content of peripheral blood progenitor transplantation in non-Hodgkin's lymphoma.
cell collectionsin 692 patients after the adminis- BoneMarrow Transplant 2004;33:291-8.
tration of myeloablative chemotherapy. Blood 87. Porrata LF, BurgstalerEA, Winters JL, et al, Im-
1995;86:3961-9. munologic autograft engineering and survivalin
77. ReiffersJ, Faberes C, BoironJM, et al, PerIpher- non-Hodgkin lymphoma. Bioi Bood Marrow
al blood progenitor cell transplantation In 118 Transplant 2016;22: 1017-23.
patients with hematological malignancies:Anal- 88. Hiwase DK, Hiwase S, Bailey M, et al. Higher
ysis of factors affecting the rate of engraftment. infused lymphocyte dose predicts higher lym-
J Hematother 1994;3: 185-91. phocyte recovery,which in turn, predicts superi-
78. Pulsipher MA, Chitphakdithai P, Logan BR, et or overall survivalfollowingautologous hemato-
al, Donor, recipient, and transplant characteris- poietic stem cell transplantation for multiple
tics as risk factors after unrelated donor PBSC myeloma. Bioi Blood Marrow Transplant 2008;
transplantation: Beneficial effects of higher 14:116-24.
CD34 + cell dose. Blood2009; 114:2606-16. 89. Page KM, KurtzbergJ. Current cord blood bank-
79. Baron F, Maris MB, Storer BE, et al, High doses ing concepts and practices. In: Horwitz M,
of transplanted CD34 + cells are associatedwith Chao N, eds. Cord blood transplantations. In:
rapid T-cell engraftment and lessened risk of Abutalib SA,ArmitageJO, series eds. Advances
graft rejection, but not more graft-versus-host and controversies in hematopoietic transplanta-
disease after nonmyeloablative conditioning and tion and cell therapy. Cham, Switzerland:
unrelated hematopoietic cell transplantation. Springer,2017:13-34.
Leukemia 2005; 19:822-8. 90. Armson BA, Allan DS, Casper RF. Umbilical
80. Keating GM. Plerixafor. Drugs 2011;71:1623- cord blood: Counselling, collection, and bank-
47. ing. J Obstet GynaecolCan 2015;37:832-44.
81. Abrahamsen JF, Stamnesfet S, Liseth K, et al, 91. Bertolini F, Lazzari L, Laurt E, et al. Compara-
Large-volumeleukapheresis yields more viable tive study of different procedures for the collec-
CD34+ cells and colony-forming units than tion and banking of umbilical cord blood. J He-
normal-volume leukapheresis, especially in pa- matother 1995;4:29-36.
tients who mobilize low numbers of CD34+ 92. Jones J, Stevens CE, Rubinstein P, et al. Obstet-
cells.Transfusion2005;45:248-53. ric predictors of placental/umbilical cord blood
82. Lee G, Arepally GM. Anticoagulation tech- volume for transplantation. Am J Obstet Gyne-
niques in apheresis: From heparin to citrate and col 2003; 188:503-9.
beyond. J ClinApher 2012;27:117-25. 93. Wagner JE, Barker IN, DeFor TE, et al, Trans-
83. Lefrere F, lohar S, Beaudier S, et al. Evaluation plantation of unrelated donor umbilical cord
of an algorithmbased on peripheral blood hema- blood in 102 patients with malignant and non-
topoletic progenitor cell and CD34+ cell con- malignant diseases: Influence of CD34 cell dose
centrations to optimize peripheral blood progen- and HLAdisparity on treatment-related mortali-
itor cell collection by apheresis. Transfusion ty and survival.Blood2002; 100:1611-18.
2007;47:1851-7. 94. Gluckman E, Rocha V, Arcese W, et al, Factors
84. Leberfinger DL, Badman KL, RoigJM, Loos T. associated with outcomes of unrelated cord
Improved planning of leukapheresis endpoint blood transplant: Guidelines for donor choice.
with customized prediction algorithm:Minimiz- Exp HematoI2004;32:397-407.
ing collection days, volume of blood processed, 95. Barker IN, Weisdorf DJ, DeFor TE, et al. Trans-
procedure time, and citrate toxicity, Transfusion plantation of 2 partially HLA-matchedumbilical
2017;57:685-93. cord blood units to enhance engraftment in
85. Porrata LF, Inwards DJ, Ansell SM, et al. Early adults with hematologic malignancy. Blood
lymphocyte recovery predicts superior survival 2005; 105: 1343-7.
C HAP T E R 2 6 TheCollection and Processingof HPCs 763
96. Scaradavou A, Brunstein CG, Eapen M, et al. 107. Rubinstein P, Dobrila L, RosenfieldR, et al. Pro-
Double unit graftssuccessfullyextend the appli- cessing and cryopreservation of placental!
cation of umbilical cord blood transplantation in umbilicalcord blood for unrelated bone marrow
adults with acute leukemia. Blood 2013; 121 : reconstitution. Proc Nat! Acad Sci USA 1995;
752-8. 92:10119-22.
97. Wagner JE Ir, Eapen M, Carter S, et al, for the 108. Laroche V, McKenna D, Moroff G, et al. Cell
Blood and Marrow Transplant Clinical Trials loss and recovery in umbilical cord blood pro-
Network. One-unit versus two-unit cord-blood cessing: A comparison of post-thaw and post-
transplantation for hematologic cancers. N Engl wash samples. Transfusion2005;45: 1909-16.
J Med 2014;371:1685-94. 109. Barker IN, Abboud M, Rice RD, et al. A "no-
98. Michel G, Galambrun C, SirventA, et al. Single wash" albumin-dextran dilution strategy for
versus double-unit cord blood transplantation cord blood unit thaw: High rate of engraftment
for children and young adults with acute leuke- and a low incidence of serious infusion reac-
mia or myelodysplastic syndrome. Blood 2016; tions. Biol Blood Marrow Transplant 2009; 15:
127:3450-7. 1596-602.
99. PolitikosI, BarkerIN. Cell dose and immunoge- 110. Regan DM, Wofford JD, Wall DA. Comparison
netic considerations in cord blood transplanta- of cord blood thawing methods on cell recovery,
tion. In: Horwitz M, Chao N, eds. Cord blood potency, and infusion. Transfusion 2010;50:
transplantations. In: Abutalib SA, Armitage JO, 2670-5.
series eds. Advances and controversies in hema- 111. Pasha R, Elmoazzen H, Pineault N. Develop-
topoietic transplantation and cell therapy. ment and testing of a stepwise thaw and dilute
Cham, Switzerland: Springer,2017:47-69. protocol for cryopreserved umbilical cord blood
100. McKenna DH. Basic cellular therapy manufac- units. Transfusion2017;57: 1744-54.
turing procedures. In: Loper K, Areman EM, 112. Haspel RL.Thawing and infusing cellular thera-
eds. Cellular therapy: Principles, methods, and py products. In: Loper K,Areman EM, eds. Cel-
regulations. 2nd ed. Bethesda, MD: AABB, lular therapy: Principles, methods, and regula-
2016:361-7. tions. 2nd ed. Bethesda, MD: MBB, 2016:
101. StaleyEM, Schwartz J, Pham HP. An update on 459-67.
ABOincompatiblehematopoietic progenitor cell 113. Shu Z, Heimfeld S, Gao D. Hematopoietic SCT
transplantation. Transfus Apher Sci 2016;54: with cryopreserved grafts:Adverse reactions af-
337-44. ter transplantation and cryoprotectant removal
102. Coelho PH, Kadidlo D, Chapman J. Umbilical before infusion. Bone Marrow Transplant 2014;
cord blood processing. In: Loper K,Areman EM, 49:469-76.
eds. Cellular therapy: Principles, methods, and 114. Sanchez-Salinas A, Cabafias-PerianesV, Blan-
regulations. 2nd ed. Bethesda, MD: AABB, quer M, et al. An automatic wash method for di-
2016:368-75. methyl sulfoxideremoval in autologous hemato-
103. SorgN, Poppe C, Bunos M, et al. Red blood cell poietic stem cell transplantation decreases the
depletion from bone marrow and peripheral adverse effects related to infusion. Transplanta-
blood buffy coat: A comparison of two new and tion 2012;52:2382-6.
three established technologies. Transfusion 115. Ca1melsB, Lemarie C, Esterni B, et al. Occur-
2015;55:1275-82. rence and severity of adverse events after autol-
104. Kim-Wanner SZ, Bug G, Steinmann J, et al. ogous hematopoietic progenitor cell infusion are
Erythrocyte depletion from bone marrow: Per- related to the amount of granulocytes in the
formance evaluation after 50 clinical-scale de- apheresis product. Transfusion 2007;47: 1268-
pletions with Spectra Optia BMC.J Transl Med 75.
2017;15:174. 116. Khera N, Iinneman J, Storer BE, et al. Limiting
105. Antonenas V, Garvin F, Webb M, et al. Fresh the daily total nucleated cell dose of cryopre-
PBSCharvests, but not BM, show temperature- served peripheral blood stem cell products for
related loss of CD34 viabilityduring storage and autologous transplantation improves infusion-
transport. Cytotherapy 2006;8: 158-65. related safety with no adverse impact on hema-
106. Kao GS, Kim HT, Daley H, et al. Validation of topoietic engraftment. Biol Blood Marrow
short-term handling and storage conditions for Transplant 2012; 18:220-8.
marrow and peripheral blood stem cell prod- 117. Cameron G, Filer K, Hall A, Hogge D. Evalua-
ucts. Transfusion2011;51(1):137-47. tion of post-thaw blood progenitor product in-
tegrity and viability over time at room tempera- Phase I trial. Lancet HaematoI2016;3(7):e330-
ture and 4°C. Cytotherapy 2013;15:S27. 9.
118. Khuu HM, Cowley H, David-OcampoV, et al. 129. VeeraputhiranM, TheusJW, PesekG, etal. Via-
Catastrophic failures of freezing bags for cellular bilityand engraftment of hematopoietic progeni-
therapy products: Description, cause, and con- tor cells after long-term cryopreservation: Effect
sequences. Cytotherapy 2002;4:539-49. of diagnosis and percentage dimethyl sulfoxide
119. Tran CA, Torres-Coronado M, Gardner A, et al. concentration. Cytotherapy 2010; 12:764-6.
Optimized processing of growth factor mobi- 130. Mitchell R, Wagner JE, Brunstein CG, et al. Im-
lized peripheral blood CD34+ products by pact of long-term cryopreservation on single um-
counterflow centrifugal elutriation. Stem Cells bilical cord blood transplantation outcomes. BioI
Transl Med 2012; 1:422-9. BloodMarrow Transplant 2015;21 :50-4.
120. Edwards J. Cell separation by counterflow cen- 131. Lecchi L, Giovanelli S, GagliardiB, et al. An up-
trifugal elutration. In: Loper K, Areman EM, date on methods for cryopreservation and thaw-
eds. Cellular therapy: Principles, methods, and ing of hemopoietic stem cells. Transfus Apher
regulations. 2nd ed. Bethesda, MD: AABB, Sci 2016;54:324-36.
2016:381-3. 132. Smagur A, Mitrus I, Ciomber A, et al. Compari-
121. Keever-Taylor CA, Devine SM, Soitter RJ, et al. son of the cryoprotective solutions based on hu-
Characteristics of CliniMACS®System CD34- man albumin vs. autologous plasma: Its effect
enriched T cell-depleted grafts in a multi-center on cell recovery, clonogenic potential of periph-
trial for acute myeloid leukemia-Bloodand Mar- eral blood hematopoietic progenitor cells and
row Transplant Clinical Trials Network (BMT engraftment after autologous transplantation.
CTN) Protocol 0303. BioIBloodMarrow Trans- Vox Sang 2015; 108:417-24.
plant 2012;8:690-7. 133. Chen G, Yue A, Ruan Z, et al. Comparison of
122. Kadidlo D. Hematopoietic progenitor cell graft the effects of different cryoprotectants on stem
modification: Cell enrichment or depletion. In: cells from umbilical cord blood. Stem Cells Int
Loper K, Areman EM, eds. Cellular therapy: 2016;2016: 1396783.
Principles, methods, and regulations. 2nd ed. 134. Kozlowska-SkrzypczakM, KubiakA, Bembnista
Bethesda, MD: AABB,2016:376-80. E, et al. Analysis of the effect of cryoprotectant
123. Geyer MB, Ricci AM, Jacobson JS, et al. T cell medium composition to viability of autologous
depletion utilizing CD34(+) stem cell selection hematopoietic cells collected by leukapheresis.
and CD3(+) addback from unrelated adult do- Transplant Proc 2014;46:2535-8.
nors in paediatric allogeneic stem cell transplan- 135. Rowley SD, Feng Z, Chen L, et al. A random-
tation recipients. Br J HaematoI2012;157:205- ized phase III clinical trial of autologous blood
19. stem cell transplantation comparing cryopreser-
124. Aversa F. T-cell depletion: From positive selec- vation using dimethylsulfoxide vs dimethylsulf-
tion to negative depletion in adult patients. oxide with hydroxyethylstarch. Bone Marrow
Bone Marrow Transplant 2015;50(SuppI2): Transplant 2003;31: 1043-51.
SII-13. 136. RodriguesJP, Paraguassa-Braga FH, Carvalho L,
125. Bleakley M, Heimfeld S, Loeb KR, et al. Out- et al. Evaluation of trehalose and sucrose as
comes of acute leukemia patients transplanted cryoprotectants for hematopoietic stem cells of
with naive T cell-depleted stem cell grafts.J Clin umbilical cord blood. Cryobiology 2008;56:
Invest 2015;125:2677-89. 144-51.
126. Horwitz ME. Ex vivo expansion or manipula- 137. Smagur A, Mitrus I, Giebel S, et al. Impact of
tion of stem cells to improve outcome of umbili- different dimethyl sulphoxide concentrations on
cal cord blood transplantation. Curr Hematol cell recovery, viability and clonogenic potential
MaligRep 2016;11:12-18. of cryopreserved peripheral blood hematopoietic
127. Horwitz ME, FrassoniF. Improvingthe outcome stem and progenitor cells. Vox Sang 2013;104:
of umbilical cord blood transplantation through 240-7.
ex vivo expansion or graft manipulation. Cyto- 138. Windrum P, Morris TCM, Drake MB, et al, for
therapy 2015; 17:730-8. the EBMT Chronic Leukaemia Working Party
128. Delaney C, Milano F, Cicconi L, et al. Infusion Complications Subcommittee. Variation in di-
of a non-HLA-matched ex-vivo expanded cord methyl sulfoxide use in stem cell transplanta-
blood progenitor cell product after intensive tion: A survey of EBMT centres. Bone Marrow
acute myeloid leukaemia chemotherapy: A Transplant 2005;36:601-3.
139. Bakken AM, Bruserud 0, Abrahamsen JF. No 149. Pamphilon D, Selogie E, McKenna D, et al. Cur-
differences in colony formation of peripheral rent practices and prospects for standardization
blood stem cells frozen with 5% or 10%dimeth- of the hematopoietic colony-forming unit assay:
yl sulfoxide. J Hematother Stem Cell Res 2003; A report by the cellular therapy team of the Bio-
12:351-8. medical Excellence for Safer transfusion (BEST)
140. StiffPJ, Koester AR, Weidner MK, et al. Autolo- collaborative. Cytotherapy 2013;15:255-62.
gous bone marrow transplantation using unfrac- 150. Slaper-Cortenbach 1. ISBT 128 coding and label-
tionated cells cryopreserved in dimethylsulfox- ing for cellular therapy products. Cell Tissue
ide and hydroxyethyl starch without controlled- Bank 2010;1 1:375-8.
rate freezing. Blood 1987;70:974-8. 151. Regan D, Yost A. Transportation and shipping of
141. Weinberg RS. Cryopreservation techniques and cellular therapy products. In: Loper K, Areman
freezing solutions. In: Schwartz J, Shaz BH, eds. EM, eds. Cellular therapy: Principles, methods,
Best practices in processing and storage for he- and regulations. 2nd ed. Bethesda, MD: AABB,
matopoietic cell transplantation. In: Abutalib 2016:483-93.
SA,Armitage JO, series eds. Advances and con- 152. Jansen J, Nolan P, Reeves M, et al. Transporta-
troversies in hematopoietic transplantation and tion of peripheral blood progenitor cell prod-
cell therapy. Cham, Switzerland: Springer, ucts: Effects of time, temperature and cell con-
2018:63-72. centration. Cytotherapy 2009; 11: 79-85.
142. Broxmeyer HE, Lee M-R, Hangoc G, et al. He- 153. AABB,America's Blood Centers, American Red
matopoietic stem/progenitor cells, generation of Cross, American Society for Apheresis, Ameri-
induced pluripotent stem cells and isolation of can Society for Blood and Marrow Transplanta-
endothelial progenitors from 21- to 23.5-year tion, College of American Pathologists, Cord
cryopreserved cord blood. Blood 2011;117: Blood Association, Foundation for the Accredi-
tation of Cellular Therapy, ICCBBA,Internation-
4774-7.
143. Winter JM, Jacobson P, Bullough B, et al. Long- al Society for Cellular Therapy, Joint Accredita-
tion Committee of ISCT and EBMT, National
term effects of cryopreservation on clinically
Marrow Donor Program, World Marrow Donor
prepared hematopoietic progenitor cell prod-
Association. Circular of information for the use
ucts. Cytotherapy 2014;16:965-75.
of cellular therapy products. Bethesda, MD:
144. Sutherland DR, Keeney M. Enumeration of
AABB,2018. [Available at http://www.aabb.
CD34+ cells by flow cytometry. In: Loper K,
org/ aabbcct/ coi/Documents/ CT-Circular-of-
Areman EM, eds. Cellular therapy: Principles,
Information.pdf (accessed December 9, 2019).1
methods, and regulations. 2nd ed. Bethesda, 154. Sauer-HeilbornA, Kadidlo D, McCulloughJ. Pa-
MD: AABB,2016:558-69. tient care during infusion of hematopoietic pro-
145. Spitzer G, Verma DS, Fisher R, et al. The my- genitor cells. Transfusion 2004;44:907-16.
eloid progenitor cell: Its value in predicting he- 155. Paulson K, Gilpin SG, Shplruk TA, et al. Routine
matopoietic recovery after autologous bone mar- filtration of hematopoietic stem cell products:
row transplantation. Blood 1980;55:317-23. The time has arrived. Transfusion 2015;55:
146. Douay L, Gorin NC, Mary JY, et al. Recovery of 1980-4.
CFU-GM from cryopreserved marrow and in 156. junior AM, Arrais CA, Saboya R, et al. Neuro-
vivo evaluation after autologous bone marrow toxicity associated with dimethylsulfoxide-
transplantation are predictive of engraftment. preserved hematopoietic progenitor cell infu-
Exp HematoI1986;14:358-65. sion. Bone Marrow Transplant 2008;41 :95-6.
147. Schwartzberg L, Birch R, Blanco R, et al. Rapid 157. Cordoba R, Arrieta R, Kerguelen A, et al. The
and sustained hematopoietic reconstitution by occurrence of adverse events during the infu-
peripheral blood stem cell infusion alone follow- sion of autologous peripheral blood stem cells is
ing high-dose chemotherapy. Bone Marrow related to the number of granulocytes in the leu-
Transplant 1993;11:360-74. kapheresis product. Bone Marrow Transplant
148. Migliaccio AR, Adamson JW, Stevens CE, et al. 2007 ;40:1063-7.
Cell dose and speed of engraftment in placen- 158. Rowley SD, Feng Z, Yadock D, et al. Post-thaw
tal/umbilical cord blood transplantation: Graft removal of DMSO does not completely abrogate
progenitor cell content is a better predictor than infusional toxicity or the need for pre-infusion
nucleated cell quantity. Blood 2000;96:2717- histamine blockade. Cytotherapy 1999; 1:439-
22. 46.
159. Milone G, Mercurio S, Strano A, et al. Adverse ment, 2014. [Available at https://bioinfor
events after infusions of cryopreserved hemato- mant.com/wp-contentluploads/20 16/09/
poietic stem cells depend on non-mononuclear FDA-Regulation-of-Stem-Cells-Draft-Guidances.
cells in the infused suspension and patient age. pdf],
Cytotherapy2007;9:348-55. 167. Food and Drug Administration.Registrationand
160. Otrock ZK, Sempek DS, Carey S, Grossman BJ. listing. SilverSpring,MD: CBEROfficeof Com-
Adverse events of cryopreservedhematopoietic munication, Outreach, and Development,
stem cell infusions in adults: A single-centerob- 2019. [Availableat https://www.fda.gov/For
servationalstudy. Transfusion2017;57:1522-6. Industry/FD ABasicsforIndustry / ucm2346 25.
161. Hoyt R, Szer J, GriggA. Neurologicalevents as- htm.]
sociated with the infusion of cryopreserved 168. Food and Drug Administration.What does FDA
bone marrow and/or peripheral blood progeni- inspect? Silver Spring, MD: CBER Office of
tor cells. Bone Marrow Transplant 2000;25: Communication, Outreach, and Development,
1285-7. 2018. [Available at https:!/www.fda.gov/
162. RowleyS, MacLeod B, HeimfeldS, et al. Severe about-fda/fda-basics/what-does-fda-inspect.]
central nervous system toxicity associated with 169. Centers for Medicare and Medicaid Services.
the infusion of cryopreserved PBSC compo- Clinical LaboratoryImprovement Amendments
nents. Cytotherapy 1999;1:311-17. (CLIA).Baltimore, MD: CMS, 2019. [Available
163. Otrock ZK,BeydounA, BaradaWM, et al. Tran-
at https://www.cms.gov/Regulations-and-
sient global amnesia associated with the infu-
Guidance/Legislation/CLWindex.html.]
sion of DMSO-cryopreservedautologousperiph-
170. Centers for Medicare and Medicaid Services.
eral blood stem cells. Haematologica 2008;93:
Proficiency testing programs. Baltimore, MD:
e36-7.
CMS, 2019. [Availableat https://www.cms.
164. Akkok CA, Liseth K,Melve GK,et al, Is there a
scientific basis for a recommended standardiza- gov/Regulations-and -Guidanc e/Legislation/
tion of collection and cryopreservation of pe- CLWProficiency_Testing_Providers.html.]
ripheral blood stem cell grafts? Cytotherapy 171. Process overview. Omaha, NE: Foundation for
2011;13:1013-24. the Accreditation of Cellular Therapy, 2014.
165. Food and Drug Administration. Tissue and tis- [Available at http://www.factwebsite.org/
sue products. SilverSpring,MD: CBEROfficeof accreditationprocess/( accessed December 9,
Communication, Outreach, and Development, 2019).]
2019. [Availableat https://www.fda.gov/Bio 172. Laboratory accreditation program. Northfield,
logicsBloodVaccines/TissueTissueProducts/de IL: College of American Pathologists, 2019.
fault.htm.] [AVailableat http://www.cap.org/web/home/
166. Food and Drug Administration. Draft guidance lab/ accreditation/laboratory-accreditation -pro
for industry and Food and Drug Administration gram (accessedDecember 9, 2019).]
staff:Minimal manipulation of human cells, tis- 173. Accreditation program. Bethesda, MD: AABB,
sues, and cellular and tissue-based products. 2019. [Availableat http://www.aabb.org/sa/
(December 2014) SilverSpring,MD: CBEROf- overview/Pages/program. aspx (accessed De-
fice of Communication, Outreach, and Develop- cember 9, 2019).]
Transfusion Sup.port for H!matopoietic
Stem Cell Transplant Recipients
James M. KelleJ" MD, PhD, and MelissaM. Cushing, MD
EMATOPOIETIC STEM CELL TRANS- from matched related donors (MRDs),who are
plantation (HSCT)involves infusing hema- typically siblings or other family members;
topoietic progenitor cells (HPCs)to replace matched or single-antigen-mismatched unrelat-
or restore the cellular components within a pa- ed donors (MUDs) recruited by organizations
tient's marrow and for reconstitution of the im- such as the National Marrow Donor Program
mune system. HSCT is often performed for pa- (NMDP); or related haploldentical donors, who
tients with hematologic malignancies, marrow are typicallyparents or children. "Matching" re-
failure states, primary immunodeficiencies, and fers to the degree of similarity between specific
congenital blood disorders, or following treat- HLAalleles of the donor and recipient and pre-
ment with chemotherapy regimens. Over 200 dicts the lil<.elihoodof tissue compatibility. Hap-
hospitals offer the approximately 20,000 HSCT loidenttcal HPCs. match only half the HLA loci
procedures performed each year in the United evaluated, as they derive from a parent or child
States, demonstrating the widespread use of this who shares half the genes with the recipient.
procedure.' HSCTpatients often pre-sent to com- (See Chapter 16for discussion of HLA.)
munity hospitals for follow-up care, and thus a BPCs canbe harvested directly by using a
broader base of the medical community needs to wide-bore needle inserted into. the donor's
understand their transfusionneeds. marrow, or through apheresis collection after
Two main categories of HSCT exist: autolo- stimulating a donor with granulocyte colony-
gous and allogeneic. The majority of HSCTs are stimulating factor [G-CSF(filgrastim)]or plerixa-
autologous (58%).l Autologous HSCTs occur for to mobilize stem cells from the marrow into
when a patient receives stored HPCs he or she the peripheral circulation (ie, peripheral blood
previously donated. Such patients undergo HPC stem cells). They can also be derived from do-
collection and storage and are subsequently nated umbilical cords/placenta after a birth (ie,
treated with high-dose chemotherapy, which cord blood). Given the low cell count of a cord
targets their underlying cancer but also ablates blood donation, HPCs from two cord blood
their marrow. Patients then receive their autolo- donations are' often used for adult patients, al-
gous HPCs to circumvent the myeloablation af- though a single donation is sufficientfor smaller
ter completing chemotherapy. AllogeneicHSCTs pediatric patients.
occur when a patient receives HPCs from anoth- Appropriate transfusion support for patients
er individual. Allogeneic HPCs can be donated receiving an HSCT is both critical (as this pa-
James M. Kelley, MD, PhD, Assistant Professor of Laboratory Medic!Il~JThe Universityof Texas MDAnderson
Cancer Center, Houston, Texas; and Melissa M. Cushing, MD, Professor of Pathology and Laboratory Medicine,
Weill Cornell Medicine, Medical Director, Transfusion Medicine and Cellular Therapy, and Associate Director of
the Clinical Laboratories,NewYork-PresbyterianlWeill Cornell Medical Campus, New York, New York
J. Kelleyhas disclosed no conflicts of interest. M. Cushing is a consultant for Cerus Corporation and Octapharrna.
161
tient population often lacks the ability to pro- tlents may receive multiple HPC products at one
duce blood cells effectively)and challenging (as time, such as through a cord blood transplanta-
their blood types can transition if receiving tion, or undergo HSCTsof different ABO groups
HPCs from an allogeneic donor with a different following disease relapse or incomplete engraft-
blood type). Complexities such as the patient's ment.
underlying condition, level of immunosuppres-
sion, additional medications, existing alloanti- ABO-Compatible Blood Components
bodies, rate of engraftment, adverse events re-
lated to transplantation, the type of HPC Patients in the pretransplantation or postengraft-
transplant, and the presence of donor passenger ment phase of HSCT can be given transfusion
lymphocytes affect the need for and the expect- based on their current forward- and reverse-
ed response to transfusion. This chapter offers typing results, as the composition of their blood
an introduction to navigating some of these should derive completely from one progenitor
complexities when providing transfusion for this source. That is, they can be assigned blood com-
unique patient population. ponents based on a hospitai's standard proce-
dures for general patients. Figure 27-1 illustrates
that in the pretransplantation and postengraft-
ment phases of HSCT, the patient produces
blood from one progenitor source. The excep-
tion to this is in the setting of nonmyeloablative
HSCT, where patients will occasionally have
ABO matching of blood components to the sustained mixed chimerism, with existing eryth-
HSCT recipient is not essential when selecting rocytes of both donor and recipient origin.
allogeneic HPC donors. Unlike solid-organtrans- Patients in the immediate posttransplantation
plants, HPCs do not express ABO antigens and period, between HPC infusion and engraftment
will not activate an immediate humoral immune (see Fig 27-1), require more careful attention to
response leading to rapid rejection. Therefore, it ABO compatibility and selection of blood com-
is possible (and common) to select an HPC do- ponents. Blood Circulatingduring this time can
nor with a different ABO group than the recipi- consist of a shifting composition of cells that
ent-meaning that samples from HSCTpatients transitions from cells arising from the patient's
can simultaneously express more than one ABO originalmarrow to those of the donor. Often this
group during immunohematology testing. can lead to identification of ABO discrepancies
Interpreting such potentially complex immu- during pretransfusion testing in the blood bank,
nohematology results and selecting appropriate which can cause delays in test result reporting.
blood components for HSCT patients requires Each sample will require manual review by
medical record investigation. The followingdata blood bank staff, rather than immediate report-
help transfusion service staff determine the ing and interfacing with the electronic medical
patient's stage of transplantation (eg, pretrans- record through blood bank automation. During
plantation, immediate posttransplantation, or the posttransplantation phase, or if complete en-
postengraftment): the originalABO group of the graftment is not achieved, the ABO groups of
HSCT patient; the ABO group, type of product, both the donor and recipient should be consid-
and date of infusion of all HPC products; any ered in selecting components for transfusion be-
ABO titer results; molecular engraftment study cause it is common for an HPC donor and HSCT
results (if available); and the pre- and posttrans- patient to have different ABO groupmgs.v' Se-
plantation transfusion histories from all centers lecting components with ABO compatibility for
at which the patient received care. Once ob- both the donor and recipient is ideal for transfu-
tained, these data should be documented in a sion in such cases. Suggestions for ABO selec-
hospital's blood bank information system be- tion are listed in Table 27-1, but specificrecom-
cause HSCT patients frequently receive transfu- mendations vary by center, particularly with
sion. Such histories can become complex: pa- platelets.
C HAP T ER 2 7 TransfusionSupport during HSCT 769
l- +-'
o
(J)
c:
w
::r: E
¢:::
m
!...
0)
c:
UJ
-c
0
0
CO
....c:
(f)
W
+=i
m
o,
!t-
O
c:
c: 0
0 (f)
E
(f) -2
0
c:
c..
E o
n,
.0
o ::r:
Although the time point of infusion is clear, compatible HSCT is more easily accomplished,
engraftment is evaluated by laboratory testing but challenges exist for the other categories.'
and defined at each center. After an expected
period has passed, the patient's blood counts Major ABO Incompatibility
(eg, white cell count, platelet count) should re- Major ABOincompatibilitymay lead to acute in-
cover in a predictable manner as engraftment travascular hemolysis if residual donor red cells
occurs. The definition of engraftment is usually in the HPC product are infused into a patient
based on the patient's absolute neutrophil count with high ABO antibody titers, This risk is miti-
(ANC),but it can also include an assessment of gated by reducing the red cell content in the
ABO typing results (to determine whether the HPC products (to generally less than 10-30 mL
ABOtype consistently matches that of the donor of red cells), using one of a number of methods,
on multiple occasions) or molecular chimerism such as hydroxyethyl starch sedimentation.
assays (to look for a sufficientpercentage of lym- Most peripheral blood stem cell products and
phoid/myeloid cells arisingfrom the donor). cord blood units contain few red cells, but mar-
row-derived products will frequently require red
Compatible Blood Components for cell depletion. If the recipient ABO antibody ti-
Non-ABO Antigens ter is high, plasmapheresis may be used before
HPC infusion, to reduce the titer of circulating
A corresponding approach of providing compo-
ABO antibodies. Hemolysis or delay of engraft-
nents compatible with both the HPC donor and ment can occur if residual host immune cells
recipient is applicable to other antigens. For ex- continue to produce antibodies against the
ample, if either the donor or the recipient has an erythroid progenitors and red cells arising from
alloantibody to another blood group, such as RH the donor HPCs. This can continue for several
or KEL,antigen-negative Red Blood Cell (RBC) months after infusion, leaving HSCT patients
units should be provided for transfusion. With transfusion dependent for long periods after
regard to RhD, the recipient can continue to retransplantation. Pure red cell aplasia (PRCA)can
ceive the same type of components transfused develop in severe cases. Although no standard
before transplantation as long as the HSCT do- treatment for PRCA is established, corticoste-
nor and recipient match. The benefit of provid- roids, rituximab, bortezomib, alemtuzumab,
ing RhD-negativeRBCsif either the recipient or rapid tapering of calcineurin inhibitors, plasma
the donor is RhD negative, to prevent allo- exchange, donor lymphocyte infusions, and da-
immunization to the highly immunogenic D an- ratumumab have been used.
tigen, has not been well studied but leads to sig-
nificant depletion of the RhD-negative RBC Minor ABO Incompatibility
inventory. RhD-negative patients with an RhD-
In minor ABO incompatibility, cells in the HPC
negative donor generally receive RhD-negative product from the donor produce antibodies
RBCs.4 In one study, 7% of RhD-negative pa- against the patient's originalblood type. This oc-
tients with hematologic malignancies formed curs when preformed antibodies from the donor
anti-D when given RhD-positiveRBCsfor trans- are present in the HPC product or when plasma
fusion." Given the minimal risk of RhD alloim- cells from the donor continue to produce anti-
munization from red cells contained in RhD- bodies after infusion. Plasma reduction of the
positive platelet units, selection of RhD-negative HPC product can decrease this effect before in-
platelets is not mandatory. fusion. Donor-origin lymphocytes can produce
The degree of ABO matching between HPC anti-Aand antl-Bthat attack the patient's residu-
donor and HSCT recipient is grouped into four al red cells, typically5 to 16 days after infusion,
categories, as shown in Table 27·1: ABO com- creating a passenger lymphocyte syndrome. The
patibility, major ABO incompatibility, minor resulting acute, immune-mediated hemolysis
ABO incompatibility, and bidirectional ABO in- can range from subclinical to severe and usually
compatibility. Transfusion management of ABO- resolves once all the patient's original red cells
CHAPTER 27 Transfusion Support during HSCT 771
en
a.:>
~
zs
~c.
E
o
(..)
oS
o
co
«
..__
o
c::
°E
""0
c::
<tS
o
0c;r
E
'6
c::
o
fJ
c::
zs
E
o
<...:l
co co co co co co co co
« « « « « « « « co
«
o o o o
o 0 o o 0 o o 0

U)
:c
Cl
c::
°C
::::l
-0
~ co o o « co co co co « co
:.0 -c 0
« « «
~0..
E
o
e...:>
o
CO
«
are cleared. In extreme and life-threatening cas- survival," acute graft-vs-host disease (GVHDJ,
es, red cell exchange with compatible units (eg, propensity for bloodstream infections," and de-
group 0) is possible. velopment of veno-occlusive disease.11 Given
the challenges with ABO compatibility, it is es-
Bidirectional ABO Incompatibility sential that transfusion services maintain a suffi-
cient supply of group 0 RBCs, as these will be
Bidirectional ABO incompatibility leads to the
required for many allogeneic HSCTpatients.
complexities of both major and minor ABO in-
compatibilities described above.
Platelets
HSCTin Non-Oncologic Patients The timing of platelet engraftment has been as-
sociated with many factors, including relation-
Extra care should be taken if the allogeneic
ship to donor (eg, MRD transplants engraft fast-
HSCT is provided for sickle cell disease or other
er than MUD transplants), conditioning
serious blood disorders. These patients often
regimen, lack of transplantation complications
have red cell alloantibodies due to frequent
such as infection or onset of GVHD, and the
transfusion, including antibodies against anti-
gens potentially expressed by the HPC donor. source and dose of CD34+ progenitor cells in
Their HSCTpreparation may alsouse a reduced- the HPC product (eg, apheresis stem cell or
intensity conditioning regimen that leaves a marrow-harvested products engraft faster than
cord-blood-derivedprcductsl.P'"
long-term mixed chimerism of red cells. Careful
attention to red cell phenotypes of the donor Platelets are often stored in plasma, which
and patient is needed to guide selection of the exposes recipients to isohemagglutinins. Al-
best-matched components. Additional discus- though the small quantity of antibodies typical-
sion of sickle cell patient management is includ- ly present in platelet components is rarely clini-
ed in Chapter 19. cally relevant, the total volume of ABO-
incompatible plasma given should be carefully
considered. Some centers transfuse platelets sus-
BlO D 0 NENT pended in platelet additive solution (PAS)rather
than plasma to reduce isohemagglutinins expo-
SUPPORT FOR HSCT
sure. Suggestions for selecting components for
PATIENTS transfusion based on ABO groups are listed in
Table 27-1. Antibodies other than those against
Red Blood Cells
ABO groups, such as those targeting HLA or
The need to transfuse RBCsafter HSCT is com- platelet antigens may occur in HSCT patients,
mon, as symptomatic anemia can occur in this contributing to platelet refractoriness. This clini-
patient population. The need for transfusion can cal situation is described in Chapters 15, 16,
be influenced by the patient's gender, underly- and 19.
ing disease, source of HPCs, and degree of mye- The transfusion threshold for platelets in
loablatlon.V Transfusion decisions are typically HSCT is determined by each center. For most
guided by standard institutional transfusion uncomplicated cases, a platelet count of
guidelines. Although MBB clinical practice 1O,OOO/flLis sufficient to prevent spontaneous
guidelines recommend restricting RBC transfu- bleeding, but higher thresholds are needed for
sion to those with a hemoglobin level <7 g/ dL those undergoing procedures or experiencing
for the general patient population, the evidence sepsis or active bleeding." Patients receiving de-
supporting these guidelines is not sufficient to fibrotide, a drug that enhances the fibrinolytic
recommend a threshold for hematology/ oncolo- activity of plasmin, used to treat veno-occlusive
gy patients with severe anemia." Transfusion- disease or sinusoidal obstruction syndrome
associated iron overload is a significant concern during HSCT, may require a higher platelet
in HSCT patients because increased serum threshold." Dosing of platelets for HSCT pa-
ferritin has been associated with decreased tients should follow guidelines similar to those
C HAP T E R 2 7 TransfusionSupport during HSCT 773
for the general patient population. Much work fections, such as cytomegalovirus (CMV), that
in recent years has evaluated standard- vs low- can be transmitted through blood components.
dose platelet transfusion strategies,'?" with CMV-reduced-risk cellular blood components
evidence suggesting that low-dose platelets for (from a CMV-seronegative donor or ~euk9cyte
prophylaxis does not affect the incidence of reduced) should be used, Irradiating compo-
bleeding. nents to prevent transfusion-associated GVHD
There is also an ongoing discussion on the (TA-GVHD), an almost universally fatal but rare
appropriateness of transfusing platelets only for complication, is also recommended for this
therapeutic intent rather than for prophylaxis immunocompromised population. Providing
before potential bleeding is present. Although a pathogen-reduced components is an alternative
therapeutic-only strategy may be appropriate for option for providing CMV-reduced-risk blood
autologous HSCT patients, data in patients with and preventing TA-GVHD. These component
acute myelogenous leukemia (AML), a condi- modifications are discussed elsewhere in this
tion frequently leading to HSCT,indicate a need manua1.
to maintain prophylactic transfusions." The Tri-
al of Prophylactic Platelets (TOPPS)also indicat-
ed benefit for prophylactic transfusion in pa- IONS
tlents with hematologic malignancies." Recent
systematic reviews support these conclusions Transfusionsupport of pediatricJISCT recipients
specificallyin HSCT patients," is similar to that for adults, as standaJ'd-of-care
has often been extrapolated from published data
Plasma, Cryoprecipitate, and Other in adult HSCTpatients." Recommendations by
Blood Derivatives one expert group for RBCtransfusion support in
children ...following HSCT who are. critically ill
There are no standard guidelines related to the but hemodynamically stable suggests a transfu-
use of plasma, cryoprecipitate, or other factors sion threshold hemoglobinlevel of 7 to 8 gI dL.25
in HSCT patients." Issues related to ABO com- Clinical decisions related to donor selection
patibility should be addressed as discussed and HPC product type may differ based on the
above. patient's age and size. For example, children
commonly receive one. cord-blood-derivedHPC
Granulocytes product compared to the. two usually infused in
Granulocytes have been used to treat neutrope- adults. Children may also be conditioned with
nic patients, typicallywith ANCs <SOO/I!L and reduced-intensity chemotherapy that allows re-
with bacterial or fungal infections that are residual production of some recipient red cells.
fractory to standard treatment. The Resolving Additional indications for HSCT include
Infection in Neutropenia with Granulocytes congenital diseases (eg, sickle cell disease,
(RING) trial, although underpowered, showed ~-thalassemia major, certain marrow failure
that granulocyte therapy did not confer benefit states); some childhood malignancies (eg, neuro-
beyond standard antimicrobial therapy.i' A Co- blastoma) are treated with high-dose chemo-
chrane review found insufficientevidence to de- therapy followed by autologous HSCT rescue.
termine an effect of granulocyte transfusion on Managing these patients Iollowing:' HSCT in-
all-causemortality in HSCTpatients." cludes addressing transfusion considerations re-
lated toti1eirllnderlying disease. For example,
in sickle cell or thalassemia patients, deterrnin-
Processing of Blood Components for
ing from molecular methods the red-cell-anti-
HSCTRecipients
gen predicted phenotype of both the donor and
Given the immunocompromised nature of recipient, as well as documenting and address-
HSCT patients, care must be taken to mitigate ing any history of existing red cell alloantibod-
risks of transfusion-related adverse events. Spe- ies, is useful when providing transfusion support
cifically,these patients are susceptible to viral in- and when selecting HPC donors. In addition,
avoiding iron overload and maintaining a high- to prevent cerebrovascular bleeding may be
er platelet count than in other HSCT recipients warranted. 24
.' l'i·.H~9E.··offerstheabili:V. t()rePlac~•••

·~r·.restotemarrow.~rl;.~~It f~ll?yn~g d:s~uctlo~bY .•~iS:~S~
.'.•. ·.·or•treattnent.
. <..•.............•......•..•.•....•.•.
\ ••.
> •.•.•..••.•.•••...••...•.•••..•..••.•...•.•.•.••.••••..•.•..•....•.••••.••..•.•••••.•••...•..••.•.•.•.....•.•.••...•.•.•.
> /'.' > •..••••••.•..•.•...••.•.•.••••.•.••••.••..••.....•••
\ •••••
.'.2{HSCTisJncreasingly being'perf()rmed in numerous centersj With patients seekihg folloW'up itt:

\j••cdlIllUtln1ty{regio~al~ospita1s'>J····.··i••·.·•.··•·•··•
...... r.·.·•.•··•.\ •....•......••.•••
3j·~SCT.patients present unique chClllenges,as these immunocbmpromised ••patients •.can P?ssess
i.'·.·.·.:<..•
i.:/\ ....,...•....•.
> i.··. . iii.. >.··::i.·.·.:\..\.,.;. ii.•·•·.·.•
blo()d cells.andilTImune ~ffect?rcells from mUltipleindiyi~~~S(redpient, mUltipleallo~e~efc.
~~rl~~k
~6~d~~dri;.~~~:~ritt~~g1~:~~,p:J~:.~~bf;!irs~if:ompliCatiOns ••due.to ••thetr .••
u y-....
4.AI~ough ABO .compatibility is not essential for selec~on?f al"lRgc donor,pr0viding ~ccegt.
. ~1~.blood.c0tl1p?n~nts~9lTI. ~gcinfusion to engraftm~~tr~q~if~s.considerationOf ~~ ?ri~.··
nal.blood .groupof·the patiental"ldthe blood group(slo{.all.~P8products.·received. Ex~~~Siy~
•,. ~~sord-ke~ping9f the transplantation and.transfusion history aids in selecting ABO-compatible
·•·••·•.·.•••
····blogqc()rn.P?ne!1ts :\"..··
••....
: •.•.•••: •.••.•...•••.•
¥n9
.':~/ ·fr,al"lsfusloI1~idelm~sal"lg;tiire 1qsSh?t1ldbe careftillycdl1sideredby each-center .forthWp'a~:'
.. ~~mpopUlati0I1'·."'••••• ·..', .. i .•••....••> ....•.... ......•. i
'.'•••••••.•
·•••••• i\ .......•
·•·••••••
·••••·••.i/ i..
6. ··TQ~se. iITltl1unocomprornise~g~v~nts require blood cornponentsthat are.irradiated.or p~~o.
~~~reducedand CM\!req1.lc~~}i~.k(ie,leukocyte reduced orCM\! seronegative) to mitigate'
the risks of transfusion-aSSOciatedgraft-vs-hostdisease and.CMV infection.
REFERENCES
1. Health Resources and Services Administration. 6. Le Viellez A, P'Ng S, Buffery S, et al. Red cell
Transplant activity report. Rockville, MD: HR- and platelet transfusion burden following mye-
SA, 2018. [Availableat https:!lbloodstemcell. loablative allogeneic haemopoietic stem cell
hrsa. govI datal donation-and- transplantation- transplantation. Intern Med J 2015;45: 1286·92.
statistics/transplant-activity-report (accessed 7. Kekre N, Christou G, Mallick R, et al. Factors
February 24, 2020).1 associated with the avoidance of red blood cell
2. Bolan CD, Leitman SF, Griffith LM, et al. De- transfusion after hematopoietic stem cell trans-
layed donor red ceil chimerism and pure red cell plantation. Transfusion2012;52:2049-54.
aplasia following major ABO-incompatiblenon- 8. CarsonJL, Guyatt G, Heddle NM, et al. Clinical
myeloablative hematopoietic stem cell trans- practice guidelines from the AABB:Red blood
plantation. Blood2001 ;98:1687-94. cell transfusion thresholds and storage. JAMA
3. KimJG, Sohn SK, Kim DH, et al. Impact of ABO 2016;316:2025-35.
incompatibility on outcome after allogeneic pe- 9. Meyer SC, O'Meara A, Buser AS,et al. Prognos-
ripheral blood stem cell transplantation. Bone tic impact of posttransplantation iron overload
Marrow Transplant 2005;35:489-95. after allogeneic stem cell transplantation. Bioi
4. Cid J, Lozano M, Klein HG, FlegelWA. Match- BloodMarrow Transplant 2013;19:440-4.
ing for the D antigen in haematopoietic progeni- 10. PullarkatV, BlanchardS, Tegtmeier B,et al. Iron
tor cell transplantation: Definition and clinical overload adverselyaffects outcome of allogeneic
outcomes. BloodTransfus 2014; 12:301-6. hematopoietic cell transplantation. Bone Mar-
5. Arora K, KelleyJ, Sui D, et al. Cancer type pre- row Transplant 2008;42:799-805.
dicts alloimmunization following Rhli-incom- 11. Maradei SC, MaiolinoA, de Azevedo AM, et al.
patible RBCtransfusions. Transfusion 2017;57: Serum ferritin as risk factor for sinusoidal ob-
952-8. struction syndrome of the liver in patients un-
C HAP T E R 27 TransfusionSupport during HSCT 775
dergoing hematopoietic stem cell transplanta- tological malignancies: An open-label, multi-

tion. Blood2009;114:1270-5. centre, randomised study. Lancet 2012;380:
12. Chang YJ,Xu LP, Liu DH, et al. The impact of 1309-16.
CD34+ cell dose on platelet engraftment in pe- 19. Stanworth SJ,Estcourt LJ,Powter G, et al. A no-
diatric patients following unmanipulated hap- prophylaxis platelet-transfusion strategy for he-
loidentical blood and marrow transplantation. matologic cancers. N Eng1J Med 2013;368:
Pediatr BloodCancer 2009;53: 11 00-6. 1771-80.
13. Pulsipher MA, Chitphakdithai P, Logan BR, et 20. Christou G, Iyengar A, Shorr R, et al. Optimal
al. Donor, recipient, and transplant characteris- transfusion practices after allogeneichematopoi-
tics as risk factors after unrelated donor PBSC etic cell transplantation: A systematic scoping
transplantation: Beneficial effects of higher review of evidence from randomized controlled
CD34+ cell dose. Blood2009;114:2606-16. trials. Transfusion2016;56:2607-14.
14. KaufmanRM, DjulbegovicB, Gernsheimer T, et 21. Roback JD, Caldwell S, Carson J, et al.
al. Platelet transfusion: A clinicalpractice guide- Evidence-based practice guidelines for plasma
line from the AABB.Ann Intern Med 2015;162: transfusion.Transfusion2010;50: 1227-39.
205-13. 22. Price TH, Boeckh M, Harrison RW, et al. Effica-
15. RichardsonPG, Smith AR,Triplett BM,et al, for cy of transfusion with granulocytes from
the Defibrotide Study Group. Defibrotide for G-CSF/dexamethasone-treated donors in neu-
patients with hepatic veno-occlusive disease/ tropenic patients with infection. Blood 2015;
sinusoidalobstruction syndrome: Interim results 126:2153-61.
from a treatment IND study. BioiBloodMarrow 23. Estcourt LJ, Stanworth SJ, Hopewell S, et al.
Transplant 2017;23:997-1004. Granulocyte transfusions for treating infections
16. Heddle NM, Cook RJ,Tinmouth A, et al. A ran- in people with neutropenia or neutrophil dys-
domized controlled trial comparing standard- function. Cochrane Database Syst Rev 2016;4:
and low-dose strategies for transfusion of plate- CD005339.
lets (SToP)to patients with thrombocytopenia. 24. Webb J, Abraham A. Complextransfusion issues
Blood2009; 113:1564-73. in pediatric heamtopoietic stem cell transplanta-
17. Slichter SJ, Kaufman RM, Assmann SF, et al. tion. TransfusMed Rev 2016;30(4):202-8.
Dose of prophylactic platelet transfusions and 25. Steiner ME, Zantek ND, Standworth SJ, et al.
prevention of hemorrhage. N EnglJ Med 2010; Recommendations on RBCtransfusion support
362:600-13. in children with hematologic and oncologic di-
18. Wandt H, Schaefer-EckartK, Wendelin K, et al. agnosesfrom the Pediatric CriticalCare Transfu-
Therapeutic platelet transfusion versus routine sion and Anemia Expertise Initiative. Pediatr
prophylactictransfusion in patients with haema- Crit Care Med 2018; 19:5149-56.
Hlunan Tissue Allografts and the
Hospital Transfusion Service
Annette J. Schlueter/MD, PhD; FranRabe, MS; and Cassandra D. Josephson, MD
HE SURGICALUSE OF HUMAN CELLS, However, because. ordering, receiving, storing,

tissues, and cellular and tissue-based dispensing (issuing), tracking, tracing, investi-
products (HCT/Ps) continues to expand. gating adverse events, and managing recalls are
Every year, member organizations of the Ameri- functions performed by both transfusion and tis-
can Association of TissueBanks (AATB)recover sue services, AABBrecommends using a central-
tissue from more than 39,000 donors and pro- ized tissue service model located within the
vide more than 3.2. million tissue grafts for transfusion service.'
transplantation. 1 These activities are increasing,
in part because of the successful clinical applica-
tion of tissue allografts from living donors (eg, N
delivery mothers donating the placenta for grafts T N
derived from the amniotic/chorionic mem-
branes). Many surgical specialties use human Allografts are selected for transplantation based
tissue allografts: orthopedic surgery, neurosur- on intrinsic qualities that meet the surgeon's
gery, cardiothoracic surgery, plastic surgery, vas- functional requirements for the patient. Bone,
cular surgery, urology, ophthalmology, burn and tendons/ligaments, and corneas are the most
other skin wound care, podiatry, sports medi- frequently implanted human tissues. Others in-
cine, trauma, and cranio/maxillofacial surgery. clude skin and associated adipose; amniotic/
Increasingly sophisticated HCTIPs continue to chorionic membranes; cartilage/meniscus (with
be developed by tissue suppliers to meet diverse or without bone); certain veins/arteries; soft tis-
clinical needs. sue, such as fascia, pericardium, nerves, rotator
Oversight of activities of a tissue service may cuff, and dura mater; semilunar heart valves;
be managed by an individual, a diverse group, and cardiac conduit grafts. Tissue recovery from
or a department within a hospital, and various deceased donors occurs within 24 hours of
departments or surgical specialties may prefer to death after appropriate consent (also known Cis
manage only the tissue their service handles, "authorization") is obtained from a designated
Annette J. Schlueter, MD, PhD, Clinical Professor of Pathology and Medical Director, DeGowin Blood Center
Patient Services and Tissue and Cellular. Therapies, Department of Pathology, University of Iowa, Iowa qty,
Iowa; Fran Rabe, MS, Quality Assurance Director, Molecular and Cellular Therapeutics, University of Minnesota,
Minneapolis, Minnesota; and Cassandra D. Josephson, MD, Professor, Pathology and Pediatrics, Emory Univer-
sity School of Medicine, Director of Clinical Research, Center for Transfusion and Cellular Therapies, and Pro-
gram Director and Medical Director, Children's Healthcare of Atlanta Blood, Tissue, and Apheresis Services,
Atlanta, Georgia
C. Josephsonhas disclosedfinancialrelationshipswith Octapharma, Immucor,Medtronic, and Terumo.A. Schlueter
and F. Rabehave disclosedno conflictsof interest.
717
legal authority (eg, the donor's next of kin) or by leads to a determination that the donor is not el-
first-person authorization, if the donor regis- igible. Allografts, similar to blood components
tered his or her wishes before death, Steps taken and cellular therapy products, are released for
to minimize tissue contamination include use in patients only after the donor has been
planning for body cooling after death; use of screened and tested for relevant infectious dis-
aseptic surgical recovery techniques, instru- eases and the results are determined to be ac-
ments, and supplies; and control of the site ceptable. (SeeTable 28-1.)
where recovery takes place.
Tissue bank personnel responsible for deter- Types of Transplantable Tissue
mining donor eligibilitydo so based on an evalu-
ation of l) the validity of the document of au- Allografts, sometimes referred to as homograjts,
thorization (or informed consent for a living are grafts transferred between members of the
donor); 2) answers provided by the donor or same species. Human allograft tissue may be
other knowledgeable person to questions about processed (including activities such as disinfect-
the donor's travel, medical history, and behav- ing, sterilizing, packaging, labeling, testing,
ior that could indicate increased risk of disease and/or preservation) to prevent or retard biolog-
transmission; 3) available relevant medical re- ic or physical deterioration and maintain tissue
cords, including circumstances surrounding quality for its intended use. Tissue allograftscan
death; 4) a physical assessment or physical ex- be derived from a single tissue or multiple tis-
amination of the donor to identify evidence of sues acting as a functional unit. During process-
high-risk behavior or active infectious disease; ing, they may be combined with other biocom-
5) relevant infusion and transfusion information patible agents to achieve desired handling and
to evaluate for potential plasma dilution of any functional characteristics. Depending on the ex-
blood samples collected for infectious disease tent of processing and the intended use and ef-
testing, as well as infectious disease test results; fect, human tissue allograftscan be regulated by
and 6) the autopsy report (ifan autopsy was per- the US Food and Drug Administration (FDA)as
formed). High-riskbehavior that is considered to human tissue, biological products, drugs, or
increase the possibility of disease transmission medical devices.'
TABLE 28·1. Required Infectious Disease Testing for Donors of Allograft Ilssue"
Infectious Agent Test perfllrrTled···FOAlh:ensedor Cleared
Cytomegalovirus· FDA-cleareddonor screeningtest for anti-CMV (total
IgGand IgM)
Hepatitis B virus (HBV) HBVsurface antigen
HBVcore antibody (lgM and IgG)
HBVnucleic acid testing
Hepatitis C virus (HCV) HCVantibody
HCVnucleic acid testing
Human immunodeficiencyvirus (HIV) HIV-1 and HIV-2 antibodies
HIV-1 nucleic acid testing
HumanT-ceillymphotropic virus (HTLV)* HTLV-Iand HTLV-II antibodies
Treponemapal/idum FDA-cleareddonor screeningtest for syphilis
West Nilevirus (WNW WNV nucleic acid testing
- ~
'Required only for tissues that are rich in viable leukocytes.
tRequiredonly for tissues from living donors.
IgM = immunoglobulin M.
C HAP T E R 2 8 TissueServicesin the Hospital 779
Autografts are implanted into the individual lations promote the expectation that tempera-
from whom they were removed. Some exam- ture, humidity, ventilation, and air filtration will
ples of autograft tissue are bone that has been be controlled in critical manufacturing areas to
surgically removed from a patient's ilium, the extent deemed necessary by the tissue estab-
shaped to desired dimensions, and implanted lishment and supported by data.
into the vertebral disk space of the same pa- During bone, soft tissue, and connective tis-
tient; skull flaps surgically removed after crani- sue processing, various solutions may be used to
al trauma, then reimplanted when cerebral reduce or eliminate bacterial contamination (the
edema recedes; and parathyroid fragments cryo- bioburden) and to remove lipids and other cellu-
preserved after parathyroidectomy that may be lar material. Antibiotics alone or in combination
reimplanted. if the patient later experiences hy- with chemicals such as alcohol, peroxide, and
poparathyroidism. surfactants can be used, and some methods are
Xenogrsfts are transplanted from one spe- patented or proprietary. Depending on the type
cies to another. Many medical products have of tissue and its clinical utility or the biomechan-
been developed from highly processed, nonvia- ical expectations for it, treatment with chemi-
ble tissue from nonhuman animals and are regu- cals and/or graft sterilization mayor may not be
lated as medical devices. For example, porcine possible. Allografts containing viable .cells or
cardiac valves are xenografts used in certain a fragile matrix (eg, fresh or cryopreserved ves-
valve-replacement surgeries. sels or cardiac grafts) cannot be subjected to
Xenotransplantation products, often con- chemicals or sterilization without an adverse af-
fused with xenografts, contain live cells, tis- fect on cellular or matrix integrity. Low-dose
sues, or organs from a nonhuman animal, or the electron beam and gamma irradiation are the
products have been exposed during manufac- most frequently used sterilization treatments in
ture to live cells, tissues, or organs from a non- the United States; however, the tissue is first
human animal. Depending on its classification treated using proprietary solutions and methods.
by the FDA, a xenotransplantation product can Claims regarding allograft sterility must be sup-
be a biologicalproduct, a drug, or a medical de- ported by validation, and although some meth-
vice. For example, xenogeneic cells contained ods inactivate viruses, claims of sterility do not
in a device used for extracorporeal hernoperfu- include viruses.
sion would be considered a xenotransplantation Various tissue preservation methods can be
product. used to extend storage of tissue allografts so an
Information concerning certain allografts, inventory (ie, a "bank") can be readily available.
such as reproductive tissues (eg, gametes) and Tissue integrity can be maintained through
cellular therapy products, as well as xenografts simple freezing or through cryopreservation, a
and xenotransplantation products, is not provid- process in Which tissues are preserved using a
ed in this chapter; nonetheless.. such allografts cryoprotectant, such as glycerol or dimethylsulf-
might be handled by a transfusion service and oxide (DMSO), and frozen at a controlled rate
tracked and maintained using procedures estab- to lower subzero temperatures that allow for
lishedIor tissue allografts. long-term storage (ie, several years). Refrigerat-
ed storage can also be used to preserve cellular
viability or slow matrix degradation, but only for
Tissue Processing
short-term storage (days to weeks); examples in;
After tissue has been recovered from a donor, elude corneas and osteochondral or osteoarticu-
various levels of tissue processing and preserva- lar allografts. The latter two grafts provide bone
tion may be accomplished. The ...tissue-process- with a small or large articulating cartilage sur-
ing facility and the processing steps (ie, manu- face. If the clinicalutility of the tissue allows, al-
facturing) are designed to prevent tissue lografts can be preserved by removing intrinsic
contamination and cross-contamination. Similar water using dehydration, dessication, or, if a
to current good manufacturing practice require- very low residual moisture level is desired,
ments, current good tissue practice (cGTP)regu- lyophilization (ie, freeze-drying). The storage
potential (the expiry) is often predicated on the burned patient, protecting the underlying tis-
ability of the packaging configuration to main- sues from dehydration and infection. Human
tain a low level of moisture over time (a year or skin and amniotic/chorionic membranes can be
more). preserved and frozen, or processed to remove
cellular elements, producing an acellular matrix
Clinical Uses of Allografts that provides a scaffoldfor revascularization and
Various human tissues are used for transplanta- cellular incorporation in soft tissue reconstruc-
tion." Examples are listed in Table 28-2. Cadav- tive surgery and wound care. In addition, donor
eric human bone can replace bone lost to corneal tissues can treat various ocular surface
degenerative disease, trauma, or malignancy.Al- diseases, such as keratoconus, other eye pathol-
logeneic bone has unique healing characteris- ogies, and trauma. Scleral, pericardial, and am-
tics, including osteoconductive and osteoinduc- nion-derived allograftscan treat glaucoma, scler-
tive properties. In vivo, it acts as a scaffoldthat al ulcers, and other eye damage.
allows recipient capillary growth into the graft Although some tissue allograftsmay provoke
(osteoconductivity) and provides stimulation for an immune response resulting in sterile inflam-
the production of new bone (osteoinductivity) mation, these do not usually result in failure of
by exposing the patient's osteogenic progenitor the allograft, presumably because of a lack of
cells to bone morphogenetic proteins (BMPs), abundant residual cellular material in processed
which are growth factors in bone that induce grafts.5,6 Therefore, it is not necessary to match
new bone formation. The result is creeping sub- most allografts to the recipient's HLA or ABO
stitution, in which bone remodeling occurs type. This is in distinct contrast to organ al-
through osteoclastic resorption of the implanted lografts, where donor-specific HLA or ABO
tissue and osteoblastic generation of new bone. blood group antibodies can pose significantchal-
Crushed bone, subjected to acid demineraliza- lenges to organ function." HLAantibodies have
tion, can be used alone or suspended in a biolog- been associated with corneal transplant rejec-
ically compatible carrier and applied to exposed tion, but HLAmatching of these allograftsis not
bone surfaces. BMPs in demineralized bone commonly performed." Although they may not
stimulate osteogenesis, the fusion of adjacent affect the function of the tissue graft, HLAanti-
bones, and healing. Precision bone grafts, in- bodies that develop as the result of a bone or
cluding a growing array of spinal implants, are skin graft may affect the patient's subsequent
shaped using sophisticated, computer-aided cut- abilityto receive an organ allograft.10, 11
ting devices to fit snugly into surgicalinstrumen- Some surgeons request ABO compatibilityfor
tation designed to allow surgeons to place the cardiovascular grafts, including cryopreserved
grafts delicately in the defect. Bone-tendon (eg,
heart valves, veins, or arteries, although the sig-
Achilles tendon) or bone-ligament-bone (eg,
nificance of this practice is unproven. Develop-
patellar-tibia ligament) grafts are routinely used
ment of RhD, Ft, and Jkb antibodies in recipi-
for anterior cruciate ligament (ACL)repair. The
implanted tendon or ligament spans the joint ents following transplantation of unprocessed
space, and the bone provides an anchor into the bone allograftshas been documented in case re-
femur and/or tibia, thereby restoring joint sta- ports": however, unprocessed allograft bone is
bility. Surgical techniques for ACL repair were no longer provided in the United States. If un-
developed using alternative fixation methods processed bone contains marrow elements that
with a combination of tendons (without a bony cannot be removed mechanically and is intend-
attachment), such as the anterior or posterior ed for use in an RhD-negative female of child-
tibialis tendons, gracilis, semitendinosus, and bearing potential, her future offspringmay be at
peroneus longus. Meniscus and joint allografts risk of hemolytic disease of the fetus and new-
are also used to regain mobility after disease or born. Thus, Rh Immune Globulin prophylaxis
trauma. Skin from a deceased donor can be used can be considered if the Rh type of the allograft
as a temporary wound dressing for a severely donor is positive or unknown.
TABLE 28·2. Examplesfor Clinical Use of Human Tissue Allografts
Amniotic/chorionic membranes Leg and foot ulcer/wound care

Conjunctiva surface repair
Corneal repair
Neurosurgery and spine surgery
Orthopedic surgery
Dental/periodontal surgery
Burns
Bone (cortical, corticocancellous, cancellous; Skeletal reconstruction
powders, pastes, putties, gels, and Spinal fusion
moldable strips) Dental implant placement
Bony defect filler
Bone-tendon (Achilles tendon) Anterior cruciate ligament repair
Bone-ligament-bone (patellar ligament) Posterior eructate ligament repair
Tendons (anterior and posterior tibialis, Rotator cuff restoration
semitendinosus, gracilis, and peroneus longus) Bicepstendon rupture repair
Card,iacvalves (aortic and pulmonary) and Replacementfor reversal of valvular insufficiency
conduits Congenital cardiac defect repairs
Cartilage (costal) Facialreconstruction
Cornea Keratoconus correction
Fuchs dystrophy repair
Traumatic scarring repair
Corneal-scleralfistula repair
Decellularizedskin (dermal matrix) Hernia repair
Soft tissue reconstruction
Gingival restoration
Demineralized bone Dental implant placement
Spinal fusion
Bony defect filler
Dura mater Dural defectlcerebrospinalleak repair
Fascialata Soft tissue reconstruction
Pelvic floor support
Meniscus Meniscus replacement
Osteoarticular/osteochondral graft (bone and Joint restoration
joint cartilage)
Pericardium Dura patch
Eyelid reconstruction
Soft tissue reconstruction
Sclera Eyeenucleation
Scleral ulcer repair
Eyelid repair
(Continued)
TABLE 28·2. Examples for Clinical Use of Human Tissue Allografts (Continued)
Leg and foot ulcer/wound care

Plastic surgery reconstruction
Veins/arteries Coronary artery bypass grafting
Tissue revascularization
Aneurysm repair
Dialysis access shunts
Disease Transmission through Tissue

1270 and 1271, of the Code of Federal Regula-
In the past, rare, sporadic transmission of infections (CFR).3Tissue banks engage in one or
tious diseases, including human immunodefi- more manufacturing functions, such as donor
ciency virus (HIV), hepatitis C virus (HCV), screening and testing and tissue recovery, pack-
hepatitis B virus (HBV),and Creutzfeldt-Jakob aging, labeling, processing, storage, andlor dis-
disease (CJD), was documented following tis- tribution for clinical use. These entities manu-
sue lmplantation." In addition, bacterial and facture what the FDA classifies as HCT/Ps.
fungal infections from allografttransmission re- Table 28-2 lists examples of common nonhema-
sulted in morbidity and mortality. Potential topoietic HCTIPs. Although a tissue distribution
sources of contaminating agents included those intermediary might perform limited manufactur-
that were donor derived or from environmen- ing functions (eg, storage, distribution), it must
tal contaminants introduced at the processing follow applicableregulations. Bydefault, an allo-
facility. Malignancy was also rarely reported to geneic HCTIP is regulated as a drug, device,
be transmitted from tissue grafts and is limited andlor biologicproduct under the Food, Drug,
to corneal allografts." Currently, disease trans- and Cosmetic Act andlor Section 351 of the
mission from transplanted tissue is rare, but Public Health Service (PHS)Act. If it meets the
vigilance and surveillance are necessary to four criteria in Table 28-3, it is regulated solely
maintain confidence that controls, such as under section 361 of the PHS Act, and regula-
cGTP measures, are working. Advances in do- tionsin21 CFRPart 1271.15
nor screening, such as for Zikavirus, effectively Three subparts of 21 CFRPart 1271 concern
reduce the risk of a donation from an infected 1) registration of tissue establishments, 2) donor
donor." Additionally, the application of newly eligibility,and 3) cGTP requirements related to
validated donor testing and tissue culture and handling of HCTIPs. Compliance with these
treatment methods, as well as the application regulations is required of tissue establishments
of quality assurance and quality control mea- to control contamination and cross-contamina-
sures, continue to improve the safety profilesof tion and avoid disease transmission. Tissue-dis-
human tissue. pensing institutions, such as hospitals, dental of-
fices, and surgical centers that provide and use
tissue within their own facility, are not subject
NlS to this oversight except under certain circum-
stances (eg,routine redistribution of allograftsor
N R autografts to other institutions, including an af-
filiate located at a different address; application
The FDAregulates the activities of tissue estab- of certain manufacturing steps that could con-
lishments (tissue banks) under Title 21, Parts taminate the tissue).
TABLE28-3. Criteria from 21 CFR 1271.1 O(a) for Regulation of an HCT/P under Section 361 of the
PHS Act
1. manipulated,and
2. The HCT/Pis intended for homologous use only, and
3. The HCT/Pis not combined with another article (with some limited exceptions), and
4. The HCT/Pdoes not havea systemic effect and is not dependent upon the metabolic activity of living cells
for its primary function
Or
The HCT/Phas a systemic effect or is dependent upon the metabolic activity of living cells for its primary
function and is for autologous use; for allogeneic use in a first- or second-degreeblood relative; or for
reproductive use
21 CFR1271.1O(a) = Code of FederalRegulations, Title Part 1271.10(a); HCT/P = human cells, tissues, and cellular and
tissue-based products; PHS = Public Health Service.
For a hospital-based tissue service that does resources for the donation and transplantation
not also qualify as a tissue establishment, com- communities. A tissue service may find MTB
pliance oversight may be provided by state laws and EBM accreditation of a supplier to be valu-
and/or accrediting organizations. The Joint able in assessing that supplier's qualifications.
Commission, AABB, the College of American The Joint Commission, CAP,MBB, and AORN
Pathologists (CAP),the Association of periOper- all have standards/guidelines that are directed
ative Registered Nurses (AORN),the MTB, and specificallyat the activities performed by hospi-
the Eye Bank Association ofAmerica (EBM) all tal-based tissue services.
publish standards or guidelines that apply to the Several states have requirements for tissue
practices of tissue services."?' The location of a banking performed within the state, and these
tissue service in a hospital or medical facilityand statutes can affect the tissue service of a transfu-
the scope of its operations dictate which stan- sion service. For example, New York State re-
dards are applicable. These standards are updat- quires tissue bank licensure when a "tissue ser-
ed regularly; therefore, periodic review of the
vice" meets the definition of a "tissue storage
most recent versions is important to guarantee
facility" or a "tissue transplantation service" as
ongoing compliance and to maintain best prac-
described in Part 52 of Title 10 (Health) of the
tices.
Both the MTB and EBM are voluntary ac- Official Compilation of Codes, Rules and Regu-
crediting organizations dedicated to ensuring lations of the State of New York In the Califor-
that human tissues intended for transplantation nia Health and Safety Code, Section 1635, a tis-
are safe, available, and of high quality. The sue bank licensing provision is included, but for
MTB's standards pertain to institutional and a transfusion service that handles tissue, it de-
quality program requirements; donor authoriza- pends, in part, on meeting criteria involving the
tion/informed consent; donor screening and storage of certain tissues.
testing; and tissue recovery/acquisition, storage, Finally,other organizations that advise or ac-
processing, release, distribution, and dispens- credit specifichealth-care services proyide direc-
ing.2oEBM's scope encompasses all aspects of tion that may affect the functions performed by
eye banking." Accreditation by these organiza- a hospital-based tissue service. For example, the
tions is based on verified compliance with estab- United Network for Organ Sharing (UNOS)has
lished standards and periodic inspections. Both policies for organ-donor vessel packaging, label-
organizations serve as scientific and educational ing, storage, shipping, tracking, and reportmg.f
H PI L u centralized process is permitted to manage these

activities. In either model, designated oversight
There is no requirement for a tissue service to is required to coordinate tissue-related activities
be managed in any particular department or by a and ensure standardization of practices through-
specificindividual. A tissue service located without the organization. The Joint Commission's re-
in a transfusion service can be accredited by quirements apply to human and nonhuman
AABB through adherence to AABB Standards cellular-based transplantable and implantable
for Blood Banks and Transfusion Services, 17 products, including tissue allografts and certain
which is based on expertise in providinghuman- medical devices, as classifiedby the FDA.Table
derived products that are perishable, potentially 28-4 lists the controls a tissue service could sup-
infectious, and sometimes in short supply, and
port.
that require bidirectional traceability between
donor and recipient. Ordering, receiving, stor-
Standard Operating Procedures and
ing, distributing (issuing), tracking, and tracing
Policies
products, as well as investigating adverse
events, including complaints, recalls, and look- Hospital tissue services must have written stan-
back investigations, are activities that are com- dard operating procedures (SOPs), either print-
mon to a transfusion service and a tissue service. ed or electronic, for all functions pertaining to
the acquisition, receipt, storage, issuance, and
Responsibility for Hospital-Based tracing of tissue grafts, as well as procedures for
Tissue Services investigating adverse events and handling re-
The Joint Commission requires organizations to calls. Manufacturers' instructions for handling
assign oversight responsibility for their tissue tissues must be followed. When the blood bank
program, use standardized procedures in tissue or transfusion service is responsible, AABBre-
handling, maintain traceability of all tissues, and quires the medical director to approve all medi-
have a process for investigating and reporting cal and technical policies and prccedures.F'r''
adverse events." Either a centralized or a de- Establishing policies that address the provision
TABLE 28·4. Overview of Controls a Tissue Service Could Support

-
1. Tissue supplier (vendor) qualification
2. Tissue graft receipt and inspection
3. Maintenanceof graft storageand monitoring, including alarms
4. Promotion of tissue recipient informed consent
5. Assurancethat allograft tissue preparationsteps (ie, instructions for use) are followed
6. Maintenanceof traceability of tissue grafts from receiptthrough storage, issue (or reissue),and final
disposition, including timely tracking to specific tissue recipients
7. Prompt handling of allograft tissue recalls or market withdrawals
8. Recognition and reporting of adverseevents in tissue recipients, and participation in investigations
9. Compliancewith applicable regulations,statutes, and/or professional standards
10. Oversight of a tissue utilization and safety committee
11. Promotion of cost-effective tissue use
12. Documentation of the above controls
of quality services to support the satisfactionand suspended or withdrawn. FDA web postings
safety of patients is also helpful. should be reviewed for information related to
closures, recalls, or MedWatch reports. Inspec-
Tiss~(!S.IIPpli.er(Vendor) Qaaalifications tion reports can be requested from the FDA
and Certification through the Freedom of Information Act. Com-
plaints from transplanting surgeons concerning
In contrast to blood banks, tissue processors and the supplier's tissue should be reviewed, along
distributors often specializein particular types of with any report of infection that might have
products (eg, allografts). Therefore, a hospital been caused by a tissue allograft. Hospital man-
tissue service may acquire human tissue prod- agement may consider establishing a committee
ucts from more vendors than are commonly of internal stakeholders, including physicians
used by a transfusion service to obtain blood who implant tissue, to provide oversight of the
components. approval of tissue suppliers, as well as tissue uti-
Tissue suppliers should be selected based on
lization and safety monitoring.
their ability to reliably provide hfgh-qualtty tis-
sues that meet expectations for availability,safe-
Inspection of Incoming Tissue
ty, and effectiveness. The tissue service should
Allografts
establish the minimal criteria for qualifyingpro-
spective suppliers. According to The Joint Com- Before being placed into inventory, tissue al-
mission's standards, accredited health-care facili- lografts must be inspected upon receipt from a
ties must confirm annually that tissue suppliers tissue supplier to ensure that packaging remains
are registered with the FDA as tissue establish- intact and the label is complete, appears to be
ments and that they maintain a state license accurate, and is adequately affixed and legible.
when required." A written process for review The incoming inspection results should be re-
and approval of suppliers is expected and may corded along with the date, time, and name of
contain the elements listed in Table 28-5. A list the staffperson conducting the inspection.
of approved suppliers that includes documenta- The Joint Commission requires hospitals to
tion of each supplier's qualifications, certifica- verify package integrity and ensure that trans-
tions, and appropriate licenses or permits should port temperature was controlled and acceptable
be developed and maintained. Tissue services (if applicable).16 Inspecting the shipping contain-
should establish procedures to receive or moni- er for evidence of residual coolant (eg, wet ice
tor evidence of compliance or noncompliance, for refrigerated grafts or dry ice for frozen grafts)
such as reviewing FDAwarning letters, tissue al- may help determine that the required tissue-
lograftrecalls, and market withdrawals. Accredi- specific storage environment was maintained
tation by AATBand/or EBAAmaybe desirable. during transport.
Qualification information for each supplier Many tissue distributors use validated ship-
should be reviewed and approved annually by ping containers that are tested to maintain re-
the hospital tissue service. During. these required. temperatures for a specified period. If
views, the performance of suppliers in meeting such a container was used, the receiver of the
the transplantation facility's needs should be tissue needs to verify only that there is no dam-
evaluated. Each year, the tissue service should age tothe conteinerand tilat it was received
determine whether the supplier remains regis- and opened within the specified time frame
tered with the FDA. Whether AATB and/or posted on the outside of the shippingcontainer.
EBAAaccreditation is current can also be con- Tissues requiring "ambient teITlperaturf (de-
firmed; it is best practice to confirm the status of fined as the temperature of the immediate envi-
this accreditation by accessing the websites of ronment) for storage and shipping do not need
AATBand EBAAto perform a real-time search. to have the temperature verified upon receipt.
PDF copies or photocopies of accreditation cer- However, the temperature of tissues requiring
tificates might provide outdated information if "room-temperature" storage should be verified
the tissue establishment's accreditation has been and documented ifthe manufacturer has specified
TABLE 28-5. Suggested Vendor Qualification Criteria for Suppliers of Human Tissue Allografts
FDAregistration Perform an eHCTERSquery (https://www.access-

data.fda.gov/scripts/cber /CFAppsPub/tiss/i ndex.cfm)
for each tissue supplier and print the report
FDAinspection findings FDAForm 483 findings, if any
Warning letters and responses, if any
Voluntary accreditation, if available Proof of current AATBaccreditation (as applicable):
access www.aatb.org and perform a search for
accredited institutions for real-time information and
produce a printout dated the day of the search
Proof of current EBAAaccreditation for ocular
tissues: access http://restoresight.org/who-we-are/
find-an-eye-bank/ to perform a searchfor real-time
information
State license, permit, or registration, if required by Proof of current status (to be reviewed annually)
state laws
Reliable supply of tissue types Adequate notification of tissue shortages
Ability to meet special requests
Suitable expiration dates for tissue products
Transparency of the organization Willingness to provide information regarding donor
selection criteria and tissue treatment methods
Medical consultatlon Accessibility of the tissue supplier's medical director
Quality assurance/regulatory resources Accessibility of the tissue supplier's quality program
staff
New or trial tissue product support Willingness to provide information on newly released
tissue products
Professionalism of sales representatives Approval sought by representativesthrough desig-
nated channels before promoting or providing tissue
within the hospital
FDA = Food and Drug Administration; AATB = American Association of Tissue Banks; EBAA = Eye Bank Association of
America; eHCTERS= electronic Human Celland Tissue EstablishmentRegistration System.
a temperature storage range on the allograft la- Hospital tissue services should store tissue al-
bel or in the package insert. lografts according to the processor's instructions
on the allograft'slabel or package insert. Storage
Tissue Storage devices can include ambient- and/or room-
As with blood components, tissue grafts are temperature cabinets, refrigerators, mechanical
stored under various conditions. (See Table 28- freezers, and liquid-nitrogen storage units. Per
6.) The appropriate storage conditions depend The Joint Commission standards," continuous
on the nature of the tissue, method of preserva- temperature monitoring of refrigerators and
tion, and type ofpackaging. freezers is required. Room-temperature storage
TABLE28-6. Storage Conditions for Commonly Transplanted Human Tissue"
Musculoskeletal and Refrigerated Above freezing (0 C) to 10 C

osteoarticular
Frozen, cryopreserved (temporary -20 C or colder to -40 C
storage for 6 months or less) ,(this is warmer than -40 C but colder
'"" than ,.".20C)
Frozen, cryopreserved (long-term -40 C or colder
storage)
Lyophilized, dehydrated Arnblentt
Birth tissue Refrigerated, frozen, cryopreserved, As established and validated by the

lyophilized, dehydrated tissue bank
skin and adipose Refrigerated Above freezing (0 C) to 10 C
Frozen, cryopreserved -40 C or colder
Lyophilized, dehydrated Arnblentt
'Warmest target temperature unless a range is listed.

'Ambient temperature monitoring not required for lyophilizedtissue.
equipment must be monitored if the allograft's the tissue must be identifiable, along with dates
package insert specifies a storage temperature and times when the tissue was accepted, issued,
range. Storage equipment Iortissuesheldat re- and prepared. Records should provide a clear
frigerated or frozen ..temperatures must have history of all the actions performed. Documenta-
functlonalalarms, and theresI)01.M~eelIlergency tion of the tissue supplier, the unique numeric
backup capabilityin case of malfunctton or dam- or alphanumeric identifier(s) of the allograft;its
age to a storage unit. Storage SOPs should ad- expiration date, and the recipient's name must
dress steps to be taken in the case of excursions be maintained for all tissue grafts used. The
from allowabletemperature limits or in the event Joint Commission also requires that documenta-
of an equipment or power failure. Lyophilized tion of the tissue type and its unique identifier
tissues with package insert instructions specify- be placed into the recipient's medical record. 16
ing storage at ambient temperature or colder Tissue service records need to permit bidirec-
can tolerate a very broad temperature ..range, tional traceability of all tissues from the donor
and monitoring during storage is not required. and tissue supplier to the recipient(s) or otherfl-
nal disposition, including the discard of tissue.
Tissue Allograft Traceability and Records must be retained for 10 years, or longer
Record-Keeping
if required by state or federal law, after distribu-
tion, transplantation, discard, or expiration
Proper management of human allografts re- (whichever occurs last).
quires that the hospital tissue service document Tissue usage information cards or other sys-
all the steps taken in tissue handling as they oc- tems supplied with the allograft by the tissue
cur and maintain comprehensive records of bank must be completed and returned to the
these steps. According to The Joint Commission tissue source facility,although the identity ofthe
standards," staff members who have handled recipient does not need to be disclosed. This
information helps maintain the traceability of tions for potential recipients of other allografts
the allograft and expedites market withdrawals affected by the incident.
or recalls should they occur. The tissue supplier State health departments have lists of infec-
may also use this information to understand al- tious diseases that must be reported when they
lograft utilization, obtain positive or negative are newly diagnosed. For example, a new diag-
feedback, and meet customer needs and expec- nosis of HIVor viral hepatitis in a tissue allograft
tations. recipient where the allograft is suspected as a
possible source may need to be reported to the
Recognizing and Reporting Adverse relevant state department of health. An epidemi-
Events Possibly Caused by Allografts ologic investigation may be needed to establish
whether the tissue allograft was the source of
Human-derived medical products, such as tissue the recipient's infection.
allografts, carry some risks that must be bal-
anced with clinical benefits. Albeit rarely, hu- Recalls and Look-Back Investigations
man tissue allograftshave transmitted disease by
bacteria, viruses, and fungi, and one tissue type A tissue product recall or market withdrawal
(dura mater) transmitted a prion disease. In ad- can occur when a tissue allograft is determined
dition, an allograft may have a structural weak- by the tissue supplier to be compromised. The
ness that can lead to an unsuccessful outcome. supplier may sequester tissues in inventory not
(SeeMethod 7-3.) yet distributed, recall all tissues from a specific
Hospital tissue services are required to have donor or processing lot, and notify hospitals that
procedures to investigate in a timely fashion any received affected allografts. Depending on the
adverse outcome suspected to be caused by a nature of the recall, it may be prudent for the
tissue allograft. The Joint Commission requires hospital to quarantine allografts in inventory,
that allograft-transmitted infections and other identify recipients, and/or notify the transplant-
severe adverse events be reported immediately ing surgeon(s) of the notification. Surgeons
to the tissue supplier." should evaluate the circumstances and notify, if
Surgeons playa critical role in identifying al- appropriate, each recipient who received a tis-
lograft-associatedadverse outcomes and need to sue graft that is being recalled.
notify the hospital tissue service immediately Look-back investigation can be triggered
when they suspect such events. Prompt notifica- when a tissue donor is found, after donation, to
tion of adverse events enables the hospital tissue have been infected with HIV,human 'l-cell Iym-
service to investigate the cause, report the issue photropic virus type I or II, HBV,HCV, or other
to the tissue supplier, and institute corrective ac- infectious disease known to be transmitted by
tion, including sequestration of any other sus- tissue grafts. Look-backinvestigations involving
pect allografts. Tissue-associatedadverse events tissue grafts are uncommon.
may also be voluntarily reported directly to the
FDAvia MedWatch, but the reporter should be Tissue Autograft Collection, Storage,
aware that FDAregulation is focused on wheth- and Use
er an infection is suspected to have been caused Surgical reconstruction using the patient's own
by the tissue allograft.The investigation of infec- tissues has advantages and disadvantages when
tions and other adverse events requires coopera- compared to the use of a tissue allograft.Advan-
tion between the tissue service, clinicians, and tages include faster incorporation/healing and
tissue supplier. Inclusion of the serial number of relative safety from transmission of viral disease
the graft with the report is essential for further or immunologic rejection. Disadvantages in-
investigation by the tissue bank. Consultation clude morbidity associated with an additional
with the hospital infection-control department surgical procedure for the patient, including
or an infectious disease specialistmay be benefi- pain and potential surgical-siteinfection. In ad-
cial. Early notification can prevent complica- dition, the quality (eg, strength) and quantity of
autologous tissue may not be adequate for the gous tissue that is intended to be shipped to a
intended use, and removing the patient's tissue different facility for autologous use does not
may adversely affect function at the site from qualify for .the FDA exemption, except under
whlchitwas removed. limited circumstances in order to accommodate
The collection, storage, and use of autolo- the medical needs of an individual patient,
gous tissues are exempted from regulation by which commonly occurs with cranial flaps and
the FDA as long as the autograft is used in the (less commonly) parathyroid tissue that may be
same surgical procedure? The same-surgical- removed at one facility and implanted in the pa-
procedure rule allows for removal of the tissue tient at another facility.23
and subsequent use in the same patieI1twithin a Written procedures should address the col-
single operation or staged follow-up surgery. lection, microbial testing, packaging, storage,
Hence, the removal and implantation may be and issuance of tissue autografts for reimplanta-
several days apart and still be considered part of tion. Appropriate cultures may be obtained af-
the same surgical procedure. Examples include ter surgicalremoval and before packaging. Auto-
skin grafting, coronary artery bypass surgery us- grafts should not be collected from patients with
ing autologous vessels, cranioplasty, and para- systemic infections or if the tissue is in close
thyroidectomy with implantation. Minimal ma- proximity to an infected area. Autografts may be
nipulation allows exceptions for tissues that are stored at the medical facilitywhere they are col-
malntalned in their original form, which gener- lected or at an off-Site, FDA-registered tissue
ally includes steps such as rinsing, cleansing, siz- bank. Procedural recommendations have been
ing, and shaping. A facility that removes autolo- published by the MTB and AORN.19,20
8. Functions performed by tissue services that mimic transfusion services include vendor qualifi-
cation; ordering, receiving, storing, distributing, and tracing products; and investigating ad-
verse events, including complaints, recalls or withdrawals, and look-back investigations.
9. Additional controls that can be undertaken by a hospital tissue service include promoting tis-
sue recipient informed consent and oversight of a tissue utilization and safety committee.
REFERENCES
1. American Association of Tissue Banks. About 12. Cheek RF, Harmon Iv, Stowell CPoRed cell allo-
us. McLean, VA: MTB, 2018. [Available at immunization after a bone allograft.Transfusion
http://www.aatb.org/?q=about-us (accessed 1995;35:507-9.
July 30,2018).] 13. Eastland T, Warwick, RM. Diseases transmitted
2. Eastlund DT, Eisenbrey AB, for the Tissue Com- by transplantation of tissue and cell allografts.
mittee. Guidelines for managing tissue ailografts In: Warwick RM, Brubaker SA, eds. Tissue and
in hospitals. Bethesda, MD: AABB,2006. cell clinical use: An essential guide. West Sus-
3. Code of federal regulations. Title 21, CFR Parts sex, UK:Wiley-Blackwell,2012:72-113.
1270 and 1271. Washington, DC: US Govern- 14. SilveiraFP, Campos SV.The Zika epidemics and
ment Publishing Office, 2019 (revised annual- transplantation. J Heart Lung Transplant 2016;
ly). 35:560-3.
4. Warwick RM, Brubaker SA, eds. Tissue and cell 15. Food and Drug Administration. Guidance for in-
clinical use: An essential guide. West Sussex, dustry: Regulatory considerations for human
UK:Wiley-Blackwell,2012. cells, tissues, and cellular and tissue-based prod-
5. Malini TI. Preparation and banking of bone and ucts: Minimal manipulation and homologous
tendon allografts. In: Sherman OH, Minkoff J, use. (December 2017) Silver Spring, MD: CBER
eds. Arthroscopic surgery. Baltimore, MD: Wil- Office of Communication, Outreach, and Devel-
liams and Wilkins, 1990:65-86. opment, 2017. [Availableat https://www.fda.
6. Fehily D, Brubaker SA, KearneyIN, Wolfinbarg- gov/ downloads/biologicsbloodvaccines/ guid
er W, eds. Tissue and cell processing: An essen- ancecomplianceregulatorylnformation/ guidanc
tial guide. West Sussex, UK: Wiley-Blackwell, es/ cellularandgenetherapy /ucm585403. pdf.]
2012. 16. Transplant safety (TS). In: Comprehensive ac-
7. McCaughan JA, Tinckam KJ. Donor specific creditation manual for hospitals. Oakbrook Ter-
HLA antibodies and allograft injury: Mecha- race, IL:The Joint Commission, 2018:TS03.
nisms, methods of detection, manifestations and 17. Gammon R, ed. Standards for blood banks and
management. Transplant Int 2018;31: 1059-70. transfusion services. 32nd ed. Bethesda, MD:
8. Subramanian V, Ramachandran S, Klein C, et al. AABB,2020.
ABO-incompatible organ transplantation. Int J 18. Standards for laboratory accreditation. North-
Immunogenet 2012;39:282-90. field, IL: College of American Pathologists,
9. Sel S, Schlaf G, Schurat 0, Altermann WW. A 2017.
novel ELISA-basedcrossmatch procedure to de- 19. Association of periOperative Registered Nurses.
tect donor-specific anti-Hl.Aantibodies responsi- Standards of perioperative nursing. In: 2015
ble for corneal allograft rejections. J Immunol Guidelines for perioperative practice. Denver,
Methods 2012;381 :23-31. CO: AORN, 2015. [Available at http://aorn.
10. Mosconi G, Baraldi 0, Fantinati C, et al. Donor- org/ guidelines/ clinical-resources/ aorn -stan
specific anti-HLA antibodies after bone-graft dards (accessed December 10,2019).]
transplantation. Impact on a subsequent renal 20. Osborne JC, Norman KG, Maye T, et al, eds.
transplantation: A case report. Transplant Proc Standards for tissue banking. 14th ed. McLean,
2009;41: 1138-41. VA: American Association of Tissue Banks,
11. Duhamel P, Suberbielle C, Grimbert P, et al. 2016.
Anti-HLAsensitization in extensively burned pa- 21. Medical standards. Washington, DC: Eye Bank
tients: Extent, associated factors, and reduction Association ofAmerica, 2017.
in potential access to vascularized composite al- 22. Organ Procurement and Transplantation Net-
lotransplantation. Transpl Int 2015;28:582-93. work. Policy 16: Organ and vessel packaging,
C HAP T E R 28 TissueServicesin the Hospital 791
labeling, shipping, and storage. Rockville,MD: regarding the scope of the exception. (Novem-
Health Resources and ServicesAdministration, ber 2017) Silver Spring, MD: CBER Office of
2018. [Availableat https://optn.transplant.hrsa. Communication, Outreach, and Development,
gov/ govemance/poltcies/'] 2017. [Available at https://www.fda.gov/down
23. Food and Drug Administration. GUidancefor in- loads/biologicsbloodvaccines/ guidancecompli
dustry: Same surgical procedure exception un- anceregulatoryinformation/ guidances/tissue/
der 21 CFR 1271.lS(b): Questions and answers ucm419926.pdf·1
Index
Page references in italics Adsorption, 418-419

refer to figures or tables. allogeneic red cells, 437-438
autologous red cells, 437
A -elutlon, 420
Aianti-A, 298-299, 298, 305-306,389 protein, nonimmunologic, 447
peripheral core chain variants, 301 red cells
serologic reactions, 303 allogeneic, 437-438, 438
A,B/anti-A,B, 305, 306 autologous, 437
plasma, 572 selective, 707, 724-727, 726-727
ABCB6,380 serum, 438-439
ABCG2,380 Adverse events/reactions
ABH, 299, 300,301,311 donor, 106, 107, 108-110,143-144
ABOblood group systems, 297-327 hemovigilance, 100-103, 101-102
antibodies, 305-306 hospital reporting to blood suppliers, 116-123
blood component requirements, 507 MERS-TM,99 ,
carbohydrate epitopes, 297-327 monitoring, CT, 99
compatibility, 523-524, 768-772, 771, near-miss events, 26
769 nonconforming events, 5, 17-20, 19,24
development and aging, 299-300, 302 reporting, 103
discrepancies, 306-310, 308-309 severity grading tool for blood donor, 124-125
genetics/genotyping, 289, 302-303, 507 transfusion-related,98
grouping, 299 see also Transfusion reactions
HOFN,305 A4GALT, 321-322
HPAs, 461-462 Agglutination
HSCT, 768-772, 771, 769 antigen and antibody concentrations, 239
incompatibility, 297-298, 770, 772 mixed-field, 273
key points, 323-324 Agglutinin syndrome, cold (CAS),318, 442-443
matching, 557 alloantibodies, 443
phenotypes, 298 autoantibodies, 318
platelets, 564-565 cold-reactive, 443
solid-organdonors, 289 specificity,442
subgroups, 303-304 serologiccharacteristics, 442-443
titers, 300, 302 Agreements, 10
typing, 299, 303, 306 AHA. See Anemia, autoimmune hemolytic
see also individualblood groups AHG testing, 318
Accidents, reporting, 38-39 AHTR. See Transfusion reactions, hemolytic
Accreditation, 21-22, 85-86, 756-757 immune-mediated acute
organizations, 2 AIHA. See Hemolytic anemia, autoimmune
regulation vs, 77, 78 AIN. See Neutropenia, autoimmune
see also Regulationsand recommendations Air embolism, 648
ACE inhibitors, in therapeutic apheresis, 712 Alarm fatigue, 600
ACHE,373 Albumin-bound drugs, in therapeutic apheresis, 712-
Acid-citrate-dextrosesolution A (ACO-A),711 713
ACKRI,370 Aliquoting, 521-522
Acute lung injury, transfusion-related, 96, 100, 128, Alleles. See Genes/genetics
156,162,472,492,640-643 Allergicreactions, in therapeutic apheresis, 711-712
ADAMTS13,714 Alloantibodies,389, 437
Adenosine diphosphate (AOP),674 autoantibody vs, 285-286
793
development, 250 blood group systems and antigens

H,312 clinicalsignificance,356-358
ALT,174 key points, 384-385
Amniocytes, 287 donor-specific(DSAs),742
Amotosalen, 216 drug-dependent
Amplificationproducts autoantibody production, 447
detection of, 234-238 drug-treated red cells and, 446-447
DNA arrays, 235, 237 laboratory investigation, 448-449
next-generation sequencing, 237-238 untreated red cellsin the presence ofdrug, 447
PCR, real-time, 234-235 exclusion, 398-401, 399-400
Anemia, 598 HLA,490
anemia, hemolytic autoimmune (AlHA),283, identification, 420-421
285,318,434-445 Luminex-based,472
adsorption, 437-439 panel, 391-392, 393, 396-397
autoantibody specificity,439, 444 red cells, 391-392, 393
blood selection for transfusion, 439-441 Insigntticant,250
DAT-negative,440-441 interpretation, 509
mixed-type, 443-444 isotypes, 244-245
patient history, 432-434 monoclonal (MoAbs),285
RBCtransfusion, 559-560 multiple, 393, 402, 406-407, 421
serologic characteristics and problems, 435- receptors (FcyRs)in target clearance, 246
437, 435, 443-444 red cells, 391-392, 393
transfusion for patients, 444 serologicreactivity, 405-406
types, 434 unexpected, 507-508, 524
anemia, hemolytic immune, drug-induced Antibodies to red cell antigens, 389-428
(DIIHA),445-449, 452-455 additive solutions, 392, 394
classification,446-447 adults vs neonates, 390, 390
mechanisms, 445,445 antigens
protein adsorption, nonimmunologic, 447 alterations, 412
anemia, hemolytic warm (WAlHA),435-442 high-prevalence,410-413, 411-412
autoantibodies, DAT-negative,441-442 low-prevalence, 413
RBCtransfusion, 560 antigiobulin reagents, 394
transfusion of patients, 441 autoantibodies
classification,554 cold,409
microangiopathic hemolytic (MARA),714, 719 warm, 408-409, 410
preoperative diagnosisand treatment, 591 biologictherapies, 395
ANR See Hemodilution, acute normovolemic diagnosisand disease, 395
Ankylosingspondylitis,HLA-B27and, 496 dosage, 390
Antibodies, 229 drug-dependent, 413-414
aggiutination reactions, 239 exclusion, 398-401, 399-400
alloantibodies, 389, 437 identification, 391-402, 393
autoantibody vs, 285-286 complex, 402-414,403,404
development, 250 considerations following,421-424
H,312 immunization to, 389
autoantibodies, 389, 436-437, 439 incubation time, 416
AIHA,mixed type, 444 interpretation of results, 397-402
alloantibodyvs, 285-286 key points, 424-425
CAS,443 medications, 395
DAT-negative,441-442 patient history, 394-395
H,313 pregnancy, 413
HI, 313 procedures, 414-421
PCH, 444-445 reactivity
specificity,439, 440, 444 positive and negative, 397
binding, red cell destruction, 247 specificity,without, 407-408
Index 795
reagents, 391-394, 393 Antihemophilic factor (AHF),688

components, 414 cryoprecipitated,157-158
red cells pooled,160
exclusion and confirmation, 401 Antihuman globulin (AHG)reaction, 506
exposure, prior, 394-395 APACHE. See Evaluation
rouleaux formation, 414 Apheresis, 141
serologicreactions, 405-406 blood components, 149-150
anomalous, 414 cytapheresis, 721-722, 721
serum-to-cellratio, 415-416 devices, 150-153
specificityassessment, 398, 401 filtration-based,707
specimen requirements, 391 selective adsorption, 707
storage, 390-391 QC,31
temperature reduction, 415 granulocyte-monocyte (GMA),707
test methods, 392 key points, 165-166
transfusion reactions, delayed serologic/ medicine, 727
hemolytic, 409 pediatric, 713
zygosity,390 plasma, 150
Anticoagulation, in therapeutic apheresis, platelet, 149
711 red cell, 149
Antifibrinolyticdrugs, 595-596, 606 therapeutic (TA),705-735
Antigens, 280 adverse effects, 711-713
absence, 487 anticoagulation, 711
agglutination reactions, 239 cardiorespiratory status, 708-709
antithetical, 265, 281-282 classification,713
blood groups, 382-383 device modalities, 706-707
without, 342-384 documentation, 727-729
Class Ill! MHC evaluation and treatment, 708
characteristics, 479-481, 480 granulocyte-monocyte, 707
configuration, 481, 481 hematologic status, 709
density, 265 indications, 713-714, 719-727
dose, 265 key points, 729-730
expressed, 265 medications, 709
high-prevalence (901 series), 383, 383 neurologic status, 708
low-prevalence (700 series), 383 patient evaluation and management, 707-
negative 711
donors, molecular testing, 288 pediatric, 713
phenotypes, 276 principles, 705-706
phenotype prediction, molecular methods, 282- renal and metabolic status, 709
283,288 replacement fluids, 709
public, 482 transfusion/apheresis history, 708
red cells, 390-391,390 vascular access, 709-711
Antigen capture assays (ACAs),468 Arboviruses, screening, 208-209
Antiglobulin testing ARDS. See Respiratorydistress syndrome, acute
direct (OAT),285, 390, 395-396, 430-435 Arterial puncture, 143
elution, 433-434, 434 Assays
evaluation, 432 ACA,468
key points, 449 antibodies to red cell antigens, 240, 240
patient history, 432-434 ChL!As, 181
positive, 430 colony-formation,753
principles, 431-432 E!As, 181
serologicinvestigation, 433 ELISA,240-242, 241, 467, 490
false-negativeresults, 534 NAT, 209, 212
false-positiveresults, 533 phenotyplng red cells, 239, 240
indirect (!AT),285 platelets, 240, 467-468
796 A A B B TE C H N I CAL MAN U A L
protein, 238-242 bags, 145

solid-phase,239-242, 490 bloodless care, 605
Assessments, external, 21-22 collection
ATAM,381 equipment, OC, 31
ATML,381 systems, 158-159
ATRs. See Transfusionreactions, allergic delivery, 606
Auditing devices, 82-83
blood utilization irradiators, OC, 30
criteria, 615-617 loss
guidelines, 614 intraoperative, 594-595
process, 614-615 g phlebotomy, 597,598
review items, 615 ordering, surgicalpractices, 528-529
review types, 617-620,615 order schedule, maximum surgical (MSBOS),
provider feedback, 600, 603 528,591-592, 592, 605
see also Monitoring rare needed, 423-424
AUG (Augustine)system (036), 381 recovery
Aurora PlasmapheresisSystem, 151-152 intraoperative autologous, 594
Autoimmune hemolytic anemia. See Anemia, postoperative, 596-597
hemolytic autoimmune selection for transfusion, 439-441
Autosomes, 256 spill cleanup, 47
supply, zero risk, 1-2
B volume, neonates, 675
B/anti-B, 305-306, 389 warmers, 540-541
antigen, 298-299, 298 OC,30
phenotype, acquired, 305 wrong blood in tube (WBIT),539
serologicreactions, 303 Bloodbanks, inventory, 527-528, 623
B(A)phenotype, 304-305 Bloodcomponents, 141-172
Babesia, 176, 179, 188,210-212 ABO, 768-772
donor testing regulations and standards, 193 administration, 537-551
transfusion-transmitted (TTB),211 collection, 149-153
Bacteria, screening, 193, 204-205 mobile Sites,35
BECs. See Computer systems, blood establishment defined, 78
Benchmarking, 621-623 discard, 4
PBM metrics, 622 dispensing, 537-543
Best practice advisory (BPA), 600 equipment, 540-542
Bg,383-384 documentation, 4, 547-548
Biohazardwaste and disposal,46-48 dosing, 680
Biologics frozen, shipping,522-523
license application (BLA),81 HSCT,772-773
product deviation (BPD),19-20,84 infectious disease screening, 173
safety cabinets (BSCs),34, 42, 43-44 irradiation, 161-162, 651
Blovigilance, 26, 99 issuing, 523-527
elements of, 99-100 key points, 165-166, 549-550
reporting, 99-100 labeling, 164-165
task force on, 99 leukocyte-reduced, 159
US, 112 manufacturers' licensure, 81-82
Bleeding non-ABO,771
donors, 134 nonemergent settings, 546
plasma transfusions, 567-568 pooled, 159-160
platelet transfusions, 563-564 preparation, 146-148, 147
in therapeutic apheresis, 712 processing and modification, 158-164
WHO scale, 562 production, automated, 148-149
Blood quarantine, 4, 164
administration, 543-547 RBCratio, 606
Index 797
return and reissue, 529 Carbohydrates

storage, 153-158,509-518,510-516 blood group systems, 297-327
expiration, 510-516 epitopes, 297
irradiation, 519-520 Cardiac. See Heart conditions
monitoring, 509-518, 510-516 CAS. See Agglutininsyndrome, cold
temperature, 509, 517 Catheter, central venous (CVC),710
testing, 523 Cautery, 595
thawing devices, QC, 30 CCls. See Corrected count increments
transportation and dispensing, 509, 510-516, CD11a1b, antigens, 471
542-543 CD34+, 746, 747
transfusion settings, unique, 548-549 HPC, 750-751
volume reduced, 162 CD38, 285, 395
washing, 162 monoclonal antibodies, RBCtransfusion, 560-561
Blooddonors. See Donors CD47,285
Bloodgroup systems, 255, 259-262, 279, 280 monoclonal antibodies, RBCtransfusion, 560-561
antigens, 355-385 CD59 system (035), 381
databases, 283 CD109, HPAs,461
naming, 280 CDI77,470-471
function, 359 Cells
history, 255 assays,490
SNV,279 counters/hemoglobinometers, QC, 31
terminology, 280-281, 281 division, 257-258
See also Genetics; particular system donor testing, 195
Bloodpressure enzyme-treated, 416-417
cuffs, QC, 31 expansion, HPC, 751
donors, 130 processing methods, 750-751
Bloodutilization, 20-21 selection systems, HPC, 750-751
auditing, 613-626 separation methods, 283
big data, 621-623 somatic, 256
CPOE, 620-621 washing/washers, 29, 520-521, 695
criteria, 615-617 see also particular cell type
high-riskpatients, 620 Cellular therapies (CT), 99
key points, 623-624 accreditation, 89
monitoring, 623 FDA,78-84
PBM,600, 604 key points, 90
process, 614-615 regulations, 77-93
review items, 615 Centrifuges/centrifugation, 146
review types, 617-620 continuous-flow, 706-707
Bodysize, neonates, 675 intermittent-flow, 707
Bombay (Oh) phenotype, 310, 312 rnicrohematocrit, 31
Bone QC, 29. 31
morphogenetic proteins (BMPs),780 Centromere, 256, 257
replacement, 780 Certification, PBM, 585, 606
Bruise, 143 Cesium irradiators, 56
BSCs. See Safety,biologicalsafetycabinets CFR. See Code of Federal Regulations
Buffy-coatpreparation, 147, 148 CH/RG (Chido/Rodgers) system (017), 377, 417-
418
C Change control, 26
C3,246-247 see also Patient blood management
C3b,250 Chemicals
C/c antigens, 344, 346, 346 fume hoods, 34, 51
Calcium, 606 hazardous, 39, 48-49, list, 67-68
Calibration, 26 hygiene plan (CHP),48, 50
Cancer, donors, 133-134 modification, 417
signage, 50 disseminated intravascular (DIC),AHTR,636

spills optimizing, 592
categorization, 52 Code of Federal Regulations (CFR),2, 78, 79, 80
managing, 52, 74-75 HCTlPs,87
work safety, 69-70 HPC, 740
Chemical safety, 48-53 HSCT, 743, 756
emergency response plan, 52 quality-related references, 28
engineering controls and PPE, 51 RTTIs,127·128
hazard identification and communication, tissue grafts, 783
50 Cold agglutinin syndrome. See Agglutinin syndrome,
labeling, 50 cold
training, 49-50 Cold stress, neonates, 676
waste disposal, 52-53 College of American Pathologists (CAP),2
work practices, 51-52 transfusion medicine checklist TRM, 86, 87
Chikungunya virus (CHIKV),208, 209 Colony-formationassay. See Assays
Chimerism, 273-274, 289 Compatibilitytesting, neonates, 677·678
TA-GVHDand, 492-493, 493 Competency assessments, 7·8
twin, 273-274 Complement
Chlamydia trschomstts, 196 activation outcomes, 247, 635
ChLIAs.See Assays,chemiluminescent AHTR,635, 636
immunoassays cascade, 636
Chlorine, spillresponse, 71 target cell opsontzation and destruction, 246·
Choosing Wisely campaign, 587, 598-599, 599 247
Chromatids, 257, 257 Complications. See Transfusion reactions
Chromosomes, 256, 257, 258 Computer systems
6,486 blood establishment (BECS),111
homologous, 256 provider order entry (CPOE)system, 599·600,
Circular of Information for the Use of Cellular 601,620·621
Therapy Products, 87, 165 validation, 13
Circulatory overload, transfusion-associated(TACO), Confidentiality,records, 16
100, 643-645 Connection devices, sterile, 30
cis, 274 Consanguineous mating, 268
cisABphenotype, 305 Consent, recipient, 537·538
Citrate Contracts, 10
reactions, 144 Controls
toxicity, 646 chart, 26
Citrate-phosphate-dextrose-adenine (CPDA)-l, in-process, 4
neonates, 676 Copper sulfate, OC, 31
Class IIII Coronary syndromes. See Heart conditions
antigens, 566 Corrected count increments for platelets, 565, 566
molecules, role, 483-484 Corrosive compounds, work safety, 69
Clinical decision support (CDS),599-600 COST collection, 382
system (CDSS),621 CPOE. See Computer systems, provider order entry
Clinical LaboratoryImprovement Amendments of Creutzfeldt·Jakobdisease (CJD),213, 214
1988 (CLIA),2,12,84,85 CROM (Cromer) system (021), 378
Clinics Crossing over/crossovers, 258, 270, 271·272, 272,
consultation, HLA,497-498 487
massive transfusion, 606 Crossmatching, 490
outcome data, PBM, 604, 604 antiglobulin, 509
utility,737-742 compatibilitytesting, 677-678
Clocks, OC, 30 computer/electronic, 508
CO (Colton) system (015), 375, 376 HLA,490
Coagulation immediate·spin (IS),508
abnormalities, AHTR, 635 interpretation, 509
Index 799
Cryoprecipitate Del types, 340

AHF,688 Dengue virus (DENV),208-209
HSCT, 773 Deoxyribonucleic acid. See DNA
neonatal and pediatric, 688 Devices, 706-707
shipping, 522 adverse effects, 20, 82-83, 711-713, 711
storage, 518 anticoagulation, 711
thawing, 518-519 classification,82
transfusion, 569-570 modalities, 706-707, 706
see also Antihemophilic factor, cryoprecipitated patient evaluation and management, 707-711
Cryopreservation, 751-752 pediatric, 713
RBCcomponents, 160-161 pressure, 541
CTL2 system (039), 342 sterile connection, 30
antigens on, 471 vascular access, 709-711
Customer feedback, 4 ventricular assist (VADs),693
focus, 5, 6, 23 see also Equipment
CVC. See Catheter, central venous, 710 DHQ. See Donor History Questionnaire
Cytapheresis, 721-722, 721 DI (Diego)system (010),372-373
Cytogenetics, 256 antigens, 373
Cytomegalovirus (CMV)testing, 196 band 3, 372, 372
immunocompromised recipients, 194-195 Dia/b(DI1/2)/anti-Dia/b, 372-373
prevention, pediatric, 694 Diagnosis-relatedgroups, 605
Cytoskeleton, 360 DIC. See Coagulation, disseminated intravascular
DIIHA. See Anemia, hemolytic immune, drug-
D induced
D/anti-D, 329, 337-342 Dimethylsulfoxide (DMSO), 748, 749, 751-752
assay, 290-291 Diploid (2N), 257
clinical considerations, 343 2,3-Diphosphoglycerate (2,3-DPG),neonates, 674,
elevated,340 677
HDFN and, 663 Disaster plan, 23
molecular testing, 288 Disinfectants, 45
partial, 339-340 Distribution, 4, 522-523
phenotypes see also Blood components, dispensing
fetal,286 Dithiothreitol (DTT), 244
negative, 340, 342 DMAIC process, 22, 23
positive, 337-340 DNA, 256, 257
pregnant women, 287 analysis, 229-238
reagent reactivity, 341 assays, alloantibody vs autoantibody, 285-286
structure, 338 chemistry and structure, 230-231, 230
testing for, 342-343 nucleotides in, 230-231
typing DO (Dombrock) system (014), 374-375
discrepancies, 343 Documentation/documents and records, 4, 5, 14-16,
donors, 342 24
patients, 342-343 creation, 14
weak, 338-339 donors, 128
DAT. See Antiglobulin test, direct electronic health records (EHR;HER), 111, 621
Data malntenance, 16
benchmark, 604-605 QC,13-14
big data, 621-623 therapeutic apheresis, 727, 728, 729
blood utilization, 614 see also Reporting!reports
collection, 600-605 Donath-Landsteiner test, 323
Databases, blood group antigen, 283 Donors/ donations
Decontamination, 45 acknowledgment, 129-130
Deemed status partners, 85 allogeneic
Deferral list, 128 criteria, 128
identification, 128-129 screening/testing, 128-129, 133,217

key points, 137 eligibility, 127-128
registration, 128-129 identification, 128-129
selection, 127-140 international variations, 195-196
autologous, 136-137 registration, 128-129
contraindications, 137 regulations and standards, 190-193
control, 395-396 selection, 742
donations, 136-137 tests, 173, 174-176,177-196,178-180,179
infectious disease testing, 195 siblings, 743
key points, 137 unit selection
selection, 127-140 antigen-negative blood, 422-423, 424
blood diseases, 134 crossmatch for compatibility, 423
blood donor room, biosafety, 46 phenotype-matched blood, 423
cancer, 133-134 Donor History Questionnaire (DHQ), 128, 132
compatibility, 491 abbreviated (aDHQ), 132-133
consent, 129-130 FDAand, 132
directed, 136 frequent donors, 132-133
deferral, 128, 129, 133 Dosage, 390
accommodations, 129-130 Drug-dependent antibodies. See Antibodies, drug-
directed, 136 dependent
educational material, 129-130, 133 DSAs. See Antibodies, donor-specific
qualification screening, 130-131
temporary, 183 E
education, 129-130,133,217 ECMO. See Oxygenation, extracorporeal membrane
eligibility,79, 130-131, 133-135, 189 E/e antigens, 344, 346
bleeding conditions and blood diseases, 134 EIR. See Reports, establishment inspection
cancer, 133-134 Electrical safety, 40-41
heart and lung conditions, 134-135 emergency response plan, 41
medications, 135 engineering controls and PPE, 40
not eligiblefor reentry, 184, 187 hazard ID and communication, 40
pregnancy, 135 training, 40
reentry, 183-187 work practices, 40
exclusion, HN, 177 see also Safety;Work environment
hemovigilance, 99, 103-106, 105 Electronic health records. See Documentation
history evaluations, 177 ELISA.See Enzyme-linkedimmunosorbent assay
identification number (DIN), 142 Elution, 419-420, 433-434, 434
preoperative autologous (PAD),592, 594 Elutriation, HPC, 750
preparation and care, 141-153 Emergencies
adverse reactions, 143-144 equipment, 541-542
consent, 141-142 exlts, 39
eligibility, 142 showers, 34, 51
fatalities, 144 transfusion, 525-526
identification, 142 Emergency response plans, 38, 40
postphlebotomy care, 142-143 biosafety, 46
vein selection and disinfection, 142 chemical,52
prior, 189, 194 electrical, 41
recipient radiation, 55
outcomes, 556 see also Safety
-specific, 135-137 Employee health services, 38
regulations, 128 Encephalomyelitis, acute disseminated, IPE, 720
repeat, 197 End-product testing and inspection, 4, 26
samples Engineering controls, 37
retention and storage, 523 biosafety, 42
RBCunit selection, 523-525 chemical safety, 51
Index 801
electrical safety, 40 FcyRIlIb,antigens, 469-470

fire prevention, 39 FDA See Food and Drug Administration
guidance, 63-65 FFP. See Plasma, fresh frozen
radiation, 55 Fibrinogen, 569, 570
see also Safety;Work environment ficin, 416
ENTl. See Nucleoside transporter 1, equilibrative Filters
Enterocolitis, necrotizing, 682 leukocyte reduction, 159
Enzyme immunosorbent assays(EIAs),181 microaggregate, 543
Enzyme-linkedimmunosorbent assay (ELISA),240- pediatric, 689
242, 241, 467 standard administration, 543-544
donor screening, 490 Fire prevention, 39-41
sandwich ELISA,241-242, 241 alarm systems, 39
technical problems, 242 emergency response plan, 40
Equipment engineering controls, 39
identification, 9 extinguishers, 39
malntenance and control, 4 hazard identification and communication,
management, 5, 8-9, 23 39
selection, 8-9 PPE,39
see also Devices training, 49
Ergonomics,35 work practices, 39-40
Erythrocyte membrane-associated protein (ERMAP), see also Safety
374 Flaviviruses,208
Erythroid phenotypes, 384 Floseal,595
Erythropoiesis-stimulatingagents (ESAs),neonates, Flow cytometry, 243
675-676 Fluorescent in-situ hybridization (FISH),257
Erythropoietin (EPO)physiologyand therapy, FNAIT,566, 567
neonates, 675-676 FNHTR. See Transfusion reactions, febrile
Ethnicity, 398, 401 nonhemolytic
Euchromatic regions, 256 Food additives, indirect, 164-165
Evaluation, 5, 554-555, 708 Food and Drug Administration (FDA),1, 2, 173
see also Monitoring; Testing Amendments Act (FDAAA),89
Events DHO, 132, 133
near-miss, 26 donor eligibility, 128, 189-194, 190-193
nonconforming, 5, 17-20, 19,24 fatalityreporting, 652
see also Adverse events guidance, 79
Expiration, components, 510-516 inspections, 83-84
Eyewashes, guidance, 65 labels, 164-165
Eyewash stations, 34 oversight, 78-84
regulations and recommendations, 78-79
F see also Code of Federal Regulations
Facilities,5, 22-23, 24, 33-35 Food, Drug, and Cosmetic (FD&C)Act, 78-79
design and workflow, 33-34 Formaidehyde, spillresponse, 72
exits, 39-40 Forms, IS
key points, 57 FORSblood group system, 323
restricted areas, 34-35 Freezers, OC, 29
visitors, 35 Frequency, alleles and genes, 275-278
see also Safety;Work environment Furosemide, AHTR,636
Failure mode effects analysis (FMEA),18 FUT, 310, 312, 313, 316
Fatalities, 19 Fy3/5/6,369
blood donation, 144 F~ (FY1I2), 368-369
FDAcontact information, 652 FY(Duffy)system (008), 279, 291,368-370
reporting, 38-39, 82, 97, 99, 652-653 clinical significanceof antibodies, 369-370
in therapeutic apheresis, 713 glycoprotein, 369
FcyR,246,250,251 phenotypes and genotypes, 369
G platforms, 237
red cell antigens, 237
G antigen, 343-344
GE (Gerbich)system (020), 377-378
bands, 256-257
Gametes, 256, 258, 264 antigens, 377
Gases phenotype prevalence, 377
compressed, work safety, 69 GIL (Gill)system (029), 380
cryogenic, spillresponse, 71 GLOBblood group system, 318-323
flammable, spillresponse, 72 antibodies to, 322
GCNT2, 316, 317 biochemistry, 319, 321
Genes/genetics, 255, 256 blood-group-carryingmolecules, 321
alleles, 256, 265, 282 genetics, 321-322
frequency, 276-278, 278 phenotypes, 319, 319
silenced, 268 prevalence, 319
arms, 256, 257 transfusion practice, 322-323
blood group systems, 255-296 Glomerulosclerosis,focal segmental (FSGS),TPE,
chimerism, 273-274 720
inheritance, 266-271 Glutaraldehyde, spillresponse, 72
key points, 292-293 Glycerol,cryopreservation and, 160, 161
modifiers, 274-275 Glycosphingolipids,synthetic pathways, 320
population genetics, 275-278 blood-group-carryingmolecules, 321
relationship testing, 278-279 Glycosyltransferase,297, 298
structural variation, 271-273 GMA. See Apheresis, granulocyte-monocyte
terminology, 280-281 Good manufacturing practice, current (cGMP),2
variation, 262-266 GPA,359
frequency, 276-278 GPB,359
incidence, 275 GPIalIIa, HPAs,460-461
inheritance, 266-271 GPIblV/IX, HPAs,460
mapping, 256-257, 258, 279 GPIIb/III, HPAs,459-460
markers, 278 GPN/CD36, HPAs,461,462
modifiers, 274-275 GPVI,HPAs,462
nomenclature, 280 Grafts, HSC, source, 742-743
nonsense coding, 264 Graft-vs-hostdisease, transfusion-associated
nonsynonymous coding regions, 264 (TA-GVHD),649-650,651
phenotype, 276 chimerism and, 492-493, 493
population, 275-278 pediatric, 694-695
position effects, 274 prevention, 161
prevalence, 275
Granulocytes
products, sequential interaction, 275
agglutinationtest, 472
recombinants, 271, 272
antigens and antibodies, 457, 469-478
regulations, 256-262
HSCT,773
silenced, 290, 291
synonymous, gene-codingregions, 264 immunofluorescence test, 472
synteny, 271 key points, 473
therapy, 89 neonatal and pediatric, 688-689
transcription factor, erythroid phenotypes, 384 shipping,522
unlinked,274-275 storage, 513, 518
variation, 262-266 testing for, 472-473
see also Polymorphisms therapeutic, 571
Genome/genomics, 255, 281-292 transfusion, 570-571
blood group systems, 281-292 Guidance, 63-65
human, 230-231, 279, 280 FDA,79
organization, 256-262 manual,83
Genotyping, 264, 265, 281-282 PBM,606
discrepancies, 289-291,290,291 GYPA/B, 359, 361, 362-363
Index 803
H allogeneic, 738-739,740,767
H systems, 310-313 autologous, 738-739,739-740,745,746,767
al1oantibodies,312 cell-processingmethods, 750-751
antibodies to, 312-313 clinical utility, 737-742
antigen, 298-299 collection, 743-748, 758-766, 767
biochemistry and genetics, 310 marrow, 744-745
genes, 302 peripheral blood, 745-747
null phenotypes, 310, 312 UCB,747-748
transfusion practice, 313 cryopreserved, 743, 751-752, 755
H(FUTl IFUT2j genes, 302 graft source, 742-743
Hand-washing, 35, 65 histocompatibility, 741-742
Haploid, 257, 264 indications, 738-739
Haplotype, 256, 273 infectious disease agents, 740
HardyWeinberg equilibrium, 276-278 infusion-related adverse events, 755
Hazards key points, 757-758
areas, 34~35 manufacturers' regulations, 88
biosafety, 42 matching, 767, 768
categories, 49, 56 mobilization, 739
chemicals, list, 67-68 patient care, 754-756
communication program, chemical, 50 processing, 748-750, 758-766
electrical safety, 40 OC,752-753
fire prevention, 39 regulatory considerations, 756-757
hazards, 49 shipping and transportation, 754
health and physical, 49 storage, 749
identification and communication, 36, 42 thawing, 749-750
shipping, 56 washing, 749
storage, 45 Hematopoietic stem cell transplantation (HSCT),
waste-reduction program, 56 493-494,737-740,757-766
HBc, 174, 183 ABOcompatibilityfor blood component, 768-772
donor testing regulations and standards, 191 blood component support, 772-773
HBsAg.See Hepatitis B surface antigen donor selection, 740, 742
donor testing, regulations, and standards, 191 key points, 774
screening for, 200 non-Hl.A,490
HBV. See Hepatitis B virus non-oncologicpatients, 772
HCTIPs. See Human cells, tissues, and cellular and pediatrics, 773-774
tissue-basedproducts platelets, 772-773
HCV. See Hepatitis C virus RBCs,772
HDFN. See Hemolytic disease of the fetus and recipients, 767-775
newborn Hemizygous, 265
Health history, DHO, 132 Hemochromatosis, 145
Heart conditions, 134-135 Hemodilution
cardiac surgery, 569-570 acute normovolemic (ANH),595
cardiovascular disease, donors, 134 isovolemicred cell exchange, 722
coronary syndromes, acute, 555-556 Hemoglobin
Heating blocks, OC, 30 donors, 130-131, 154-155
Heavy chalns, 243-244, 244 reports, 603
Helgeson phenotype, 378 thresholds, 597-598, 602
HELLPsyndrome (hemolysis,elevated liver enzymes, Hemoglobinuria, paroxysmal cold (PCH),434-435,
low platelet count), 719 444-445
Hematocrit, 130, 131,154 Hemolysis
Hematoma, 143 autoimmune, 434-445
Hematopoietic growth factors, myeloid, 745-746 classification,435
Hematopoietic progenitor cells (HPCs), 737, 767 drug-induced, 445-449
accreditation considerations, 756-757 extravascular, 247-248
hyperhemolysis, antibody-negative DHTRsand, C virus (HCV), 174, 175, 179, 185, 196
249 donor testing, 196
immune-mediated, 429-455 exposure, 38
instrument-related, 713 regulations and standards, 191
intravascular, 248 screening for, 201
nonimmune-mediated, 637 transmission, 201, 217
types, 434-435, 435 E virus (HEV),202-203
Hemolytic disease of the fetus and newborn (HDFN), non-A, non-B (NANB),173
286,329,355,389-390,434,659-665 posttransfusion (PTH), 173
ABO-associated,305 Heredity,255
antt-D, 329 Heterochromatic regions, 256
DAT,434 Heterozygous, 265
diagnosis and monitoring, 660-661 Hil, 361, 363
maternal allolrnmunization, 660 Histocompatibility, 741-742
molecular testing, 286-287 determinants, non-HLA,489-490
pathophysiology, 659-660 HIT_See Thrombocytopenia, heparin-induced
prevention, 662-665 HN/ AIDS, 1, 175, 177, 179, 185, 196
Rh testing, 348 donor testing, 196
thrombocytopenia, pregnancy-related, 665-667 regulations and standards, 190
treatment, 661-665 screening, 173, 174, 177, 199-200
Hemorrhage, 596, 620 transmission, I99
Hemostasis HLA-B27,ankylosing spondylitis and, 496
abnormalities, 647-648 HLAsystem, 479-501
laboratory tests, 674 alleles
Hemotherapy, decisions and outcomes, 553-581 Class Iregion, 484
cryoprecipitate, 569-570 hypersensitivity reactions, 496
granulocyte, 570-571 nomenclature, 482-483, 483
key points, 573 antibodies
massive transfusion, protocols, 571-573 detection, 490
plasma, 567-569 development, 491
platelets, 561-567 antigens
RBCs,553-561 Class Iregion, 484
Hemovigilance, 95-126, 627 nomenclature, 481-482
adverse events, 100 public, 482
donor, 99,103-106, 105 red cells, 383-384
goal,97 biochemistry, tissue distribution, and structure,
international, 96-97 479-484
key points, 112-113 clinical aspects, 495-498
modules, 100 complex, 485
next steps, 106-112 crossmatching and antibody detection, 490
recipients, 100-103 disease association, 495, 496
reporting, 100 exposure, 38
standards, 106 function, 483-484
US, 97-100,105 future, 498
Hepatitis HPAs, 462
Bvirus (HBV),173, 175, 179, 182, 183-184, immune responses, 498
190,191,196-201 key points, 499
donor testing, 196 matched unrelated donor (MUD), 740
exposure, 38 MHC genetics, 484-488
regulations and standards, 190, 191 non-HLAhistocompatibility determinants, 489-
transmission, 200, 217 490
B surface antigen (HBsAg), 183 platelet transfusion, 565-567
prophylaxis, 38 regulatory aspects, 498
screening, 173, 174 role, 498
Index 805
splits and cross-reactive groups, 482 denaturation, 418

testing, 493-495 19A,245, 245, 246
transfusion and, 491-493 IgD,245
transplantation, 493-495 IgE, 245, 245
typing, 488-489, 741 IgG, 244-245, 245, 249, 303, 436
HNA. See Neutrophil alioantigen, human IgM, 244, 245, 246, 248, 305-306, 436
Homozygosity,265, 290 intravenous (WIG), 395
Hospitalregulations, 85-86 papaln and, 244
Housekeeping, 34 Rh,329,395
HPA. See Platelets, alloantigens, human Immunohematology reference laboratories (IRLs),.424
HPC. See Hematopoietic progenitor cells Immunology, 244-251
HSCT. See Hematopoietic stem cell transplantation key points, 251
HTLV. See Human T-celllymphotropic virus red cells and, 250-251
HTRs. See Transfusion reactions, hemolytic transfusion medicine, 229-254
Human cells, tissues, and cellular and tissue-based Immunomodulatory therapy, 429-430
products (HCT/PS), 82, 86-89 Immunosuppression, adjunctive (AMR),TPE, 720
donors, 86, 86, 195, 196 Immunotherapy, 355, 356-359
manufacture, 87 agents, 395
Human genome, 230-231 IN system (023), 379
Human Resources, 5, 7-8, 23 Incubation time, 416
aids, 15 Infants. See Pediatrics
competency assessments, 7-8 Infections/infectious diseases, 606
hiring, 7 emerglng, 111
jobs descriptions, 7 markers, 197
see also Training; Work environment rates, 198
Human T-celllymphotropic virus, Types I and II relevant transfusion-transmitted (RTTIs),127-
(HTLV-IiII),174,175,179,187, 196 128, 129
donor testing, 196 residual risks of transfusion, 196-199
regulations and standards, 192 agents for which blood is tested, 196-198
screening, 201-202 agents with no donor screening tests, 198-199
Hyperhemolysis,antibody-negative DHTRsand, 249 unknown risks, 196
Hyperkalemia, 646-647 sepsis, transfusion-related, 637-638
Hypersensitivityreactions, HLAalleles, 496 threat, 111
Hypokalemia, 646-647 transfusion-transrnitted, estimated risks, 198
Hypotension vector-borne, 205-212
intraoperative blood loss, 595 Infectious waste, 46-47
in therapeutic apheresis, 712 disposal, 48
Hypothermia, 648 Infectious disease screening, 173-227, 174-176
neonates, 676 donors, 177-199
HyTRs. See Transfusion reactions, hypotensive key points, 217-218
overview, 173-177
I pathogen inactivation technology, 215-217
I system (027), 315-318 residual risks of transfusion, 196-199
antibodies to, 317-318, 317 screening, 199-217
genetics, 317 transmission risks, 217
transfusion practice, 318 Information management, 5,17,24
lAT. See Antiglobulintest, indirect see also Computer systems; Data
Ii blood group collection, 315-318 Information systems, laboratory (LIS),617
Immune effector cells (IECs),89 Informed consent, HSCT, 743-744
Immunization, alloimmunization, 389 Infusions
Immunocompromised recipients, CMV testing, 194- pumps, 541
195 rapid,548
Immunoglobulins sets, 542-543
AHTR,635 syringes, 541
Inheritance L
autosomal Labeling, 15
codominant, 266-267 blood components, 164-165
dominant, 266, 267 HSCs, 754
recessive, 267-268 RBCs,517
dominant, sex-linked, 269-270, 269 whole blood, 145
genetic traits, 266-270 Laboratories, 34
patterns of, 486-488, 486 biosafety, 46
recessive, 270, 271 design, 39
sex-linked, 268-270 medical, legalities, 84-85
Injuries, 583-584 see also Work environment
reporting, 38-39, 82 Laboratory coats, guidance, 63
Inspections LANsystem (033), 380·381
blood component units, 522 Landsteiner, Karl, 255, 297
incoming supplies, 10 Laser desorption/ionization tfme-of-flight,matrix-
shipping, 522-523 assisted (MALDHOF), 282
INTERCEPT,163, 199,216,567 Latex allergy, 39
Intravenous access, 542 LE (Lewis)system, 311, 313-315, 417
Intravenous solutions, compatible, 544 biochemistry and synthesis, 314·315
Inventory management, 527-529 phenotypes, 314-315, 314
Iron transfusion practice, 315
overload, 652 Lean Six Sigma, 22
supplementation, 131 Leukemia, myelogenous, cytapheresis, 721-722
Irradiation. See Radiation Leukocytapheresis, 743
Ischemia, decreased risk, 605 Leukocytes
ITP. See Thrombocytopenia, immune filtration, 159, 543·544
IVIG. See Immunoglobulins, intravenous pediatric, 694
prestorage, 158·159
J reduction, 159,520
Light chains, 243-244, 244
Jehovah's Witnesses, 584
Linkage, 270-272, 273
JK (Kidd)system (009), 291, 370-372
disequilibrium, 273,487-488
antibodies to, 371
Lipoprotein,low-density (LDL),727
glycoprotein, 370 Liquids,flammable, spillresponse, 72
phenotypes, 371
LISS,415
urea transport, 371-372 Liverfailure, acute, TPE, 720
JMH (John Milton Hagen) system (026), 379-380 Locus, 256
Jobs. See Human Resources reference genomic (LRG)record, 280
JR system (032), 380 LU (Lutheran) system (005), 363-364, 364
rare phenotypes, 364
K
LW (Landsteiner-Wiener)system (016), 375-377
KANNOsystem (037), 382 Lyonization, 258, 262
Karyotype,256
KELsystem (006), 364-368 M
antibodies/antigens, 365, 366-367, 367 M (MNSl), 361-362,362
glycoprotein and gene, 365 MAHA. See Anemia, microangiopathic hemolytic
null (Ko) phenotype, 367-368 Malaria, transfusion-transmitted (TTM),212-213
phenotype prevalence, 366 FYsystem and, 370
protein, 365 MALDI-TOF.See Laser desorption/ionization time-
Kidneytransplants, 494-495 of-flight,matrix-assisted
Kleihauer-Betketest, 664 Management, 38
KN (Knops)system (022), 378-379 change control, 26
K.nOd phenotype, 367-368 see also Work environment
Kx Antigen (XKl), 368 Manufacturers, HCT/Ps, 757
Index 807
Markers, 255 N
Marrow
N (MNS2), 361-362, 362
collection guidelines, 744
NANB. See Hepatitis, non-A, non-B
donation, 89
National Committee for ClinicalLaboratory
harvesting, 744-745
Standards (NCCLS),2
Mass spectrometry, 282
National Electrical Code, 34
Massive transfusion. See Transfusions,massive
National Healthcare SafetyNetwork (NHSN) 99
Materials management, 5, 23 100 ' ,
McLeod phenotype, 270, 271, 368
adverse reactions, 101-102
Medical devices. See Devices
incident codes, 104
Medical event (incident) reporting system for
National Marrow Donor Program (NMDP), 88-89
transfusion medicine (MERS·TM),99 89 '
Medical need, exceptional, 135-136
NEC. See Enterocolitis, necrotizing
Medical waste, 46, 48
Neonates. See Pediatric transfusions; Perinatal issues
Medications, transfusion deferral, 135
Nerve injury, local, 143
Meiosis, 256, 257-258, 279, 272
Neuromyelitis optica (NMO), TPE, 720
Membrane attack complex (MAC),247, 247, 250
Neutropenia
MER2,379
autoimmune (AIN),472
Mercury, spill response, 73
MHC genetics, 484-488, 485 neonatal alloimmune (NAN),471-472
HLAgenetic regions, 484, 486 Neutrophils
restriction, 483 alloantigen, human (HNA),472-473
Minisatellite, 279 antigens, human (HNA),469-470, 470,471
MINY,361, 363 immune disorders, 471·472
Mirasol, 163, 216 NGS. See Sequencing, next-generation
Mitosis, 256, 257, 263-264 Nitrogen, liquid, work safety, 70
MNS system (002), 359-363, 360 Nucleic acids
antigens and antibodies, 362-363 analysis, 229
phenotype prevalence, 362 chemistry and structure, 230·231, 230
Molecular biology, 229-254 isolation, 231
genetics, 255 testing (NAT), 181-182,233
immunohematology, 283 Nucleic acid sequence-based amplification(NASBA),
key points, 251 233-234
Molecular testing Nucleoside transporter 1, equilibrative (ENT!), 381
antigen-negative donors, 288 Nucleotides, single nucleotide variants (SNVs),263·
blood groups, 283-291 264,279
D type, 288 blood group system, 279
discrepancies, 289-291,290,291 genotyping, red cell phenotypes, 289-291
HDFN,286·287 o
paternal,287-288
prenatal,286-288 o allele, 302·303
red cell antigens, 284 Octaplas, 163·164, 199
resolution 282-283 OK system (024), 379
Rh variant antigen, 288 Organization. See Work environment
Monitoring, 5, 20·22, 24 Outcomes measurement, 4
blood ordering and usage, 614 Oxygenation, extracorporeal membrane (ECMO),
components, 509-518,510·516 690-691
look-back, 189, 194 blood component preparation protocols, 692
radiation, 54
p
see also Audits
MSBOS.See Bloodorder schedule, maximum surgical P1Pksystem (003), 318
MTPs. See Transfusions, massive, protocols PI substance, 417
Mur antigen, 363 PAD. See Blooddonation, preoperative autologous
Mutations, 263 Papain, 416
Para-Bombayphenotype, 312 indications and thresholds, 678-679

ParvovirusB19, 193,214-215 key points, 695-696
Paternity, 279 neonates, 687-688, 687
zygositytesting, 287-288 orders to prepare and to transfuse bloodand blood
Pathogens components, 538-539
blood-borne,41 pretransfusionsample, 539
inactivation, 162-164, 199 prophylacticmedications, 540
pediatric, 695 out-of-hospital,548-549
technologies, 215-217, 216 thrombocytopenia,fetal and neonatal alloimmune
reduction technology (PRT),566 (FNAIT),464-465, 667-668
Patient blood management (PBM),20-21, 583-612, vascular access and technique, 681
613,615-616 ventricular assistdevices (VADs), 693
certification,585, 606 WB,684-685
data collection, 600-605 Pedigrees, 266, 266
definition and scope, 583-584 PEG,415
education, 585, 587-588 Peptide-bindingspecificity,vaccine development,
guidance, 606 496
implementation, 588, 606 Perinatal issues, 659-672
key points, 607 HDFN,659-665
methods, 585-600 key points, 668
performancemeasures, 614 thrombocytopenia,pregnancy-related,665-
program levels, 586-587 667
qualification/competence, 588 see also Neonatal and pediatric, transfusions,
resources, 584-585 Peripheral blood
standards and certification,585 HPC collection, 745-747
strategies stem cells (PBSCs),87
intraoperative, 594-596 Peripheral nervous system diseases,TPE, 720
postoperative,596-598 pH,416
preoperative, 588, 591-592, 594 meters, QC, 30
transfusion extremes, 605-606 Phenotypes, 265-266, 275, 282
Patient care, 754-756 ABO, 298, 299
controls chart, 26 antigen-negative,276
evaluation and management, 707-711 null, 310, 312
high-riskpatients, 620 prevalence, 276
specimens, shipping,56 secretor, 297
in therapeutic apheresis, 707-709, 708, 709 Photopheresis, extracorporeal, 723-724, 724
PCR. See Polymerasechain reaction Phlebotomy, therapeutic, 144-145
Pediatrictransfusions, 673-703, 713 Physicalexamination, donors, 130
administration, rate of, 689-690 Pipette recalibration, QC, 30
adverse effects,694-695 Plasma
body size and blood volume, 675 AB,572
cell washing, 695 AHF, cryoprecipitated, 687
cold stress, 676 apheresis, 150
compatibilitytesting, 677-678 component storage, 155-158
component choice, 680-681 cryoprecipitatereduced, 158
conditions specificto, 681-682 derivatives,screening, 214-215
cryoprecipitate,688 exchange, therapeutic (TPE),705-720, 715-719
erythropoietin physiologyand therapy, 675-676 fresh frozen (FFP),155-156,568-569,606, 687
exchange, 680-681 cryoprecipitate-depleted,158
granulocytes, 688-689 within 24 hours after phlebotomy (PF24),
guidelines,679 156-157
hematopoiesis, coagulation, and physiology,673- HSCT,773
675,674 liquid,157
HSCT,773-774 neonatal and pediatric, 687-688, 687
Index 809
platelet-rich (PRP), 147, 148 transfusions, 561-567, 685-687

pooled, 159-160 ABO, 564-565
quarantined, 156 bleeding, 563-564
recovered, 158 components, 686-687
reduction, HPC, 748-749 dose, 562-563, 686-687
shipping, 522 guidelines, 686
storage, 513-515,518 HLA,565-567
testing problems, 307, 309 invasive procedures, 563
thawed, 157,518-519 neonatal and pediatric, 685-687
in therapeutic apheresis, 712 prophylactic, 561-563, 564
transfusions, 567-569 RhD matching, 564-565
bleeding, 567-568 VOlume-reduced,162
prophylaxis for invasive procedures, 567 washing, 162
vitamin K antagonist reversal, 568 Pollution, minimizing, 56
types, 568-569 Polycythemia, neonatal, 681-682
usage, 616-617 Polymerase chain reaction (PCR),231-233, 232,
whole blood, 572-573 279,283
Plasmodium spp., 212 contamination, 233
Platelets inhibitors, 232
additive solution (PAS),150, 155, 160 primer design, 232-233
alloantigens, human (HPAs),457-461,458-459, real-time, 236, 282
566-567 reverse transcriptase, 233
ABOand other blood groups, 461-462 specimen processing and template degradation,
CD109,461 232
GPla/IIa, 460-461 Polymorphisms, 262, 278
GPlblV/IX, 460 restriction fragment length (RFLP),282
GPIIb/lII, 459-460 short tandem repeat (STR),492, 496-497
GPW/CD36, 461,462 single nucleotide (SNPs),457
GPVI,462 Pooling, 159-160, 521
HLA,462 PPE. See Protective equipment, personal
nomenclature, 457 PI PKblood group system, 318-323
testing, 466-469 antibodies to, 322
antigens and antibodies, 457-469, 473-478 biochemistry, 319, 321
drug-dependence, 469 blood-group-carryingmolecules, 321
key points, 473 genetics, 321-322
testing, 466-469 phenotypes, 319, 319
apheresis, 149-150 prevalence, 319
screening for bacteria, 204-205 transfusion practice, 322-323
assays, 240, 467-468 Pragmatic Randomized Optimal Platelet and Plasma
autoantibodies, 468 Ratios (PROPPR)trial, 572
bacterial contamination, 175 Prenatal practice/pregnancy, 420, 665-667
corrected count increment, 565-566 donors, 135
cryopreservation and cold storage, 161 fibrinogen, 569
disorders, alloimmune, 462-465 molecular testing, 286-288
genotyping, 468 see also Pediatric transfusions
HSCT,772-773 Pressure devices, 541
inventory shortage, 527-528 Pretransfusion. SeeTransfusions, pretransfusion
pooled, 159-160 Primer
shelf life, 154 design, 232-233
storage, 154-155, 511-513,518 sequence-specific(SSP),488, 489
gel production, 519 Prions, 213-214
OC,29 Proband, 266
shipping, 522 Process, 26
volume reduction, 520 computer system, 13
control and management, 3, 5,10-14,23-24,26 Quarantine

documents, 15 blood components, 164
improvement, 22, 24 errors, 197
interaction, 4 failure, 197
protocols, 12
test methods, 13 R
validation, 10-13, 12 Radiation,53-57, 519-520
Processing, 748-750 biologicaleffects, 53
Proficiencytesting (PT),22, 84 cesium irradiators, 56
Protective equipment, personal (PPE), 63-65 dose, 161
biosafety, 42, 45 emergency response plan, 55
chemical safety, 51 engineering controls and PPE, 55
electrical safety, 40 exposure limits, 54
fire prevention, 39 irradiators
laboratory coats, 63 QC,30
masks, 64-65 replacement, 56
PBM,606 measurement units, 53
radiation, 55 monitoring, 54
see also Safety,37; Work environment neonatal and pediatric, 694-695, 694
Protein assays regulations, 53-54
flow cytometry, 243 training, 54-55
fluid-phase,238-239 waste management, 55
less common, 242-243 work practices, 55
protein microarrays, 242 RAPHsystem (025), 379
solid-phase,239-242 Reaction rates, 107, 108-110
suspension array technology (SAT),243 Reagents
Western blotting (WB),242-243 high-flow-protein, 348
Proteins QC,31
analysis,238-244 Recalls,managing, 84
nomenclature, 280 Receipts, 10
PRT. See Pathogens, reduction technology Receiving,supplies, 523
PublicHealth Service(PHS)Act, 78, 79, 82, 84, 86, 87 Recipients
Purpura, posttransfusion (PTP),465, 650-652 baseline assessment, 538
consent, 537-538
Q
designated/directed donation, 135-137
Qualification,26 hemovigilance, 100-103
Quality history and education, 538
assurance, 2, 26 identification and labeling, 503-504, 525
factor (OF),53 prior, 189, 194
glossary,26-27 testing, pretransfusion, 504-509
indicators, 20, 26 transfusion risks, 111
planning, 3, 4-5, 27 verification at administration, 544
Quality control (QC), 2-3,13-14,26,752-753 Records, 5, 16-17,24
documentation, 13-14 changes to, 16
frequency, 13 confidentiality, 16
intervals, 29-31 error correction, 17
objectives, 614-615 locus reference genomic (LRG),280
Quality management systems (OMS), 1-31 retention policy, 16-17
approach, 4-5 storage, 17
concepts, 2-4 see also Documentation
evaluation, 5 Red Blood Cell (RBC)transfusions, 553-561, 598-
key points, 23-24 599,682-685
practice, 5-23 additive solutions, 676
systems and processes, 3, 27 age, 676-677, 684
Index 811
AlHA, 559-560 SCD,682·683

alternatives, 676-677 survival, 389·390
components, 153-154 testing problems, 307, 309
cryopreservation, 160-161 washing, 162
emergency, 556-557 Refrigerators,OC, 29
HDFN,665 Reglstration,blood establishments and device
HSCT,772 manufacturers, 79, 81
infants and children, 682·685 Regulationsand recommendations, 21, 61-62,498,
irradiation, 162 756·757
neonates, 675·682 accreditation vs, 77, 18
recipients of monoclonal antibodies, 560-561 allografts,782-783, 183
shipping, 522 clinicalhistocompatibility,498
single-unit, 598·599 donor eligibility,128
storage, 510-511,517,556,676·677 FDA,81
storage lesion, 676·677 radiation, 53-54
thalassemia and sickle cell disease, 557· see also Standards
559 Relationshiptesting, 278-279
thawing and deglycerolizing,519 Relativerisk (RR),HLAtype and disease, 496
thresholds, 589-590 Renal failure, acute, TPE, 720
unit selection, 523·525 Replacement fluids,in therapeutic apheresis, 109
Red cells Reporting/reports
alloimmunization, 682·683 adverse events, 111
anion exchanger, 372 biovigilance,99-100
antibodies establishment inspection (EIR),83
binding, 241 hemovigllance, 100
nonhemolytic, 249·250 SDS,50
antigens, 255 transfusion-related adverse reaction, 103
expression differences, 255 virus test results, 111
molecular testing, 284 see also Documentation/documents; Records
apheresis, 149 Requirements, 27
autologous phenotype, 397 Respiratorydistress syndrome, acute (ARDS),640·
adsorption, 418·419 641
adsorption-elution, 420 Reviews. See Audits; Evaluation;Monitoring
antibodies, 421·422 RFLP.See Polymorphism,restriction fragment length
chemical modification,417 RHAGblood group system (030), 347, 380
donor unit selection, 422-423 RHCE, 332, 334, 335, 345
elution, 419·420 RhCE,347
enzyme modification, 416·417 D epitopes, 340
immunoglobulins, denaturation, 418 RHD, 286, 290-291, 292, 334, 335, 337, 345
inhibition techniques, 417·418 genotyping, 287
obtalning, 415 molecular testing, 288
pH,416 nonfunctional, 340
reactivity, 405·406 zygosity,287-288, 346
serologlc clues, 410 RhD,347
titration, 420·421 -likeloops, 339
drugs, therapeutic, 285 matching, platelets, 564-565
exchange, 722·723, 122 model, 339
IgG,285 typing, 507, 524
immunologlc responses, 250·251 RHNULL syndrome, 347
membrane, 301 Rh system, 329-354
phenotypes antibodies to, 347-348
molecular testing, 283-285, 284 antigens, 329, 330-332, 337-346
SNVgenotyping, 289·291 high-prevalence, 348
reduction, HPC, 748·749 molecular testing, 288
antisera, 336-337 incidental spillresponse, 71-73

chromosomal structure, 333-334 key points, 57
gene products, 334 levels, 41-42, 43-44, 66
genotyping, 346-347 mobile blood-collectionsites, 35
confirming, 346-347 nitrogen, liquid, 70
fetal, 346 plan, 35-38, 46
RHDzygosity testing, 346 precautions, 41-42, 66
SCD patients, 347 program, 33, 35-39
transfusion recipients, 346 radiation, 53-57
glycoprotein, associated (RhAG),380 solvents, flammable, 70
haplotypes prevalence, 333 training, 36, 37, 42
immune globulin (RhIG),329, 395 work practices, 37, 45-46
HDFN, 662-665,665 see also Fire prevention; Protective equipment,
key points, 349-350 personal; Work environment
locus, 333-334, 335 "SafetyReporting Requirements for Human Drug and
phenotypes, 337-340, 342 BiologicalProducts," 97
proteins, 334 Samples, 503-504
terminology332-333 donors, retention and storage, 523
typing pretransfusion, 539
discrepancies, resolving, 349 recipients, 503-507, 523
false-positiveand false-negative,348-349 SC (Scianna)system (013), 374
technical considerations, 348-349 SCD, RH genotyping, 347
Riboflavin(vitamin B2), 216, 217 Screening
Riskassessment, 11, 12 infectious diseases, 174-176
Rituxlinab, TPE, 720 specificagents, 199-217
Root cause analysis, 18, 19 Sd' substance, 417
Rosette test, 664 Secretor phenotype, 297
Rouleaux formation, 414 Sepsis. See Infections/infectious diseases
Sequencing, 489
s massivelyparallel (MPS),283
S (MNS3)/s (MNS4), 361-362,362 next-generation (NGS),237-238, 279, 489
antibodies to, 362 Serologictesting, 181, 504
Safety,S, 22-24, 33 discrepancies, molecular testing, 288, 289-291,
acids, 69, 71 289, 290, 291
acrylamide, 69 principles, 504-505, 505, 506
alkalis, 69 Serum
bases, spillresponse, 71 testing problems, 307, 309
biosafety, 41-48 -to-cellratio, increasing, 415-416
cabinets, 42, 43-44 Sex chromosomes, 256
caustics, spill response, 71 see also Genes/genetics
chemicals, 48-53 Shipping and transportation, 522-523, 754
data sheet (SDS),49, 50, 51 blood components, 509,510-516,542-543
disaster plan, 23 containers, QC, 31
electrical, 40-41 hazardous materials, 56
emergency response, 38, 46 HSC, 754
engineering controls and PPE, 37, 42 Shock, AHTR,635
eyewashes, 65 Short tandem repeat (STR),279
eyewash stations, 34 SHOT. See Transfusions, serious hazards of
face shields, 64-65 Siblings,268
federal regulations and recommendations, 33 Slcklecell disease (SCD),286, 682-684
fire prevention, 40-41 RBCtransfusion, 557-558, 559
goggles, 64 red cell exchange, 722-723
hazard identification and communication, 36, 42 transfusion complications, 683-684
HBV,HCV, or HlV exposure, 38 WBC filtration, 159
Index 813
SID system (038), 323, 382 (Sba) Suspension array technology (SAT),243
Skin Syphilis. See Treponema pallidum
donors, lesions, 131 Syringe set, neonatal, 692
uses, 780 System, 27
SLC14Al,371-372
SMIMl, 381 T
SNP. See Polymorphisms, single nucleotide TACO. See Circulatory overload, transfusion-
SNV. See Nucleotides, single gene associated
Solid-organtransplantation, 495 TA-GVHD.See Graft-vs-hostdisease, transfusion-
consultation, 497-498 associated
donors, ABO genotyping, 289 Tandem repeats, variable number of (VNTR),
non-HLA,490 279
Specification, 27 T cells, 483
Spills Temperature reduction, 415
bases, 71 Templates, 15
blood,47 degradation, 232
caustics, 71 Terminology, 106, 280-281
chlorine, 71 Fisher-RaceDCE, 333
formaldehyde, 72 Testing
gases, 71, 72 cases, 11-12
glutaraldehyde, 72 logistics, 178
incidental, 71-73 method validation, 13
liquids, flammable, 72 molecular, 288, 289-291, 289, 290, 291
mercury, 73 nucleic acid (NAT), 181-182,233
SSP. See primer, sequence-specific point-of-care, 596
S-s-U- phenotype, 362 recipients, pretransfusion, 504-509
Standard operating procedures (SOPs), 10, 15 relationship, 278-279
Standards results, reactive 182, 189
clean spaces/rooms, 34 RHD zygosity, 346
hemovigilance, 106 rosette, 664
PBM,585 serologic, 181,504,504-505,505,506
see also Regulations and recommendations serum, 307, 309
Staphylococcus eureus, contamination, 205 surrogate, 173-174
Statistical tools, 3 see also Evaluation
Storage Thalassemia, 286, 684
antibody changes, 390-391 RBCtransfusion, 557, 558
bags, 156 Thermometers, QC, 30
components, 153-158,509-518,510-516 Thrombocytopenia
hazardous materials, 45 drug-induced, 465-464
lesion, 517 fetal and neonatal alloimmune (FNAlT),464-465,
plasma, 156 667-668
WB, 146 heparin-induced (HIT),465-466
STR. See Polymorphisms, short tandem repeat immune (ITP), 466, 667
Structural variation, 263-264, 271-273 pregnancy-related,665-667
Sulfhydrylreagents, 418 therapy-induced hypoproliferative, 561-563
Suppliers, 4, 5, 9-10, 23 transfusion guidelines, 686
contracts and agreements, 10 Thrombocytopenic purpura, thrombotic (TTP), 567-
factors, 9 568
°
qualification, 9-1
receipt and inspection of incoming supplies,
in therapeutic apheresis, 711-712
Thrombocytosis, massive, cytapheresis, 722
10 Thromboelastography (TEG), 21
Surgery, minimally invasive, 595 Thromboelastometry (TEM), 21
see also Blood, ordering Thrombosis, 606
Surveillance terms, reconciled, 106 Timers, QC, 30
Tissue key points, 90

allografts, 778 regulatory considerations, 77-93
clinical use, 780, 781-782 Transfusion reactions, 634-648
disease transmission, 782 allergic (ATRs),639-640
donation, 777-782 categories, 628-632
donor eligibility,778 clinical evaluation and management, 633
hospital tissue services, 784-790 fatality reporting, 651-652
hospital transfusion service and, 777-791 hemolytic (HTRs),355, 389-390, 434
infectious disease testing, 778, 778 acute (AHTR),247-248, 634-636
key points, 789-790 antibody-negative (AN-DHTRs),249
laws, regulations, and standards, 782-783,783 delayed (DHTRs),248, 286, 648-652
preservation, 779-780 hemovigilance, 627
processing, 779 hypotensive (HyTRs),645
transplantation, 777-782 identification, 627, 632
autografts, 788-789 key points, 653
donor testing, 195 laboratory investigation, standard, 633-634
homografts, 778 nonhemolytic, febrile (FNHTR),491-492, 540,
processing, 779-780 638-639
storage, 516, 786-787,787 noninfectious, 627-657
adverse events, 788 pediatric, 694
recalls and look-back investigations, 788 purpura, posttransfusion (PTP), 465, 650-652
record-keeping, 787-788 surveillance, 623
traceability, 787-788 suspected, 547, 627-634
transplantable, 778 Transfusions
xenografts, 779 blood selection, 439-441
Tissue-basedproducts, testing donors, 195 blood utilization, 613-626
Tissue services, hospital-based, 784-789 chain, 95
incoming tissue, inspection, 785-786 completion, 547
responsibility for, 784, 784 components, issuing, 523-527
standard operating procedures and policies, 784- deferral list, 135
785 distribution, 522-523
supplier qualifications and certification, 785, 786 documentation, 525, 547-548
Titration, 420-421 emergency, 525-527
Tracking, 9, 14 extremes, 605-606
Training, 7 fetal, 661-662
biosafety, 42 guidelines, 599-600
electrical safety, 40 hemogiobin, 602
fire safety, 39 HLAand, 491-493
new employees, 14 infections, transmitted (TTl), 100
nonconformances, 17 monitoring system (TTIMS), 111-112
radiation, 54-55 sepsis, transfusion-related, 637-638
restricted areas, 34 inventory management, 527-529
safety program, 36, 37 key points, 529-530
see also Human Resources massive, 526-527, 571-573, 605-606
Traits, 256 complications, 647-648
recessive, 268 formula-based, 571
see also Genes/genetics; Inheritance hemostatic abnormalities, 647-648
TRALI.See acute lung injury, transfusion-related liberal vs restrictive strategies, 553-556
Tranexarnic acid, 596 pediatric, 693
trans, 274 protocols (MTPs),572
Transcription-mediated amplification(TMA),233- monitoring, 509-518,510-516,545,547
234,234 nonemergent, 546
Transfusion medicine, 1 pediatric, 693
FDA,78-84 PI PKblood group system, 322-323
Index 815
preparation of ordered units, 539-540 V

pretransfusion, 537
Vaccine development, peptide-binding specificity
processing, 518-522
and,496
recipient's blood, 504-509, 505 Validation, 10-13,27
testing, 504-509 Variants of unknown significance (VUS),283
test results, positive, 535 Vascular access
rate of, 545 central catheters, 710
readiness, 542. pediatric, 689
recent recipients, 283, 285 subcutaneous ports, 710
reduction, 177 in therapeutic apheresls, 709-711
refractoriness, 462-464, 491, 565-567 Vasovagal reactions, 143-144
requests, 503 VEL (Vel)system (034), 270, 279, 381
safety officer (TSO), 21 Venipuncture site, 142
samples and requests, 503-504 Venous access
second, 21, 21 . devices, 710
serious hazards of (SHOT), 96-97, 96 HPC collection peripheral blood, 74S
senrice-related activities, 503-535 Ventilation, 34
FDA and, 81 Ventricular assist devices (VADs),pediatric, 693
registration, 81 Verification, 27, 544-545
small-volume aliquoting, 690, 691 View boxes, OC, 30
starting, 545 Vitamin K antagonist reversal, plasma, 568
storage, 509-518,510-516 VNTR. See Tandem repeats, variable number of
threshold, 597-598 Volume reduction, 520
platelets, HSCT, 772-773 pediatric, 695
RBC, 589-590 von Willebrand factor (vWF)
unique settings, 548-549 plasma and, 567-568
warm-reactive patients, 441 TPE, 714
see also Plasma, transfusions; Platelets,
transfusions; RBC transfusions w
Transplantation WAlBA. SeeAnemia, hemolytic warm
haploidentical, 742 Waste disposal/management, 56-57
HLA testing and, 493-495 biohazard, 46-48
stage of, 768, 769 chemical, 52-53·
Transportation, See Shipping infectious, 46
Trending, 14 medical, 46
Treponema psllidum, 180, 186, 196 radiation, 55
donor testing, 192 Waterbaths, ~C, 30
screening for, 203- West Nile virus (WNV), 180, 188
Trypanosoma cruzt, 175, 180, 187, 209-210 donor screening/testing, 195, 196
regulations and standards, 192 NAT, 175
screening, 173, 174 regulations and standards, 193
Trypsin, 416 screening for, 206
Tubing welder, sterile, 691 Western blotting, 242-243
Typing, sequence-based, 489 Whole blood (WB), 141-172
cold-stored, 153
U collection, 145-148, 147
Ultraviolet light filtration, 147, 148
A (UVA),216, 217 key points, 165-166
B (UVB), 216 labeling, 145
Umbilical cord blood (UCB), 87, 737, 740, 741-742 resuscitation, 572-573
HPC collection, 747-748 shipping, 522
Universal Protocol, preprocedure verification process, storage, 146, 510, 518
86 transfusion, pediatric, 684-685
Utilization review, 617-620 See also Blood components; Donors
Withdrawals, managing, 84 x
Work environment,S, 22-23, 24, 33-75
XCFR,hemovigilance, 96, 99
change control, 26
X chromosome inactivation, 258, 262
ergonomics, 35
Xenotransplantation products, 779
gloves, 63-64
XG (Xg')system (012), 374
guidance, 63-65
XKsystem (019), 364-368
hand-washing, 35, 65
protein, 365
instructions, 15
inventory management, 527-529 y
key points, 57
leadership,S, 6, 23 YTEG(YT3),373
management, 38 YTLl(YT4),373
organization,S, 6, 7, 23 YTOT(YT5),373
radiation, 55 YTsystem (011),373
uniforms, 63
Z
see also Facilities; Human Resources; Safety
Wralb, 373 Zikavirus (ZIKV),176, 180, 188
Wrong blood in tube (WBIT),539 donor testing, regulations and standards, 193
screening for, 206-208
Zygosity, 265, 390
paternal samples, 287-288

AABB Technical Manual 20th Edition

Uploaded by

Copyright:

Available Formats

AABB Technical Manual 20th Edition

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

AABB Technical Manual 20th Edition

Uploaded by

Copyright:

Available Formats

20TH EDITION 20TH EDITION

4550 Montgomery Avenue. 700N • Bethesda,MD "lInII'I,&_'lIY,

Phone: + 1.301.907.6977• www.aabb.org

Transfusion Therapy: Clinical Principles and Practice, 4th edition

Claudia S. Cohn, MD, PhD

Meghan Delaney, DO, MPH

Susan T. Johnson, MSTM, MT(ASCP)SBB

AABB ISBNNo. 978-1-56395-370-5

Technical manual / editor, Claudia S. Cohn-20th ed.

QUALITY AND RELATED ISSUES

1. Quality Management Systems: Principles and Practice •••••••• 1

2. Facilities, Work Environment, and Safety •••••• • • •• • • • • • • •• 33

Appendix 2-3. BiosafetyLevel 2 Precautions 66

3. Regulatory Considerations in Transfusion Medicine and

4. National Hemovigilance: The Current State ••••..•••.•••••. 9S

BLOOD COLLECTION AND TESTING

S. Allogeneic and Autologous Blood Donor Selection .•••..•. 127

6. Whole Blood and Apheresis Conection of Blood

7. Infectious DiseaseScreening••••••••••••••••••••••••••••• 173

8. Molecular Biology and Immunology in Transfusion

9. Blood Group Genetics •••.••.•••••.••••••.•.•..•..••••.•• 255

Genetic Modifiers of Blood Group Antigen Expression 274

11. The Rh System • • . • • . . . . • . • • • • • • • • • . . . • • • • • • • • • . • • . • • • • •. 329

12. Other Blood Group Systems and Antigens ••••••••••••••••• 355

The SC System (013) 374

13. Identification of Antibodies to Red Cell Antigens •••••••••• 389

14. The Positive Direct Antiglobulin Test and Immune-

Key Points 449

15. Platelet and Granulocyte Antigens and Antibodies •••••••• 457

16. The HLA System•.•••••..••••••••••••..•••.•..••••••••.•• 479

ESSENTIALS OF TRANSFUSION PRACTICE

17. Transfusion-Service-Related Activities: Pretransfusion

18. Administration of Blood Components •••••••••••••••••••• 537

19. Hemotherapy Decisions and Their Outcomes •.•••••••••••• 553

20. Patient Blood M&ll'Iilg4!rnent •••••.••••••••••••••••••••••••

Steven M. Frank, MD, and Nicole R. Guinn, MD

21. Approaches to Blood Utilization Auditing ••••••••••••••••• 613

The Role of a Computerized Provider Order Entry System in Blood Utilization

22. Noninfectious Complications of Blood Transfusion •••••••• 627

SPECIAL PATIENTS AND SITUATIONS

23. Perinatal Issues in Transfusion Practice ••••••••••••••••••• 659

24. Neonatal and Pediatric Transfusion Practice •••••••••••••• 673

25. Therapeutic Apheresis •••.••.•••••••••••••••••••••••••••• 705

Vascular Access. . . . . . . .. . . . . . . . . . .. . ; 709

26. The Collection and Processing of Hematopoietic

27. Transfusion Support for Hematopoietic Stem Cell

28. Human Tissue Allografts and the Hospital

1. General Laboratory Methods

Method 1-1. Shipping Hazardous Materials

2. Red Cell Typing Methods

Method 2-1. Determining ABO Group of Red Cells-Slide Test

3. Antibody Detection, Identification, and Compatibility Testing

Method 3-1. Using Immediate-Spin Compatibility Testing to Demonstrate ABO

4. Investigation of a Positive DAT Result

Method 4-1. Cold-AcidElution Procedure

5. Hemolytic Disease of the Fetus and Newborn

6. Blood Collection, Component Preparation, and Storage

7. Transplantation of Cells and Tissue

Method 7-1. Infusing Cryopreserved Hematopoietic Cells

8. Quality Control Methods

1. ;e~ ~:: :~~:ti:do~::;~:a~~~PY are highly regulated, involving multiple