Classic Papers in Rheumatology (PDFDrive)

Classic Papers in Rheumatology
Classic Papers in Rheumatology
Edited by
Paul Dieppe
Director, MRC Health Services Research Collaboration,
Department of Social Medicine, University of Bristol, Bristol, UK
H Ralph Schumacher, Jr
Professor of Medicine, University of Pennsylvania,
Chief of Rheumatology, VA Medical Center, Philadelphia, USA
Frank A Wollheim
Professor, Lund University Hospital,
Lund, Sweden
MARTIN DUNITZ
© 2002 Martin Dunitz Ltd, a member of the Taylor & Francis group
First published in the United Kingdom in 2002

by Martin Dunitz Ltd, The Livery House, 7--9 Pratt Street, London NW1 0AE
Tel: +44 (0) 20 74822202

Fax: +44 (0) 20 72670159
E-mail: info@dunitz.co.uk
Website: http://www.dunitz.co.uk
This edition published in the Taylor & Francis e-Library, 2003.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of the publisher or in accordance with the provisions of the Copyright,
Designs and Patents Act 1988 or under the terms of any licence permitting limited copying issued by the
Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP.
Although every effort has been made to ensure that all owners of copyright material have been
acknowledged in this publication, we would be glad to acknowledge in subsequent reprints or editions any
omissions brought to our attention.
Although every effort has been made to ensure that drug doses and other information are presented
accurately in this publication, the ultimate responsibility rests with the prescribing physician. Neither the
publishers nor the authors can be held responsible for errors or for any consequences arising from the use
of information contained herein. For detailed prescribing information or instructions on the use of any
product or procedure discussed herein, please consult the prescribing information or instructional
material issued by the manufacturer.
A CIP record for this book is available from the British Library.
ISBN 0-203-21423-4 Master e-book ISBN
ISBN 0-203-27085-1 (Adobe eReader Format)

ISBN 1–901865–48–7 (Print Edition)
Distributed in the USA by
Fulfilment Center
Taylor & Francis
7625 Empire Drive
Florence, KY 41042, USA
Toll Free Tel: +1 800 634 7064
E-mail: cserve@routledge_ny.com
Distributed in Canada by
Taylor & Francis
74 Rolark Drive
Scarborough, Ontario M1R 4G2, Canada
Toll Free Tel: +1 877 226 2237
E-mail: tal_fran@istar.ca
Distributed in the rest of the world by

ITPS Limited
Cheriton House
North Way
Andover, Hampshire SP10 5BE, UK
Tel: +44 (0)1264 332424
E-mail: reception@itps.co.uk
Project management, design and layout: Top Draw Design

Contents
Contributors vii
Foreword ix
Preface xi
Section 1: Diagnosis
1 Clinical outcome measurement 2
Nicholas Bellamy and W Watson Buchanan
2 Imaging 15
Inga Redlund-Johnell
3 Serological tests 33
Josef Smolen
4 Other investigations: synovial fluid analysis, arthroscopy and
blood biochemistry 48
Paul Dieppe
5 Clinical genetics 64
Sophia Steer and Gabriel S Panayi
Section 2: Diseases
6 Rheumatoid arthritis 81
Frank A Wollheim
7 Chronic arthritis in children 108
Jacqui Clinch and †Barbara Ansell
8 Ankylosing spondylitis 118
Muhammad Asim Khan
9 Psoriatic arthritis 134
Dafna D Gladman
10 Reactive arthritis and Reiter’s syndrome 148
Henning K Zeidler
11 Lupus 164
Graham RV Hughes and Munther Khamashta
12 Scleroderma (systemic sclerosis) 176
E Carwile LeRoy
13 Systemic vasculitis 192
David GI Scott
14 Sjögren’s syndrome 202
Roland Jonsson
15 Myositis 222
Frederick W Miller and Paul H Plotz
16 Arthritis and infection 236
Andrew Keat and Ralph C Williams, Jr
17 Osteoarthritis 256
Paul Dieppe
18 Crystal deposition diseases 275
Daniel J McCarty
19 Osteoporosis 290
Cyrus Cooper
Section 3: Regional and Miscellaneous Disorders

20 Back pain 308
Malcolm IV Jayson
21 Soft tissue disorders 325
Cathy Speed and Brian Hazleman
22 Fibromyalgia 346
Hugh Smythe
23 Geographical disorders 359
Adewale Adebajo
24 Associations of other systemic diseases with arthropathies 375
Section 4: Therapy
25 Education 391
Ylva Lindroth
26 Key drug developments 407
Peter M Brooks
27 Exercise and rehabilitation in arthritis 428
Michael V Hurley
Index 441
vii
Contributors
Adewale Adebajo
Consultant Rheumatologist and Honorary Senior Lecturer, Division of Genomic Medicine,
The University of Sheffield, Royal Hallamshire Hospital, Sheffield, UK
†Barbara Ansell
Nicholas Bellamy
Professor, Director of CONROD and Chair of Rehabilitation Medicine, Department of
Medicine, University of Queensland, Brisbane, Queensland, Australia
Peter M Brooks
Professor, Executive Dean (Health Sciences), University of Queensland, Royal Brisbane
Hospital, Herston, Queensland, Australia
W Watson Buchanan
Emeritus Professor of Medicine, Sen. William Osler Health Institute, Hamilton, Ontario,
Canada
Jacqui Clinch
Rheumatology Department, Southmead Hospital, Westbury on Trym, Bristol, UK
Cyrus Cooper
Professor of Rheumatology, MRC Environmental Epidemiology Unit (University of
Southampton), Southampton General Hospital, Southampton, UK
Paul Dieppe
Professor, Director of MRC Health Services Research Council, University of Bristol, Bristol, UK
Dafna D Gladman
Professor of Medicine, Centre for Prognosis Studies in the Rheumatic Diseases, Toronto
Hospital Western Division, Toronto, Ontario, Canada
Brian Hazleman
Rheumatology Research Unit, Addenbrooke's Hospital, Cambridge, UK
Graham RV Hughes
Consultant Rheumatologist and Head of the Lupus Unit, Department of Rheumatology,
St Thomas Hospital, London, UK
Michael V Hurley
ARC Research Fellow and Reader in Physiotherapy, Rehabilitation Research Unit, King's
College London, Dulwich, London, UK
Malcolm IV Jayson
Professor of Rheumatology, University of Manchester, Manchester, UK
Roland Jonsson
Professor of Medicine (Immunology), Broegelmann Research Laboratory, University of Bergen,
Bergen, Norway
Andrew Keat
Consultant Rheumatologist, Arthritis Centre, Northwick Park Hospital, Harrow, Middlesex, UK
Munther Khamashta
Department of Rheumatology, St Thomas Hospital, London, UK
viii
Muhammad Asim Khan

Professor, Metro Health Medical Center, Cleveland, Ohio, USA
E Carwile LeRoy
Professor of Medicine, Department of Microbiology and Immunology, College of Medicine,
Medical University of South Carolina, Charleston, South Carolina, USA
Ylva Lindroth
Rheumatology Consultant, Slottsstadens Laekarhus, Malmoe, Sweden
Daniel J McCarty
Will and Cava Ross Professor of Medicine Emeritus, Division of Rheumatology, Department of
Medicine, Milwaukee, Wisconsin, USA
Frederick W Miller
Chief, Environmental Autoimmunity Group, Office of Cinical Research, National Institute of
Environmental Health Sciences, National Institutes of Health, Bethesda, Maryland, USA
Gabriel S Panayi
Professor of Rheumatology, Arthritis Research Campaign, Department of Rheumatology, Guy's
Hospital, London, UK
Paul H Plotz
Arthritis and Rheumatism Branch, National Institute of Health NIAMS, Clinical Center,
Bethesda, Maryland, USA
Department of Radiology, University Hospital MAS, Malmo, Sweden
Professor of Medicine, University of Pennsylvania, Chief of Rheumatology, VA Medical Center,
Philadelphia, USA
David GI Scott
Honorary Professor – School of Health Policy and Practice UEA, Department of Rheumatology,
Norfolk and Norwich University Hospital, Norwich, UK
Josef Smolen
Professor of Medicine, Department of Rheumatology, University of Vienna, Vienna General
Hospital, Vienna, Austria
Hugh Smythe
University Health Network, The Toronto Western Hospital, Toronto, Ontario, Canada
Cathy Speed
Rheumatology Research Unit, Addenbrooke's Hospital, Cambridge, UK
Sophia Steer
ARC Clinical Lecturer, Department of Rheumatology, Guy’s Hospital, London, UK
Ralph C Williams, Jr
Master ACR, Santa Fe, New Mexico, USA
Frank A Wollheim
Professor, Department of Rheumatology, Lund University Hospital, Lund, Sweden
Henning K Zeidler
Professor of Internal Medicine and Rheumatology, Medical Highschool Hannover, Division of
Rheumatology, Hannover, Germany
ix
Foreword
New paradigms drive out the old. We have had enough of the old history – Hippocrates,
Sydenham, Garrods, Pierre Marie and others. This excellent book covers 27 aspects of
rheumatology, summarizing what used to be called ‘recent advances’. Each of the editorial
teams cover up to 14 classic papers each from the 1940s to the present, under the analytic
headings ‘Summary’, ‘Key message’, ‘Why it’s important’, ‘Strengths’, ‘Weaknesses’ and
‘Relevance’, making them easy to understand. They include a few old classics from 1907, 1912,
1940, and Steinbrocker et al. (1949) (see page 8), which have been missed.
Choice is always disputatious, but the contributors here are outstanding and important
authorities in their own fields. The theories of ‘phlogistom’ have perished and, in more recent
times, the theories of focal sepsis and the role of physiotherapy versus rehabilitation have
undergone drastic criticisms. This book is easy to read, covers the main topics with which
rheumatologists (and GPs!) have to deal, and is wholly recommended.
EGL Bywaters
Beaconsfield, UK
xi
Preface
It has been said that we are currently ‘drowning in information but thirsting for knowledge’.
Through electronic databases we have immediate access to all the recently published literature
on any subject we care to think of. But in medicine, in spite of its immediacy, seduction and
popularity, current journal literature may not provide us with very much understanding of a
subject. An historical perspective and knowledge of the early descriptions of things we now take
for granted can add greatly to our understanding and enthusiasm for our subject.
Rheumatology is a relatively young sub-specialty of medicine, which has only come into its
own over the last 30 or 40 years. However, we are already drowning in a mass of current
literature, with dozens of journals and hundreds of reviews competing for our attention. And
we have already forgotten the older literature that set the scene for this present deluge of
information. This book attempts to redress the balance by providing readers with access to
some of the classic papers – the early descriptions of key findings, diseases and phenomena of
relevance to clinical rheumatology.
Each chapter starts with a short ‘scene-setting’ introduction and rationale for the choice of
papers that follow, and then each chosen classic paper is presented according to a common
template to make it easy to follow and understand. Related references are usually included to
help interested readers delve further into the literature.
The contents are split into four sections. In rheumatology, as in any science, technological
developments have been of crucial importance to the growth of our subject. In Section 1 we
cover the impact of technology on investigative rheumatology; this spans the development of
simple clinical measures (metrology) to the impact of modern clinical genetics on our subject.
Section 2 covers each of the major disease entities, providing readers access to early descriptive
accounts and crucial developmental literature in each area. In Section 3 we address the very
common and important regional musculoskeletal disorders, such as back pain, soft tissue
disorders and fibromyalgia, as well as miscellaneous disorders. Finally, Section 4 is concerned
with the therapy of the rheumatic diseases, and includes key developments in educational and
physical interventions, as well as drug therapy.
The authors are all experts in their field, and therefore are well-placed to select the most
important papers in their area. Nevertheless, many of them have told us how difficult they
found it to make a small selection, and several pleaded with us to allow them to have more
papers in their chapter. We were tough, limiting their choices for the sake of space and
discipline. We appreciate that others may have made different choices and that some readers
may disagree with the choices made. Perhaps we should open a competition for the ‘all time
top five papers in each category.
We believe that this book should be an important addition to the library of everyone
interested in clinical rheumatology. However, it is perhaps of most importance to trainees, as
they must surely be expected to develop a greater depth of understanding of their subject while
training than can be obtained from modern books and journals. But, of course, we are all
trainees. And awareness of history adds good quality to all teaching.
Paul Dieppe
Frank A Wollheim
Dedication
The editors dedicate this book to the late Barbara Ansell (1923–2001).
1
Section 1
Diagnosis
2
CHAPTER 1
Clinical outcome
measurement
Nicholas Bellamy and W Watson Buchanan
Introduction
Validity, reliability and responsiveness are the quintessential requirements of a health status
or outcome measure (1). In the last century there has been a steady evolution in the sophisti-
cation of measurement techniques. While there remains much to achieve, and accepting that
many of the remaining problems are the same problems confronting early pioneers, never-
theless there has been substantial progress in the discipline of musculoskeletal clinical
metrology, and there are currently available several advanced high performance techniques
with which to quantitate health status in rheumatology. A knowledge of the early work of
Taylor, Keele, Steinbrocker and Lansbury, is germane to a comprehensive understanding of
the field.
The early pioneers created a foundation for subsequent work in the areas of pain mea-
surement (Keele), functional assessment (Taylor, Steinbrocker), articular index and compos-
ite index construction (Lansbury). Their work was without precedent, and unassisted by desk
top computers, sophisticated statistical packages and high speed communication. Today,
remnants of their early work, can be recognized in the ARA core set measures for rheuma-
toid arthritis (RA) (9), the ACR response criteria (10), the OARSI core set measures for OA
(11) and the OARSI responder criteria (12). The measurement of pain and function
remains a challenge but modern instruments, such as the Health Assessment Questionnaire
(5), Arthritis Impact Measurement Scales (6, 7), Western Ontario and McMaster Universities
Osteoarthritis Index (13) and the Australian/Canadian Osteoarthritis Hand Index (14) are
on the leading edge of the development. The problem of weighting and aggregating data
from different dimensions, that challenged Lansbury, remains complex, controversial, and
without final resolution. Musculoskeletal metrology remains a dynamic and evolving field
(1), which owes much to these early pioneers.
Clinical outcome measurement 3
References
1. Bellamy N. Musculoskeletal Metrology. Lancaster, Kluwer Academic Publishers. 1993: 1–367.
2. Felson DT, Anderson JJ, Boers M, et al. The American College of Rheumatology
Preliminary Core Set of Disease activity measures for rheumatoid arthritis clinical trials.
Arthritis and Rheumatism 1993; 36: 729–740.
3. Felson DT, Anderson JJ, Boers M, et al. The American College of Rheumatology
Preliminary Definition of Improvement in rheumatoid arthritis. Arthritis and Rheumatism
1995; 38: 727–735.
4. Osteoarthritis Research Society (OARS) Task Force Report: Design and conduct of clini-
cal trials of patients with osteoarthritis: recommendations from a Task Force of the
Osteoarthritis Research Society. Osteoarthritis and Cartilage 1996; 4: 217–243.
5. Dougados M, Le Claire P, van der Heijde X, et al. Response criteria for clinical trials on
osteoarthritis of the knee and hip. Osteoarthritis and Cartilage 2000 (In Press).
6. Fries JF, Spitz P, Kraines RG, et al. Measurement of patient outcomes in arthritis. Arthritis
and Rheumatism 1980; 23: 137–145.
7. Meenan RF, Gurtman PM, Mason JH. Measuring health status in arthritis: The Arthritis
Impact Measurement Scales. Arthritis and Rheumatism 1980; 23(2): 146–152.
8. Meenan RF, Jason JH, Anderson JJ, Guccione AA, Kazis LE. AIMS2. The content and
properties of a revised and expanded arthritis impact measurement scales health status
questionnaire. Arthritis and Rheumatism 1992; 35(1): 1–10.
9. Bellamy N, Buchanan WW, Goldsmith CH, et al. Validation study of WOMAC: A health
status instrument for measuring clinically important patient relevant outcomes to anti-
rheumatic drug therapy in patients with osteoarthritis of the hip or knee. Journal of
Rheumatology 1988; 15: 1833–1840.
10. Bellamy N, Campbell J, Haraoui B, et al. Development of the Australian/Canadian AUS-
CAN Osteoarthritis (OA) Hand Index. Arthritis and Rheumatism 1997; 40(9) (Suppl): S110.
11. Likert R. A technique for the measurement of attitudes. Archives of Psychology 1932; 140:
44–60.
12. Huskisson EC. Measurement of pain. Lancet 1974; 2: 1127–1131.
13. Hochberg MC, Chang RW, Dwosh I, et al. The American College of Rheumatology 1991.
Revised criteria for the classification of Global Functional Status in rheumatoid arthritis.
Arthritis and Rheumatism 1992; 35: 498–502.
14. Lansbury J, Baier HN, McCracken S. Statistical study of variation in systemic and articular
indexes. Arthritis and Rheumatism 1962; 5(5): 445–456.
4
Paper 1
A table for the degree of involvement in

chronic arthritis
Author
Taylor D
Reference
Canadian Medical Association Journal 1937; June, 608–610
Summary
Taylor reported his recommendations regarding a classification table for chronic arthritis at
the Medico-Chirurgical Society of Montreal on 22 January 1937, while Demonstrator in
Medicine at McGill University. The Taylor Index was published 5 months later in the
Canadian Medical Association Journal. Given the relatively limited distribution of the journal at
that time, it is not surprising that the paper escaped world attention. In retrospect, Taylor’s
paper marks the beginning of quantitative measurement in rheumatology. His was the first
attempt to rationalize the categorization of musculoskeletal health status. In essence, Taylor
described, what he termed four divisions or ‘degrees’ varying from minimal to advanced
stages, in the form of a ‘table of degree’. The table was designed for both rheumatoid
(atrophic) arthritis and osteoarthritis (hypertrophic), although other applications were
envisaged. Indeed Taylor reported two components to the table, one based on x-ray changes,
the other based on clinical findings.
The radiographic component provides four gradations within each of which are separate
criteria for atrophic and hypertrophic arthritis. The terms commonly used today, such as ero-
sions, osteopenia, bone cysts, joint space narrowing, sclerosis, receive no mention. Each grade
is defined by one or more non-exclusive descriptors. The basic radiographic categories are:
Stage I: Soft tissue changes.

Stage II: Joint space is definitely altered.
Stage III: Destruction of joints.
Stage IV: Extreme widespread joint changes.
The clinical component likewise uses combinations of descriptors from the pain stiffness and
function domains, creating categorizations that are not mutually exclusive, and which are
based on symptom severities across several domains. The basic clinical categories are:
Stage I: Slight symptoms.

Stage II: Moderate symptoms.
Stage III: Severe symptoms.
Stage IV: Extreme symptoms.
The significant feature is the separation of the early or mild cases from those in the advanced stage.
No actual data on reliability, validity or responsiveness are included in the original report.
The Taylor Index is important as the first attempt to categorize patients with either RA or
OA according to a complex measure based on various radiographic and clinical features.
Key message
The key message was that the severity of musculoskeletal disease could be quantitated, and
that quantitative information could play a key role in the evaluation of therapeutic interven-
tions. Taylor noted that detailed clinical and laboratory studies were essential in problem
solving and might be facilitated by the use of his table.
Why it’s important

Taylor was probably the earliest pioneer, in musculoskeletal clinical metrology. His report
not only details the origins of the discipline, but also highlights the challenge of defining a
standard, in an era devoid of precedent. Taylor recognized that the rheumatism literature
was complex and controversial, and had failed to account adequately for inherent between-
patient and within-patient variation. He also recognized the disjunction that can occur
between radiographic and clinical findings.
While recognizing the descriptive nature of his table, Taylor also envisaged therapeutic
applications. He also recognized that his table would require modification for use in diseases
other than RA and OA. The report contains extensive description of the profiles of patients
who might fall into the different categories, and in places provides advice as to the preferred
method of caring for certain types of patients. It therefore provides a perspective on thera-
peutic management in the early 1900s.
Taylor recognized the constitutional reaction that can accompany RA, and suggested that it
be graded (mild, moderate, severe, extreme). Perhaps most importantly Taylor recognized that
through his table, the important observations made on patients differing in disease severity
could be more readily communicated to others, and thereby facilitate decision making.
Strengths
1. The first standard method for classifying patients with chronic arthritis according to their
radiographic and functional status.
2. Provided a methodology that was easily comprehended and readily applied.
3. Influential in the subsequent development of the Steinbrocker Functional Classification
and the ACR Functional Classification.
4. Recognized that radiographic and clinical features should be graded separately.
Weaknesses
1. From a clinimetric perspective, the Taylor Index combines measurements on different
clinical domains in the assignment of a single category. The descriptors of pain, stiffness
and functional capacity given are not mutually exclusive and therefore may conflict.
2. A lack of data in the original report on validity, reliability and responsiveness is a signifi-
cant limitation.
3. There are reasons to believe that the Taylor Index may have lacked responsiveness,
because of its scale construction. The presence of a restricted number of non-exclusive
categories would tend to mitigate against index responsiveness.
Relevance
The Taylor Index is relevant because it was the first of its kind, and formed the foundation of
the Steinbrocker Functional Classification and the ACR Functional Classification. In addition
Taylor’s publication provides insight into concepts and strategies prevalent in the early
1900s.
6
Paper 2
The pain chart

Author
Keele KD
Reference
Lancet 1948; 255: 6–8
Summary
Keele described ‘a simple method of minimizing errors of description by standardizing ter-
minology, and recording intensity of pain in a pain chart’. His work was inspired by Sir
Thomas Lewis and placed priority on the graphic representation of the time intensity of pain
by pain charts. Keele described five hierarchical categories of pain: nil, slight, moderate,
severe and agony. His seminal paper described temporal variation not only in articular pain,
but also in angina pectoris, gastric ulcer, tetany, tuberculous cystitis, Banti’s disease and rec-
tal cancer. Patients self-completed the pain charts which were subdivided according to the
24 hours of the day. The descriptors were explained to the patient and were also written on
the chart for reinforcement, and reference. Keele’s report contains information on the valid-
ity and responsiveness of the method but not its reliability. Indication is given regarding
potential applications of the method.
Key message
Keele was the first to suggest an adjectival or descriptive Likert scale based on the English
adjectives of degree for quantitating changes over time in pain intensity.

Pain is the key symptom in most rheumatic diseases. It is entirely subjective and only appre-
ciable to the sufferer. As such it represents an important but complex metrologic challenge.
The classification systems proposed by Taylor and Steinbrocker, encompassed pain but were
functionally based measures and of questionable responsiveness. Keele’s paper serves as a
bridge between the earlier work of Likert in 1932 (reference 11, Introduction) on the mea-
surement of attitude using descriptive terms, and the later work of Huskisson who intro-
duced the visual analogue scale into rheumatology in the 1970s (reference 12, Introduction).
To a variable extent, similar descriptors of none, mild, moderate, severe and extreme remain
in use today (reference 1, Introduction).
Scaling is a critical element of any health status measure since it is a major determinant of
the capacity of a measure to detect change (reference 1, Introduction). Keele had the fore-
sight to recognize the importance of being able to represent pain graphically on a pain
chart, in a way that is now commonplace in publications on interventional research in
rheumatology.
Figure 1.1 Keele described the

Five different types of pain scale early use of a Likert scale for
a Likert scale the measurement of pain
None Mid Moderate Severe Extreme severity in arthritis. The Likert
scale is just one of many
b Visual analog scale methods of recording the
No pain Extreme pain severity of pain – this
illustration shows five of them.
c Numerical rating scale (Adapted from Bellamy N,
Principles of outcome
0 1 2 3 4 5 6 7 8 9 10 measurement, in Klippel J,
Dieppe P (eds) Rheumatology,
d Continous chromatic analog scale
2nd edition, 1997, Mosby
International, London.
e Pain faces scale for children
Strengths
1. A simple method for quantitating longitudinal changes in pain intensity.
2. Established a precedent for future pain scales.
3. Emphasized issues of validity and responsiveness.
4. The diagnostic value of the pain chart was exemplified by observations on placebo injec-
tions in psychogenic pain.
Weaknesses
1. Report based on only seven selected non-random clinical cases.
2. Statistical methods not employed to establish validity, reliability and responsiveness.
3. Agony might be considered a qualitative aspect of pain rather than a simple measure of
intensity.
4. No diagnostic or demographic information were provided for the articular pain case,
which is believed to have been RA.
Relevance
The quantitation of pain by a reliable, valid and responsive method facilitates improved diag-
nostic, prognostic and evaluative activities, and provides a rational basis for developing scal-
ing priorities for unidimensional and multidimensional health status questionnaires in
diverse musculoskeletal diseases. Keele established a precedent for standardization in scaling
terms and demonstrated its applicability in several different and unrelated conditions.
8
Paper 3
Therapeutic criteria in rheumatoid arthritis

Authors
Steinbrocker O, Traeger CH, Battereman RC
Reference
Lancet 1949; 140: 659–662
Summary
Steinbrocker, et al. reported criteria for the Committee for Therapeutic Criteria of the New
York Rheumatism Association and the Subcommittee for Therapeutic Criteria of the
American Rheumatism Association. The report contains a description of three sets of criteria
for: (a) Classification of Rheumatoid Progression; (b) Classification of Functional Capacity,
and (c) Response of Rheumatoid Activity to Therapy. The criteria for Classification of
Functional Capacity are the only ones that remain in use and are generally referred to as the
Steinbrocker Functional Classification. Steinbrocker, et al’s classification of functional disabil-
ity in 1949 was similar to that of Taylor, which is not entirely surprising given that Taylor was
a member of the aforementioned Committee for Therapeutic Criteria. The Steinbrocker
Functional Classification contains four classes defined by whether an individual has complete
function, or has function adequate for normal activity, or has limited function, or is incapaci-
tated. The system has been used widely over the last half century, but is acknowledged to lack
some sensitivity for use as an outcome measure to detect differential change over time. This
problem has been partially addressed by revision of the wording of this simple four point
scale by a subcommittee of the American College of Rheumatology (reference 13,
Introduction). The Steinbrocker Functional Classification was endorsed by a major interna-
tional rheumatology association and provided a standard method for classifying functional
status in RA. It is of note that the report does not contain any data on the reliability, validity
or responsiveness of the classification system.
Key message
The key message is summed up by the authors themselves: the criteria ‘provide the common
language so sorely needed in the therapeutic investigation of rheumatoid arthritis’.

Prior to 1949, and with the exception of the Taylor Index, there had been no standard sys-
tem for classifying functional status. While Taylor made the first footsteps on the bare clini-
metric landscape, his contribution did not immediately receive endorsement by any major
international society. However, Taylor’s inclusion in the Criteria Committee of the New York
Rheumatism Association and the Criteria Subcommittee of the American Rheumatism
Association, is an indication that his earlier contribution had been noted and his opinion
sought and respected regarding the exact wording and formulation of the Steinbrocker
Functional Classification. The Steinbrocker Functional Classification allowed patients to be

classified as follows:
Class I: Complete. Ability to carry on all usual duties without handicaps.

Class II: Adequate for normal activities, despite handicap of discomfort or limited motion of
one or more joints.
Class III: Limited. Able to perform little or none of the duties of usual occupation or self care.
Class IV: Incapacitated, largely or wholly. Bedridden or confined to wheelchair; little or no
self care.
While developed for RA patients, the categories appear applicable to many forms of
chronic arthritis. The revision recommended by the ACR in a publication by Hochberg, et al.
(reference 13, Introduction), resulted again in four categories, but using different defini-
tions. The publication on the revised Steinbrocker Functional Classification provided data
on validity and reliability, but not on the key requirement for detecting differential change,
that of responsiveness (reference 13, Introduction).
Strengths
1. A simple standard method for classifying patients with rheumatoid arthritis according to
their functional status.
2. A method endorsed by a major international society and therefore likely to be employed.
3. Established a precedent for future functional scales.
Weaknesses
1. The classification systems proposed by both Taylor and Steinbrocker, are of questionable
responsiveness. Individuals categorized in Class IV in 1949 would have been unlikely to
improve, while relatively few patients with active RA would have fallen into Class I. This
would have effectively restricted the index to a two-point scale for many sufferers.
2. The lack of intermediate gradations between Class II and Class III would have further lim-
ited the capacity of the classification index to detect small but important alterations in
health status. The more recent availability of superior symptom-modifying and disease-
modifying therapies for RA may have altered the situation somewhat, but nevertheless, the
principle applies. For RA clinical trials and clinical practice applications, in which sensitiv-
ity to change is important, there are better instruments including the Health Assessment
Questionnaire (HAQ)(reference 6, Introduction), and the Arthritis Impact Measurement
Scales (AIMS, AIMS2) (references 7 and 8, Introduction).
3. The lack of data on the validity, reliability and responsiveness contained in the original
report of the Steinbrocker Functional Classification is a significant limitation. The authors
noted that the criteria were based entirely on objective information, although none was
provided in the report.
Relevance
The ability to classify patients according to standard criteria represented a significant devel-
opment. In particular the Steinbrocker Functional Classification was a response to the influ-
ence of subjective factors in quantitating patient status, the lack of separation between esti-
mates of disease activity and functional status, and the lack of a common language with
which to express the results of an assessment. Steinbrocker and colleagues defined a refer-
ence standard, which despite its limitations, remains in use today in a modified form.
10
Paper 4
Quantitation of the activity (manifestations) of

rheumatoid arthritis
Author
Lansbury J
References
1. Method for recording its systemic manifestations. American Journal of the Medical Sciences
1956; 231: 616–621
2. Recession of morning stiffness as patients go into remission. American Journal of the Medical
Sciences 1956; 232: 8–11
3. The maximum 5-minute Cutler sedimentation rate as an index. American Journal of the
Medical Sciences 1956; 232: 12–16
4. Area of joint surfaces as an index to total joint inflammation and deformity. American
Journal of the Medical Sciences 1956; 232: 150–155
5. A method for summation of the systemic indices of rheumatoid activity. American Journal of
the Medical Sciences 1956; 232: 300–310
6. Correlation of the systemic and joint findings. American Journal of the Medical Sciences 1957;
233: 375–378
7. A numerical method for summing up total deformity. American Journal of the Medical
Sciences 1958; 235: 154–156
Summary
Between 1956 and 1958 John Lansbury of Temple University School of Medicine in
Philadelphia, published seven of the most important papers in clinical metrology. Because of
their extreme value, they will be reviewed as a single entity. Lansbury was professor of clinical
medicine when he published methods for the quantitation of the activity of RA. Clearly he
was a master of his art, establishing not only a standard in the field, but noting many of the
same complexities and conflicts that metrologists still seek to address adequately at the pre-
sent time. His first paper defined 16 domains for measurement: rest pain, motion pain,
morning stiffness, diurnal gelling, fatigue, ASA tablets/day to control pain, 4 pm tempera-
ture, basal body weight, grip strength, steps up/down, timed rises from chair, haemoglobin,
ESR, serum albumin, serum globulin and CRP. Lansbury also noted the convenience of
using a rubber stamp form for recording purposes.
This first paper is conceptual. While it contains no data, it provides the foundation for the
remaining papers all of which contain quantitative information. The measurement system
was based on the premise that vague impressions conveyed in the general course of a clinical
interview were deficient and liable to error. By breaking down the measurement problem
into its basic components, and measuring each in relevant units, a multidimensional profile
could be constructed, that conveyed the detail in a standardized and structured fashion, and
facilitated communication and decision making.
The second paper provides a method for quantitating morning stiffness based on its dura-
tion rather than its intensity. Lansbury contended that most patients could distinguish stiff-
ness from pain (an issue that remains controversial to the present day). He noted an associa-
tion between morning stiffness and Cutler sedimentation rate which he quantitated (con-
struct validity), although he did not express the association using a correlation coefficient.
Lansbury described the dynamic profile of morning stiffness in six patients experiencing a
treatment-induced complete remission, and thereby demonstrated the responsiveness of the
measure. Interestingly, he regarded morning stiffness as a ‘valid index to the total amount of
systemic or fibrositic activity of RA in the cases studied’.
The third paper examined the value of the Cutler sedimentation rate as a quantitative
measure of rheumatoid activity. No claim was made for superiority over the methods
described by Westergren and Wintrobe, and the time-consuming nature of this method,
requiring taking readings every 5 minutes, was duly noted. Lansbury described the curvilin-
ear dynamics of the sedimentation rate in 10 patients experiencing a complete remission,
and demonstrated the responsiveness of the measure. He observed between-subject variabili-
ty and described the average sedimentation rate in a further 100 patients.
The fourth paper is extremely important since it describes his development of an articular
index, the Lansbury Index. With the collaboration of an anatomist Professor Donald Hart,
Lansbury meticulously measured the surface area of the bony surfaces of each individual
joint by exactly applying aluminium foil to ‘the surface which in life is covered with articulat-
ing cartilage’. For each joint the foil was then weighed repeatedly and the average weight (to
two decimal places), referred to the weight of a standard square centimeter of the foil used,
and the surface area of each joint deduced. It should be noted that all the measurements
were conducted on a single skeleton of unknown morphology. Nevertheless, the relative
weights would probably be the same or similar in different skeletons. The total joint area in
the skeleton studied was 1090.74 cm2. From the area Lansbury simply rounded to the nearest
whole number and in Table 2 of the publication gave values for individual joints. The total
amount of inflammation was described by summing the individual values for all affected
joints. Using this approach Lansbury created not a direct measure of inflammation, but an
index of joint size that could be used to estimate disease activity. In Table 4 of the publica-
tion a four point grading system, based on the joint values or weights, is illustrated.
Lansbury divided the total score for each joint into four equal grades: Grade I: Minimal
(0–25%), Grade II: Slight (26–50%), Grade III: Moderate (51–75%) and Grade IV:
Maximum (76–100%). The value for Grade IV was the actual value or weight for that joint
that had been determined by the aluminium foil method. Lesser grades were direct percent-
ages as outlined previously. Lansbury was an astute researcher and cautioned that Grade IV
inflammation did not imply four times greater inflammation than Grade I. He also appreciat-
ed that while clinicians might agree on what constituted minimal or extreme inflammation,
they would be less likely to agree on the intermediate grades.
The fifth paper concerns a method of summating the systemic indices of rheumatoid activ-
ity. With humility, Lansbury noted that he was aware that the method had ‘been applied to
too few cases to permit a conclusive statement to be made as to its ultimate value’. He went
on to describe a method of compiling a single composite score based on five subscores
(Anaemia, ESR, Grip strength, Fatigability and Aspirin need). Lansbury reasoned that to be
acceptable in a composite index, each component must be ‘present in the majority of cases
and must exhibit statistically significant regularity of trend which correlates with the general
trend of the disease’. He proceeded to describe these statistical relationships, the exact meth-
ods of ascertainment of the measurements, and highlighted some of the operational difficul-
ties of the methods. Lansbury rationalized that the complete absence of abnormality on any
of the indices constituted a clinical remission. The composite score was calculated as follows:
The value for the average degree of severity in a series of patients was assigned a value of 100,
so that gradations above and below that point could be expressed in percentages. Equivalent
percentages for the full range of probable observed data were displayed in a table for ready
reference. The status of an individual patient was defined by summing the percentages
describing each of the component elements into a so-called Index of Systemic Activity. The
12
exact percentages are shown in Table 2 and a worked example in Table 3 of the original pub-
lication. The responsiveness of the index was established in five cases of RA experiencing a
remission.
The sixth paper examines the correlation between Articular and Systemic Indices. Based
on 80 unselected cases a correlation of 0.78 (p = 0.01) was observed. Lansbury also provided
evidence for parallelism, that is synchronous fluctuations, between the Articular Index and
Systemic Index scores in a single patient. He recommended that the indices be kept separate
and cautioned that parallelism might not be invariable.
The seventh paper concerns a numerical method for summing up total deformity. The so-
called Motion Index was based on the Articular Index, and expressed total lost motion as a
percentage of total possible lost motion. Lansbury clearly differentiated between this method
of quantitating joint motion and functional capacity in the social sense. He acknowledged
that the latter depended on other factors such as disease distribution and activity, the
patient’s courage and type of employment. The Motion Index score was calculated from the
sum of all contributing joints. For each joint the difference between the normal and the
observed motion (in degrees) was multiplied by the joint size in square centimeters. The val-
ues for 28 knuckles, 24 small toe joints and two of all the peripheral movable joints were
summed and then divided by 1000. In order to save time, Lansbury accepted that range of
movement in small joints might be visually estimated rather than be measured goniometri-
cally. He envisaged that the Motion Index would not be used routinely, but rather be used in
long-term studies and rehabilitation environments.
Key message
Lansbury demonstrated that disease activity in RA was amenable to quantitation using stan-
dard methods. Although no longer in use, his Articular Index and Systemic Index were con-
ceptually advanced.

Lansbury’s seven papers represent a conceptual leap in musculoskeletal clinical metrology.
He had a comprehensive understanding of the multidimensional and dynamic nature of the
measurement problem in RA. The Articular Index provided a rational basis for weighting
and aggregating together unidimensional information from joints of different size, while his
Systemic Index attempted to address the complexities of weighting and aggregating together
diverse information from different domains, measured on scales differing in length. The
weighting and aggregation of multidimensional data remain an on-going challenge.
Strengths
1. The strength of these papers lies in Lansbury’s capacity to convey the essence of the
metrologic problem succinctly and with balance. He realized the complexity of the prob-
lem, was always cautious regarding the generalizability of his observations and acknowl-
edged the necessity for more extensive evaluation. While none of his indices remains in
use, they offered partial solutions, and provided a foundation for later developments.
2. Lansbury understood the need to establish the validity and responsiveness of his methods.
As a consequence his papers contain more statistical information than others from the
same period.
Weaknesses
1. Lansbury’s work was constrained by the existent level of understanding regarding the
nature of RA.
2. His work preceded the personal computer and the general use of statistical methodology
referable to evaluating the clinimetric properties of measuring instruments. Nevertheless,
he progressed the field using a combination of refined observational skills and intellectual
brilliance.
3. Perhaps the only area of weakness was in his treatment of the pain and function dimen-
sions, which he approached quite superficially by modern standards. While Lansbury’s
early work was based on complex concepts but simple statistics, his later work showed an
increasing sophistication with respect to his use of statistical methodology (reference 14,
Introduction).
Relevance
Lansbury’s papers are essential reading. For students they provide the necessary background
for interpreting progress in the assessment of RA throughout the remainder of the century.
For the developers of new measurement techniques they provide reassurance that others
have traveled the same path, and in some instances encountered the very same challenges.
Finally, they serve as a record of the achievements of an outstanding clinical metrologist and
physician.
15
CHAPTER 2
Imaging
Introduction
Imaging in medicine began soon after the production of the famous x-ray picture of the nor-
mal hand of Roentgen’s wife in 1895. The first decades of the 20th century was an epoch
when radiology was an issue of concern for every open-minded physician rather than a mat-
ter for a few specialized radiologists. At that time the population statistics in western coun-
tries showed that there were very few older people compared to the present population
demography. For example in the city of Malmö, there were only 5997 people aged over 65
years in 1920 (5% of the population), and 129 (2.1%) of these people had been examined
radiologically (according to my survey of the register of the Radiological Department from
1920, Malmö General Hospital). In comparison, in 1995 the corresponding figures were
49 616 persons over 65 years (20% of the population), and 24 057 (48.5%) of them have had
skeletal radiological examinations. It is thus not surprising that diseases of older people
(such as osteoarthritis (OA)) were not well reported during the first years of radiology. In
the patient material of 1920 in Malmö, very few had the diagnosis of arthritis, and in those of
1945, before the introduction of antibiotics, most cases of arthritis were probably of infective
origin.
Imaging has an important place in the diagnosis of rheumatic diseases. I have tried to find
some of the first published papers that demonstrated the characteristic imaging features of
different kinds of arthritis, starting with x-ray papers from the beginning of the 20th century
(Papers 1 and 2), and going on to examples of important applications of the newer tech-
niques of computerized tomography (CT) and and magnetic resonance imaging (MRI)
(Papers 3 and 4). It has not been easy. The early papers seldom had more than one author
and the papers were written in a different style from that of today: the introduction was inter-
mingled with the discussion and case reports were also mixed with discussion points. There
were never any presentations of large groups of patients in the early radiological (or in other
medical) papers, rather an eagerness to be the first one to present cases with this new diag-
nostic method. That eagererness was repeated during the 1970s and 1980s when CT and
MRI were introduced respectively. One of the differences between these papers, and the
early ones on the plain radiograph, is that the number of authors of a paper today are nearly
as many as the number of patients studied!
Imaging is also of value in sorting out some of the complications of the rheumatic dis-
eases, and its impact has been particularly important in the recognition of cervical subluxa-
tion in rheumatoid arthritis (RA) (Paper 5). Finally, imaging can also help with assessing
severity and measuring the progression of the rheumatic diseases, so I have also included the
classic papers introducing systems for assessment of the two commonest forms of arthritis,
RA (Paper 6) and OA (Paper 7).
16
Paper 1
The skiagraphy of rheumatoid arthritis

Authors
Bowker GE, Lindsay J
Reference
Edinburgh Medical Journal 1907; X:317–320
Summary
Earlier reports on RA have discussed the probability of a specific bacillus responsible for RA.
Some infective process may give rise to a general toxaemia and so influence the joint struc-
tures that it could produce the features of RA. Many infective processes like gonorrhoea or
rectal ulcers may cause acute RA with the same clinical findings, but with the help of x-rays it
is possible to differentiate among these cases. In cases of acute RA there has never been any
evidence of any change in the configuration of the bony parts of the joints involved but soft
tissue swelling may occur. Two case reports, one concerning a man with gonorrhoeal arthri-
tis and one child with acute RA confirm these observations.
In cases of chronicity, the disappearance of cartilage allows a closer approximation of the
joints. In further advanced cases the softened joints of the proximal interphalangeal joints
are eroded, as are the heads of the metacarpals and the carpal bones. With progression, atro-
phy of the shafts will appear and the destruction may lead to cupping, and finally a trumpet-
shaped appearance may appear as shown in the x-ray illustration. Finally bony ankylosis fre-
quently supervenes.
Related references (1) Osgood RB. Differential diagnosis of the chronic non-tubercular
joint diseases by means of the roentgen ray. American Quarterly
Roentgen 1906–07; 4:1. (Summary in Röfo 1908; 12:77).
(2) Reuss E. Über einen fall von knochenatrophie nach gelenkreuma-
tismus (One case of bone atrophy associated with joint rheuma-
tism). Röfo 1912–1913; 19:430–437.
Key message
Joint arthritis can be divided into chronic or RA with x-ray changes (Figure 2.1) and acute
(reactive) arthritis without x-ray changes.

This and the related references were amongst the first to show that radiology could discrimi-
nate between different kinds of arthritis. It is interesting to note, from the summary, that at
this time RA was regarded as likely to be infective in origin. However, the typical x-ray
changes of the disease are well described and illustrated, including the erosions.
Imaging 17
Figure 2.1 A radiograph of the hand in RA

showing a characteristic bone erosion at
the index finger metacarpophalangeal joint.
Strengths
1. The findings are supported with reproduction of the x-rays.
2. The main features of rheumatoid radiographs, including the characteristic erosions, are
well documented.
Weakness
The observations are grounded in only three cases.
Relevance
The typical radiographic features of RA are now well described, but at this time it was not
clear that RA was separate from other forms of arthritis. These observations help show con-
temporary rheumatologists that RA has distinctive features.
18
Three early descriptions of the radiology of arthritis

Paper 2a
Typical alterations of the bones in gout

Author
Köhler A
Reference
Archives of the Roentgen Ray 1912; 17:330–333
Summary
According to Huber (1) the bubble-like cavities in the phalanges of the fingers in gout ought
to be filled with uric acid salts. These changes would thus not be due to atrophic changes but
to the substitution of urates for the natural phosphates of the affected bone. Five cases are
presented with x-rays. Typical changes at the articular ends of the phalanges are sharply
defined transparent areas. The symmetrical arrangement of the transparent foci seems to
point to some connection between the position of the foci and the distribution of the blood
vessels. The same destructive lesions may be found in the foot, espcially in the first metatarsal
joint. Typical changes have so far not been found in hundreds of x-rays of the larger joints in
patients with gout. Along the shafts of the long bones an irritation or stimulus may produce a
gouty deposit and erosion of the bone, but the appearance is not diagnostic or typical of
gout.
Related reference (1) Huber X. Zur verwerthung der röntgen-strahlen im gebiete der
inneren medicin. Deutsche Medizinische Wochenschrift 1896;
22:182.
Imaging 19
Paper 2b
A case of arthropathia psoriatica

Authors
Ström S
Reference
Acta Radiologica 1920–21; 1:21–25
Summary
This case-report concerns a 50-year-old man with psoriasis for 30 years in whom treatment
with arsenic and chrysarobin had only led to temporary improvement. At the age of 28 years
he got thickening of the nails and swelling of the thumbs and big toes, and later on the little
fingers and all toes became swollen and shortened without any real pain. He was referred for
x-ray with the suspicion of his having syphilis. The x-ray showed a total destruction of several
of the affected joints with a tapering towards the ends of the phalanges. The sesamoid bones
of the thumbs were three times enlarged with irregular form. Spurs, as in OA, were occasion-
ally found. In some of the toes a total phalangeal destruction was found, resembling that
seen in cases with Raynaud’s disease. In all the toes there was some bony destruction.
Different aetiologies have been proposed in psoriasis, and the findings here support rather
the theories of neuropathic origin or that of a disturbance in the internal secretion.
Related reference (1) Belot X, Chaperon MMJ. Radiographies de lesions et deformations

dans le psoriasis arthropathique. Bulletin Memoires Soc Radiol Med
Paris 1909; 1:269–271.
20
Paper 2c
Abnormal calcium deposition inside the knee,

a contribution to the question of the primary
‘meniscopathy’
Authors
Werwath K
Reference
RÖFO 1928; 37:169–171
Summary
A 45-year-old, earlier healthy man attended the surgical clinic because of pain and hydrops
of the left knee. X-ray revealed calcification of the cartilage and menisci. At operation the
diagnosis of calcium deposition was confirmed. A meniscectomy was performed and synovial
material was cut out. Microscopy showed spots of calcification of the cartilage as well as
scanty cartilage cells and a predominance of fibrous tissue. The findings differed partly from
that of OA. The differential diagnosis ought to be a metabolic disturbance or a post-traumat-
ic disorder, but there were neither trauma nor acute arthritis in the anamnesis. The probable
diagnosis ought to be a disturbance of the calcium metabolism with secondary deposition in
the cartilage.
Related reference (1) McCarthy DJ, Haskin ME. The roentgenographic aspects of
pseudogout (articular chondrocalcinosis). An analysis of 20
cases. American Journal of Roentengenology 1963; 90:1248–1257.
Key message of the three papers

The key message behind these papers is that different types of rheumatic disease lead to
characteristic changes on the plain radiograph (Figure 2.2).
Why they are important

These are early descriptions of the radiological features of gout, psoriatic arthritis (PsA) and
chondrocalcinosis respectively. The typical changes found on a plain radiograph in each of
these conditons are described and illustrated. The important finding coming from a combi-
nation of such studies is that each rheumatic disease results in quite different changes in the
bones and spaces of synovial joints, allowing the plain radiograph to be used as a powerful
diagnostic tool.
Strengths
1 These papers are amongst the first to describe and illustrate the different radiographic
changes of the rheumatic diseases.
2 The descriptions and illustrations are often good, particularly in the gout paper.
Imaging 21
Figure 2.2 Radiographs showing the typical

gouty erosion at the first metatarsophalangeal
joint (left) and chondrocalcinosis in the knee
joint (right).
Weaknesses
1. All papers are of single or a few case reports.
2. The speculation about the nature of the conditons is sometimes misplaced.
Relevance
The authors provide early descriptions of the plain radiographic changes that characterize
the different rheumatic diseases.
22
Paper 3
Computerized tomography of sacroiliitis

Authors
Carrera CT, Foley WD, Kozin F, Ryanand L, Lawson TL
Reference
American Journal of Roentengenology 1981; 136:41–46
Summary
Sacroiliitis is often difficult to diagnose using conventional radiographs and radionuclide
scanning due to low sensitivity and/or specificity for the early diagnosis of sacroiliitis.
Computed tomography was used in a study of eight normal volunteers and 20 consecutive
patients with a clinical suspicion of sacroiliitis in a double-blind pilot study. The gantry was
tilted 30o caudally, and 5mm CT images as well as the plain films were studied for joint nar-
rowing, subchondral sclerosis, erosion and ankylosis.
Three patients did not fulfil clinical criteria of spondyloarthritic syndromes and appeared
to be cases with degenerative joint disease, old fracture, and osteitis condensans ilii respec-
tively. Twelve among the other 17 patients had CT changes diagnostic of sacroiliitis but only
five had changes at the conventional radiographic examination. Another four patients had
equivocal plain film changes and three were normal. The radiation dose was 2.5 times higher
with CT but the testicular dose was zero compared to 0.08 rad with plain films. There was no
interobserver inconsistency in the interpretation of the CT images but some uncertainty in
that of the plain films.
Related references (1) Calin A. Abuse of computed tomography. Arthritis and Rheumatism
1982; 25:147–148.
(2) Murphey MD,Wetzel LH, Bramble JM, et al. Sacroiliitis: MR imag-
ing findings. Radiology 1991; 180:239–244.
Key message
CT is superior to conventional radiography in diagnosing arthritis of the sacroiliac joints and
the gonadal dose is low with the described technique.

The sacroiliac joints are orientated oblique to both the conventional coronal and sagittal
planes of the body, and are thus not easy to depict with x-ray. Conventional x-ray tomography
with several exposures was formerly a good diagnostic help, but was giving the young patients
a high dose of gonadal radiation. This technique with tilting of the gantry is still the best one
when examining the sacroiliac joints with CT. MRI is more sensitive, but not more specific.
Strengths
1. This was a double-blind pilot study (the radiologists were not blind but blinded of the pos-
sibilities of the new technique).
2. The patient material is compared to a normal material.
Imaging 23
Weakness
There were only eight normal controls.
Relevance
This paper shows how to use the new technology with a lower radiation dose to the patient.
24
Paper 4
Early detection of carpal erosions in patients with

rheumatoid arthritis: A pilot study of magnetic
resonance imaging
Authors
Gilkeson G, Polisson R, Sinclair H, Voglier J, Rice J, Caldwell D,
Spritzer C, Martinez S
Reference
Journal of Rheumatology 1988; 15:1361–1366
Summary
Standard radiographs taken early in the course of joint arthritis are usually normal or show
unspecific changes. This pilot study consisted of 10 patients with classical or definite RA and
their radiography and MRI were evaluated by radiologists without knowledge of any connec-
tion between the x-ray and MRI films. The MRI appearance of the wrists was also compared
with 39 non-rheumatoid patients. MRI was superior in detecting synovial inflammation and
erosions in the form of pannus invading carpal marrow. Early erosions were found on the
junction between the hamate, capitate and bases of the 3rd and 4th metacarpals. If the
expense of MRI was not so great it could be a valuable tool in therapeutic trials.
Related reference (1) Ostergaard M, Hansen M, Stoltenberg M, Lorenzen I. Magnetic

resonance imaging-determined synovial membrane volume as a
marker of disease activity and a predictor of progressive joint
destruction in the wrists of patients with rheumatoid arthritis.
Arthritis and Rheumatism 1999; 42:918–929.
Key message
MRI is superior to standard radiography in detecting early RA.

The use of MRI in medicine started at the beginning of the 1980s. For the diagnosis of
rheumatic diseases it is a sensitive but expensive technique. Even if the time that the patient
has to spend in the machine is reduced in the future, film-reading and interpretation will
still take a lot longer than with conventional x-rays. However, the cost of MRI is only a frac-
tion of that of some of the new interventions being used to treat RA, so MRI may prove cost-
effective in assessing such therapies and monitoring patient progress (1).
In patients with inflammatory arthritis suspected of having RA, it can be very valuable to
obtain early evidence of the characteristic anatomical changes, confirming the diagnosis and
allowing early use of suppressive therapy. However, it takes several months for changes to
appear on a plain radiograph. This paper shows the MRI changes of RA appear earlier in the
course of the disease.
Imaging 25
Strength
The findings in patients with suspected RA were compared to a normal material.
Weakness
A pilot study with only 10 patients.
Relevance
Early evidence of the potential value of MRI in the early detection of joint damage in RA.
26
Paper 5
Rheumatoid arthritis with death from

medullary compression
Authors
Davis FW Jr, Markley H
Reference
Annals of Internal Medicine 1951; 35:451–454
Summary
A 58-year old woman with a 25 year history of classical RA and confined to a wheel chair for 6
years was admitted to hospital because of dyspnoea. Three months prior to admission she
had been treated for an upper respiratory infection with dysphagia and stridor. A laryn-
goscopy revealed limitated abduction of the vocal cords. Her neck was unduly short. She
gradually deteriorated with increasing stridor, dysphagia and dyspnoea and she died due to
what was interpreted as acute bulbar failure. An autopsy revealed an intruding odontoid
process through the foramen magnum, impinging on the inferior surface of the medulla. A
perforated oesophageal ulcer with subacute osteochondritis of cricoid cartilage and wide-
spread RA deformity were also found.
Related references (1) Mikulowski P, Wollheim FA, Rotmil P, Olsen I. Sudden death in
rheumatoid arthritis with atlantoaxial dislocation. Acta Medica
Scandinavica 1975; 198:445–451.
(2) Redlund-Johnell I. Dislocations of the cervical spine in rheumatoid
arthritis. Thesis, Lund University, 1984.
Key message
Atlanto-axial and other cervical dislocations are important sources of morbidity in RA and
may even be a threat to the life of the patient.

Inflammatory processes such as retropharyngeal abscess as well as odontoid fracture or con-
genital anomalies of the craniovertebral region were well known causes of fatal medullary
compression, but this was the first reported case due to rheumatoid involvement of the cran-
iovertebral region. It was also the first case of laryngeal arthritis to be reported.
Strength
Clear description of cervical disease in RA leading to upward subluxation of the odontoid
peg and consequent medullary pressure.
Imaging 27
Weaknesses
1. A single case description.
2. The autopsy findings of oesophageal ulcer with cricoarytenoid arthritis were incorrect and
were cricoarytenoid arthritis with per continuum spreading into the oesophagus.
Relevance
This paper and the related references led to important changes in radiological examinations
in RA, especially in the pre-operative evaluation of the neck, which subsequently included
lateral examinations in flexion and extension to demonstrate cervical dislocations.
28
Paper 6
Radiographic evaluation of rheumatoid arthritis

and related conditions by standard reference films
Authors
Larsen A, Dale K, Eek M
Reference
Acta Radiologica (Diagnosis) 1977; 18:481–491
Summary
A review of radiographic evaluation of RA is given. Standard reference films are introduced
for evaluation of RA and related conditions in the extremity joints. In this system numerical
evaluation of arthritis is given for individual joints in a patient.
Related references (1) Larson A. Radiological grading of rheumatoid arthritis, an interob-

server study. Scandinavian Journal of Rheumatology 1973; 2:136.
(2) Larson A, Thoen J. Hand radiography of 200 patients with
rheumatoid arthritis after an interval of one year. Scandinavian
Journal of Rheumatology 1987; 16:395–401.
(3) Larson A. How to apply the Larson score in evaluating radiograph-
sof rheumatoid arthritis in longterm studies. Journal of
Rheumatology 1995; 22:1974–1975.
(4) Kellgren JH, Lawrence JS. Radiological assessment of rheumatoid
arthritis. Annals of Rheumatic Diseases 1957; 16:485–493.
(5) Steinbrocker O, Traeger GH, Battermann RC. Therapeutic criteria
in rheumatoid arthritis. Journal of the American Medical Association
1949; 140:659–662.
(6) Sharp JT, Lidsky MD, Collins LC, Moreland J. Methods of scoring
progression of radiologic changes in rheumatoid arthritis.
(7) Van der Heijde DM, van Riel PL, Nuver-Zwart HH. Effects of
hydroxychloroquine and sulphasalazine on progression of joint
damage in rheumatoid arthritis. The Lancet 1989; 1:1036–1038.
Key message
With the help of standard reference films it is possible to evaluate the severity of joint dam-
age of peripheral joints in RA.
Imaging 29

Radiographic evaluation of the severity of RA began in 1949 when Steinbrocker, et al. divided
the changes into four stages according to occurrence of osteoporosis, cartilage and bone
destruction and ankylosis. That method only took consideration of a single joint, not the sta-
tus of, for example, the whole hand. The present method with standard reference films
reflects the severity of the whole hand, foot and other regions. An interobserver analysis
demonstrated that nine out of 10 arthritis films were staged uniformly by six observers. The
method is not specific for RA and can also be used in psoriatic arthropathy and ankylosing
spondylitis. Some problems may occur in cases with OA secondary to arthritis. The method is
purely a radiographic evaluation and should not be considered as a general measure of the
severity of disease.
The ‘Larson method’ described in this paper became one of the standard methods for
evaluation of joint damage and of radiological progression of RA. This has became very
important in the evaluation of new treatments thought to slow disease progression, as well as
for studies of the natural history of the disease. Larson has continued to refine and develop
the methodology, as outlined in the related references. The main alternative is the Sharp
method, and more recently that described by van der Heijde, et al. (8).
Figure 2.3 Representation of the

Larson grades used to assess
rheumatois arthritis (as illustrated
in Larson (3).
Grade 0 Grade 1 Grade 2
Grade 3 Grade 4 Grade 5
Strengths
1. The method described is simple and easy to learn.
2. The reliability of the method is high.
Weaknesses
1. The most severe form of joint destruction seen in RA (arthritis mutilans), is not included
in the Larson grades.
2. Some progression of joint damage can occur without a change in the grade.
Relevance
The described method is still one of the standard techniques used to assess the radiographic
progression of RA. Recently, MRI based techniques have been introduced, and they may
supercede the Larson technique for the evaluation of new disease-modifying anti-rheumatic
drugs.
30
Paper 7
Radiological assessment of osteo-arthrosis

Authors
Kellgren JH, Lawrence JS
Reference
Annals of Rheumatic Diseases 1957; 16:494–502
Summary
A series of 510 x-rays from 85 persons in the age group 55–64 years chosen at random from
an urban population was graded for osteo-arthrosis by two observers on four occasions to
determine the extent of observer difference. Standard films for four grades of osteo-arthrosis
for each of eleven joints were chosen. A significant correlation between the two observers was
obtained for all joints except the wrist. The stimates of prevalence, however, varied widely
because of the cumulative effect of observer bias (+/-31%). It is concluded that comparison
of prevalence estimates by different observers could have little value in population studies.
Two readings by the same observer gave only a slightly better correlation on the reading of
individual x-rays, but, by excluding observer bias they gave a much closer estimate of preva-
lence (+/-5%). These two readings however, differed substantially from the mean value for
all readings (-8% and -17%). A combined reading by two observers reduced the influence of
personal bias and differed little from the mean value (-3%). It is suggested that, where possi-
ble, in all population studies the x-rays should be read by either the same observer or prefer-
ably by two observers in consultation.
Key message
It is possible to grade x-rays for OA into five categories in a reproducible manner, appropri-
ate for population studies of prevalence.
Related references (1) Cooper C. Radiographic atlases for the assessment of osteoarthri-
tis. Osteoarthritis and Cartilage 1995; 3(suppl. A):1–2.
(2) Spector T, Cooper C. Radiographic assessment of osteoarthritis in
population studies: whither Kellgren and Lawrence?
Osteoarthritis and Cartilage 1993; 1:203–6.
(3) Buckland-Wright JC. Quantitative radiography of osteoarthritis.
Annals of Rheumatic Diseases 1994; 53:268–275.
Imaging 31

The Kellgren and Lawrence system of grading an OA joint into one of five grades has been
the ‘gold-standard’ for reporting and assessing the severity of OA for over 40 years. It is a sim-
ple system applicable to population studies and to patient groups, and has been used to
assess progression and outcome as well as the presence (or absence) of OA and its severity.
There have been however, a number of criticisms of the system, which assumes linear pro-
gression between grades and lumps together bone changes such as osteophyte formation
with joint space narrowing, which is a surrogate for cartilage loss. These criticisms have led to
the development of new atlases which allow individual features to be scored and to the intro-
duction of quantitative methods (see related references). However, in spite of that, most cur-
rent studies of OA still use or refer to the Kellgren and Lawrence system.
Strengths
1. A simple, plausible system which has four grades of severity and which has an atlas to help
reproducibility.
2. Widely used.
Weaknesses
1. The different grades are not well described or clear.
2. There are slightly different descriptions in different publications from the authors.
3. Assumes linearity of progression between grades.
4. Assumes that bone and cartilage changes go together.
Relevance
Introduction of the system which has dominated OA research ever since. However, it could
now be overtaken by quantitative MRI.
33
CHAPTER 3
Serological tests
Josef Smolen
Introduction
It is now difficult to conceive of a time when systemic rheumatic disorders were not thought
to have an immunological component. However, prior to the Second World War this was the
case. Infections (such as chronic tonsillitis) were thought to be the likely causes of conditions
such as rheumatoid arthritis (RA). Furthermore, at that point, there was little or no differen-
tiation made between RA and other systemic rheumatic disorders, and no blood tests avail-
able to aid diagnosis. The discovery of a variety of factors (subsequently defined as auto-anti-
bodies) in the serum of patients with rheumatic diseases changed all that. The discovery of
these antibodies also led to the development of diagnostic serology, and to a continuing new
chapter of research into the etiopathogenesis of rheumatic diseases.
The starting point was the discovery of rheumatoid factor. Waaler’s important paper of
1940 (Paper 1) provides the first description of the sheep cell agglutination test for the iden-
tification of this anti-immunoglobulin. The test was subsequently developed independently
by Rose, and the ‘Waaler-Rose test’ became the first diagnostic assay for an antibody used in
rheumatology. Subsequent technological developments have changed and refined our ways
of detecting and measuring rheumatoid factor, but the test remains widely used today.
One of the most active and productive areas of rheumatological research over the last 50 years
has centred on the classification and etiopathogenesis of the connective tissue disorders. This all
started with Hargreaves’ discovery of the ‘LE cell’ in 1948 (Paper 2). This led to the understand-
ing that in systemic lupus erythematosus (SLE) there were auto-antibodies directed against
nuclear antigens. This test was initially also of some value diagnostically, although it has subse-
quently been overtaken by the more specific and sensitive assays of individual nuclear antigens.
Following these early discoveries, there was then an explosion of serological work in the
1960s, 70s and 80s, leading to the huge spectrum of diagnostic auto-antibody tests available
to us today, and stimulating profitable research into the mechanisms of tissue damage in
connective tissue disorders. Some of the most important of these discoveries have been
picked out. The first non-DNA-non-histone antigen to be identified was the Sm factor found
by Tan and colleagues in 1966 (Paper 3). In 1972 Sharp and colleagues described extractable
nuclear antigen (ENA) and linked it to a specific phenotype of connective tissue disease
expression (mixed connective tissue disease). This was the first disease to be defined serolog-
ically (Paper 4). Subsequently a group of other antibodies were discovered which not only
aid the diagnosis and classification of connective tissue diseases, but which also provide
direct links to tissue damage. These include the Ro and La antibodies found in SLE and
Sjögren’s syndrome (Paper 5), Scl-70, the first antibody linked to scleroderma (Paper 6) and
anti-Jo-1, the first dermatomyositis specific antibody (Paper 7).
Technology has been an important driver of this area of discovery. At the time of writing
we are moving into another technologically-derived new era of medicine, the ‘post-genome’
era, and with this change we are starting to see a shift of interest from the serology to the
genome and the production of the proteins that for the antigens and antibodies that are so
important in systemic rheumatic diseases.
34
Paper 1
On the occurrence of a factor in human serum

activating the specific agglutination of sheep
blood corpuscles
Author
Waaler Erik
Reference
Acta Pathologica Microbiol Scandinavica 1940; 17:72–l88
Summary
Human serum can augment the agglutination of sheep red blood cells (SRBC) coated with
anti-SRBC serum and can lead to agglutination of SRBC coated with sub-agglutinating doses
of the anti-SRBC serum. The factor, which was independent of natural heteroagglutinins and
thus depended on the presence of agglutinins to SRBC, was called agglutination activating
factor. Although most human sera contained little or no such factor, some sera were highly
active. Both in terms of titre as well as frequency, the most active sera were from rheumatoid
arthritis (RA) patients, although 50 of 77 patients did not have the factor and sera from a few
patients with non-rheumatic diseases also carried this activity. The activity was not associated
with the sedimentation rate nor the presence of anti-streptococcal agglutinins and was ther-
mostable to heating to 60°C for 30 min. The agglutination activating factor was contained in
the globulin and not the albumin fraction of serum.
Related reference (1) Rose HM, Ragan C, Pearce E, Lipman MO. Differential agglutina-
tion of normal and sensitised sheep erythrocytes by sera of patients
with rheumatoid arthritis. Proceedings of the Society for Experimental
Biology and Medicine 1948; 68:1–6.
Key message
A description of a serum factor (later named rheumatoid factor (RF)) in patient and normal
human sera, found most frequently and in highest titres in patients with RA.
Serological tests 35

Prior to this publication RA was not regarded to be an autoimmune disease and there were
no serological tests available to support the differentiation of RA from other rheumatic con-
ditions. Thus, this paper became the starting point for diagnostic tests for RA as well as the
search into the immunological etiopathogenesis of the disease. Although the first descrip-
tions of the reported ‘agglutination activated activity’ was provided by Olsen and by Meyer,
Waaler’s major contribution was its demonstration in human sera and the observation of its
link to RA. Several years later, Rose, et al. (without quoting or knowing the paper by Waaler)
developed the method for the sheep cell agglutination test widely known and practised until
recent years as Waaler–Rose test. In contrast to Waaler, who felt that ‘the reaction can not be
of any diagnostic value’, Rose, et al. found a predominance of high titres in patients with RA
as well as an association of the magnitude of the reactivity with RA disease activity and sug-
gested that ‘the sheep cell differential agglutination test may be of value in determining the
activity status of RA’ and may ‘be helpful as an aid in the differential diagnosis between RA
and other diseases with arthritic manifestations’. Singer and Plotz then exchanged SRBC by
Latex particles and anti-SRBC serum (from rabbits) with human gamma globulin prepara-
tions for reasons of reproducibility and stability (latex fixation test). Although RF is not
pathognomonic for RA, it is a helpful diagnostic test and became part of the classification cri-
teria of the American College of Rheumatology. The sheep cell agglutination and the latex
fixation test have been now widely replaced by nephelometric techniques.
It has later been shown that the RF activity was an auto-antibody to human IgG (or IgG
from other species). The specificity of RF as a diagnostic test is not high, particularly since its
frequency in early RA is relatively low. This has prompted several groups to search for other
auto-antibodies more specifically associated with RA and has led to interesting and in part
diagnostically helpful findings.
Strengths
1. Good example of thorough follow-up of a serendipitous observation with a single serum.
2. Good application of ‘classical’ serological methods, clear and relevant experimental
design.
3. First unequivocal report on anti-protein auto-antibodies in rheumatic diseases.
Weaknesses
1. The nature of the agglutination activating activity as an immunoglobulin was not elucidated,
but the methods of those days were relatively poor: electrophoresis had not yet been invent-
ed and thus gamma globulin not yet been separated from other globulins, immunoglobu-
lins had not yet been detected and therefore not the immunoglobulin classes either.
2. The conclusion that the RF activity was of no diagnostic value is not easy to follow on the
basis of the data presented; it appears to be superficial and partly due to a lack of using
statistical analyses and of following patients longitudinally.
3. The diagnostic criteria for RA used in this study are not explained and many patients pre-
sumed to have RA may not have had the disease.
Relevance
The discovery of RF led to the search for immunological mechanisms in RA.
36
Paper 2
Presentation of two bone marrow elements:

The ‘tart cell’ and the ‘LE’ cell
Authors
Hargraves MM, Richmond H, Morton R
Reference
Mayo Clinic Proceedings 1948; 23:25–28
Summary
In smears of bone marrow from patients with acute disseminated lupus erythematosus (SLE),
a cell type has been found which was called an LE cell and represented a mature neu-
trophilic polymorphonuclear leukocyte containing chromatin material within a large phago-
cytic vacuole. The nuclear nature of the material was shown by specific nuclear stain
(Feulgen’s stain). The LE cell is suggested to be the result of a lytic-phagocytic phenomenon
in which neutrophils are chemoattracted by nuclear material released as a consequence of
cell destruction and phagocytose this material. Several neutrophils can surround the same
mass of material, but once one of the cells has succeeded in ingesting it, the chemotactic
action is lost and the other cells move off. The phenomenon has been found only in bone
marrow in certain cases of SLE, although the diagnosis may be challenged in some of the
more than 25 cases observed with the LE cell phenomenon.
Key message
First description of a reactivity to nuclear material in patients with SLE which has led to a
diagnostically valuable test and was the starting point for all subsequent investigations into
anti-nuclear antibodies, i.e. auto-antibodies to nucleic acids and nucleic acid binding pro-
teins, as well as the recognition of the autoimmune nature of SLE.

Before the description of the LE cell phenomenon, insight into the pathogenesis of SLE
stemmed solely from clinical and histopathological observations. The implication of the
description that the LE cell in SLE was somehow associated with a ‘lytic-phagocytic’ process
and that nuclear material in SLE could be modified to become chemoattractive was a first
indication of an autoimmune phenomenon (aside from the anti-erythrocyte antibodies).
The description of the LE cell phenomenon was followed by the findings that it could be also
reproduced in peripheral blood, that the inducing factor (‘LE cell factor’) was a gamma
globulin and that the phenomenon required the presence of nuclear material, the LE cell
factor and complement. This laid the grounds for (a) all subsequent laboratory tests in the
differential diagnosis of. SLE and other connective tissue diseases, (b) the research in the
field of anti-nuclear auto-antibodies, (c) the findings of nuclear antigens as targets of these
auto-antibodies, (d) the detection of many nuclear antigens and their biological functions,
(e) the recognition of SLE as an immune-mediated and autoimmune disease, and (f) the
finding of immune complexes as decisive contributors to the pathology and clinical symp-
toms of the disease.
The LE cell phenomenon was helpful in the diagnosis of SLE and became one of the clas-
sification criteria of the disease. Later, after Coons invented indirect immunofluorescence
microscopy, the fluorescent anti-nuclear antibody (FANA) test was described and became a
helpful tool in diagnosis and research of connective tissue diseases. ANA patterns soon
became discernible and were finally associated with particular diseases.
The nature of the auto-antigen targeted by the LE cell factor had not been unequivocally
elucidated until 50 years after the description of the phenomenon, when it could be defined
as histone H1 by immunoblotting and various inhibition experiments. The presence of
nucleic acids was not necessary to elicit the phenomenon. The LE cell phenomenon and, in
particular, the anti-histone H1 auto-antibodies, are associated with more severe courses of
the disease. Nevertheless, neither the LE cell phenomenon nor anti-histone H1 antibodies
are pathognomonic for SLE.
Strengths
1. Very well observed abnormality in SLE bone marrow and definition of the target moiety as
nuclear material.
2. Excellent description of the events governing the phenomenon using only conventional
light microscopy and initiating the concept of SLE as characterized by auto-antibodies to
nuclear antigens and immune complexes.
3. Diagnostically valuable and stimulating for subsequent research.
Weaknesses
1. The nature of the target antigen remained elusive.
2. Simple observational description without experimental investigations into the basis of the
phenomenon.
3. Lack of systematic investigation of different patient cohorts.
Relevance
The discovery of the LE cell began the investigation of anti-nuclear antibodies and autoim-
mune mechanisms in connective tissue disorders.
38
Paper 3
Characteristics of a soluble nuclear antigen

precipitating with sera of patients with systemic
lupus erythematosus
Authors
Tan EM, Kunkel HG
Reference
Journal of Immunology 1966; 96:464–471
Summary
Precipitating antibodies which reacted with soluble extracts of different human tissues or
nuclei including nuclei from calf thymus were found in sera of patients with systemic lupus
erythematosus. The antigen was different from DNA and histone, but was associated with
protein fractions and insensitive to deoxyribonuclease, ribonuclease and trypsin, but sensi-
tive to periodate treatment. The reactivity was detected by Ouchterlony double-immunodiffu-
sion and by immunoelectrophoresis techniques. The reactive serum factor was shown to
migrate in the gamma globulin fraction. The antigen was provisionally termed ‘Sm’ and the
reactivity was described to be found in approximately 75% of SLE patients but only very few
patients with other disorders.
Key message
The first description of an auto-antibody reactivity to a non-DNA-non-histone auto-antigen in
SLE. This antigenic moiety was termed Sm and the auto-antibodies appeared to be highly
specific for SLE.

Prior to this study, only DNA, histones and their complexes had been identified as antigens
eliciting auto-antibody reactivities to nuclear antigens in SLE (see chapter on LE cell). The
technological advancements of the 1950s and early 1960s, which included the Ouchterlony
double-immunodiffusion technique, immunoelectrophoresis and chromatography allowed
precipitating antibodies to be distinguished and reactive structures to be better character-
ized, particularly proteins. The fact that the material extracted from different tissues and
nuclei reactive in the precipitation reaction was insensitive to deoxyribonuclease, ribonucle-
ase and trypsin characterized it as clearly different from DNA and histone, but it also gave no
line of identity in double-immunodiffusion with the other antigens. In contrast, there were
lines of identity when a given serum (from patient Sm) was reacted with soluble extracts
from tissues and nuclei of different origin and species. This indicated that the Sm antigen
was widely distributed and immunologically identical between species.
Using an anti-Sm positive serum, Steitz and her colleagues were able to identify the anti-
gen as a group of proteins involved in splicing. Thus, the development of auto-antibodies in
patients with SLE and the detection of the anti-Sm activity was pivotal for the discovery of
hitherto unknown, highly conserved and ubiquitous RNA-binding proteins involved in vital
cellular activities. The RNA and protein components of this protein complex (as well as
many of their functions) have been well characterized meanwhile, using newer techniques
such as immunoblotting and RNA-precipitation, and the characteristic antigenic reactivity
was found to reside in the D-antigen. The Sm particle is associated with the 70 kD-U1-RNA
particle (see subsequent chapter), and there are certain immunological cross-reactivities
between the different components. The antibodies inhibit the splicing process in vitro, but
their penetration into living cells, although possible, is still debated and does not appear to
have major consequences in vivo. The proteins are modified during apoptosis.
Strengths
1. Detection of a new reaction with high specificity for SLE.
2. Use of extracts from tissues and nuclei of different organs and different species.
3. Excellent application of existing techniques to define a new auto-antibody, setting the
stage for subsequent similar investigations which led to the detection of other auto-anti-
bodies and for investigations into the nature of the nuclear antigens.
Weaknesses
1. The nature of the target antigen remained elusive due to the lack of more sophisticated
techniques.
2. Lack of systematic investigation of different patient cohorts (however, this was not the
main purpose of the study).
3. Over-estimation of the frequency of anti-Sm antibodies in SLE.
Relevance
The first discovery of an antibody to a non-DNA-non-histone auto-antigen, which was found
to be a protein involved in splicing.
40
Paper 4
Mixed connective tissue disease – an apparent

distinct rheumatic disease syndrome associated
with a specific antibody to an extractable nuclear
antigen (ENA)
Authors
Sharp GC, Irvin WS, Tan EM, Gould RG, Holman HR
Reference
American Journal of Medicine 1972; 52:148–159
Summary
A nuclear antigen extracted from calf thymus nuclei at low ionic strength, precipitated by
acetate and alcohol and solubilized in sodium chloride, containing mainly protein and
ribonucleic acid (RNA), was named extractable nuclear antigen (ENA) and used in a
haemagglutination test. The antigen was recognized strongly by sera from 25 patients with a
mixture of symptoms of SLE, polymyositis and scleroderma (termed mixed connective tissue
disease, MCTD) as well as by approximately 50% of SLE patients but not patients with other
CTDs. Recognition of the antigen was very sensitive to treatment with RNase, slightly sensi-
tive to trypsin and resistant to DNA. In contrast to SLE, the MCTD sera did not react with Sm
and not, or only at low titres, with DNA. The clinical and serological characteristics of
MCTD, which included a speckled pattern on indirect immunofluorescence (IIF) and a
favourable prognosis, are described.
Related reference (1) Mattioli M, Reichlin M. Characterization of a soluble nuclear

ribonucleoprotein antigen reactive with SLE sera. Journal of
Immunology 1971; 107:1281–1290.
Key message
Demonstration of an anti-nuclear auto-antibody reactivity distinct from Sm but also eliciting
a speckled pattern in patients with SLE and an SLE-like syndrome, which led to the descrip-
tion of a new disease entity, mixed connective tissue disease.

The authors described an auto-antibody to an extractable nuclear antigen (ENA) which con-
tained protein and RNA, but in contrast to Sm was RNase sensitive. Their antigen was used in
a haemagglutination test and reactivity was found almost exclusively with sera from patients
with SLE or SLE-like disease. It was recognized that all 25 patients suffering from a disorder
which had features of SLE but also of polymyositis and dermatomyositis, some of whom had
also rheumatoid arthritis-like joint involvement, had auto-antibodies to ENA. These patients,
in contrast to SLE patients with anti-ENA antibodies, lacked reactivity to Sm and DNA. The
reactivity to ENA had been described by the authors previously. A similar antigen named Mo
was salt-extracted from thymus nuclei and defined by immunoprecipitation by Mattioli and
Reichlin at the same time, but only related to SLE. Sera from patients with MCTD had high
titres of both ANA and anti-ENA antibodies. Sera with anti-Sm antibodies had been shown to
induce a speckled nuclear fluorescence upon IIF, and so did anti-ENA positive sera. It was
later shown that ENA contained mainly 70 kD-U1-RNP and that Sm and U1-RNP were part of
the same particle involved in RNA splicing.
MCTD became the first disease which was defined serologically. During the subsequent
decades a discussion continued as to the nature of this disease as a clinical entity. However,
additional characteristics were found supporting the unique nature of MCTD, such as differ-
ences in major histocompatibility antigens when compared to SLE and scleroderma and a
highly specific auto-antibody reactivity to a particular epitope of hnRNP-A2, the RA33 anti-
gen. A number of investigators have defined classification criteria for MCTD.
Strengths
1. Description of a new anti-nuclear auto-antibody reactivity.
2. Excellent clinical skills leading to recognition of a new connective tissue disease.
3. First disease defined well by the presence of a particular anti-nuclear auto-antibody.
Weaknesses
1. The nature of the target antigen remained elusive due to the lack of more sophisticated
techniques.
2. Antigen only recognized by haemagglutination test.
3. Missed that many patients with SLE also had anti-U1RNP without concomitant anti-Sm
and also some scleroderma patients had the antibody.
Relevance
The discovery of MCTD: the first connective tissue disorder to be defined serologically.
42
Paper 5
Characterization of a soluble cytoplasmic

antigen reactive with sera from patients with
systemic lupus erythematosus
Authors
Clark G, Reichlin M, Tomasi TB
Reference
Summary
A serum of a patient (Ro) with an SLE-like syndrome was shown by complement fixation
reaction and double-immunodiffusion technique to react with a cytoplasmic protein con-
tained in several human tissues, but not red blood cells. This protein was distinct from
known nuclear antigens such as DNA, histone, nucleohistone, Sm and related extractable
antigens. The antigen was found to be an acidic macromolecule containing sulfhydryl
groups. Antibodies to this antigen were found in 40% of SLE patients and also in patients
with Sjögren’s syndrome, but not patients with other connective tissue diseases.
Related references (1) Alspaugh M, Tan EM. Journal of Clinical Investigation 1975; 55:1067.
(2) Lerner MR, Boyle JA, Hardin JA, Steitz JA. Two novel classes of
small ribonucleoproteins detected by antibodies associated with
lupus erythematosus. Science 1981; 211:400–402.
Key message
Description of the reactivity of SLE (and Sjögren’s syndrome) sera with a cytoplasmic antigen
named Ro. The antigen itself and its cytoplasmic nature made it distinct from the previously
recognized nuclear antigens targeted by SLE sera.

While a number of nuclear antigens had been characterized as targets of SLE sera, patient
Ro had auto-antibodies directed to a cytoplasmic antigen in her serum. This new reactivity
was found in 40% of SLE and in Sjögren’s syndrome sera. A similar reactivity was later found
primarily in patients with Sjögren’s syndrome (and termed SS-A) as a nuclear antigen, but
SS-A and Ro were shown to be identical. The riddle’s resolution lay in the fact that the anti-
gen shuttled between cytoplasm and nucleus. The Ro antigen turned out to be an RNA-bind-
ing protein, but the RNAs that were bound by Ro were hitherto unknown small RNAs,
termed Y RNAs. Another antigen which was also recognized by sera from patients with
Sjögren’s syndrome and SLE and had been termed La or SS-B was associated with yet anoth-
er group of RNAs, namely 4.5S RNA species which, among other origins, can also be virus
associated. Importantly, Ro and La have been found to be associated with each other in a dis-
tinct particle. However, the function of the Ro-RNAs as well as the particle have not yet been
fully elucidated.
Anti-Ro and anti-La are used in the differential diagnosis of connective tissue diseases, and
La is relatively typical (though not pathognomonic) for Sjögren’s syndrome. On the other
hand, anti-Ro, even though a more promiscuous auto-antibody, has been associated with a
series of subtypes of SLE. This includes mild subtypes, such as the so called ANA-negative
SLE (which lacked ANA on conventional rat liver sections due to the low concentration of
the Ro antigen in rat liver) and subacute cutaneous lupus erythematosus, but also neonatal
LE and congenital heart block, both on the basis of transplacentar passage of these (and
other) auto-antibodies and their binding to particular organs in the neonate.
Thus, the characterization of the Ro antigen has led to (a) the recognition of the underly-
ing cause of neonatal lupus and congenital heart block, (b) the characterization of particu-
larly mild forms of SLE, (c) the detection of a new ribonucleoprotein particle and (d) the
detection of hitherto unknown RNA species.
Strengths
1. Characterization of a new target antigen for auto-antibodies of patients with connective
tissue diseases, particularly SLE and Sjögren’s syndrome.
2. Demonstrated of the cytoplasmic nature of this antigen as opposed to the nuclear anti-
gens known in those times.
3. Stimulation of further research into cytoplasmic auto-antigens.
Weaknesses
1. Technical limitations of those days precluded better characterization of the antigen.
2. The presence of Ro in a particle was not recognized.
3. Patients were not well characterized clinically.
Relevance
The discovery of Ro antigen which has been found to be associated with a variety of clinical
syndromes including neonatal lupus with congenital heart block and mild forms of SLE.
44
Paper 6
Identification of a nuclear protein (Scl-70) as a

unique target of human antinuclear antibodies
in scleroderma
Authors
Douvas AS, Achten M, Tan EM
Reference
Journal of Biological Chemistry l979; 254: 10514–10522
Summary
Using biochemical and immunological techniques, a 70 kD protein contained in rat liver
nuclei was characterized to be reactive with sera from five scleroderma patients. The antigen
was termed Scl-70. Absorption experiments revealed that this was the major or only antinu-
clear reactivity contained in these sera. The preparation of several distinct fractions of
nuclear extracts allowed the recognition that the antigen was a basic non-histone nuclear
protein associated with the chromatin fraction and localized on the chromosomes of cells in
mitosis.
Key message
Characterization of a nuclear, DNA-associated protein as target for sera from several patients
with scleroderma. This antigen, Scl-70, is a major, if not sole, target in the sera of these
patients.

This is the first nuclear antigen ever characterized as target antigen for patients with systemic
sclerosis. Scl-70 was later revealed to be a degradation product of the protein topoisomerase
(also named Scl-100 by virtue of its molecular mass). The antigen is found in about 40% of
patients with systemic sclerosis and is mainly associated with the diffuse form, while limited
systemic sclerosis is associated with auto-antibodies to centromere proteins which typically is
present in the CREST syndrome. Other auto-antibody systems also occur in scleroderma.
Interestingly, all these auto-antibodies can be localized to the nucleolus, although the exquis-
ite specificity for nucleolar auto-antigens in systemic sclerosis is unresolved.
Anti-Scl-70 auto-antibodies are pathognomonic for scleroderma, but it is uncommon for
patients to have more than one scleroderma-typical auto-antibody specificity. The different
auto-antibody systems are commonly related to distinct clinical subtypes of the disease and
anti-Scl-70 is related to diffuse cutaneous involvement but also pulmonary fibrosis.
The recognition of Scl-70 as a target antigen has significantly advanced the diagnostic
armamentarium in the differentiation of the connective tissue diseases and, together with
the subsequent detection of other scleroderma-typical auto-antigens, also allowed distinct
subsets to be discerned which are characterized by different clinical and particularly prog-
nostic features. All these auto-antibodies are highly characteristic for systemic sclerosis, while
other anti-nuclear auto-antibody subsets are virtually never found (with the exception of
occasional anti-Ro reactivities).
Strengths
1. Characterization of a new target antigen for auto-antibodies of patients with systemic scle-
rosis.
2. Use of available techniques to optimize the characterization of the antigen and definition
as a non-histone chromosomal protein.
3. Definition of the specificity and selectivity of the immune response in scleroderma
patients.
Weaknesses
1. The higher molecular weight parent protein was not detected.
2. Sera from only five scleroderma patients were tested.
3. No description of the patients’ clinical features.
Relevance
The discovery of anti-Scl-70, the first of a group of auto-antibodies which are associated with
distinct subsets of scleroderma.
46
Paper 7
Myositis autoantibody inhibits histidyl-tRNA

synthetase: a model for autoimmunity
Authors
Mathews MB, Bernstein RM
Reference
Nature 1983; 304:177–179
Summary
IgG fractions from sera of patients with myositis and anti-Jo-1 antibodies, but not myositis
and other auto-antibodies or other autoimmune rheumatic disorders, were used for
immunoprecipitation of HeLa cell extracts labelled with 35-S-methionine or 32-P-phosphate.
A protein of 50,000 kD molecular mass and an RNA species of tRNA size were detected by
polyacrylamide gel electrophoresis. All anti-Jo-1-positive sera induced the same reaction.
RNase treatment revealed that the anti-Jo-1[ reactivity was retained indicating that the major
epitope(s) resided on the protein. Further electrophoretic analyses, including RNA finger-
printing, suggested that the RNA species was a tRNA containing the anticodon for histidine.
While it had no effect on the stability of the complex between histidine and its cognate
tRNA, anti-Jo-1 IgG inhibited 3-H-histidine joining to its tRNA, but not the charging of
tRNAs with other amino acids. This inhibition was highly specific for anti-Jo-1-positive sera,
since other sera did not inhibit 3-H-histidine-tRNA interaction. Removal of the Jo-1 antigen
from the aminoacyl-tRNA synthetase preparation specifically eliminated the capacity to
charge tRNA with histidine. It was concluded that the Jo-1 antigen was histidyl-tRNA syn-
thetase or a functional subunit of this enzyme. Importantly, the antigen appeared to be con-
tained only in the cytoplasm. Finally, the authors speculate on the association of autoimmu-
nity to Jo-1 with infections with myotropic viruses capable of carrying a histidine residue
attached to its RNA.
Related references (1) Nishikai M, Reichlin M. Heterogeneity of precipitating antibodies

in polymyositis and dermatomyositis – Characterization of the
Jo-1 antibody system. Arthritis and Rheumatism 1980; 23:881–888.
(2) Ross MD, Hendrick JP, Lerner MR, Steitz JA, Reichlin M. Nucleic
Acids Research 1983; 11:853–870.
Key message
Characterization of the protein reactive with the most common auto-antibody in sera from
patients with poly- and dermatomyositis improve diagnostic approaches and possible search
into the etiopathogenesis of myositides.

Poly- and dermatomyositis have been found associated with a large number of precipitating
auto-antibodies, anti-Jo-1 being the most common one. While it had been noted at about the
same time that anti-Jo-1-positive sera precipitated histidyl-tRNA, the study by Mathews and
Bernstein revealed that the reactivity was directed primarily to the enzyme that catalyzes the
charging with histidine of the specific tRNA. This finding led to the recognition that a variety
of other myositis-specific auto-antibodies were directed to tRNA-synthetases, albeit different
species of these enzymes. Although anti-Jo-1 is the most common of these auto-antibodies,
the clinical features of patients with auto-antibodies to the different synthetases are similar.
One of the features of this ‘anti-synthetase syndrome’ is involvement of the lung. Thus, the
definition of histidyl-tRNA-synthetase as a major auto-antigen in myositis has allowed better
diagnostic and prognostic approaches to this disease. Interestingly, the antibodies are not
only directed to the proteins but also to the tRNA species themselves. However, completely
different auto-antigens are recognized by sera from other patients with myositides, and many
of them have other clinical characteristics, frequently with quite aggressive disease.
There are two other important aspects of this study: First, the authors showed that the
auto-antibody inhibited the function of the target protein (in this case the charging of his-
tidyl-tRNA with histidine which is catalyzed by the target synthetase); thus, similar to anti-Sm
antibodies, the targeted epitope ought to be the functional site of the molecule. Second, this
and subsequent investigations also revealed the exquisite specificity of certain auto-antibody
systems for particular disorders: anti-Jo-l and the other anti-synthetases are pathognomonic
auto-antibodies, suggesting a common underlying cause for at least one large subset of the
myositides. However, despite these highly specific serologic characteristics, the cause(s) of
the disease have not yet been elucidated.
Strengths
1. Characterization of the major target antigen for myositis-specific precipitating anti-Jo-1
antibodies.
2. Demonstration of the interference of the auto-antibody with the function of the target
enzyme.
3. Demonstration of the cytoplasmic rather than nuclear nature of this auto-antigen.
Weaknesses
1. Other anti-synthetases not described as targets for myositis sera.
2. Target epitope(s) not characterized.
3. Induction of anti-Jo-1 discussed in a speculative way without further data (nor subsequent
experimental approach).
Relevance
Anti-Jo-1 was the first specific auto-antibody to be associated with myositis. As outlined in the
chapter on polymyositis, the different auto-antibodies found in this family of diseases offer a
means of classification.
48
CHAPTER 4
Other investigations:
synovial fluid analysis,
arthroscopy and blood
biochemistry
Paul Dieppe
Introduction
Rheumatology is a subspeciality of medicine in which diagnoses still depend more on the his-
tory and physical examination than on special investigations. Indeed, the relative lack of spe-
cial diagnostic and therapeutic procedures has been seen as a problem for the discipline.
However, the fact that rheumatology is concerned with diseases of joints, particularly synovial
joints, means that access to products of the joints, as well as visualization of joint structures
have been explored for their diagnostic potential, as well as the insights they provide on dis-
ease mechanisms.
The most obvious place to look is in the synovial fluid. The first, definitive studies of
human synovial fluid were published as monographs rather than scientific papers. In 1938
Kling published a book describing the synovial membrane and synovial fluid (1), but the
most important and ‘classic’ contribution was that of Marion Ropes and Walter Bauer, pub-
lished in 1953 (2), a masterful contribution. Hollander was the early proponent of synovial
fluid analysis as a diagnostic aid to arthritis, describing it as a ‘liquid biopsy’ of the synovial
joint, and documenting his personal experience of some 100 000 aspirations and injections
(Paper 1). The most important development in the diagnostic examination of synovial fluid
samples came with Dan McCarty’s introduction of polarized light microscopy for the identifi-
cation of urate (Paper 2) and subsequently pyrophosphate (Paper 3) crystals. Polarized light
microscopy is now a routine part of rheumatological practice, and a little known contribu-
tion of Ralph Schumacher’s group has shown that it is one of the few tests in our armamen-
tarium that actually changes clinical practice (Paper 4). Orthopaedic surgeons, not content
with putting small needles into joints to aspirate fluid samples, subsequently introduced
arthroscopic examination of joints, allowing direct visualization of structures and biopsy of
any structures that appear abnormal (Paper 5). Subsequently mini-arthroscopes have been
introduced for office use by rheumatologists, although their value remains disputed.
Joint diseases are characterized by two different pathological processes, first, inflammation
of the synovial lining, and second, destruction or altered turnover of structural elements
such as the articular cartilage. Both types of process lead to the release of products into the
bloodstream, leading to the possibility that simple serum assays might be of diagnostic value.
It has long been known that most inflammatory processes in the body (including synovitis)
lead to a raised erythrocyte sedimentation rate (ESR). We now know that this is mediated by
Other investigations 49
the release of cytokines from the inflamed tissue, and resulting changes in the synthesis of a
range of proteins in the liver. Serum levels of one of these ‘acute-phase proteins’, C-reactive
protein (CRP) are particularly elevated in inflammatory arthritis, and is regularly used to
assess disease activity. A particularly important development in this story came with the
recognition that CRP levels also had predictive value for subsequent erosive damage to joints
in rheumatoid arthritis (RA) (Paper 6) indicating that the measurement of CRP could (and
perhaps should) be used to monitor RA and to assess the degree of success of suppressive
therapies. More recently, the development of extremely sensitive immunological assays has
made it possible to measure serum levels of a variety of products of connective tissue
turnover, some of which are relatively specific to joint pathology. Our final selection in this
group of papers illustrates the potential value of one of these assays in predicting joint dam-
age in a variety of types of arthritis (Paper 7). Only time will tell what impact the use of these
‘biochemical markers’ is destined to have on rheumatological practice (3).
References
1. Kling DH. The synovial membrane and synovial fluid. Los Angeles Medical Press, 1938.
2. Ropes MW, Bauer W. Synovial fluid changes in joint disease. Harvard University Press, 1953.
3. Myers SL. Synovial fluid markers in osteoarthritis. Rheumatic Diseases Clinics of North America
1999; 25:443–449.
50
Paper 1
Examination of synovial fluid as a diagnostic

aid in arthritis
Authors
Hollander JL, Reginato A, Torralba TP
Reference
Medical Clinics of North America 1966; 50:1281–1293
Summary
From a wide experience with synovianalyses we have concluded that careful examination of
joint fluid is the most definitive diagnostic laboratory test for differentiation of the various
forms of arthritis. It is also the diagnostic aid least frequently utilized by physicians. Study of
the joint fluid is as important in arthritis as urinalysis in renal disease. Synovial fluid is actual-
ly a ‘liquid biopsy’ from the site of inflammation. Many new facts about arthritis are being
added to our knowledge each year from studies on the synovial fluid. Some of these advances
have been described in this report, together with a detailed description of techniques and
findings in various forms of arthritis.
Related references (1) Hollander JL. The most neglected differential diagnostic test in
arthritis. Arthritis and Rheumatism 1960; 3:364–367.
(2) Kling DH. The Synovial membrane and Synovial Fluid. Los Angeles
Medical Press, 1938.
(3) Ropes MW, Bauer W. Synovial fluid changes in joint disease. Harvard
University Press, 1953.
(4) Hollander JL. In: Textbook of Rheumatology.
Key message
Synovial fluid examination is a very important aid to the diagnosis of the different forms of
arthritis.

The monograph of Ropes and Bauer (3) is the most important contribution to this field, as it
documents synovial fluid findings in health as well as disease, and provides a comprehensive
overview of preceding literature as well as the extensive work of the authors. However, it was
Hollander, rather than Ropes and Bauer, who championed the routine use of synovial fluid
analysis in rheumatological practice. Hollander and colleagues report on their unrivalled
experience of synovial fluid analysis in this paper, reporting on findings of gross appearance,
mucin clot formation, cell counts, rheumatoid factor levels, culture, and polarized light
microscopy. They provide a table of the main findings in different diseases.
Hollander reports on normal synovial fluid, and on total and differential cell counts in
inflammatory and non-inflammatory forms of arthritis, findings that have remained largely
unchallenged and believed in ever since by subsequent generations of rheumatologists.
Strengths
1. Built on vast experience (Hollander claims that his unit has personal experience of nearly
100 000 joint aspirations and injections).
2. Clear documentation of the main findings of synovial fluid in different forms of arthritis.
3. Comments on all the major available assays of synovial fluids.
Weaknesses
1. No data is presented, just what the authors believe to be the synovial fluid findings in dif-
ferent conditions.
2. It is unclear on what the report of normal synovial fluid findings is based on.
3. This is a review which contains as much propaganda (in favour of aspirating joints) as it
does science.
Relevance
Became the established accepted view on synovial fluid findings in different forms of arthri-
tis, and the value of analysis in rheumatological practice.
52
Paper 2
Identification of urate crystals in gout

synovial fluid
Authors
McCarty DJ, Hollander JL
Reference
Summary
Urate crystals in synovial fluid and the crystals from a subcutaneous tophus were found to
have identical optical properties by polarized light microscopy. They were negatively birefrin-
gent with extinction on the long axis. Urate crytals were identified by polarized light
microscopy in the aspirated synovial fluid obtained from 15 of 18 patients with clinical gout.
The crystals were specifically digested by uricase in all 15 samples and in two instances disap-
peared in the control specimens as well. Urate crystals were also identified by ordinary light
microscopy in 11 of the same 18 synovial fluid samples.
Comparison of the results of ordinary light microscopy with those seen by polariscopic
observation revealed that the percentage positive identification was greater when the latter
method was used. In two instances, crystals were seen by ordinary light which did not have
the characteristics of urate crystals. One of these was from an otherwise typical case of acute
gouty arthritis. The concentration of urate in the synovial fluid during the acute attack is
probably higher than that found in the serum.
Related reference (1) McCarty DJ. The inflammatory reaction to microcrystalline sodium
urate. Arthritis and Rheumatism 1965; 8:726–735.
Key message
Polarized light microscopy leads to more accurate identification of urate crystals in the syn-
ovial fluid than ordinary light microscopy, which had been used previously. Ordinary light
microscopy results in both false negative and false positive results. The highly characteristic
properties of urate crystals viewed by polarized light (needle shape and strong negative bire-
fringence) allows easy identification.

Polarized light microscopy of synovial fluid for the identification of urate and pyrophosphate
crystals in order to diagnose gout and pseudogout respectively, is now a routine part of our
clinical practice. However, prior to the publication of this paper that was not the case. As the
authors say in their introduction ‘urate crystals have been noted in the synovial fluid from
gouty patients in the past, but this finding was believed to be infrequent and of little assis-
tance to the clinician in his attempt to establish a diagnosis’.
With this paper McCarty and Hollander showed that polarized light microscopy allowed
the easy identification of urate crystals in synovial fluids, and that they were almost always
present in gout. This finding, which has stood the test of time, led to the widespread use of
the technique in clinical practice. The paper illustrates the characteristic appearances of
urate crystals viewed by polarized light microscopy, as shown in Figure 4.1.
Figure 4.1 Monosodium urate

monohydrate crystals from a gouty
tophus viewed by compensated
polarized light microscopy. Note
the needle shaped morphology and
strong negative birefringence
(resulting in a yellow appearance
when the long axis of the crystal
is aligned with the axis of the
compensator). (Courtesy of Dr
Angela Swan.)
Strengths
1. Introduction of a new, reliable technique for the identification of urate crystals.
2. Clear, concise, well illustrated description of the characteristics of the crystals when viewed
by polarized light microscopy.
3. Check on the validity of the technique by using uricase to digest crystals.
4. Comparison with the only other method available at the time, i.e. ordinary light
microscopy.
5. Discovery of two false positives using ordinary light microscopy pre-empts their discovery
of pyrophosphate crystals.
Weaknesses
1. The authors mention, but do not present proper results of the examination of synovial flu-
ids from non-gouty patients. Therefore they were not able to say with certainty that the
test was specific for gout.
2. The paper is generally lacking in methodological detail and is rather brief.
3. They do not describe how samples were searched for crystals under either ordinary or
polarized light, or how easy or difficult it was to find the particles.
4. Their speculation on urate concentrations in synovial fluid, which reached the summary,
has no foundation in data.
Relevance
The paper that led to the introduction of the most important laboratory investigation specif-
ic to rheumatology.
54
Paper 3
The significance of calcium phosphate crystals

in the synovial fluid of arthritic patients: the
‘pseudogout syndrome’. II. Identification of crystals
Authors
Kohn NN, Hughes RE, McCarty DJ, Faires JS
Reference
Summary
In our study of a series of synovial fluids obtained by aspiration from a group of patients with
arthritis, we observed seven fluids that contained significant amounts of crystalline material.
In five instances, these crystals were shown to be identical crystallgraphically and distinct
from apatite and tophaceous material. Physical and chemical studies of these crystals led to
the conclusion that they were a form of calcium pyrophosphate. The utilization of x-ray dif-
fraction in obtaining powder pattern photographs proved to be especially useful in this prob-
lem. The applicability of this method in the investigation of normal and abnormal biological
processes involving the crystalline state is briefly discussed.
Related references (1) McCarty DJ, Kohn NN, Faires JS. The significance of calcium phos-
phate crystals in the synovial fluid of arthritic patients: the
‘pseudogout syndrome’. I. Clinical aspects. Annals of Internal
Medicine 1962; 56:711–736.
(2) McCarty DJ. Calcium pyrophosphate dihydrate deposition disease.
Arthritis and Rheumatism 1976; 19(Supplement Volume).
(3) McCarty DJ, Hogan JM, Gatter RA. Studies on pathological calcifi-
cations in human cartilage. Journal of Bone and Joint Surgery 1966;
48(A):309.
(4) Dieppe PA, Crocker PR, Huskisson EC, Willoughby DA. Apatite
deposition disease, a new arthropathy. The Lancet 1976; 1:266.
(5) Schumacher HR, Sonlyo AP, Tse RL. Arthritis associated with
apatite crystals. Annals of Internal Medicine 1979; 87:411.
Key message
Calcium pyrophosphate crystals are found in the synovial fluid of some patients with acute
arthritis of the knee joint.

This paper on the ‘pseudogout syndrome’ should be read alongside the related contribution
quoted above (1). The pair make fascinating reading, and represent a major milestone in
the development of rheumatology. In the first paper, Dan McCarty and colleagues provide
us with the first description of pseudogout. Their paper, as often seems to be the case with
original descriptions of diseases, covers all the major features of the condition, including its
predilection for the knee and wrist joints of older people, and the presence of radiographic
chondrocalcinosis.
The second paper, featured here, is also truly ground breaking, because of the use of com-
plex physical and chemical techniques for the identification of the crystals found in these
cases. They did four key, original things: first, they extracted the mineral from the synovial
fluids using hyaluronidase digestion (a difficult procedure); second, they synthesized pure
crystals of calcium pyrophosphate dihydrate to act as a positive control (an extremely diffi-
cult procedure); third, they used the then little known technique of infra-red spectroscopy to
compare the extracted mineral with other samples, including their synthetic pyrophosphate
crystals, and finally they used x-ray powder diffraction to provide a definitive identification of
the crystals. They also compared their ‘new crystals’ with urates using polarized light
microscopy, and describe the weak positive birefringence of pyrophosphate crystals
(Figure 4.2). This is painstaking and superb work. They got it right and the work has never
been bettered. Some 15 years later other groups used these techniques, along with the new
approach of analytical electron microscopy, to identify basic calcium phosphates and other
forms of mineral in joint samples, and only then were the findings reported here repeated
and confirmed.
The relevance of the finding, as we now know and as was flagged in these papers, is that
calcium pyrophosphate dihydrate crystals can cause inflammation and are responsible for
the arthritis of pseudogout.
Figure 4.2 Calcium pyrophosphate

dihydrate crystals that have been
extracted from joint tissues are
seen here under compensated
polarized light microscopy. Note
the varying morphology and weak
positive birefringence that help to
distinguish them from urate
crystals. (Courtesy of
Dr Angela Swan.)
Strengths
1. This is a detailed, careful and well presented study using excellent methodology.
2. The use of control samples, including bone apatite and synthetic calcium pyrophosphate
crystals (which are very difficult to make).
3. This is one of the first studies in medicine to use physical techniques such as x-ray dif-
fraction.
Weaknesses
1. The numbers of patients studied was small.
2. Only five samples were examined by the physical methods used.
Relevance
The first and definitive, description of pseudogout and of the crystals that cause it.
56
Paper 4
Usefulness of synovial fluid analysis in the evaluation

of joint effusions. Use of threshold analysis and
likelihood ratios to assess a diagnostic test
Authors
Eisenberg JM, Schumacher HR, Davidson PK, Kaufmann L
Reference
Archives of Internal Medicine 1984; 144:715–719
Summary
This study applied threshold analysis and likelihood ratios to determine the usefulness of a
diagnostic test. Eleven staff rheumatologists or rheumatology fellows provided probability
estimates for the most likely diagnoses both before and after synovial fluid analyses were per-
formed on 180 patients with joint effusions. They also indicated whether the planned thera-
py was altered by test results. The therapeutic thresholds and log likelihood ratios were
derived for the six most frequent diagnoses. Synovial fluid analysis was most useful for
patients likely to have gout, pseudogout or infectious arthritis. The derived therapeutic
thresholds were consistent with recommended medical practice, for example, with a lower
threshold for possible septic arthritis (20%) than for possible gout (65%). This study demon-
strates that threshold analyses and likelihood ratios can be used to assess the clinical contri-
bution of diagnostic tests.
Related references (1) Pauker X, Kassirer X. The threshold approach to clinical decision
making. New England Journal of Medicine 1975; 293:229–235.
(2) Schumacher HR. Analyzing synovial fluid: a useful diagnostic aid
for practitioners. Modern Medicine 1977; 45:58–63.
Key message
Knowledge of the results of synovial fluid analysis can lead to changes in diagnosis and treat-
ment of arthritis.

This paper was well ahead of its time. In recent years great effort has been put into the analy-
sis of data from clinical trials of interventions, and in finding ways of putting such data to use
in routine clinical practice. Similar work is just beginning to appear in relation to diagnostic
tests, although very few of the investigations used in arthritis have been subjected to rigorous
tests of their validity or value. Yet, as illustrated by this paper, the theoretical basis for such
work has been available for a long time, and it is relatively easy to put it into practice.
This paper describes the decisions made by clinicians as to the diagnosis and treatment of
arthritis before and after the results of synovial fluid analysis was made available to them. It is
one of very few contributions to the rheumatological literature which assesses the value of
diagnostic investigations in routine clinical practice. The six most common diagnoses are
presented in the paper, and the data show that synovial fluid findings change the diagnosis
and treatment in a significant proportion of cases (Table 4.1). Gout was the diagnosis most
likely to be changed, and synovial fluid findings were of most value to clinicians in diagnos-
ing septic arthritis and gout. The authors also analysed the data to look at therapeutic thresh-
olds (i.e. the degree of diagnostic certainty required by physicians to institute specific thera-
py for a condition), and found this was lowest for septic arthritis (20%) and highest for gout
(65%).
The paper is important not only because it shows the importance of synovial fluid analysis
in clinical practice, but also for the way it applies sound theory and mathematical principles
to the analysis of clinical decision-making in rheumatology.
Table 4.1 Data from the paper showing the number of times in which the final diagnosis made by
clinicians (after they had access to synovial fluid data) was the same or different from that which
they made before aspiration of synovial fluid.
Disease Same final diagnosis Different final diagnosis % changed
Osteoarthritis 31 6 16
Rheumatoid arthritis 24 5 17
Gout 25 9 26
Sepsis 11 3 21
Pseudogout 9 1 10
Traumatic arthritis 7 2 22
Strengths
1. Innovative use of theory and mathematical techniques for decision making and the value
of diagnostic tests in rheumatology.
2. Analysis of a relatively large number of clinical cases (180).
3. Comprehensive reporting of the data.
Weaknesses
1. This is a difficult paper to read, which does not explain the theoretical background of the
approach very well.
2. Clinicans taking part were asked to list their four most likely diagnoses and to suggest a
therapy, which made it difficult to understand what they thought they were treating or
why.
3. It was not possible to assess the importance of other types of information, such as serum
uric acid levels, in clinical decision making.
Relevance
One of the few papers in the rheumatology literature which show that an investigation can
alter clinical practice.
58
Paper 5
Arthroscopy of the knee in rheumatic diseases

Authors
Jayson MIV, Dixon ASTJ
Reference
Summary
Arthroscopy offers a safe and simple method of examining the morphology of the synovial
membrane of the knee joint and of obtaining biopsies under direct vision. The technique is
described in detail and typical results are presented.
Related references (1) Burman MS, Finkelstein H, Mayer L. Arthroscopy of the knee
joint. Journal of Bone and Joint Surgery 1934; 16:255.
(2) Johnson LL. Comprehensive arthroscopic examinantion of the knee.
Mosby: St Louis. 1977.
(3) Halbrecht JL, Jackson DW. Office arthroscopy: a diagnostic alter-
native. Arthroscopy 1992; 8:320–326.
(4) Ike RW, O’Rouke KS. Detection of intra-articular abnormalities in
osteoarthritis of the knee: a pilot study comparing needle
arthroscopy with standard arthroscopy. Arthritis and Rheumatism
1993; 36:1353–1363.
Key message
Biopsies taken under direct vision via an arthroscope, so that areas of abnormality are sam-
pled, are more likely to provide a diagnosis than a blind biopsy.

The authors describe the technique of arthroscopy done under local anaesthetic (as a day
case procedure) in detail. They then describe and illustrate the results of a series of 23 cases
performed by themselves. They pay particular attention to their findings on examination and
biopsy of the synovium, pointing out that the changes in early inflammatory arthritis are very
patchy (Figure 4.3). They show that using this technique they are able to visualize and biopsy
abnormal areas, whereas a blind biospy might easily result in areas of normal synovium being
sampled. They also describe cartilage changes, point out that it is difficult to see the menisci
using this technique, and record the fact that the attending irrigation of the joint seemed to
result in a clinical improvement in the joint in many cases.
This paper was the first one to describe the use of arthroscopy as a part of rheumatological
practice, rather than by orthopaedic surgeons (who had been using arthroscopy for some
years). It pre-empted the extensive use of the technique over the following two decades, after
which the office mini-arthroscopy approach was introduced. They were also amongst the first
people to describe the improvement in joints that can result from lavage.
Figure 4.3 Typical appearance of inflamed

synovium seen through an arthroscope.
While this area of synovium show
extensive hypertrophy and inflammation
other areas in the same joint looked
relatively normal.
Strengths
1. It was highly innovative to introduce arthroscopy under local anaesthetic into rheumato-
logical practice.
2. The illustrations of the synovial and cartilage changes that can be seen using this tech-
nique are extensive, in colour and of a high quality.
3. The technique is described in sufficient detail for others to be able to reproduce it.
Weaknesses
1. They only describe 23 cases.
2. Most of the cases described had RA, so the range of changes and uses for arthroscopy that
subsequently became apparent (such as its value in picking up pigmented vilonodular syn-
ovitis) were not apparent.
3. They do not discuss the extent to which the findings changed their diagnosis or treatment
of these 23 cases.
4. There was no formal assessment of outcome resulting from lavage.
Relevance
The introduction of arthroscopy into rheumatological practice.
60
Paper 6
The assessment of rheumatoid arthritis: a study

based on measurements of the serum acute-phase
reactants
Authors
McConkey B, Crockson RA, Crockson AP
Reference
Quartely Journal of Medicine 1972; 41:115–125
Summary
We have studied 187 patients with RA for periods ranging from 3 months to 6 years; 55 of
them were observed for more than 3 years. The aim of the study was to compare changes in
clinical state with changes in measurements of the blood levels of two acute-phase proteins
(CRP and haptogolobin).
The results showed that measurements of the serum acute-phase reactants reflected exac-
erbation or remission of arthritis when such changes occurred over short periods of time, for
example a few months. The serum acute-phase reactants also reflected the course of RA over
longer periods of time; in patients studied for more than 3 years there was a close correlation
between the course of the disease and the levels of the acute-phase reactants. We suggest that
measurements of the serum acute-phase reactants provide an accurate and objective way of
assessing the progress of RA and of its response to treatment.
Related references (1) Lansbury J. Report of a three-year study on the systemic and articu-
lar indexes in rheumatoid arthritis. Arthritis and Rheumatism
1958; 1:505–522.
(2) Crockson RA. A gel diffusion precipitin method for the estimation
of c-reactive protein. Journal of Clinical Pathology 1963;
16:287–289.
Key message
C-reactive protein levels in the serum reflect disease activity in RA; sustained elevation of
CRP levels is a poor prognostic factor. In the long-term therapy of RA one should aim to
reduce CRP levels to as near normal as possible.

The authors of this paper note that in 1958 Lansbury had pointed out that what was needed
in RA was a test which tells us how well the disease is controlled (1). Before the pioneering
work of this group on CRP the ESR had been the standard blood test used to assess disease
activity. In addition, clinical assessments, such as the ‘Lansbury Index’ were used.
Crockson showed that CRP was a good acute-phase reactant and developed an accurate
measurement of serum levels. He then teamed up with McConkey to apply his assay to the
assessment of RA. This paper summarizes their work, which clearly shows that CRP measure-
ments are superior to ESR assays in both short-term and long-term assessments of RA.
Although there was a strong statistical association between ESR and CRP in the whole group,
in many individual instances the ESR did not correlate with the CRP, and in these cases it was
the CRP which correlated with clinical activity and progression and not the ESR.
In addition, they showed that serial measurements have a great advantage over single
point measurements, and that there were correlations between long-term disease progression
and the general levels of CRP in the serum. This led them to conclude that regular measure-
ment of CRP levels should provide the test that Lansbury wanted. Now, most rheumatologists
do use serial assays of CRP to help them monitor the activity and response to therapy of their
patients with RA. Thus this paper represents the initial evidence base for one of the currently
accepted approaches to the management of this disease.
Strengths
1. The authors applied their extensive experience of the measurement of acute-phase pro-
teins to a large number of patients with RA studied over many years.
2. The hypothesis is clear and well-founded.
3. Clear presentation of a lot of data.
Weaknesses
1. The clinical assessments against which the CRP was correlated are very poor, relying large-
ly on physician and patient assessments of change (better, same or worse).
2. They say that they also measured x-ray progression, but do not present data relating CRP
to x-ray changes.
3. The patients are a highly selected group from one hospital base, and may not be represen-
tative of RA as a whole.
Relevance
Led to the introduction of CRP as the key serum assay for the assessment of disease activity
and responses to treatment in RA.
62
Paper 7
Cartilage oligomeric matrix protein: a novel

marker of cartilage turnover detectable in
synovial fluid and blood
Authors
Saxne T, Heinegard D
Reference
British Journal of Rheumatology 1992; 31:583–591
Summary
Cartilage oligomeric matrix protein (COMP) is a tissue-specific non-collagenous protein. We
have developed an enzyme-linked immunosorbent assay for the detection of this protein in
the synovial fluid and serum. The protein has been quantified in these fluids in patients with
RA, reactive arthritis (ReA), juvenile chronic arthritis, osteoarthritis (OA) and in sera of con-
trol subjects. The protein was detectable in all fluids and the synovial fluid levels were always
higher than in serum in paired samples. The highest knee joint synovial fluid levels were
found in ReA and the lowest in RA patients with advanced destruction of the knee joint.
However, the relative synovial fluid content of COMP was higher in these RA patients than in
patients with advanced OA. In patients with long-standing reactive synovitis the concentra-
tions were decreased. This decrease, however, was less marked than for proteoglycan concen-
trations. The serum concentrations were low in patients with juvenile chronic arthritis and in
patients with RA with advanced cartilage destruction of the studied knee joint. In the other
groups serum levels did not differ between groups or from controls.
Related references (1) Saxne T, Heinegard D, Wolheim FA, Pettersson H. Differences in

cartilage proteoglycan level in synovial fluid in early rehuma-
toid arthritis and reactive arthritis. The Lancet 1985;
8447:127–128.
(2) Petersson IF, Boegard T, Svensson B, Heinegard D, Saxne T.
Changes in cartilage and bone metabolism identified by serum
markers in early osteoarthritis of the knee joint. British Journal of
Rheumatology 1998; 37:46–50.
(3) Lohmander LS, Saxne T, Heinegard DK. Release of COMP into
joint fluid after knee injury and in osteoarthritis. Annals of
Rheumatic Diseases 1994; 53:8–13.
(4) Sharif M, Saxne T, Shepstone L, Kirwan J, Elson C, Heinegard D,
Dieppe P. Relationship between serum COMP levels and dis-
ease progression in osteoarthritis of the knee joint. British
Journal of Rheumatology 1995; 34;306–310.
Key message
That serum and synovial fluid levels of a cartilage-specific protein reflect both normal
turnover and different pathological processes occurring in different forms of joint disease.

Prior to this work there had been other attempts to use the measurement of cartilage macro-
molecules released into the synovial fluid and blood as potential markers of disease process
and progression. Most of these had concentrated on proteoglycans or their glycosaminogly-
can side chains, such as keratan sulphate, and the results had, in general, been disappoint-
ing. In this paper Saxne and Heinegard describe the application of an assay of a different
cartilage macromolecule, COMP, which at that time was thought specific to articular carti-
lage. They describe a careful set of experiments which first of all provide the characteristics
and validation of the COMP assay, and then apply it to relatively large groups of patients with
a variety of different joint diseases, whose disease status had been carefully documented both
clinically and radiographically.
What they found was that synovial concentrations of COMP showed similar differences to
those seen when proteoglycans are assayed, suggesting that release is due to changes in the
articular cartilage. They also found that COMP was readily detectable in the serum of normal
controls, suggesting that there is normal turnover of the protein. They also found marked
differences in the levels found in different disease states, suggesting that turnover and
release differs in the different types of joint pathology. In a thoughtful discussion, the
authors consider the difficulties in interpretation of data of this sort, dependent as it is on
synthesis as well as release of COMP, and, in the case of serum assays, on what is happening
in multiple joints.
This paper is important, not only because of the high standard of the work, but also as it is
the original description of this group’s work on COMP, which has proved to be one of the
more promising ‘biochemical markers’ of joint disease. Subsequent work (cited in the relat-
ed references) showed that COMP levels were predictive of joint damage.
Strengths
1. Careful characterization of the COMP assay.
2. Application to large numbers of well characterized patients.
3. Inclusion of normal control subjects.
4. An excellent discussion on the difficulties of interpreting marker assays.
Weaknesses
1. The data are cross-sectional and not prospective.
2. No normal synovial fluid data are included.
Relevance
COMP is one of the few biochemical markers that is both joint-specific and showing signs of
being of value in the prediction of outcome in arthritis, as well as indicative of different
pathological processes.
64
CHAPTER 5
Clinical genetics
Sophia Steer and Gabriel S Panayi
Introduction
In the last two decades the field of human genetics has advanced and expanded perhaps
more than any other in human biology. The realization that the presence of genetic markers
and the study of their transmission through disease pedigrees could localize chromosomal
regions linked with disease (from where positional cloning could be used to identify the pre-
cise disease gene) drove the initial attempts to establish a genetic map. The recognition of
DNA polymorphisms as genetic markers allowed this to start in earnest in the 1980s. The first
generation of markers were restriction fragment length polymorphisms, the typing of which
was laborious. These were superceded by microsatellites which are amenable to typing by
polymerase chain reaction (PCR), and these in turn are being replaced by single nucleotide
polymorphisms. The application of this reverse genetics philosophy (i.e. isolating a disease-
causing gene without knowing its function) coupled with the advance of recombinant DNA
technology has resulted in the identification of genes responsible for many Mendelian dis-
eases. This progress has been relevant to the field of rheumatology with disease loci located
for Marfan’s syndrome, Ehlers–Danlos syndrome and osteogenesis imperfecta, but we have
chosen to focus the papers in this chapter on the more common inflammatory rheumatic
diseases which remain a substantial challenge in terms of gene localization.
The application of similar techniques to these diseases has been less rewarding than ini-
tially hoped. Their pattern of familial clustering suggests that they are polygenic or oli-
gogenic in origin with each contributing susceptibility locus potentially having only a small
effect. Approaches to localizing such disease genes include the candidate gene approach,
where a suitable gene is tested for association with disease, usually on the basis of a putative
pathogenic role, and the use of linkage analysis in family studies.
The first two papers in this chapter are examples of the candidate gene approach. They
both look for the association of an human leucocyte antigen (HLA) gene with disease in a
cohort of unrelated cases and compare the prevalence of the allele with that in a control
population. Both are very successful studies demonstrating a strong association that has in
each case been replicated on many occasions. Before the 1980s the direct analysis of candi-
date genes was the only approach available and the major histocompatibility complex
(MHC) antigens were one of the few genetic determinants known. These studies demon-
strate the power of this approach when there is a good a priori hypothesis for the gene in
question and when cases and controls are well characterized and matched.
Recently candidate gene studies have fallen out of favour largely because findings in one
disease cohort have often failed to be replicated in others. This has been attributed to popu-
lation stratification occurring as a result of inadequate matching of cases and controls.
However candidate gene studies offer huge advantages in terms of power and case ascertain-
ment and are likely to be the future for gene localization studies.
Clinical genetics 65
The use of multicase pedigrees for classical linkage analysis is difficult in polygenic multi-
factorial diseases where the precise mode of inheritance is not known. A best fit model of
transmission of disease is used in the analysis. Errors in this model can lead to false linkage
assignment. Family studies do however have the advantage of internal genetic homogeneity
thus avoiding the potential for false conclusions being drawn due to inadequately matched
case and control populations. Methods that do not rely on models of disease transmission
(nonparametric methods) have been developed to take advantage of this, and the transmis-
sion disequilibrium test described in the fourth paper in this chapter is an example of one of
these. These methods are being applied more widely with promising results but are less pow-
erful than candidate gene studies in case-control populations, and have the inherent difficul-
ties of family finding which can be particularly difficult in late- onset diseases. Methods using
siblings rather than parents in the transmission disequilibrium test have been developed to
make this less of a problem.
The final two papers of the chapter illustrate how the molecular structure of disease sus-
ceptibility genes can both lend support to existing hypotheses of disease causation and pro-
voke further discussion of pathological mechanisms. One of the most persuasive reasons for
investing so much in discovering the genes determining susceptibility and severity in
rheumatic diseases is the elucidation of their underlying aetiology and pathogenesis. This of
course requires not only gene localization but much more in the form of functional in vitro
and in vivo studies.
The third paper in this chapter introduces the field of pharmacogenetics. The ultimate
aim is the development of specific targeted therapy that cures disease (assuming that disease
cannot be averted) without adverse effect in each and every affected individual. The sequenc-
ing of the the entire human genome, combined with the development of genome wide sin-
gle nucleotide polymorphism maps and chip-based technologies, will allow the scale of cur-
rent approaches to increase by several orders of magnitude. Parallel advances in statistical
techniques and approaches that allow case-control studies to emerge as robust and powerful,
combined with functional studies, may allow us to reach this goal.
66
Paper 1
HLA-Dw4 in adult and juvenile rheumatoid

arthritis
Authors
Stastny P, Fink CW
Reference
Transplantation Proceedings 1977; 9(4):1863–1866
Summary
HLA-D typing of patients with adult onset erosive seropositive rheumatoid arthritis (RA) was
performed using the mixed lymphocyte reaction between the patients’ cells and homozygous
stimulating cells for several HLA-D loci. HLA-Dw4 was present in 16% of normal white con-
trols and in 59% of the white RA patients. This difference was highly significant (p<0.001)
and suggested that HLA-Dw4 conferred a relative risk of 7.5 times that of other HLA-D anti-
gens. The frequency of HLA-Dw4 in black Dallas residents with RA was 14% and in patients
of Mexican origin it was 10%, but there were no non-white control groups for comparison.
Related references (1) Stastny P. Association of the B-cell alloantigen DRw4 with rheuma-
toid arthritis. New England Journal of Medicine 1978;
298(16):869–871.
(2) Panayi GS, Wooley P, Batchelor JR. Genetic basis of rheumatoid
disease: HLA antigens, disease manifestations, and toxic reac-
tions to drugs. British Medical Journal 1978; 2(6148):1326–1328.
Key message
The frequency of HLA-Dw4, later reclassified within HLA-DR4, in white patients with adult-
onset RA was significantly increased at 59% compared to a rate of 16% in a white control
population. This increase was not apparent in non-white adults with the disease.

By the early 1970s there was some evidence implicating the immune system in RA, such as
rheumatoid factor production and the development of lymphoid nodules in the synovial
membrane of affected joints. An association of the disease with the HLA antigens seemed
likely but previous studies looking at class I antigens using serological methods had been
negative. The poor response of peripheral blood lymphocytes from patients with RA in
mixed lymphocyte culture (MLC) against cells from other RA patients led Stastny to investi-
gate their lymphocyte defined (LD) histocompatibility determinants, subsequently known as
HLA-D. He demonstrated that the low MLC reactions were due to the sharing of an LD
determinant among RA patients and defined this further as LD-104 which subsequently
became HLA-Dw4.
This paper set out to confirm the preliminary work outlined above and was the first case-
control study of a candidate gene in RA to show a positive association. It demonstrates the
importance of ethnicity in population-based case-control studies as with a less homogeneous
group of patients the result would have been much less significant. Panayi, et al. (2) replicat-
ed these findings using serological methods for detecting HLA-DR antigens and as did
Stastny in the same year (1).
The location of this gene close to the HLA region, a region known to be analogous to the
immune response gene in mice and monkeys, allowed further speculation as to the role of
the immune system in RA. Thus this finding lent support to the rationale for the extensive
immunologically-based research that followed.
Strengths
1. The first example of an association study in RA, a technique that continues to play a
major role in the analysis of complex genetic disease.
2. It placed the genetic basis for RA on a firm footing.
Weakness
None.
Relevance
The association of RA with DR4 remains the strongest genetic association for this disease and
has been pivotal to our understanding of its pathogenesis and the involvement of T cells in
particular.
68
Paper 2
Ankylosing spondylitis and HL-A27

Authors
Brewerton DA, Hart FD, Nicholls A, Caffrey M, James DC, Sturrock RD
Reference
The Lancet 1 (7809) 1973; 904–907
Summary
Sixty-five caucasian men and 10 caucasian women attending hospital rheumatology clinics
with undoubted classical ankylosing spondylitis (AS) underwent HL-A typing using a lympho-
cytotoxicity method. The HL-A27 antigen was identified in 72 out of 75 patients (96%) and
in 3 out of 75 controls (4%). To investigate the possibility that this high frequency might
result from the disease or its treatment 60 first degree relatives were also HL-A typed. Thirty-
one of the 60 (52%) possessed the HL-A27 antigen.
Related references (1) Schlosstein L, Terasaki PI, Bluestone R, Pearson CM. High associa-
tion of an HL-A antigen, W27, with ankylosing spondylitis. New
England Journal of Medicine 1973; 288(14):704–706.
(2) Rubin LA, Amos CI, Wade JA, Martin JR, Bale SJ, Little AH, et al.
Investigating the genetic basis for ankylosing spondylitis.
Linkage studies with the major histocompatibility complex
region. Arthritis and Rheumatism 1994; 37(8):1212–1220.
(3) Taurog JD, Richardson JA, Croft JT, Simmons WA, Zhou M,
Fernandez-Sueiro JL, et al. The germfree state prevents develop-
ment of gut and joint inflammatory disease in HLA-B27 trans-
genic rats. Journal of Experimental Medicine 1994;
180(6):2359–2364.
Key message
HL-A27, later reclassified as HLA-B27, is present in 96% of caucasian patients with AS and in
52% of their first degree relatives versus a control population frequency of 4%.

The suggestion of a genetic contribution to AS was first made in the 1950s following reports
of multicase families and increased disease concordance in monozygotic twins. These provid-
ed the rationale for examining the association of genetic determinants with AS. This paper
reports a highly significant result which was replicated at the same time by Schlosstein, et al.
(1). Claims that this association would lead to discovery of the pathogenic mechanisms
underlying disease and the rational design of treatment were made then but have yet to be
realized.
Despite the strong association of AS with B27 only a small proportion of B27 positive indi-
viduals develop disease and the explanation for this remains elusive. It seems unlikely to be
due to a rare environmental trigger as the disease is geographically widespread and varia-
tions in prevalence seem to reflect variations in B27 gene frequency. Furthermore, twin con-
cordance studies suggest the disease is strongly genetic with a heritability of 90%. The contri-
bution of B27 to the genetic susceptibility of AS is estimated to be 20–50% of the total (2),
with the remaining components coming from a number of weaker susceptibility loci. This is
supported by B27 transgenic rat models of disease where the background strain of the rat
influences the expression of disease. The location of the non-MHC genetic contributions to
disease susceptibility are not known. The first genome wide screen in AS was in 105 affected
sibling pair families and demonstrated several regions of linkage in addition to that at the
MHC locus.
However the absence of disease in transgenic HLA B27 positive rats raised in a germ free
environment (3) emphasizes the importance of interplay between genes and the environ-
ment. Disease causation hypotheses building on the normal role of B27 in its presentation of
antigenic peptide to cytotoxic T cells and natural killer (NK) cells are numerous and include
molecular mimicry and arthritogenic peptide theories, but the pathogenesis of this disease
remains unknown more than 25 years on from this important paper.
Strengths
1. Clear significant result.
2. Well designed with respect to cases and controls.
Weakness
None.
Relevance
This robust association has provided the stimulus for further family and genetic studies of
AS. These are likely to suggest additional candidate genes and improve our understanding of
the disease pathogenesis.
70
Paper 3
HLA-DR antigens and toxic reactions to sodium

aurothiomalate and D-penicillamine in patients
with rheumatoid arthritis
Authors
Wooley PH, Griffin J, Panayi GS, Batchelor MD, Welsh KI, Gibson TJ
Reference
New England Journal of Medicine 1980; 303(6):300–302
Summary
Ninety-one consecutive patients selected because they had classic or definite RA (according
to the American Rheumatism Association criteria) which was severe enough to warrant treat-
ment with sodium aurothiomalate or D-penicillamine, were HLA typed. Their notes were
reviewed for evidence of drug toxicity, classified as rashes, proteinuria or haematological
problems (eosinophilia, thrombocytopenia). Seventy-one patients had toxic reactions to
either or both drugs; the total number of toxic episodes was 95. The remaining 20 patients
took one of the drugs for at least 6 months without toxicity. Nineteen of 24 patients in whom
proteinuria developed were positive for HLA-B8 and HLA-DRw3; 14 of 15 episodes of auroth-
iomalate-induced proteinuria and 9 of 13 episodes of penicillamine-induced proteinuria
occurred in patients with these antigens. The relative risk of proteinuria during aurothioma-
late therapy was increased 32 times in patients with the DRw3 antigen but there was no statis-
tically significant association between HLA-DRw3 and proteinuria during penicillamine ther-
apy. No significant associations were found between any HLA antigen and the development
of skin rashes or haematological complications.
Related references (1) Panayi, GS, Wooley P, Batchelor JR. Genetic basis of rheumatoid
disease: HLA antigens, disease manifestations, and toxic reac-
tions to drugs. British Medical Journal 1978; 2(6148):1326–1328.
(2) Sinigaglia F, Scheidegger D, Garotta G, Scheper R, Pletscher M,
Lanzavecchia A. Isolation and characterization of Ni-specific T
cell clones from patients with Ni-contact dermatitis. Journal of
Immunology 1985; 135(6):3929–3932.
(3) Romagnoli P, Spinas GA, Sinigaglia F. Gold-specific T cells in
rheumatoid arthritis patients treated with gold. Journal of
Clinical Investigation 1992; 89(1):254–258.
Key message
Toxicity during aurothiomalate or penicillamine treatment for RA may be under genetic
control.

A previous study (1) had found a significant association between HLA-DRw2 and DRw3 and
toxic effects of therapy with these two drugs. This study was undertaken to confirm or refute
that association. In confirming an association with HLA-DRw3 the principle that susceptibility
to drug toxicity may be genetically determined was established for rheumatic disease. This
study also implied that mechanisms of toxicity may be under the control of the major histo-
compatibility system and this has since been confirmed in work by Sinigaglia, et al. and others
(2, 3). They have demonstrated the important role of HLA class II molecules in the stimula-
tion and proliferation of nickel and gold specific T cell clones.
The first clinical observations of inherited differences in drug effects such as haemolysis
after antimalarial therapy and inherited levels of glucose 6-phosphate dehydrogenase activity,
and peripheral neuropathy with isoniazid and inherited differences in the acetylation of the
drug were reported in the 1950s. The molecular basis for these began to be elucidated in the
1980s with the discovery of a polymorphic locus encoding the enzyme CYP2D6 and since
then many such pharmacogenetic traits have been discovered. For drugs that have a narrow
therapeutic index and are metabolized by a polymorphic enzyme, for example azathioprine
metabolized by thiopurine methyltransferase, or warfarin metabolized by cytochrome P450,
the influence of such a monogenic trait can be profound in terms of toxicity. The impor-
tance of this was highlighted recently by the high incidence of serious and fatal adverse drug
reactions reported in US hospitals.
Genetic polymorphisms have now been identified in more than 20 drug-metabolizing
enzymes and exist for almost every gene involved in drug metabolism and disposition as well
as for many drug receptors, transporters and cell signalling pathways. The overall effect of a
drug is therefore not a monogenic trait but determined by the interplay between several
genes. The rapidly expanding knowledge of the genetic sequence of target genes combined
with the development of single nucleotide polymorphism (SNP) maps and automated
screening for SNPs will allow the identification of functionally important polymorphisms.
This approach should lead to the development of individual pharmacogenetic profiles and
prescription advice based on genotype. In the longer term it may lead to the development of
new drug regimes specifically for genetically identifiable subgroups of the population with
the aim of improving outcome.
Strength
Established the role of genetics in drug toxicity in rheumatic disease.
Weaknesses
1. Skin rashes not precisely delineated which may explain lack of association with HLA.
2. Relatively small number of cases.
Relevance
The first report of a genetic association with drug toxicity in RA, which has been pivotal in
the development of thinking about tailoring therapies to the genetic backgrounds of individ-
ual patients.
72
Paper 4
Transmission disequilibrium as a test of linkage

and association between HLA alleles and
pauciarticular-onset juvenile rheumatoid arthritis
Authors
Moroldo MB, Donnelly P, Saunders J, Glass DN, Giannini EH
Reference
Arthritis and Rheumatism 1998; 41(9);1620–1624
Summary
This study aimed to establish whether HLA class I and II alleles previously found to be asso-
ciated with pauciarticular-onset juvenile rheumatoid arthritis (POJRA) in population associ-
ation studies are transmitted from heterozygous parents to affected offspring to an extent dif-
ferent from the expected 50%. One hundred and one white North American families that
had been serologically HLA typed during earlier studies were used in the analysis. Each family
had a single child with POJRA and serological HLA data were available for both parents, the
proband, and unaffected siblings in 95% of families. HLA-A2, B27 and B35 were the class I
alleles that showed a significant increased frequency of transmission to affected offspring, as
did the class II alleles HLA-DR5 and DR8. HLA-DR4 was transmitted to affected subjects sig-
nificantly less frequently than expected. When the data were stratified by age and sex, HLA-
A2, HLA-B35, DR5 and DR8 were all transmitted at higher than expected rates to female
patients, particularly among those with a young age at disease onset. HLA-B27 showed strong
transmission disequilibrium among male patients with disease onset after the age of 8 years.
HLA-DR4 was transmitted less frequently than expected to females younger than 8 years at
disease onset.
Related references (1) Stastny P, Fink CW. Different HLA-D associations in adult and
juvenile rheumatoid arthritis. Journal of Clinical Investigation
1979; 63(1):124–130.
(2) Glass D, Litvin D, Wallace K, Chylack L, Garovoy M, Carpenter CB,
et al. Early-onset pauciarticular juvenile rheumatoid arthritis
associated with human leukocyte antigen-DRw5, iritis, and anti-
nuclear antibody. Journal of Clinical Investigation 1980;
66(3):426–429.
Key message
HLA alleles A2, B27, B35, DR5 and DR8 are transmitted from heterozygous parents to affect-
ed offspring significantly more frequently than expected, establishing linkage and associa-
tion between the MHC and POJRA.

In the Western world POJRA is the most common of the group of phenotypically heteroge-
neous childhood arthritides previously classified as juvenile chronic arthritis (JCA) in Europe
and juvenile rheumatoid arthritis (JRA) in the US, now classified as juvenile idiopathic
arthritis (JIA). Although the observed familial occurrence of JIA is low and no large twin
study has been performed for any subgroup, the evidence in support of a genetic basis came
from the observation that first degree relatives can present with the same phenotypic features
of JIA.
The HLA system and the POJRA subgroup have been researched the most intensively. The
first association study in JCA was reported by Stastny and Fink in 1979 (1) and showed a posi-
tive association with HLA-DR8. This was followed shortly by Glass, et al.’s report of association
of POJRA with HLA-DR5 (2). Subsequently associations with HLA-DR6, HLA-DPw2, HLA-A2,
and a specific HLA-DP beta allele were reported in POJRA. All these studies were population-
based association case-control studies which may give false positive results as they are vulnera-
ble to population stratification, particularly in outbred populations, which may in theory
have genetically heterogeneous cases and controls.
By using the transmission disequilibrium test this problem is avoided as the controls are
within family and therefore genetically homogeneous. The disadvantages apart from the
obvious difficulties in recruiting families, include the loss of power due to overmatching of
controls and the test’s ability to detect linkage between a marker locus and a disease locus
only if association due to linkage disequilibrium is present. It is a particularly valid way of
designing a replication study, which is how it has been used here, as it provides definite
confirmation of the previous findings in a way that another population association study
could not.
The strong evidence of linkage and association between multiple HLA alleles and POJRA
from this study leads one to question why this should be. It is not explained by linkage dise-
quilibrium either between different HLA alleles or with other closely related genes. There is
some evidence to suggest that these alleles share a consensus amino acid motif and could
thereby influence disease pathogenesis possibly by affecting the development of the T cell
repertoire, or by binding specific ‘arthritogenic’ peptides. Another explanation would be
that within the subgroup of POJRA there is in fact further clinical heterogeneity and this has
been supported by studies looking at outcomes of this group.
Strengths
1. Excellent study design.
2. Confirms and clarifies the previous population-based association data.
Weakness
Relatively small number of families which may result in real effects being missed.
Relevance
This paper demonstrates the genetic heterogeneity that exists amongst POJRA which raises
questions regarding the clinical homogeneity of the disease. The confirmation of linkage
and association with the MHC region suggests that immune mechanisms are important for
the pathogenesis of POJRA; the involvement of T cells seems likely given the association with
specific HLA-DR molecules.
74
Paper 5
The shared epitope hypothesis. An approach

to understanding the molecular genetics of
susceptibility to rheumatoid arthritis
Authors
Gregersen PK, Silver J, Winchester RJ
Reference
Arthritis and Rheumatism 1987; 30(11);1205–1213
Summary
Much work investigating the inheritance of RA has focussed on the associations between
HLA class II serological specificities and disease. This has demonstrated association with dif-
ferent DR4 serological subtypes in some populations, lack of association with DR4 in others
and association with certain other DR alleles, particularly DR1 in DR4 negative patients.
Analysis at DNA sequence level revealed that the DR4 serological subtypes differed only in
the DRB1 gene and that these differences were restricted to the codons surrounding position
70 of the N-terminal domain of the molecule, the region corresponding to the third hyper-
variable region of the DRB1 molecule. The differences in sequence lead to amino acid sub-
stitutions with substantial implication in terms of charge and therefore protein binding. The
substitutions in the DR4 subtypes associated with RA (Dw4, Dw14, Dw15) and in DR1 are of
similarly charged amino acids, whereas those not associated (Dw10) differ markedly. These
regions of shared sequence therefore have a similar conformation when expressed as pro-
tein, with similar properties in terms of antigen binding, presentation and immune regula-
tion. The protein epitope is thereby shared by serologically distinct HLA haplotypes.
Related references (1) Wordsworth BP, Lanchbury JS, Sakkas LI, Welsh KI, Panayi GS,
Bell JI. HLA-DR4 subtype frequencies in rheumatoid arthritis
indicate that DRB1 is the major susceptibility locus within the
HLA class II region. Proceedings of the National Academy of Sciences
USA 1989; 86(24):10049–10053.
(2) Nepom GT, Seyfried CE, Holbeck SL, Wilske KR, Nepom BS.
Identification of HLA-Dw14 genes in DR4+ rheumatoid arthri-
tis. The Lancet 1986; 2(8514):1002–1005.
(3) Cairns JS, Curtsinger JM, Dahl CA, Freeman S, Alter BJ, Bach FH.
Sequence polymorphism of HLA DR beta 1 alleles relating to T-
cell-recognized determinants. Nature 1985; 317(6033):166–168.
(4) Brown JH, Jardetzky TS, Gorga JC, Stern LJ, Urban RG,
Strominger JL, et al. Three-dimensional structure of the human
class II histocompatibility antigen HLA-DR1 (see comments).
Nature 1993; 364(6432):33–39.
Key message
The seemingly conflicting serological data on HLA class II disease associations in RA can be
explained by analysis of RA populations at the sequence level; the lack of association with the
Dw10 subtype of DR4 and the positive association with DR1 are both accounted for by this
hypothesis.

This hypothesis is based on the immune response being dependent on antigenic peptide
presented in the groove of the class II MHC molecule to the T cell receptor. This concept
suggests that one specific HLA molecule or protein conformation should be associated with
a disease and did not fit easily with the multiple reports of different HLA-DR4 subtypes being
associated with RA in various populations. Previous work involving alloreactive T cell clones
had provided evidence of shared T cell recognition sites amongst class II molecules from
DR4 positive and DR4 negative patients with RA (3), but it was the sequencing of the differ-
ent HLA-DR4 subtypes (1) that finally demonstrated that they shared a common antigen
binding structure.
Further support for the relationship between structure and immune response has since
been derived from experiments involving site directed mutagenesis of the DRB1 gene, and
by the confirmation of the HLA class II molecule crystal structure which has localized the
shared sequence motif in the antigen binding groove (4). Furthermore, since the publica-
tion of this hypothesis the results of studies testing it in patients of different ethnic back-
grounds have lent it additional support (2).
The mechanism through which the shared epitope might determine the immune
response has been debated at length and may be through presentation of arthritogenic pep-
tide to T cells. However there are variations in sequence between DR4 subtypes outside the
shared sequence that may also influence antigen binding which would not be taken into
account by this model. Alternatively the shared epitope may play a role in the selection of the
T cell repertoire via thymic education.
Strength
Explains the majority of the population association data in RA.
Weakness
Does not provide an explanation for DR4/DR1 negative RA.
Relevance
This hypothesis provides an explanation for the majority of population association data in
RA, lends great support to a central component of immunology and provides a framework
for further analysis of class II association data in other immunological diseases.
76
Paper 6
HLA heterozygosity contributes to susceptibility

to rheumatoid arthritis
Authors
Wordsworth P, Pile KD, Buckely JD, Lanchbury JS, Ollier B, Lathrop M, Bell JI
Reference
American Journal of Human Genetics 1992; 51:585–591
Summary
The HLA-DR genotypes of 184 patients with severe RA and of 46 patients with Felty’s syn-
drome were analysed with the aim of establishing the relative contribution of the RA associat-
ed subtypes of DR4 (Dw4, Dw14, Dw15). The relative risks for RA associated with particular
genotypes were calculated using the observed frequencies of these genotypes in patients and
controls; they were calculated relative to DRX/DRX where DRX is any antigen except DR4
or DR1. There was an excess of DR4 homozygotes, particularly Dw4/Dw14 compound het-
erozygotes. The risk associated with the Dw4 subtype of DR4 depended on which other allele
was present. It was highest for the Dw4/Dw14 genotype (RR 49) and also high for the
Dw4/DR1 genotype (RR 21). Dw4/Dw4 homozygotes were at intermediate risk (RR 15)
while Dw4 combined with a non-DR4, non-DR1 allele was associated with the lowest RR of 6.
Four cases of the rare Dw4/Dw15 genotype were found (expected <0.5).
The results were compared with the genotypes of 63 patients with RA who were ascer-
tained on the basis of being known DR4 homozygotes without reference to disease severity.
The excess of Dw4/Dw14 in this group was less apparent and not statistically significant
(RR1.4), suggesting that this genotype may be associated with severe disease.
Related references (1) Nepom BS, Nepom GT, Mickelson E, Schaller JG, Antonelli P,
Hansen JA. Specific HLA-DR4-associated histocompatibility
molecules characterize patients with seropositive juvenile
rheumatoid arthritis. Journal of Clinical Investigation 1984;
74(1):287–291.
(2) Nepom GT, Seyfried CE, Holbeck SL, Wilske KR, Nepom BS.
Identification of HLA-Dw14 genes in DR4+ rheumatoid arthri-
tis. The Lancet 1986; 2(8514):1002–1005.
(3) Weyand CM, Hicok KC, Conn DL, Goronzy JJ. The influence of
HLA-DRB1 genes on disease severity in rheumatoid arthritis
(see comments). Annals of Internal Medicine 1992;
117(10):801–806.
(4) Weyand CM, Xie C, Goronzy JJ. Homozygosity for the HLA-DRB1
allele selects for extraarticular manifestations in rheumatoid
arthritis. Journal of Clinical Investigation 1992; 89(6):2033–2039.
Key message
Combinations of the alleles Dw4, Dw14, Dw15 and DR1 predispose to severe RA.

Prior to this study HLA-D heterozygous genotypes had been identified in seven out of 12
patients with JRA (1) and in six out of seven adult patients with RA who were DR4 homozy-
gotes (2). However, neither the high prevalence of compound heterozygotes amongst DR4
homozygotes nor the significance of the effect of heterozygosity on disease severity were well
established. Much of the previous work addressing HLA association with disease had used
differences in allele frequencies between populations rather than analysis of individual geno-
types. However, shortly after this publication work by Weyand, et al. emphasized the impor-
tance of this approach. They demonstrated that clinical subsets with different disease mani-
festations could be characterized through HLA-DRB1 genotyping. Extra-articular disease was
strongly associated with Dw4 homozygosity (4).
This paper and the work by Weyand et al. (3,4) demonstrate that both HLA haplotypes
contribute to the phenotypic expression of the disease and to the genetic risk of RA. The
mechanism is not clear but some revision of the shared epitope hypothesis is obviously neces-
sary to take account of these findings. It is probably not explained by the concentration of
HLA class II antigens on the cell surface as there was only a limited increase in Dw4/Dw4
and Dw14/Dw14 homozygotes in this study.
The increase in all of Dw4/DR1 and Dw4/Dw15 and Dw4/Dw14 genotypes where Dw14,
Dw15 and DR1 all share identical amino acid sequences differing from Dw4 by a single
amino acid substitution, strongly suggests a role for synergy in mediating the heterozygote
effect. Dw4 and Dw14 haplotypes may act synergistically possibly by enabling more efficient
binding of arthritogenic peptides. The above work suggests that whatever the underlying
mechanism it does not function only at the initiation of disease but also in its progression.
Perhaps by playing a role in T cell selection in the thymus the HLA molecules determine an
increased repertoire capable of recognizing a specific antigen and regulate the overall com-
position and diversity of the T cell compartment. An understanding of this mechanism may
shed more light on the pathogenesis of RA.
Strengths
1. Demonstrates the importance of analysing clinical subsets of disease to obtain clear evi-
dence of a genetic effect.
2. Unequivocally confirms the importance of HLA heterozygosity.
Weakness
None.
Relevance
This paper demonstrates how advances in molecular methods can provoke further question-
ing of central immunological mechanisms and challenge prevailing views of disease patho-
genesis.
79
Section 2
Diseases
81
CHAPTER 6
Rheumatoid arthritis
Frank A Wollheim
Introduction
In several countries Rheumatology was in the past very much focussed on rheumatoid arthri-
tis (RA), which was the major cause of disability. Although the picture has changed RA
remains a major problem for many sufferers amounting to 0.5% to 1% of the adult popula-
tion in most populations. Expanded knowledge of the pathophysiology of joint inflammation
and destruction, as well as therapeutic advances occur at a fast pace. It is possible to control
disease activity and help patients to a better or even normal life. However, we are still igno-
rant regarding aetiology and no cure is in sight.
The selection of ‘the’ most important papers in the vast field encompassed under the title
of RA is an impossible task. I have tried to select papers dealing with both fundamental
issues, such as the nature of RA, when does it start, and what makes it chronic, as well as the
more practical issues regarding classification, clinical assessment and outcome measures.
Each of these topics could easily be dealt with in one whole chapter. Several important
papers regarding genetics, rheumatoid factors (RFs), imaging and pharmacotherapy will be
covered in other chapters. I hope the reader will forgive my biases and find some interest in
the selections.
The selected papers range from very pragmatic ones, like the American Rheumatism
Association (ARA) criteria from 1958 and the Health Assessment Questionnaire (HAQ) from
1980, to important clinical discoveries exemplified by the description of low complement lev-
els in synovial fluid made in 1964 and the description of the nailfold vasculitis in 1956. More
basic classics constitute the discovery of the GM system in 1956 and that of microchimerism
in RA in the 1990s. Classics in drug therapy is covered in another chapter, but one therapy
paper dealing with joint surgery is included.
I would like to dedicate this chapter to two of the great scholars and inspirators, who
helped to make rheumatology academically strong by training generations of young rheuma-
tologists from several countries, Eric GL Bywaters (1) and Morris Ziff (2). The cited papers
bear witness of their stringent and to the point style from which we can still learn.
References
1. Bywaters EGL. Historical aspects of the aetiology of rheumatoid arthritis. British Journal of
Rheumatology 1988; 27(suppl. 2):110–5.
2. Ziff M. Rheumatoid arthritis – Its present and future. Journal of Rheumatology 1990;
17:127–133.
82
Paper 1
1958 revision of diagnostic criteria for

Authors
Ropes MW, Bennet GA, Cobb S, Jacox R, Jessar RA
Reference
Abstract
A committee chaired by Marion Ropes was appointed by the ARA to define criteria best suited to
define RA. A number of physicians ‘particularly interested in rheumatic diseases in various sections
of the United States and Canada’ were asked to each contribute data from their clinics of the five
most recent patients with definite RA, five recent patients with probable RA and five patients with
no evidence of RA. In all 332 patients were analysed for presence of 11 criteria that had been
selected on the basis of clinical experience by the authors. Sensitivity and specificity was calculated
for presence of five or more criteria (definite disease), 3–4 criteria (probable disease), and less than
3 of the 11 criteria (no disease). In order to avoid inclusion of false positives, 19 exclusions were
also listed.
Summary
The 11 selected criteria vary individually with regard to specificity and sensitivity (Table 6.1),
but the combination gave a sensitivity of 0.70 and specificity of 0.91 for greater than five cri-
teria for definite RA, and a sensitivity of 0.88 and a specificity of 0.77 for probable RA. It is
not clear from the publication how much was added when applying the 19 exclusions. It was
pointed out that one of the exclusions, antinuclear antibodies, would remove 3–5% of RA
patients. It was stressed that objective joint changes such as swelling, tenderness and nodules,
had to be verified by a physician.
Table 6.1 Selected criteria to define RA.
1. Morning sickness
2. Pain on motion or tenderness in at least one joint
3. Swelling of one joint, representing soft tissue or fluid
4. Swelling of at least one other joint
5. Symmetrical joint swelling
6. Subcutaneous nodules
7. Typical radiological arthritic changes
8. Positive test for rheumatoid factor in serum
9. Poor mucin precipitate from synovial fluid
10. Characteristic histological changes in synovial membrane
11. Characteristic histopathology of rheumatoid nodules
Rheumatoid arthritis 83
Related references (1) Ropes MW, Bennet GA, Cobb S, Jacox R, Jessar RA. 1958 revision
of diagnostic criteria for rheumatoid arthritis. Bulletin of
(2) Arnett FA, Edworthy SM, Bloch DA, et al. The American
Rheumatism Association 1987 revised criteria for the classifica-
tion of rheumatoid arthritis. Arthritis and Rheumatism 1988;
31:305–314.
Key message
RA can be defined with the help of standardized criteria and further subdivided into cate-
gories of certainty.

Proper classification of a condition is an absolute requirement for prognostic and therapeu-
tic studies. Pathogenetic as well as aetiological research must be based on precise clinical
diagnosis.This was a simple instrument allowing stringent classification of a disease of
unknown cause and variable expression. This was a prerequisite for epidemiological investi-
gations and geographical comparisons between different populations.
Strengths
1. The simple nature of the instrument facilitated universal testing and use.
2. The instrument could be used even if not all criteria were available.
3. The list of criteria was didactically useful in teaching students what RA is about.
Weaknesses
1. The long list of exclusions require extensive work up.
2. The exclusions could falsely exclude patients with features of overlap symptoms.
3. The criteria are not really diagnostic in the clinical setting.
4. The category of possible RA included a substantial number of patients who never devel-
oped established disease.
Relevance
The 1958 revision (1) actually only added one additional exclusion, agammaglobulinaemia.
This revised form was then in universal use for three decades and must be one of the most
cited papers in rheumatological literature. It was replaced in 1988 (2) by what is now called
the 1987 ACR criteria, which are actually only moderately better, although simpler to use.
Many investigators have questioned whether RA is one disease or a designation for a mixed
bag of different conditions. A further signum of relevance was accumulating evidence, that
the number of positive criteria correlated to disease severity and indeed to mortality. The
robust nature of the criteria tell the message, that RA is until proven wrong, indeed one
condition.
84
Paper 2
Prevalence of rheumatoid arthritis

Author
Lawrence JS
Reference
Summary
A random sample of 751 males and 814 females in the town of Leigh in Lancashire and an
area sample of 485 males and 540 females in Wensleydale in Yorkshire, UK, have been inves-
tigated clinically, radiologically, and serologically to determine the prevalence of RA. The
examination was completed in 86% of the Leigh sample and in 87% of the Wensleydale sam-
ple.Using the ARA criteria, the minimal prevalence of ‘definite’ disease was 0.4% in males,
and 1.4% in females, and that of ‘probable’ disease was 1.7% in males and 3.8% in females.
Radiological evidence of erosive arthritis was present in 8% of all those x-rayed, both in
males and females, but the disease was more severe in females. Changes were most frequent-
ly encountered in the cervical spine. A positive sheep cell agglutination test was found in 4%
of males and in 5% of females.
It is estimated that, in Great Britain in 1959, approximately 377 000 males and 1 034 000
females had ‘probable’ or ‘definite’ RA.
Related references (1) Silman AJ. Has the incidence of rheumatoid arthritis decline in
the United Kingdom? British Journal of Rheumatology 1988;
27:77–79.
(2) Gabriel SE, Crowson CS, O’Fallon WM. The epidemiology of
rheumatoid arthritis in Rochester, Minnesota. Arthritis and
Rheumatism 1999; 42:415–420.
(3) Wiles N, Symmons DP, Harrison B, et al. Estimating the incidence
of rheumatoid arthritis: trying to hit a moving target? Arthritis
and Rheumatism 1999; 42:1339–1346.
Key message
RA is a common disease. Its prevalence is 2–3 times higher in women, and the prevalence
increases with age, in particular in women (Figure 6.1). RA is more severe in women.

This was probably the first population-based application of the ARA criteria of 1958. It also
compared a rural and an urban population. Although it involved clinical examination of
some 2500 individuals, it was the work of one author, showing firstly this man’s assiduity and
second, giving proof that the method was practicable. The author states in all modesty, that
his data might be of use to other UK investigators who would like ‘a standard to compare
their own findings’. In fact, solid data on prevalence in a defined population and point in
time, are universally useful.
19 Males 19 Females
18 18
17 17
16 Clinical criteria 1950 16
15 Clinical criteria 1954–58 15
14 ARA criteria 1954–58 14
13 13
Percentage with RA
Percentage with RA
12 12
10 10
9 9
8 8
7 7
6 6
5 5
4 4
3 3
2 2
1 1
0 0
15-14 -34 -44 -54 -64 65+ 15-14 -34 -44 -54 -64 65+
Age (years) Age (years)
Figure 6.1 RA diagnosed clinically in the 1950 complaints survey and the 1954–58 x-ray servey in Leigh,
and by the ARA criteria in the latter; this figure is based on unpublished material. The Wensleydale
data have been excluded.
Strengths
1. Application of new methodology to population-based samples.
2. High response rate strengthens reliability of the observations.
3. Same observer compares different populations.
Weaknesses
1. The radiological data are incomplete and confusing. Probably they do not distinguish RA
from osteoarthritis lesions.
2. The prevalence of individual criteria is not presented.
3. No data on disease duration or functional status are provided.
4. No incidence data are provided.
Relevance
This study, simple as it may seem, has withstood the test of time. Many more recent studies
have confirmed much of the Lawrence data. It is not possible to summarize all important
progress in a few sentences that followed this early study. One area of controversy is whether
or not RA is becoming less severe or less prevalent (1). Only solid population-based prospec-
tive studies are helpful to address this problem (2, 3). Several recent prevalence studies indi-
cate lower than 1% prevalence and higher age of onset in the adult population, but it has
not been convincingly shown that the disease is actually diminishing in defined populations.
86
Paper 3
The ameliorating effect of pregnancy on chronic

atrophic (infectious rheumatoid) arthritis,
fibrositis, and intermittent hydrarthrosis
Author
Hench PS
Reference
Proceedings of the Staff Meetings of the Mayo Clinic 1938; 13:161-167
Summary
‘Rheumatic women have long been devotees of the sun; perhaps it is important for them to
invoke also the kindly and regular auspices of the moon. At least many physicians during the
last century have believed that a normal regular menstrual cycle is good insurance against
rheumatism, and have listed menstrual irregularities among the common predisposing caus-
es of the disease. Some still hold that (1) chronic arthritis is etiologically related to defective
catamenia or other uterine disorders, the menopause, too rapid child-bearing or prolonged
lactation and that (2) pregnancy is dangerous for arthritic women because the joints are apt
to flare up after parturition. I have studied the effect of thirty-seven pregnancies on twenty-
two women with chronic articular disease, and have concluded that, regardless of its after-
math, pregnancy (like jaundice) involves a physiologic state which is decidedly antagonistic
to the continuation of symptoms of certain articular disease, especially chronic atrophic
(infectious rheumatoid) arthritis.’
Key message
The author cites a large number of authorities who have pointed out the dangers of female
reproduction to the joints, including Charcot, AR and AE Garrod, Bannantyne, Kersley and
van Breemen. In contrast only few authors had observed beneficial effects of pregnancy, and
these were mostly related to intermittent hydrathrosis. Although some authors had hinted at
beneficial effects even in ‘atrophic’ arthritis, Hench was impressed by the absence of real
observations and set out to produce just this. 20 of the 22 patients experienced relief, usually
during every pregnancy, and most often within the first month. He also noticed that relapse
was the rule, usually within one or a few months post-partum. Furthermore it was stated that
the two women who did not improve during pregnancies, also had failed to improve during
unrelated episodes of jaundice.
Why it is important
Like so many landmark observations, this one went counter to current dogma, although in
retrospect, there were several hints around pointing to the benefit of pregnancy. The docu-
mentation that symptoms of RA could be influenced in a predictable way created, as Dr
Hench points out a possibility to investigate pathogenesis and search for therapies. On
another note, as also recognized by the author, it was important to be able to inform
patients with RA what a pregnancy could mean to them, and such information did not exist.
Furthermore RA was then and even much later, considered as a relentlessly progressive con-
dition, ‘chronic progressive polyarthritis’, and this paper indicated that this perhaps was not
entirely true.
Strengths
1. This is patientside research asking a distinct question and arriving at a clear-cut result.
2. The paper is based on an original hypothesis and follows a logic from a previous study of
jaundice.
3. The author proves that he is both an independent thinker and has a balanced knowledge
of previous work in the field.
4. The author realises the importance of the observation.
Weakness
1. No predefined definitions of improvement or relapse.
2. The observations seem to be gathered both prospectively and retrospectively, and the mix
could have introduced bias.
3. The results are presented in a verbal and subjective way, without attempt to quantify
results or present objective measures.
4. The follow up period is not standardized, one pregnancy was still ongoing.
Relevance
The implication of this paper are far reaching. The author makes the interesting statement
that it would seem likely, that the agent responsible for improvement would be similar in
jaundice and pregnancy, and that therefore neither bile salts or female sex hormones would
be a likely cause. This, in fact, was the seed for the later discovery of the effect of cortisone
on RA. The relevance of the paper can best be appreciated by considering the 1000 or so
hits relating to pregnancy and RA in Pub Med since 1965, 50 of which were published in the
last year.
88
Paper 4
Participation of monocyte-macrophages and

lymphocytes in the production of a factor that
stimulates collagenase and prostaglandin
release by rheumatoid synovial cells
Authors
Dayer JM, Breard J, Chess L, Krane SM
Reference
Journal of Clinical Investigation 1979; 64:1386–92
Summary
Cultured mononuclear cells from human peripheral blood produce a soluble factor (MCF)
that stimulates collagenase and prostaglandin E2 (PGE2) release by cultured rheumatoid
synovial cells up to several hundred-fold. These target rheumatoid synovial cells lack conven-
tional macrophage markers. To determine which mononuclear cells are the source of MCF,
purified populations of monocyte-macrophages, thymus-derived (T) lymphocytes, and bone
marrow-derived (B) lymphocytes were prepared. The monocyte-macrophages alone pro-
duced levels of MCF that were proportional to cell density but unaffected by phytohaemag-
glutinin or pokeweed mitogen. No detectable collagenase activity was produced by the cul-
tured monocyte-macrophages or lymphocytes. Purified T lymphocytes produced levels of
MCF approximately equal to 1–3% those of purified monocyte-macrophages in the presence
or absence of the above lectins. Purified T lymphocytes modulated the production of MCF by
the monocyte-macrophages, however, in a manner dependent upon relative cell densities
and the presence of lectins. For example, at optimal ratios of T lymphocytes : monocyte-
macrophages, MCF production was markedly stimulated by pokeweed mitogen. Thus, inter-
actions of T lymphocytes and monocyte-macrophages could be important in determining lev-
els of MCF, which regulate collagenase and PGE2 production by target synovial cells in
inflammatory arthritis.
Related references (1) Dayer JM, Robinson DR, Krane SM. Prostaglandin production by
synovial cells: stimulation by a factor from human mononuclear
cells. Journal of Experimental Medicine 1977; 145:1399–1404.
(2) Mizel SB, Dayer JM, Krane SM, Mergenhagen SE. Stimulation of
rheumatoid synovial cell collagenase and prostaglandin produc-
tion by partially purified lymphocyte-activating factor (inter-
leukin 1). Proceedings of the National Academy of Sciences USA 1981;
78:2474–2477.
(3) Burger D, Rezzonico R, Li JM, et al. Imbalance between interstitial
collagenase and tissue inhibitor of metalloproteinases 1 in syn-
oviocytes and fibroblasts upon direct contact with stimulated T
lymphocytes: involvement of membrane-associated cytokines.
Key message
Macrophages interact with synovial fibroblasts and stimulate them to produce mediators of
inflammation. This effect is due to a soluble factor, later called IL-1beta and it can also be
effectuated through direct cell-cell interaction.

This report clearly demonstrated a key interaction between macrophages and synovial cells,
where IL-1 is involved, and which can be modulated by T lymphocytes, and which results in
chronic destructive synovitis. Understanding more about this process could be expected to
identify possible targets for therapeutic intervention.
Strengths
1. The highly original approach to study cell interactions in an ex vivo system.
2. The combination of experimental and analytical skills with biological fantasy.
Weakness
Molecular biology was not yet available to the author in the 1970s. It would have allowed the
author to characterize the factor or indeed factors involved in signalling. This work was left
to other laboratories, and is still not complete.
Relevance
The relevance of this early observations has become increasingly obvious with the evolving
explosion in cytokine research. One need only mention the IL-1 family of cytokines and their
receptors and natural inhibitors, among them IL-1Ra as well as IL-18 and IL-18 binding pro-
tein. The demonstration of stimulation of PGE2 secretion by a factor of 200 antedated the
discovery of cyclooxygenase 2 by a decade. Dr. Dayer was involved in the discovery of IL-1Ra
and of the natural tumour necrosis factor inhibitors, which now are transforming rheumato-
logical therapy.
90
Paper 5
Appearance of anti-HLA-DR reactive cells in

normal and rheumatoid synovium
Authors
Klareskog L, Forsum U, Malmnäs Tjernlund UK, Kabelitz D, Wigren A
Reference
Scandinavian Journal of Immunology 1981; 14:183–192
Summary
The reactivity of rabbit anti-HLA-DR antigen antibodies with cells in normal and rheumatoid
synovial tissue was investigated by indirect immunofluorescence on frozen sections of tissue.
The antibodies reacted with a significant portion of the synovial lining cells of both normal
and rheumatoid synovial tissue, with endothelial cells, and with a number of, most probably,
migratory cells. After dispersion of cells from rheumatoid synovial tissue by digestion with
collagenase and DNase, adherent cells of both a macrophage-like and a dendritic appear-
ance reacted with the HLA-DR antigen antibodies. The adherent cells were also found to be
potent stimulators in the allogeneic MLR. In addition, it was found that a high proportion of
T lymphocytes from both peripheral blood and synovial tissue from rheumatoid patients
bound anti-HLA-DR antibodies. The present data suggest a role for synovial lining cells in
HLA-D-locus-dependent events of importance in the pathogenesis of RA and other joint dis-
eases, and point to the need for further investigations on T lymphocytes derived from the
site of inflammation in the study of RA.
Related references (1) Janossy G, Panayi G, Duke O, et al. Rheumatoid arthritis: a disease
of T-lymphocyte/macrophage immunoregulation. The Lancet
1981 Oct 17; 2(8251):839–844.
(2) Klareskog L, Forsum U, Scheynius A, et al. Evidence in support of a
self-perpetuating HLA-DR-dependent delayed-type cell reaction
in rheumatoid arthritis. Proceedings of the National Academy of
Sciences USA 1982; 79:3632–3636.
(3) Kingsley GH, Panayi GS, Lanchbury O. Immunotherapy of
rheumatic diseases – practice and prospects. Immunology Today
1991; 12:177–179.
(4) Burmester GR, Yu DT, Irani AM, Kunkel HG, Winchester RJ. Ia+ T
cells in synovial fluid and tissues of patients with rheumatoid
Key message
Macrophage/monocytes arriving in the joint of RA patients can activate T lymphocytes by
means of their cell surface HLA-DR expression.

This report in conjunction with The Lancet publication (1), which appeared a few months
later focussed interest on the involvement of antigen presenting cells (APC, Figure 6.2) inter-
acting with resident lymphocytes as a central pathogenetic event in RA. The nature of this
interaction, and ways to influence the interaction or its consequences, are still hot items after
two decades in basic and applied RA related research. The activated T cell presence had
been reported earlier (4).
Figure 6.2 Pathogenesis of RA

Antigen presentation (from Related reference 3).
T cell APC
Migration 'Activated' Regulatory

CD4+ T cells mechanisms
Effector mechanisms
Cytokines, growth factors,
mediators
Immune and non-immune cells
Antibodies
Joint Joint
inflammation destruction
Strengths
1. These investigators built the bridge between established bench technology and the oper-
ating room.
2. They asked the right question – which cells are involved in RA synovitis?
3. They used the right reagents to identify them.
Weaknesses
1. The patients’ clinical data are largely absent.
2. The definition of ‘normal’ tissue can be questioned.
Relevance
This observation stimulated scientists world wide in attempts to identify the antigen. This has
proved frustrating, perhaps because there is not one antigen but several. It has furthermore
motivated therapeutic efforts targeting T lymphocytes, which have been unsuccessful.
Nevertheless, it has given rise to the paradigm which is still useful 20 years later (3).
However, therapies targeting macrophage function are now in the forefront of RA therapy.
Both the Swedish and the British groups of investigators extended observations from other
fields of pathology to RA. It shows how important interfaces and bridges are for progress in
science.
92
Paper 6
Measurement of patient outcome in arthritis

Authors
Fries JF, Spitz P,Kraines RG, Holman HR
Reference
Arthritis and Rheumatism 1980; 23:137–145
Summary
A structure for representation of outcome is presented, together with a method for outcome
measurement and validation of the technique in RA. The paradigm represents outcome by
five separate dimensions: death, discomfort, disability, drug (therapeutic) toxicity, and dollar
cost. Each dimension represents an outcome directly related to patient welfare. Quantitation
of these outcome dimensions may be performed at interview or by patient questionnaire.
With standardized, validated questions, similar scores are achieved by both methods. The
questionnaire technique is preferred since it is inexpensive and does not involve interobserv-
er validation. These techniques appear extremely useful for evaluation of long-term outcome
of patients with rheumatic diseases.
Related references (1) Meenan RF, Gertman PM, Mason JH. Measuring health status in
arthritis. The arthritis impact measuring scales. Arthritis and
Rheumatism 1980; 23:145–152.
(2) Steinbrocker O, Traeger CH, Batterman RC. Therapeutic criteria
in rheumatoid arthritis. Journal of the American Medical Association
1949; 140:659–662.
(3) Fries JF. Toward an understanding of patient outcome measure-
ment. Arthritis and Rheumatism 1983; 26:697–704.
Key message
A simple self administered patient questionnaire was proven to give robust, relevant and
reproducible quantitative information concerning physical health in patients suffering from
established RA.

The HAQ proved for the first time that patients could respond to the right questions in a
reproducible fashion, that the test could be repeated over time and could be used to mea-
sure changes. Furthermore the scale was more continuous than the classical functional class-
es of Steinbrocker (2). The HAQ index allows characterization of the severity of disability in
individual patients and comparisons between patient groups. In contrast to laboratory mea-
sures such as CRP, which relate to the process rather than to the outcome, and surrogate
measures, such as grip strength, which may or may not be critical for function. HAQ is a
direct measure of functional qualities with high patient relevance (3).
Strengths
1. Limited number of questions.
2. Simple standard for answers.
3. Concerns functions of obvious relevance to most patients and may alert health care
providers to important disabilities.
4. Minimal cost to administer.
Weaknesses
1. Patient’s mood may influence answers.
2. Requires translation and validation in different settings.
3. Measures only physical consequences of RA.
Relevance
Every rheumatologist treating RA patients may benefit from the routine use of the HAQ in
daily practice. It is in wide use in therapeutic trials and modifications are developed for other
diseases. HAQ impairment measured in early disease has been shown to predict mortality. It
should be said that the AIMS instrument (1), which was published in the same issue of
Arthritis and Rheumatism as the HAQ, is more comprehensive and measures social perfor-
mance, pain, anxiety and depression as well. This more complex approach is no doubt an
advantage in several situations, but also explains the more limited use.
94
Paper 7
When does rheumatoid arthritis start?

Authors
Aho K, Palosuo T, Raunio V, Puska P, Aromaa A, Salonen JT
Reference
Summary
Some 10 000 sera had been collected and stored in connection with a community-based pop-
ulation survey in eastern Finland in 1972 and 1977. Tests for RF were performed on a repre-
sentative sample amounting to 1500 sera. Thirty of these patients had developed RA as diag-
nosed by specialist physicians between a few months and 9 years after sampling. Nine of these
pre-onset sera were Waaler-Rose positive and 16 latex test positive. The proportion of positive
sera was higher the closer to onset of RA the sera had been drawn.
Related references (1) Hench PS, Rosenberg EF. Palindromic rheumatism: a ‘new’ oft
recurring disease of the joints (arthritis, periarthritis, para-
arthritis) apparently producing no articular residues: report of
34 cases; its relation to ‘angioneural arthrosis’, ‘allergic rheuma-
tism’, and rheumatoid arthritis. Archives of Internal Medicine
1944; 73:293–321.
(2) Williams MH, Sheldon PJDS,Torrigiani G, et al. Palindromic
rheumatism. Annals of Rheumatic Diseases 1971; 30:375–380.
(3) Schumacher HR, Kritridou RC. Synovitis of recent onset. A clinico-
pathologic study during the first month of disease. Arthritis and
Rheumatism 1972; 15:465–485.
(4) Schumacher RH. Palindromic onset of rheumatoid arthritis.
Clinical, synovial fluid, and biopsy studies. Arthritis and
Rheumatism 1982; 25:361–369.
(5) Gonzales-Lopez L, Gamez-Nava JI, Jhangri G, et al. Decreased pro-
gression to rheumatoid arthritis or other connective tissue dis-
eases in patients with palindromic rheumatism treated with
antimalarials. Journal of Rheumatology 2000; 27:41–46.
(6) Gran JT, Husby G, Thorsby E. HLA antigens in palindromic
rheumatism, nonerosive rheumatoid arthritis and classical
rheumatoid arthritis. Journal of Rheumatology 1984;
11(2):136–140.
(7) Aho K, Heliövaara M, Knekt P, et al. Serum immunoglobulins and
the risk of rheumatoid arthritis. Annals of Rheumatic Diseases
1997; 56:351–356.
(8) Masi AT. Sex hormones and rheumatoid arthritis: cause or effect
relationships in a complex pathophysiology. Clinical and
Experimental Rheumatology 1995; 13:227–240.
Key message
Rheumatoid factor positivity is a common finding in asymptomatic patients who later devel-
op RA. In a later extended report increased serum immunoglobulin (Ig) G and IgA levels
were found predictive of RA (7).

Hench first described a dramatic episodic arthritis termed palindromic rheumatism in the
1940s (1), and later it was shown sometimes to develop into RA (2). Although no definite evi-
dence of RA was identified on careful histological examination in patients with very recent
onset of synovitis (3) or palindromic rheumatism later developing RA (4), therapy with anti-
malarials had recently been shown to delay or prevent development of RA in palindromic
rheumatism (5). Decreased levels of male hormone had also been found in stored pre-dis-
ease onset sera (8), pointing to early abnormalities in body fluids which may influence dis-
ease susceptibility. The unique Finnish studies present convincing evidence of ongoing
pathology sometimes preceding RA, even in apparently asymptomatic individuals. These
observations open the possibility to explore pre-clinical events that may pertain to the aetiol-
ogy of RA. The identification of individuals like those in Finland could clearly help to identi-
fy triggering events in the environment.
Strength
The utilization of a unique serum collection to study pre-clinical serology of RA.
Weaknesses
1. Clinical data regarding disease evolution or severity are not presented.
2. It was not clear whether any of the patients had suffered from palindromic rheumatism
before disease onset.
Relevance
The material from the so-called Mini-Finland study has shown that serologic events precede
clinical disease in many patients. This study opens important approaches. It was later shown,
using samples from 124 RA patients from the same material, that hyper IgG and IgA was also
common before disease onset. It is not known whether tissue types were predictive in these
patients. It could be of interest to study the samples of these patients with regard to evidence
of microbial infections or other environmental events. Perhaps one study could try an inter-
ventional approach in line with that in reference (4). It has been established in a Norwegian
study, that HLA-DR 4 is common in patients with palindromic rheumatism developing RA (6).
96
Paper 8
Agglutination of erythrocytes coated with

‘incomplete’ anti-RH by certain rheumatoid
arthritic sera and some other sera
Author
Grubb R
Reference
Acta Pathol Microbiol Scand 1956; 39:195–197
Summary
Some pathological sera, notably certain RA sera, agglutinate Rh-positive cells coated with
‘incomplete’ anti-Rh despite dilution. Human serum can be grouped with the aid of some
rheumatoid sera.
Related references (1) Grubb R, Laurell A-B. Hereditary serological human serum
groups. Acta Pathologica et Microbiologica Scandinavica 1956;
39:390–398.
(2) Eberhardt K, Grubb R, Johnson U, Pettersson H. HLA-DR anti-
gens, Gm allotypes and antiallotypes in early rheumatoid arthri-
tis – their relation to disease progression. Journal of Rheumatology
1993; 20:1825–1829.
(3) Grubb R, Grubb A, Kjellén L, et al. Rheumatoid arthritis – a gene
transfer disease. Experimental and Clinical Immunogenetics 1999;
16:1–7.
(4) Bjarnadottir M, Nathansson C, Balbin M, et al. Nucleotide
sequences specific for nonnominal immunoglobulin allotypes
in rheumatoid arthritis patients and in normal individuals and
their expression in synovial tissue of rheumatoid arthritis
patients. Experimental and Clinical Immunogenetics 1999; 16:8–16.
(5) Nelson JL. Maternal-fetal immunology and autoimmune disease: is
some autoimmune disease auto-alloimmune or allo-autoim-
mune? Arthritis and Rheumatism 1996; 39:191–194.
Key message
Sera of patients with RA have potency to agglutinate sheep erythrocytes coated with sub-
agglutinating doses of human gamma globulin depending on genetic variations of the
employed source. Thus is was possible for the first time to show genetic polymorphism
among human gamma globulins, the Gm (Gamma gobulin marker) system was discovered.

The observation that immunoglobulin allotypes in man display mendelian inheritance
changed the fundamental understanding of immunoglobulin control. The Gm system
became a powerful instrument in population genetics. The significance of anti-Gm antibody
formation in RA is yet to be fully understood. The intriguing observation of the presence of
‘forbidden’ or nonnominal allotypes in RA, made using homozygous RA patients blood and
PCR amplification (4), is still a puzzle which Professor Grubb was trying to elucidate at the
time of his untimely death in 1998. He interpreted the results as indicating transfer of genetic
material by means of viral infection, called transduction. Accordingly he proposed the
hypothesis that RA is a gene transfer disease (3).
Strengths
1. An unexpected observation of antibody activity in RA sera opened a new field of genetics.
2. It has given a handle for exploration into the aetiology of RA.
Weakness
Understanding the full implication of the observation requires a good amount of basic
knowledge. Or as Professor Grubb liked to phrase it: ‘He who is not confused is not
informed’.
Relevance
The Gm system was the first genetic gammaglobulin allotypic variation, several other have
followed. Gm antibodies and the Gm allotypes offer opportunities to test the gene transfer
hypothesis of RA. This involves a virus vector which incorporates foreign antigenic material
into a susceptible recipient, who under the right conditions develops RA. Support for the
hypothesis has been published recently (5). The observation of microchimerism among
patients with RA is another supporting observation. Although no direct toxic effect has been
linked to Gm antibodies in RA, occurrence of anti-G1m(a) antibodies in early disease was
highly predictive of small joint erosions (2).
98
Paper 9
Identification of immunoglobulins and complement

in rheumatoid articular collagenous tissues
Authors
Cooke TD, Hurd ER, Jasin HE , Bienenstock J, Ziff M
Reference
Summary
Ninety-three patients with a variety of joint diseases were studied for evidence of immune
complexes in articular collagenous tissue. Frozen sections of freshly obtained biopsies of hya-
line articular cartilage and menisci were stained with fluoresceinated monospecific antisera
for evidence of human Igs (IgG, IgM,IgA) and the beta1c component of complement. The
criterion for the presence of complexes was the staining of two or more Igs and beta1c in an
identical location of sequentially-cut sections. Of the 42 patients with RA 83% were positive
by this criterion. In those with classic RA the incidence was 92%. Sixteen patients with fresh
joint trauma or non-arthritic disease had negative findings. Among 26 patients with non-
inflammatory disease, four of eight with polyarthritis whose features suggested primary
degeneration, one of 11 patients with secondary degenerative arthritis, and a single case of
synovial osteochondromatosis also had positive findings. Among nine patients with miscella-
neous inflammatory arthritides, all of three with psoriatic arthritis were negative; however
two of six with other inflammatory arthritides were positive. The findings in classic RA sug-
gest that immune complexes are deposited in the articular collagenous tissues. The persis-
tence of these complexes may play a significant role in the chronicity of the synovitis.
Related references (1) Cooke TD, Hurd ER, Ziff M, Jasin HE. The pathogenesis of chron-
ic inflammation in experimental antigen-induced arthritis II.
Preferential localization of antigen-antibody complexes to col-
lagenous tissue. Journal of Experimental Medicine 1972;
135(2):323–338.
(2) Smiley JD, Sachs C, Ziff M. In vitro synthesis of immunoglobulin by
rheumatoid synovial membrane. Journal of Clinical Investigations
1968; 47:624–632.
(3) Ishikawa H, Smiley JD, Ziff M. Electron microscopic demonstra-
tion of immunoglobulin deposition in rheumatoid cartilage.
Key message
The paper establishes the presence of immune complexes in articular cartilage from patients
with inflammatory joint disease. The presence of immunoglobulin in RA cartilage was also
confirmed using electron microscopy (3). This indicates that this tissue may be an important
reservoir of disease perpetuating material.

This paper brings convincing proof of penetration of immune complexes as defined by Ig
plus complement in identical location. The authors had previously shown such deposits in
the antigen-induced model of monarthritis (1). If large proteins like these can penetrate car-
tilage, this tissue could also harbour other macromolecules and these could contribute to
chronic inflammation by immunological mechanisms. An early important paper from Dr.
Ziff’s laboratory (2) had shown that synovial tissue has the same capacity to produce Ig as
lymph nodes.
Strengths
1. Large and well characterized clinical material.
2. Patient application of careful previous animal experiments.
Weaknesses
1. No identification of the putative antigen. However this has still to be found 25 years later.
2. The ‘control’ group is not clearly defined. It included post-traumatic joint problems, but
also ‘other conditions’.
Relevance
These observations clearly implicated cartilage as a possible reservoir of substances attracting
what appeared to be complement-binding immune complexes, and gave rise to an attractive
hypothesis of persisting antigen as a driving force for chronicity in RA. This was before the
cytokine era, and today one would probably implicate cytokines and proteases to have done
the ground work, opening the entrance for the gamma globulin and complement. However
the electron microscopic deposits (3) speak for themselves, and it should not be ruled out
that gamma globulin and complement interact with pro-inflammatory mediators to support
inflammatory tissue damage.
100
Paper 10
Prognosis of rheumatoid arthritis. A prospective

survey over 11 years
Authors
Jacoby RK, Cosh JA, Jayson MI
Reference
British Medical Journal 1973; 2(858):96–100
Summary
One hundred patients with ‘definite’ or ‘classical’ RA were followed in a hospital clinic from
within 1 year of the onset of the arthritis. The average interval between onset and first atten-
dance was 3.7 months. Onset was commoner in the winter, transient prodromal symptoms
being noted in 23 patients, with possible precipitating factors in 14. The serum RF test was
positive at some time in 88 patients.
The patients were reassessed between 8 and 14 years later. Seventeen died during this
period, five possibly as a result of the disease or its treatment. The remaining patients had
improved as a whole in terms of the blood sedimentation rate, haemoglobin, titre of RF test,
and status of the disease, but there was an overall deterioration in functional capacity. Both
the RF titre and the functional capacity at an earlier review could be directly correlated with
the outcome, but other factors were not found to influence the ultimate prognosis.
Related references (1) Reilly PA, Cosh JA, Maddison PJ, et al. Mortality and survival in
rheumatoid arthritis: a 25 year prospective study of 100
patients. Annals of Rheumatic Diseases 1990; 49:363–369.
(2) Corbett M, Dalton S, Young A, et al. Factors predicting death, sur-
vival and functional outcome in a prospective study of early
rheumatoid disease over fifteen years. British Journal of
(3) Fex E, Jonsson K, Johnson U, Eberhardt KB. Development of radi-
ographic damage during the first 5–6 years of rheumatoid
arthritis. A prospective follow-up study of a Swedish cohort.
British Journal of Rheumatology 1996; 35:1106–1115.
Key message
RA is a heterogeneous disease with modestly increased mortality. Survivors have a reduced
inflammatory activity contrasting to progressive joint damage.

This was one of the first long-term cohort studies of RA, which together with its follow-up
report (1) contains a number of important observations. The patients were all seen by Dr.
Cosh within 1 year from onset, and all could be traced for decades or until death. It gives the
causes of death for the 17 patients, four of which were directly related to disease manifesta-
tions: vasculitis, renal amyloidosis, constrictive pericarditis and ‘rheumatic’ heart disease.
The authors also try to identify precipitating events, such as trauma or infection. This
Bath/Bristol cohort, together with a similar one from London (2) form the background to
which later cohort studies can be compared if one wants to address questions related to
change in disease pattern or severity.
Strengths
1. ‘He has the best view who has seen the thing from its start’ (Aristoteles).
2. The comprehensive inclusion of relevant variables.
3. The complete tracing of the patients.
Weaknesses
1. The selection of classical and definite RA only selected for severe disease.
2. No attempt was made to separate natural course from influence of therapy.
Relevance
The two British studies are evidence against some claims of the association between more
benign nature and later age of onset of RA. The radiological data, although qualitative, are
of interest in assessing the more recent detailed quantitation of progression rate (3). Of par-
ticular interest are the mortality data. It is possible, that some of the cardiovascular mortality
may be related to a chronic inflammatory state. With the advent of early aggressive interven-
tion with potent therapeutic agents in RA, it needs to be remembered what the disease evolu-
tion would be without these measures. Such evidence is never easy to identify. We should
perhaps not forget, that many RA patients even before the methotrexate era fared rather
well.
102
Paper 11
Studies on the depressed hemolytic complement

activity of synovial fluid in adult rheumatoid
arthritis
Author
Hedberg H
Reference
Acta Rheumatologica Scandinavica 1963; 9:165–193
Summary
Synovial fluid from the knee joints and serum of 68 patients with various joint diseases were
analysed by measuring the concentration of haemolytic complement (CH50). Decreased
synovial fluid levels were found in patients with SLE and active RA. Lower levels in RA
correlated with disease activity and presence of subcutaneous nodules and rheumatoid fac-
tor. The low levels were found in both fresh and stored fluids. The findings were also consis-
tent in sequential samples. Fluids from patients with seronegative RA had intermediate CH50
levels and those from osteoarthritis patients were not depressed. Serum CH50 levels were
normal in RA but low in SLE.
Related references (1) Pekin JT, Zvaifler NJ. Hemolytic complement in synovial fluid.
Journal of Clinical Investigations 1994; 43:1372–1382.
(2) Hedberg H. Studies on synovial fluid in arthritis. I. The total com-
plement activity. Acta Medica Scandinavica 1997;
479(suppl.):1–137.
(3) Franco AE, Schur PH. Hypocomplementemia in rheumatoid
(4) Winchester RJ, Agnello V, Kunkel HG. The joint fluid gammaG-
globulin complexes and their relationship to intraarticular com-
plement diminution. Annals of the New York Academy of Sciences
1969; 168:195–203.
Key message
Synovial fluids collected from patients with RF positive RA are characterized by a diminished
concentration of haemolytic complement in contrast to fluids from patients with OA.

The observation of unexpectedly low total complement levels in the synovial fluid was made
independently in another laboratory (1). It pointed at local complement consumption in the
joint and thus was an important clue to pathogenesis of seropositive RA. Serum complement
levels have been studied by a number of investigators and often considered unaffected.
However, this is not the case. If one considers that two dominating components of comple-
ment, C3 and C4, are regulated as acute phase components, one would expect elevated
rather than normal or low levels, as are often observed (2, 3). Thus it seems evident that
complement consumption is a feature of active RA. A direct correlation of low synovial fluid
levels of complement and amount of immune complexes supports a role for immune com-
plexes in the pathogenesis of RA.
Strengths
1. A sound methodology approach allowing comparison between individuals.
2. Looks at the site of pathology rather than just in the blood.
3. Realization of the importance of the observation.
Weakness
Paper has a dull lay out and was not published in a high impact journal.
Relevance
Initial hopes that testing for synovial fluid complement levels would become a diagnostic tool
to distinguish different forms of arthritis have not held up. Complement is important in the
normal immune defence, and genetic deficiency is often correlated to lupus-like syndromes
or susceptibility to infection. The precise role of low complement in RA has not been fully
worked out. Figure 6.3 clearly documents selective local complement consumption.
Figure 6.3 Ratio of

C'SF/C's C'SF/C'S synovial fluid/serum
1.0 1.0 RA class.
Controls (24 cases)
concentrations of
(17 cases) CH50 and alpha-
globulin in controls
and RA patients (from
0.5 0.5 Heberg [1963]).
0 0
0 0.5 1.0 0 0.5 1.0
αSF/αS αSF/αS
104
Paper 12
Orthopedic surgery in rheumatoid arthritis

Authors
LaineV, Vainio K
Reference
Bulletin of Rheumatic Diseases 1964; 15:360–361
Summary
Since 1952, when close cooperation was established between the rheumatologist and
orthopaedic surgeon, more than 6500 operations have been performed on patients with RA
at the Rheumatism Foundation Hospital (317 beds) in Heinola, Finland. In summary our
experience shows that patients with RA tolerate orthopaedic surgery very well and with few
complications. The results are best when surgery is performed early in the disease before
severe deformities are produced. However, much can be offered to the severely deformed
patient by surgical procedures.
Related references (1) Vainio K. History of surgery of rheumatoid arthritis in Europe.

Scandinavian Journal of Rheumatology 1983; 12:65–68.
(2) Laine K. General discussion. In: Early Synovectomy in Rheumatoid
Arthritis, W Hijmans, WD Paul (eds.). Amsterdam: Excerpata
Medica Foundation, 1969:214–215.
Key message
Patients with active RA tolerate joint surgery and benefit in the short-term from so-called
‘early synovectomy’.

As Vainio points out (1) the history of orthopaedic surgery in general was rather short and
surgery was mostly used as a last resource in rehabilitation of the patient with advanced
‘burn out’ disease. This was to a large extent due to fears that the surgical procedure could
somehow activate the rheumatoid process. It took some courage to challenge this view, but
starting in the early 1960s, rheumatologists and orthopaedic surgeons in large numbers
visited Heinola and exported the message. This led to the establishment of several centres
modelled after the Heinola experience in Europe and overseas. In particular great hopes
were put on early synovectomy as a means to halt the disease process or at least ameliorate
its consequences. Dr. Laine (2) realized the difficulties relating to controlled trials of the
outcome of surgical interventions. Time has shown that early synovectomy is effective to
relieve pain but does not halt progression of destruction. Nevertheless, the Heinola
experiences formed a foundation upon which later surgical rehabilitation was built, largely
consisting of total joint replacement.
Strengths
1. The establishment of interdisciplinary cooperation around the RA patient.
2. The appreciation of difficulties of standardizing surgical technique.
3. The coining of stages of joint damage where ‘early’ and ‘late’ refer to severity rather than
to duration of the disease.
Weaknesses
1. No controlled trials emerged from Heinola regarding effects of surgery.
2. Failure to promote the idea that outcome assessment should be performed by the
rheumatologist or some other neutral assessors, rather than by the orthopaedic surgeon
responsible for the procedure.
Relevance
The introduction of joint surgery as a routine therapy in management of RA (Figure 6.4) was
an immense step forward. Although synovectomies did not stop progression, which was the
initial hope, it provided some years of relief to a large number of patients. It also stimulated
widespread interest in the development of better procedures for joint replacement, which
continues to be a mainstay of successful rehabilitation in RA. It has a long history.
Figure 6.4 Number of admissions and number

2000
of operations (from Laine and Vainio [1964]).
1500
1000
500
1953 1958 1963

Rheumatism Foundation Hospital
Admissions
Operations
106
Paper 13
Peripheral vascular obstruction in rheumatoid

arthritis and its relation to other vascular leisons
Author
Bywaters EGL
Reference
Summary
Ten cases of RA have been described, which showed vessel obliteration associated with symp-
toms ranging from small infarcts, presenting as brown lesions of the nailfold, nailedge, or
digital pulp, to gangrene of all four limbs with visceral lesions. This is not thought to repre-
sent polyartheritis nodusa or Buerger’s disease, but to resemble more closely the arterial
changes seen in scleroderma and SLE, and perhaps also the changes described as ‘remittent
necrotizing acrocyanosis’ (1) or ‘non-specific obliterative angiitis’ (2). None showed the ordi-
nary symptoms of Buerger’s disease, although all except one were ambulant. It is not felt that
cortisone played any important part, since only three of the 10 had had such preceding treat-
ment.
Related references (1) Edvards EA. Journal of the American Medical Association 1956;
161:1530.
(2) Pennoch LL, Primas DH. Angiology 1956; 7:32.
Key message
Occluding vascular lesions causing nailfold infarcts or major gangrene occur in rare cases of
RA and can be distinguished from polyangiitis nodusa.

The realization that RA is a systemic disease was based on careful clinical and pathological
examinations and only individuals with expertise in both fields could see relevant connec-
tions. Professor Bywaters was both pathologist and clinician and had the insight to distin-
guish the novel type of nailfold necrotic pathology now called ‘Bywater’s lesion’(Figure 6.5).
Classification of conditions with vascular occlusion was never easy and this paper does not
pretend to give a final answer. However it points to a new type of pathology easily recognized
based on the careful description and illustration in this article. The observation that corti-
sone therapy was not a likely cause is also important, and has been amply supported by
numerous later observations.
Strengths
1. The stringent clinical description, relevant pathology, and the realization that nailfold
necrosis and pulp lesions were related.
2. A large number of very fine illustrations.
3. The balanced discussion of published and unpublished observations.
4. The realization that the real knowledge of the nature of the lesions was not at hand.
5. The realization of the existence of rare cases of ‘malignant’ RA.
Weaknesses
1. Perhaps giant cell artheritis (GCA) should have been discussed as a differential diagnosis.
Some of the arterial lesions bear similarities to GCA.
2. The statement that the lesions were not specific for RA.
Relevance
The observations have since been included in medical student teaching, and it appears that
Bywater’s lesions are indeed specific for RA. The observation that rheumatoid nodules were
present in all cases indicated the possibility that vascular changes could have pathogenetic
importance for the formation of rheumatoid nodules.
Figures 6.5 Nailfold lesions (from Bywaters [1957]).

108
CHAPTER 7
Chronic arthritis in
children
Jacqui Clinch and †Barbara Ansell
Introduction
Juvenile idiopathic arthritis is the most common rheumatic condition of childhood, with an
annual incidence of 1.4 cases per 10 000 under the age of 16 years and a prevalence of 1.0
per 1000. It was in 1897 that George Frederic Still described a series of 22 cases of polyarthri-
tis in childhood. Twelve cases had fever and florid lymphadenopathy; with these cases he
brought together almost all the distinctive features of systemic-onset childhood arthritis
except for the rash which was subsequently investigated by Isdale and Bywaters. At the end of
the Second World War, the Canadian government presented the hospital at Taplow,
England, to the people of Britain to serve as a memorial to the work of the Canadian Red
Cross Society in this country during the War. It was established as a special unit for juvenile
rheumatism under the directorship of Professor Eric Bywaters. Initially most children were
admitted with rheumatic fever, in later years chronic arthritis became more prevalent.
Still considered that there were different entities of childhood arthritis and this view has
been confirmed and many different classification systems have ensued. The terms ‘juvenile
rheumatoid arthritis’ (JRA) and ‘juvenile chronic arthritis’ (JCA), both major sources of dis-
agreement in the past, were discarded. In 1997 Petty and Southwood proposed the
International League of Associations for Rheumatologists (ILAR) classification for juvenile
idiopathic arthritis (JIA) (Paper 2). This was adopted as an umbrella term to indicate disease
of childhood-onset, characterized primarily by arthritis persisting for at least 6 weeks, and
currently having no known cause. The ILAR classification divides the disease into seven cate-
gories; systemic-onset, oligoarthritis (persistent or extending), polyarthritis (rheumatoid fac-
tor (RF) negative), polyarthritis (RF positive), psoriatic arthritis, enthesitis-related arthritis
and ‘other arthritis’.
The aetiology of JIA remains elusive. Many factors have been implemented but none
proven. Possible causes include infection, autoimmunity, trauma and stress, with genetic pre-
disposition having an important role. There is mounting evidence that viral and bacterial
infections are associated with both acute and chronic childhood arthritis. In 1985 Chantler
and co-workers described the finding of persistent rubella infection in children with JIA.
More recently streptococcal infections have become the target of much research. The associ-
ation between JIA and iridocyclitis was noted in the 19th century where children with seem-
ingly low-grade arthritis developing ocular inflammation that led to irreversible visual impair-
ment. Smiley and co-workers described a large cohort of children with Still’s disease in
Taplow (Paper 3) who underwent opthalmological investigations regularly over a set time
period.
Treatment up until recent years concentrated on first-line anti-inflammatory medications
such as non-steroidal anti-inflammatory drugs (NSAIDS) and corticosteroids. Approximately
one-third of all patients with JIA achieve adequate control of their disease with NSAIDS.
However, these medications do need to be monitored particularly with regard to gastric irri-
Chronic arthritis in children 109
tation and, in the higher doses, renal impairment. Corticosteroids allow rapid reduction in
inflammation while second-line agents have time to reach therapeutic levels. The side-effects
of steroids in children (including the marked reduction in growth velocity) render their
long-term use undesirable. More recently the treatment of difficult JIA has been revolution-
ized by the anti-metabolite methotrexate. In 1992 a large multicentred trial concluded that
methotrexate given weekly in low doses was extremely effective for children with resistant JIA
(Paper 1). Later in that year Wallace and Sherry reported their encouraging experience with
higher doses of methotrexate, particularly in the oligo- and polyarticular subtypes. Over the
past 2 years much research has been published in the adult literature regarding Cox-2
inhibitors, tumour necrosis factor (TNF)-alpha receptor inhibitors and other biological ther-
apies. Long-term outlook with these therapies in children is currently unknown.
In three of the papers reviewed the new ILAR classification was not in existence. Hence
the terms JRA, JCA and Still’s disease are used. These terminologies are now combined
under the umbrella term JIA.
110
Paper 1
Methotrexate in resistant juvenile rheumatoid

arthritis: Results of the USA–USSR double-blind,
placebo-controlled trial
Authors
Giannini EH, Brewer EJ, Kuzmina N, Shaikov A, Maximov A, Vorontsov I,
Fink C, Newman A, Cassidy J, Zemel L
Reference
New England Journal of Medicine 1992; 326:1043–1049
Summary
One hundred and fourteen children were entered into a multicentre, randomized, double-
blind clinical trial designed to evaluate the effectiveness and safety of orally administered
methotrexate in JRA. The children, with a mean duration of disease of 5.1 years, were ran-
domized to receive 10 mg/m2 methotrexate, 5mg/m2 methotrexate, or placebo weekly for 6
months. Two NSAIDs and prednisolone less than 10 mg daily were allowed. The authors
showed that the higher dose of methotrexate was extremely effective in the treatment of
resistant JRA. Only three children had the drug discontinued because of mild side-effects.
Key message
First large randomized trial showing that low-dose methotrexate has anti-inflammatory activi-
ty and clinical effectiveness in resistant JRA.

Prior to methotrexate the choice of second-line agents in children with resistant JRA was lim-
ited both by clinical effectiveness of the medications and some severe toxic side-effects.
Cyclophosphamide, cyclosporin, penicillamine and sulphasalazine are all used in resistant
cases but many are discontinued due to side-effects or no perceived benefit. High-dose
steroids, previously used as long-term treatments, have proved disastrous to many children
and young adults who have markedly delayed growth velocity, severe osteoporosis and
changes in body appearance that affect morale and social integration.
Methotrexate has distinct advantages over other second-line agents, including its oral or
subcutaneous route, once weekly dosage and lack of known long-term side-effects (such as
oncogenicity and fertility). This trial also illustrated a trend towards a dose-response relation
with the higher dose proving more effective. Methotrexate is now the standard second-line
treatment of pauci- and polyarticular JRA that is not controlled by NSAIDs alone. Subsequent
papers have gone on to show increased effectiveness in higher doses than used here and also
benefit from the subcutaneous route of administration.
Strengths
1. Large randomized controlled trial.
2. Wide range of outcome measures.
3. Clearly shows the benefit of methotrexate in children with resistant disease.
Weakness
No long-term data on clinical effectiveness and toxicity.
Relevance
The treatment of resistant JIA has always involved weighing the risks of second-line treat-
ments against benefits they may provide. Methotrexate has provided physicians with an effec-
tive, apparently safer option in these difficult cases.
112
Paper 2
Revision of the proposed classification criteria for

juvenile idiopathic arthritis: Durban 1997
Authors
Petty RE, Southwood T, Baum J, Bhettay E, Glass D, Manners P, Maldonado-
Cocco J, Suarez-Almazor M, Orozco-Alcala J, Prier A-M
Reference
The Journal of Rheumatology 1998; 25:10 1991–1994
Summary
For many years a united international classification for chronic childhood arthritis has been
absent. The lack of consensus has led to difficulties in trial structure and data interpretation.
The authors put forward the new ILAR classification in which the term JIA replaces JRA and
JCA. The classification is based on clinical characteristics during the first 6 months of disease
and divided into seven distinct categories. The ILAR classification aspires to define cate-
gories of arthritis that are clinically homogenous such that to a degree this may predict
response to therapy and outcome. A detailed account of the new classification and possible
pitfalls is given.
Key message
The ILAR classification arguably provides an international classification system for chronic
childhood arthritis that allows professional unity and a structured approach to ongoing and
future trials.

The classifications of childhood arthritis represent attempts of many generations of physi-
cians to introduce order in what is often viewed as heterogeneity of chronic inflammatory
joint disease. Previously there have been two different classifications (North American (ACR)
and European (EULAR)) which have made comparability of clinical trials difficult and inter-
pretation of serological, epidemiological and genetic data almost impossible.
Using an international assembly of paediatric rheumatologists unambiguous terminology
was developed to describe seven separate categories of childhood arthritis. These categories
are based on clinical description in the hope that clinical clarity may contribute to under-
standing of the disease processes. The ILAR classification highlighted psoriatic arthritis as a
separate entity, whereas previously this heterogeneous form of childhood arthritis was
included in the spondyloarthropathy category and felt to be a ‘contaminant’.
Oligoarticular diseases also received attention. Many paediatric rheumatologists have
reported initial oligoarticular arthritis extending into a more difficult polyarticular form that
is RF negative. By reclassifying this as an extending oligarticular JIA alternate, more aggres-
sive therapies can be introduced earlier and multidisciplinary input ‘stepped up’.
Strengths
1. Provides a clear, primarily clinical classification of chronic childhood arthritis.
2. Devised by international working party thus enables comparability of outcome data and
further research progress.
Weakness
There remains an ‘other’ category that serves as a potentially confusing area for collections
of symptoms that do not fit elsewhere. A caveat to this however is that, with careful evalua-
tion, these children could enable other categories to be developed or existing ones broad-
ened.
Relevance
Paediatric rheumatology is an expanding speciality with increasing numbers of paediatricians
and rheumatologists becoming involved in the care of children with complex forms of arthri-
tis. It is important that all physicians involved use the same classification internationally so
that we can work together to further our understanding of chronic childhood arthritis.
114
Paper 3
Iridocyclitis in Still’s disease: its complictions

and treatment
Authors
Smiley WK, May E, Bywaters EGL
Reference
Annals of Rheumatic Diseases 1957; 16:371
Summary
This paper is one of the original, comprehensive descriptions of iridocyclitis associated with
JIA (referred to here as Still’s disease). The authors looked at their own population in
Taplow and also comprehensively reviewed the literature. One hundred and eighty-three
children with Still’s disease were examined regularly over a range of 1–9 years after the first
appointment for ocular abnormalities. Ten cases were found. The clinical characteristics and
complications of iridocyclitis are described. The insidious nature of iridocyclitis is stressed,
particularly as this often leads to progressive impairment of vision unless treated. Band ker-
atopathy, secondary glaucoma, cataracts and eventual blindness were all described in their
cohort. Treatment with mydriatics, steroids (topically and systemically) and surgery are
described.
Key message
There is a strong association between Still’s disease and iridocyclitis. The onset of this ocular
inflammation may be silent and, unless recognized and treated early, leads to permanent
visual disability.

This paper is the first large study of children with chronic arthritis that particularly investi-
gates ocular inflammation. Previous to this study there had been many reports of associated
eye disease but no prospective data on larger numbers. Smiley and co-workers concentrate
on the complications of iridocyclitis in Still’s disease and the importance of diagnosing and
treating as early as possible. Many of the complications they describe have not previously
been associated with Still’s disease, in particular band keratopathy, a condition not associated
with iridocyclitis in adults.
Iridocyclitis has been the subject of much research in paediatric rheumatology since these
reports from Taplow and other institutions. We now know that chronic anterior uveitis (iri-
docyclitis) is strongly associated with oligoarticular disease, anti-nuclear antibodies and the
female sex (approximately 20% incidence in this group). It is less common in polyarticular
disease and very rarely seen in the systemic disease. The importance of early investigations
has been further supported in subsequent studies. In almost half of patients, iridocyclitis is
present at the time the arthritis is diagnosed. The risk of developing ocular inflammation
continues for at least 5 years even if the joint disease is quiescent.
Strengths
1. Original paper looking prospectively at a large cohort of children with arthritis and their
risk of developing iridocyclitis.
2. Detailed descriptions of iridocyclitis and its many complications seen in children with
chronic arthritis.
3. Comprehensive review of other literature published in this period.
Weaknesses
1. The importance of long-term ocular follow-up in chronic arthritis was not stressed at this
time. The incidence of 5.5% quoted is low compared to current figures and probably rep-
resents shorter follow-up and the heterogeneity of the group they observed.
2. The accounts of their cases are occasionally repetitive.
Relevance
Chronic anterior uveitis continues to cause significant morbidity in children who often suffer
very mild episodes of joint inflammation. These children may present late to specialist cen-
tres with advanced eye disease that proves resistant to the treatment that earlier would have
restored normal vision.
116
Paper 4
Prognosis in juvenile chronic polyarthritis

Authors
Ansell BA, Wood PH
Reference
Clinics in Rheumatic Diseases 1976; 2:397–412
Summary
This paper documents the clinical characteristics and outcome of 243 children who fulfilled
the criteria for Still’s disease (then relating to a juvenile chronic polyarthritis) when first
seen. The children were reviewed regularly during a 15 year period. The authors divide the
children into five main clinical disease categories; JRA (RF +ve), monoarticular disease, sys-
temic disease, juvenile ankylosing spondylitis, and chronic iridocyclitis Still’s disease (in this
last group the authors recognize the association of iridocyclitis in young females with
monoarticular disease).
There is a thorough review of outcomes looking at disability, mortality and behaviour.
Much of the data presented highlights the improved outcome of many forms of childhood
arthritis when compared to adult disease. Overall mortality was 7%; disability measures
showed that 40% recovered with no limitation and 30% with slight limitation, those that pre-
sented later in their disease had a less favourable outcome. Quality of life in these children at
the end of their paediatric follow-up is discussed.
Key message
Childhood arthritis, when diagnosed and treated early, has an encouraging prognosis in
many children.

This series (collected between 1947 and 1958) enabled the comprehensive study of the char-
acteristics and outcome of childhood chronic arthritis. Previous reports from orthopaedic
centres painted a depressing picture; in contrast the experience at Taplow showed the out-
comes of these diseases to be improved (particularly when compared to adult disease).
Prognosis can only be assessed when different outcome measures are known for each ‘set’ of
characteristics. The authors here were able to describe prospectively the identity of variants
in childhood polyarthritis as the disease progressed and group them accordingly. From this
data they were able to look at outcome measures for each group and thus develop prognostic
indicators. The delay in time-to-diagnosis and treatment in its relation to ultimate outcome is
highlighted; particularly recognizing the ‘reversibility’ of many forms of juvenile arthritis
when dealt with in the early stages.
From this series Ansell and Wood are able to describe five main syndromes. Of particular
interest are the specific characteristics they found with JRA, iridocyclitis and systemic disease.
They show the association between clinical differences in different groups and correlation
with the differences in ultimate outcome. Their mortality and long-term disability figures
provided early evidence that juvenile chronic arthritis had a better overall outcome (without
subdividing into disease categories) than adult disease.
Strengths
1. Original paper prospectively studying a large series of children over a long period (15
years).
2. Detailed clinical descriptions of clinical variants within this initially homogenous group.
3. Relates outcome measures to presentation, clinical developments of the disease and envi-
ronment.
4. Comprehensive literature review.
Weaknesses
1. Series of children and their long-term outlook in period of time when therapeutic possi-
bilities more limited.
2. Unable to assign a significant minority of children to a diagnostic group at this time (not
able to study human leucocyte antigen (HLA) or anti-nuclear antibody (ANA) status),
thus degree of difficulty in assessing their outcomes.
Relevance
Many forms of childhood arthritis are now recognized to have a favourable outlook, particu-
larly when diagnosed and treated early. In a majority of cases we are able to demonstrate clin-
ical signs and symptoms that fit a category in which we are able to advise the child and par-
ents on ultimate outcome of the disease. This early work of Ansell and Wood is of utmost
importance in developing the diagnostic and outcome measures we have today.
118
CHAPTER 8
Ankylosing spondylitis
Muhammad Asim Khan
Introduction
Ankylosing spondylitis (AS) has a long and interesting history. There is evidence favouring
antiquity, although there can be some difficulty differentiating AS from DISH (diffuse idio-
pathic skeletal hyperostosis) when examining paleopathological specimens (1). The skeletal
radiograph of the famous Egyptian Pharoah Rameses II (1290–1221 BC) shows evidence of
the disease (2). However, recognition of AS as a clinical entity came in the 19th century;
Brodie (1841) is sometimes credited with the definitive description, although the disease has
eponyms that come from the later work of Strümpell (1897), Marie (1898), and von
Bechterew (1893) (3).
Developments in the 20th century started with the application of clinical radiography,
allowing the detection of the characteristic sacroiliitis and spondylitis. In the imaging chapter
an early example (Fraenkel, 1907) is reviewed (chapter 2), a later more comprehensive
description of the imaging changes came from Buckley in 1931 (4). However, in spite of
these early descriptions, it was not until the second half of the century that AS became clearly
differentiated from rheumatoid arthritis (RA), and as recently as 1960 Graham was having to
argue that the term ‘rheumatoid spondylitis’ was unhelpful (5).
Our selection of papers spans the second half of the 20th century, which has seen major
developments in our understanding of AS, and are confined to key clinical developments.
Paper 1 is the definitive description of the major characteristic pathological feature of the
condition, i.e.enthesitis, and the next two papers (Papers 2a and b) are the definitive clinical
descriptions, both of AS and of the family of disorders known as the ‘spondyloarthritides’,
more often called ‘spondyloarthropathies’. The ground breaking work relating AS to the
HLA-B27 gene was published in 1973, and is described in the clinical genetics chapter of this
book (chapter 5). This was a great stimulus to research on AS, which had been relatively
neglected up to that point. It has led to major recent developments in our understanding
through highly sophisticated work on the structure and function of HLA-B27, and with trans-
genic mice (6). The third selection (Paper 3) describes the more clinically orientated work
that led to an association being made between the ‘HLA-B27 positive spondyloarthropathies’
and gastrointestinal inflammation, even if there were no signs or symptoms of inflammatory
bowel disease (although it transpires that the gut inflammation is not directly associated with
HLA-B27). This type of work has fuelled the continuing debate that gut infections and
inflammation are critical in the pathogenesis of AS and related disorders. In this context it is
of interest that sulphasalazine can be of therapeutic value in AS, but only for peripheral
arthritis, and not the axial disease.
Ankylosing spondylitis 119
The final selections are concerned with the very important issues of classification and
assessment. There have been a variety of systems used as potential diagnostic or classification
criteria, but the two best appear to be those of the European Spondyloarthropathy Study
Group (ESSG) and of Bernard Amor, as described in Papers 4a and 4b. The assessment of
disease activity and progression has been a major obstacle to clinical research in AS.
However, recent work, described in Papers 5a and 5b have led to the development of useful
assessment instruments, now called the WHO/ILAR core sets and remission criteria that
have been found to be very helpful in fully assessing the dramatic efficacy of the TNF-alpha
blocking therapy in AS (7).
References
1. Rogers J, Watt I, Dieppe P. Paleopathology of spinal osteophytosis, vertebral ankylosis,
ankylosing spondylitis and vertebral hyperostosis. Annals of Rheumatic Diseases 1985;
44:113–120.
2. Dastugue J. Les maladies de nos ancestres. La Recherche 1982;13:980–988.
3. O’Connell D. Ankylosing spondylitis. The literature up to the close of the 19th century.
4. Buckley CW. Spondylitis deformans. British Medical Journal 1931; 1:1108-12.
5. Graham W. Is rheumatoid arthritis a separate entity? Arthritis and Rheumatism 1960; 3:
88–90.
6. Taurog JD, Maika SD, Satumtira N, et al. Inflammatory disease in HLA-B27 transgenic rats.
Immunological Reviews 1999; 169:209–213.
7. Brant J, Haibel H, Cornely D, et al. Successful treatment of active ankylosing spondylitis
with the anti-tumor necrosis factor alpha monoclonal antibody infliximab. Arthritis and
Rheumatism 2000; 43:1346–1352.
120
Paper 1
Enthesopathy of rheumatoid and ankylosing

spondylitis
Author
Ball J
Reference
Summary
This paper, which is the Heberden Oration in 1970, for the first time clearly demonstrated
the differential pathology of AS and RA. It is a histological description of autopsies and biop-
sy data, and represents changes observed in relatively advanced AS. The involved sites in AS
show multiple focal inflammatory lesions, primarily localized to the ligamentous attachments
with active erosive lesions, and also lesions in various stages of healing, that may culminate in
new bone formation and progressive ankylosis of the axial skeleton.
Ball emphasized a striking and characteristic inflammatory enthesopathy (enthesitis) in
AS, and pointed out that the natural history or course of this enthesitis could explain some
of the clinical and pathological features of this disease. He demonstrated that the enthesitis
observed in the extraspinal sites was also present in the axial skeleton of AS patients, involv-
ing the sacroiliac and apophyseal joints, and the intervertebral discs. The enthesis is the
point of bony attachment of the tendon, ligament, joint capsule, or fascia to bone. It is a
metabolically active site, especially during the growth phase, and it is of interest that AS
oftens begins in adolescence and young adults. Enthesopathy is seen in many arthropathies,
but inflammatory enthesopathy is the cardinal feature and possible unifying inflammatory
lesion of AS and related spondyloarthropathies (SpA). This enthesitis often leads to new
bone formation but the underlying mechanism is not yet understood.
Related references (1) Braun J, Bollow M, Neure L, Seipelt E, Seyrekbasan F, Herbst H,

Eggens U, Distler A, Sieper J. Use of immunohistologic and in
situ hybridization techniques in the examination of sacroiliac
joint biopsy specimens from patients with ankylosing spondyli-
tis. Arthritis and Rheumatism 1995; 38:499–505.
(2) McGonagle D, Khan MA, Marzo-Ortega H, O’Connor P, Gibbon
W, Emery P. Enthesitis in spondyloarthropathy. Current Opinion
in Rheumatology 1999; 11:244–250.
Key message
There is a striking and characteristic inflammatory enthesopathy (enthesitis) in AS, and the
natural history or evolution of this enthesitis can explain many of the clinical and pathologi-
cal features of this disease. The enthesitis observed in the extraspinal sites also involves the
axial skeleton, including the sacroiliac and apophyseal joints, and the intervertebral discs.

Enthesitis is now considered a characteristic feature of AS and related SpA, but because of
the relative inaccessibility of the enthesis, the inflammatory, microbiological, and immuno-
logical events at that site have not as yet been well defined. Recent magnetic resonance imag-
ing studies have drawn attention to the ubiquitous nature of enthesitis in AS and associated
SpAs, especially adjacent to synovial joints, and have delineated the early changes, including
the oedematous bone marrow spaces in the immediate vicinity of the enthesitis that Ball
described in his seminal paper.
This paper also popularized the use of the term enthesopathy/enthesitis in the English
language. The word enthesopathy came into use to signify traumatic changes to tendon
insertions, e.g. tennis elbow, because the word enthesis means ‘insertion or implantation’.
Czech physician Niepel and his colleagues first used the term in the 1960s to describe the
inflammatory symptoms at insertional sites that are an important feature of AS and related
SpA. Although the term enthesopathy or inflammatory enthesopathy was initially used, the
term enthesitis was subsequently preferred. Enthesopathy is seen in many arthropathies, but
inflammatory enthesopathy (enthesitis) is the cardinal feature and possible unifying inflam-
matory lesion of AS and related SpA. Clinically, the concept of enthesitis had gained suffi-
cient recognition to be included as one of the classification criteria for SpA and for subsets of
juvenile chronic arthritis.
Needle biopsies of the sacroiliac (SI) joints in five AS patients with active disease have now
been performed by Braun, et al. for immunohistological studies. Dense cellular infiltrates of
T cells and macrophages were found in the synovial portion of the SI joints of all patients. In
these infiltrates a high amount of tumor necrosis factor alpha (TNF-alpha) messenger RNA
(mRNA) was detected by in situ hybridization, and near the site of new bone formation, a
lower amount of transforming growth factor beta (TGF-ß) mRNA were detected, while no
message for interleukin-1 was found in the three patients examined by this technique.
Further investigation of such biopsy specimens showed the absence of DNA of reactive arthri-
tis-associated bacteria. The authors suggested that their finding of abundant TNF-alpha mes-
sage in the SI joints could have implications regarding potential immunotherapeutic
approaches to this disease. In fact they have now reported in an open label pilot study a dra-
matic improvement in more than 90% of active AS patients treated with anti-TNF-alpha anti-
body infliximab (Remicade) (reference 7, Introduction).
Strength
Demonstrates clearly for the first time that the natural history or course of enthesitis could
explain many of the clinical and pathological features of AS. The other two papers have taken
the next steps to understand the mechanism of enthesitis and its more effective therapy.
Weaknesses
1. A histological description of autopsies and biopsy specimens.
2. Represents changes observed in relatively advanced disease.
Relevance
The involved sites in AS show multiple focal inflammatory lesions, primarily localized to the
ligamentous attachments, with active erosive lesions, and also lesions in various stages of heal-
ing, that may culminate in new bone formation and progressive ankylosis of the axial skeleton.
Understanding the mechanism of enthesitis will lead to its more effective therapy. The mecha-
nisms responsible for the increased bone formation in AS is not yet fully understood.
122
Paper 2a
Ankylosing spondylitis
Authors
Hart FD, Robinson KC, Allchin FM, Maclagan NF
Reference
Quarterly Journal of Medicine 1949; 18: 217–238
Paper 2b
Associations between ankylosing spondylitis,

psoriatic arthritis, Reiter’s disease, the intestinal
arthropathies, and Behçet’s syndrome
Authors
Moll JM, Haslock I, Macrae IF, Wright V
Reference
Medicine 1974; 53:343–364
Summary
Hart, et al. provided the first complete and most accurate clinical description of AS, and
defended the name AS rather than rheumatoid spondylitis. Describing the early lumbo-sacral
backache they noted many patients complained of pain in the buttock or buttocks, while in
some the first complaint was mid-lumbar ache, or mid-thoracic or cervical pain. They point-
ed out that ‘at the onset such aches and pains were usually trivial and often ignored by the
patient’s medical officer, or resulted in the patient being treated as a case of “fibrositis”,
“rheumatism”, or occasionally “psychoneurosis”.’ Many of their patients had wrongly been
given psychiatric treatment, and many had been suspected of malingering. The most effec-
tive treatment those days was deep x-ray therapy.
The term ‘spondylarthritides’ was proposed by Moll, et al. for the group of related inflam-
matory joint diseases, the centre piece of which is AS, and they clarified the shared character-
istic clinical features that they emphasized so well in their paper. Clinical and radiological
involvement of the SI joint is an outstanding feature of the SpA, especially AS. Patients with
Reiter’s syndrome/reactive arthritis, psoriasis, ulcerative colitis and Crohn’s disease can
often develop AS. The reverse has also been known, that is, AS patients are more often found
to have Crohn’s disease, ulcerative colitis or psoriasis. Reiter’s syndrome/reactive arthritis
can sometimes be difficult to differentiate from psoriatic arthritis.
Related references (1) Hersch AH, Stecher RM, Solomon WM, Wolpow R, Hauser H.
Heredity in ankylosing spondylitis. A clinical description of fifty
families. American Journal of Human Genetics 1950; 2:391–408.
(2) McEwen C, DiTata D, Lingg C, Porini A, Good A, Rankin T.
Ankylosing spondylitis and spondylitis accompanying ulcerative
colitis, regional enteritis, psoriasis and Reiter’s disease. A com-
parative study. Arthritis and Rheumatism 1971; 14:291–318.
Key message
These two papers provide the basic foundation for the more accurate and detailed clinical des-
cription of AS and the diseases that are grouped with it under the term ‘spondyloarthritides’.

The clinical features of AS that should provide the earliest diagnostic clues were emphasized
by Hart, et al.: AS should not be diagnosed if SI joints were radiologically normal, unless
other evidence was strong, and they described one of the two such patients they had diag-
nosed when their SI joints were as yet normal. They also pointed out that the disease was very
often not diagnosed until it was relatively far advanced, and that it ‘can be diagnosed with
greater ease than most conditions in medicine’.
The concept of SpAs was put forward by Moll, et al., in which AS formed the centre piece
of this group of diseases that share clinical features and genetic predisposing factors; the
remarkable association with the major histocompatibility complex class I molecule HLA-B27
that was discovered while their paper was in press, confirmed the validity of their proposal.
Five subgroups can be now differentiated: AS, reactive arthritis (ReA) that encompasses
Reiter’s syndrome, psoriatic arthritis, arthritis associated with inflammatory bowel disease,
and undifferentiated SpA (including enthesitis related arthritis).
Moll et al. also emphasized that extra-articular organ systems may also be involved in these
diseases, especially the gastrointestinal tract. Mielants, et al. (see Related reference (1), Paper
3) were subsequently able to demonstrate on ileocolonoscopic studies 11 years later the pres-
ence of inflammatory gut lesions in all the diseases in the SpA group, even in the absence of
gastrointestinal symptoms. Bowel inflammation is present in a large number of patients with
SpA who had no intestinal symptoms or overt inflammatory bowel disease (IBD).
Enteropathic arthritis develops in up to 20% of patients with chronic IBD, and most of these
patients have peripheral synovitis that correlates with flare up of bowel disease, but one-
fourth have axial disease which does not fluctuate with bowel disease activity.
Strengths
1. Hart, et al. provided one of the first accurate and comprehensive clinical description of
the early stages of AS, and emphasized the clinical features that should provide the earli-
est diagnostic clues.
2. Moll, et al. proposed the term ‘spondylarthritides’ for the group of related inflammatory
joint diseases, the centre piece of which is AS, and they clarified the shared characteristic
clinical features of this group of diseases.
Weaknesses
1. The paper by Hart, et al. did not contain any pathological findings but it was because none
of their patients had come to autopsy.
2. The paper by Moll, et al. included Behçet’s syndrome in the spondyloarthritides group but it
is not generally considered to belong to this group of diseases.
Relevance
Hart, et al. emphasized more than 50 years ago the clinical features of AS that should lead to
early diagnosis. It took more than 40 years for these clinical features to be incorporated in
the classification criteria. Moll, et al. emphasized the shared characteristic clinical features of
this group of diseases, and this concept was further reinforced by the discovery of the genetic
marker, HLA-B27, that shows a remarkable association with this group of diseases.
124
Paper 3
HLA-B27 related arthritis and bowel inflammation.

Part 2 Ileocolonoscopy and bowel histology in
patients with HLA-B27 related arthritis
Authors
Mielants H, Veys EM, Cuvelier C, De Vos M, Botelberghe L
Reference
Summary
The concept of SpAs put forward by Moll, et al. gathers together a group of chronic rheuma-
tological diseases, where extra-articular organ systems may also be involved, especially the
gastrointestinal tract. Mielants, et al. (1, 2) were the first to demonstrate on ileocolonoscopic
studies the presence of inflammatory gut lesions in all the diseases in the SpA group, in the
absence of gastrointestinal symptoms. The prevalence of these inflammatory gut lesions var-
ied between 20% and 70% in the different diseases. Their recent data are based on perfor-
mance of biopsies of the terminal ileum and proximal colon in 108 patients with SpA (55
had ReA and 53 had AS), and 56 controls (47 patients with other rheumatic diseases, and 19
patients with chronic constipation, colonic polyps or adenocarcinoma). There was histologi-
cal evidence of IBD with features either of acute enterocolitis, or early Crohn’s disease in 30
of 53 (56.6%) AS patients, 37 of 55 (67%) patients with ReA, and in only one of 56 controls
(2%). These lesions were histologically ‘acute’ in approximately 25% and ‘chronic’ in 30%.
Only 18 of the 67 (27%) patients with histological gut inflammation, however, had intestinal
symptoms.
Related references (1) Mielants H, Veys EM, Cuvelier C, De Vos M. Ileocolonoscopic find-
ings in seronegative spondylarthropathies. British Journal of
(2) Mielants H, Veys EM, Cuvelier C, De Vos M, Goemaere S, De
Clercq L, Schatteman L, Elewaut D. The evolution of spondy-
larthropathies in relation to gut histology. Journal of
Rheumatology 1995; 22:2266–2284. (This work is in 3 parts: Part I
covers clinical aspects (pp. 2262–2272). Part II covers histologi-
cal aspects (pp. 2273–2278). Part III covers the relationship
between gut and joint (pp. 2279–2284).)
Key message
Subclinical inflammatory lesions in the gut have been observed on ileocolonoscopic mucosal
biopsy in SpA patients without gastrointestinal symptoms, or clinically obvious IBD or psoria-
sis. Persistent subclinical gut inflammation was observed in active joint disease.

They have hypothesized that subclinical inflammatory bowel disease is present in some
patients with SpA, in which the joint inflammation can be the only clinical manifestation.
Their studies support the aetiopathogenetic role of the gut in SpA. The link between gut
inflammation and arthropathy has also been demonstrated in the HLA-B27 transgenic rats,
that make a nice animal model for SpA. The development of gut inflammation precedes the
development of SpA, and if these animals are born and raised in a germ-free environment
they do not develop gut inflammation or SpA, although they may develop psoriasiform skin
and nail lesions. Thus presence of gut flora are needed for the SpA, except the psoriasiform
skin and nail lesions.
Presence of chronic (Crohn-like) gut inflammation and mild complaints of diarrhoea
implies a high risk of evolution to AS and IBD. The presence of these gut lesions is not
associated with HLA-B27, and the data support the existence of a B27-independent com-
mon pathogenic link between gut inflammation and SpA. SpA patients with acute or
chronic inflammatory gut lesions seem to respond better to sulphasalazine treatment than
those with normal gut histology, suggesting that sulphasalazine may have a beneficial effect
on the patients with SpA by healing their gut inflammation. Sulphasalazine has a beneficial
effect on articular activity by controlling gut inflammation, but it does not prevent evolu-
tion to IBD.
Mielants, et al. have also studied long-term evolution of gut inflammation in patients with
SpA (1, 2). Evolution to IBD was observed in 7% of these patients. Follow-up studies of SpA
patients with the subset with chronic inflammatory gut lesions suggest that anywhere
between 15–25 % of them will develop clinically obvious Crohn’s disease, suggesting that
these patients had initially a subclinical form of IBD. The presence of these Crohn-like gut
lesions is not associated with HLA-B27. Persistent gut inflammation was observed in active
joint disease. Presence of chronic gut inflammation and mild complaints of diarrhoea
implies a high risk of evolution to AS and IBD.
Strengths
1. Their studies suggest the aetiopathogenetic role of the gut in SpA.
2. The link between gut inflammation and arthropathy has been further supported by the
HLA-B27 transgenic rat model for SpA (reference 6, Introduction).
Weakness
Unable to pinpoint the precise bacterial trigger for AS, if there is one.
Relevance
These authors have identified a group of SpA patients with inflammatory chronic, or relaps-
ing acute arthritis who have histological evidence of ileocolitis, even in the absence of any
gastrointestinal symptoms. The presence of these gut lesions is not associated with HLA-B27.
Evolution to IBD can occur in some of these patients on follow-up. Sulphasalazine has a ben-
eficial effect on articular activity (peripheral arthritis) by controlling gut inflammation, but it
does not prevent evolution to IBD and does not benefit axial disease.
126
Paper 4a
A wider spectrum of spondyloarthropathies

Authors
Khan MA, van der Linden SM
Reference
Seminars in Arthritis and Rheumatism 1990; 20:107–113
Paper 4b
The European Spondyloarthropathy Study

Group preliminary criteria for the classification
of spondyloarthropathies
Authors
Dougados M, van der Linden S, Juhlin R, et al.
Reference
Summary
The clinical spectrum of these diseases is now understood to be much wider and no longer
implies necessarily the presence of sacroiliitis or spondylitis. For example, the disease may
start with features such as enthesitis (inflammation of the sites of attachment of ligaments
and tendons to bone), dactylitis, or oligoarthritis that in some cases may progress to sacroili-
itis and spondylitis, with or without any extra-articular features such as acute anterior uveitis
or muco-cutaneous lesions. The various forms of SpA show a strong association with the his-
tocompatibility antigen HLA-B27; however, the strength of this association varies markedly
not only between the various SpA but also among racial and ethnic groups. Khan and van
der Linden, primarily based on their earlier studies, pointed out that features typical of SpA
may occur in different combinations so that the existing criteria for disease classification and
diagnosis were inappropriate or too restricted, and that these criteria do not recognize the
existence of a much wider disease spectrum.
The disease spectrum of Reiter’s syndrome has also been broadened considerably, and the
clinical spectrum of psoriatic SpA has been better clarified. ‘Incomplete’ forms of Reiter’s
syndrome were found to be much more common than the classical triad of arthritis, conjunc-
tivitis, and urethritis. The term ‘B27-associated ReA’ has been used in recent years to refer to
SpA following enteric or urogenital infections, and the disease spectrum includes the clinical
picture of typical Reiter’s syndrome. Some of the less well defined B27-associated clinical syn-
dromes that have been clarified included seronegative oligoarthritis, polyarthritis, or dactyli-
tis (‘sausagelike’ toes) of the lower extremities, and heel pain caused by enthesitis (calcaneal
and tarsal periostitis, Achilles tendonitis, and plantar fasciitis). These and other undifferenti-
ated forms of SpA had been ignored in previous epidemiological studies because of the inad-
equacy of the existing classification criteria.
Epidemiological studies, therefore, had been hampered by incomplete definition of the
clinical disease spectrum, and there was a need to use broad criteria that will include the
patients that had been ignored because their diseases do not fit the current criteria.
The European Spondyloarthropathy Study Group (ESSG) provided the first widely used
classification criteria for the whole group of SpA patients, and pointed out their usefulness in
encompassing the wider disease spectrum, including those with undifferentiated SpA. These
ESSG preliminary criteria for the classification of SpAs are listed in Table 8.1.
Table 8.1 ESSG criteria for spondyloarthropathy
Inflammatory spinal pain or Synovitis

Asymmetrical or
Predominantly in the lower limbs
And one or more of the following:
Alternate buttock pain
Sacroiliitis
Calcaneal enthesitis
Positive family history
Psoriasis
Inflammatory bowel disease
Urethritis or cervicitis or acute diarrhoea occuring within 1 month before arthritis
Related references (1) Amor B, Dougados M, Listrat V, et al. Are classification criteria for
spondylarthropathies useful as diagnostic criteria? Rev de
Rheumatisme (English edition) 1995; 6:10–15.
(2) Khan MA, van der Linden SM, Kushner I, Valkenberg HA, Cats A.
Spondylitic disease without radiological evidence of sacroiliitis
in relatives of HLA-B27 patients. Arthritis and Rheumatism 1985;
28:40–43.
Key message
The difficulty in clearly categorizing some patients with SpA occurs most often in early stages
of these diseases, because their characteristic signs and symptoms may only become manifest
after a long follow-up, and may sometimes be aided by family history and examination of the
patient’s family members. Khan and van der Linden not only pointed out this wider spec-
trum of SpA, but also emphasized that features typical of SpA may occur in different combi-
nations, and that the existing criteria were too restricted and therefore unable to recognize
the wider disease spectrum. The utility of the ESSG criteria have been validated in many
countries and ethnic/racial groups.
Why it is important
Spondyloarthropathy (SpA) is a heterogeneous clinical entity that includes AS, the prototype
of this group of interrelated disorders, ReA (including Reiter’s syndrome), psoriatic arthritis,
and arthritis associated with chronic IBDs. It may not always be possible to differentiate clear-
ly between the various forms of SpA, especially in early stages of the disease, because they
generally share many clinical features, both skeletal and extra-skeletal. As in other diseases of
128
undetermined aetiology, the diagnosis of these diseases is based on clinical and roentgeno-
graphic features.
The studies of Moll, et al. and the discovery in 1973 of the association of HLA-B27 with AS
and related SpA had clearly demonstrated that these diseases share clinical and radiographic
features as well as genetic predisposing factors. HLA-B27 was helpful as a genetic marker for
these diseases and the work of many investigators led to the observations that raised ques-
tions about our perceptions of the clinical spectrum of SpA. Khan and van der Linden, based
on their study, pointed out the wider spectrum of SpA, and that features typical of SpA may
occur in different combinations, and that the existing criteria for disease classification and
diagnosis were too restricted. Their previously published studies had demonstrated that
although radiographically detected sacroiliitis is extremely frequent in AS, it may not be an
obligate manifestation, especially in early or atypical forms of the disease. Moreover, arthritis
involving the axial skeleton, including the SI joints, can be present in some patients without
evidence of erosive disease roentgenographically. The recent advances in radiographic imag-
ing, especially dynamic magnetic resonance imaging, can now help overcome some of the
diagnostic difficulties of early roentgenographic sacroiliitis, since these new techniques can
visualize both acute and chronic changes in the SI joints at a very early stage. These authors
emphasized the usefulness of the ESSG criteria in encompassing the wider spectrum of SpA
patients, including those with undifferentiated SpA.
The ESSG criteria are now very widely used and they have been validated in many diverse
SpA populations in the world. For example, the clinical utility of these criteria, that were
developed based on European patients, has been validated in Mexicans, Brazilians, Turks,
Eskimos, Inuits, and other native populations of north America. The ESSG criteria, however,
are unable to recognize those patients with SpA who have neither chronic inflammatory back
pain nor any oligoarthritis. Some of these patients can be recognized by the Amor’s criteria
to have SpA (Table 8.2). These two criteria tend to complement each other, and therefore
could be used together in tandem to encompass the widest spectrum of SpA patients.
Strengths
1. These papers have provided better disease characterization and definition.
2. They have improved the classification of the diseases grouped under the term SpAs.
Weaknesses
The ESSG criteria are unable to recognize those patients with SpA who do not suffer from
chronic inflammatory back pain or any oligoarthritis, yet such patients may meet the Amor’s
criteria. These two criteria tend to complement each other, and should therefore be used
together to encompass the widest possible disease spectrum.
Relevance
The HLA-B27-associated undifferentiated forms of SpA had been ignored in previous epi-
demiological studies because of the inadequacy of the existing classification criteria, but
these clinical syndromes have now been clarified, and the classification criteria for SpA now
encompass a much wider spectrum of this group of diseases.
Table 8.2 Amor criteria for spondyloarthropathy
Parameters Scoring
A. Clinical symptoms or past history of:

1. Lumbar or dorsal pain at night or morning stiffness of lumbar or dorsal pain 1
2. Asymmetrical oligoarthritis 2
3. Buttock pain 1
or if alternate buttock pain 2
4. Sausage-like digit(s) (fingers or toes) 2
5. Heel pain or other well defined enthesitis pain 2
6. Iritis 2
7. Non-gonococcal urethritis, cervicitis within 1 month before the onset of arthritis 1
8. Acute diarrhoea within 1 month before the onset of arthritis 1
9. Psoriasis, balanitis, or inflammatory bowel disease (ulcerative colitis or Crohn’s disease) 2
B. Radiological findings
10. Sacroiliitis (at least bilateral grade 2 or unilateral grade 3) 2
C. Genetic background
11. Presence of HLA-B27 and/or family history of ankylosing spondylitis,
reactive arthritis, uveitis, psoriasis or inflammatory bowel disease 2
D. Response to treatment
12. Clearcut improvement within 48 hours after non-steroidal anti-inflammatory drugs
(NSAID) intake or rapid relapse of the pain after their discontinuation 2
A patient is considered to be suffering from a spondyloarthropathy if the sum is >6.

130
Paper 5a
A new approach to defining functional ability in

ankylosing spondylitis: the development of the Bath
ankylosing spondylitis functional index
Authors
Calin A, Garrett S, Whitelock H, Kennedy LG, O’Hea J, Mallone P, Jenkinson T
Reference
Paper 5b
Selection of instruments in the core set for DC-ART,

SMARD, physical therapy, and clinical record
keeping in ankylosing spondylitis. Progress report
of the ASAS Working Group
Authors
Van der Heijde D, Calin A, Dougados M, Khan MA, van der Linden S,
Bellamy N, and on behalf of the ASAS Working Group
Reference
Summary
The clinical assessment of disease activity in patients with AS is often not easy. A few function-
al indices have now been developed for longitudinal follow-up of AS patients, and for moni-
toring their response to therapy. The most widely used measures of functional capacity in AS
is the Bath ankylosing spondylitis functional index (BASFI) (score 0–10). Disease activity has
been assessed by using a self-administered composite index called the Bath ankylosing
spondylitis disease activity index (BASDAI) that consists of an evaluation on a visual analogue
scale (from 0–10) focusing on axial pain, peripheral joint pain, fatigue/tiredness, enthesitis,
and morning stiffness). BASDAI has been found to be a valid and appropriate composite to
define disease activity in AS.
There is also a need to develop a reliable, reproducible, and simple radiological scoring
method to document radiographic changes in AS. The Bath ankylosing spondylitis radiologi-
cal index (BASRI) is one such index and it has a moderate-to-good reliability. The radi-
ographs for BASRI are scored using the New York criteria for the SI joints, while the lumbar
and cervical spine, and the hip joints are graded 0–4. These scores are added together to give
BASRI-t (total) and if the hips are excluded one gets BASRI-s (spine). There are thus three
types of BASRI scores (total, spinal, and hip).
The process of selecting the most appropriate indices needed further standardization.
Therefore, a working group called Assessments in Ankylosing Spondylitis (ASAS) was set up
for the selection of core sets for the following three different settings: (a) for disease control-
ling anti-rheumatic therapy (DC-ART); (b) for therapy with symptom modifying anti-
rheumatic drugs and for physical therapy (SM-ARD/PT); and (c) for clinical record keeping
in daily practice. The instruments that were finally chosen are listed in Table 8.3.
Table 8.3 Specific instruments for each domain in core sets for DC-ART, SM-ARD/PT, and clinical
record keeping
Domain Instrument
Physical function BASFI or Dougados functional index

Pain Visual analogue scale (VAS)-last week-spine-at night-due to AS
and VAS-last week-spine-due to AS
Spinal mobility Chest expansion test
and modified Schober’s test
and occiput-to-wall distance test
Patient global VAS-last week
Stiffness Duration of morning stiffness-spine-last week
Peripheral joints Number of swollen joints (44 joint count)
Entheses No preferred instrument is yet available
Acute phase reactants Erythrocyte sedimentation rate
X-ray spine AP+ Lateral lumbar spine
and lateral cervical spine
and x-ray pelvis (sacroiliac and hip joints)
X-ray pelvis (see under x-ray spine above)
Fatigue No preferred instrument is yet available
The SM-ARD/PT core set comprises the following five components:

Physical function
Pain
Spinal mobility
Patient global
Stiffness
The additional components for clinical record keeping include:
Peripheral joints
Entheses
Acute phase reactants
The core set for DC-ART, in addition to the above mentioned 8 components, requires:
X-ray spine
X-ray pelvis
Fatigue
132
Related references (1) Calin A, Nakache J-P, Gueguen A, Zeidler H, Mielants H,

Dougados M. Defining disease activity in ankylosing spondylitis:
is a combination of variables (Bath ankylosing spondylitis dis-
ease activity index) an appropriate instrument? Rheumatology
1999; 38:878–882.
(2) Calin A, Mackay K, Brophy S. A new dimension to outcome: The
Bath ankylosing spondylitis radiology index (BASRI). Journal of
(3) Garrett S, Jenkinson T, Kennedy LG, Whitelock H, Gaisford P,
Calin A. A new approach to defining disease status in ankylos-
ing spondylitis: the Bath ankylosing spondylitis disease activity
index. Journal of Rheumatology 1994; 21:2286–2291.
Key message
It is more difficult to assess disease activity of AS than of RA, but the recent studies discussed
above provide the proper instruments for assessing disease and functional activity and
response to therapy, as well as for clinical record keeping in daily practice.

The clinical status of a disease such as AS can be defined by several components, e.g. pain,
function, metrology, and laboratory and radiographic findings. Both erythrocyte sedimenta-
tion rate and C-reactive protein are moderately correlated with disease activity and also with
damage, but they do not comprehensively represent the disease process in AS. Radiographic
evidence of structural damage is an important endpoint in the assessment of AS, and
presence of sacroiliitis is a prerequisite for classification according to the (modified) New
York criteria.
This process of selecting functional and radiological indices needed further standardiza-
tion, and there is a general consensus that a clinical outcome measure for use in clinical
practice should be simple, brief, easy to score, reliable, valid, and sensitive to change. The
OMERACT (Outcome Measures in Rheumatology) group has recently proposed a novel
user-friendly paradigm to capture the essential elements of discriminative power (including
sensitivity to change), truth, relevance, and feasibility (e.g. applicability and costs). This para-
digm is termed the OMERACT filter. The specific instruments selected by the ASAS working
group were therefore subjected to the OMERACT filter test for relevance and feasibility.
The ASAS working group has proposed the instruments that need to be used in all
research projects in AS. These instruments have now been approved by the OMERACT, the
World Health Organization (WHO), and the International League of Associations for
Rheumatology (ILAR) for use in clinical research projects in AS, as well as for clinical record
keeping in daily practice. They are now called the WHO/ILAR core sets and remission
criteria.
Strength
The first set of properly validated instruments to be used in clinical research projects in AS,
as well as for clinical record keeping in daily practice.
Weakness
The next step will be to assess the validity of the two measures, entheses and fatigue, which
were not accepted by OMERACT because currently no preferred instrument are available for
them.
Relevance
A properly validated set of instruments have now been developed for longitudinal follow-up of
AS patients, for monitoring their response to therapy, and for clinical record keeping in daily
practice. They have been found to be very helpful in fully assessing the dramatic efficacy of
anti-TNF-alpha therapy (infliximab and etenercept) in AS and related spondyloarthropathies.
134
CHAPTER 9
Psoriatic arthritis
Dafna D Gladman
Introduction
The association between arthritis and psoriasis was first made in the 1850s by Alibert (1).
However, subsequently it was thought that any inflammatory arthritis that occurred in people
with arthritis was a form of rheumatoid arthritis (RA).
The first paper in this selection (Paper 1) is a true ‘classic’, being the paper in which the
late Verna Wright described psoriatic arthritis (PsA) as a separate entity. Baker and col-
leagues made similar observations a few years later (2), and in 1964 the American
Rheumatism Association (ARA) included the disease within its classification of rheumatic dis-
eases for the first time (3). Verna Wright, in conjunction with John Moll and other col-
leagues made many important contributions to the field, including the discovery of the famil-
ial nature of the condition (Paper 2). However, in spite of all their work, a number of
authors have continued to question the existence of PsA as a separate entity (4).
Studies on the pathogenesis and genetics have helped to distinguish it as a distinct condi-
tion. For example, Veale and colleagues (Paper 3) showed that the reaction in the synovium
differs in PsA and RA, and Espinoza and colleagues (Paper 4) found that the fibroblast reac-
tion was abnormal in both the skin and synovium of patients with psoriatic arthropathy.
Finally Gladman and her colleagues (Paper 5) have shown that there is a relationship
between the expression and severity of PsA and genetic markers in the human leucocyte anti-
gen (HLA) region, although these are different from those seen in rheumatoid disease.
Wright’s early descriptions of the disease suggested that it was a relatively mild condition
(5), but, as discussed in the section on Paper 1, more recent studies from other countries
have suggested that this is not the case (6, 7). There may be genuine differences in expres-
sion in different countries, but it seems likely that this is not as benign a disease as was first
thought, and patients certainly have problems with disability and employment. Furthermore,
Gladman’s group have shown that life expectancy is shortened by the condition (Paper 6).
Reasons for early mortality, given that this is not a systemic disease, remain to be elucidated.
Psoriatic arthritis 135
References
1. Wright V, Moll JHM. Seronegative polyarthritis. Amsterdam: North Holland Publishing
Company, 1976.
2. Baker H, Golding DN, Thompson M. Psoriasis and arthritis. Annals of Internal
Medicine1963; 58:909–925.
3. Blumberg DS, Bunim JJ, Calkins E, Pirani CL, Zaifler NJ. ARA nomenclature and classifi-
cation of arthritis and rheumatism. Arthritis and Rheumatism 1964; 7:93–97.
4. Cats A. Is psoriatic arthritis an entity? In: Rheumatology/85 (Brooks P, York J, (eds.)).
Amsterdam: Elsevier, 1985:295–301.
5. Roberts MET, Wright V, Hill AGS, Mehra AC. Psoriatic arthritis. A follow-up study. Annals
of Rheumatic Diseases 1976; 35:206–212.
6. Alonso J, Perez A, Castrillo J, Garcia J, Noriega J, Larrea C. Psoriatic arthritis: a clinical,
immunological and radiological study of 180 patients. British Journal of Rheumatology 1991;
30:245–250.
7. Gladman D, Stafford-Brady F, Chang C, Lewandowski K, Russell M. Longitudinal study of
clinical and radiological progression in psoriatic arthritis. Journal of Rheumatology 1990;
17:809–812.
136
Paper 1
Psoriasis and arthritis

Author
Wright V
Reference
Summary
This study compares 34 patients with psoriasis and erosive arthritis with 55 unselected
patients with seropositive RA and with 310 patients with uncomplicated psoriasis. The paper
notes the equal male to female ratio of PsA which contrasts with the female preponderance
in both RA and uncomplicated psoriasis. The psoriasis usually precedes the arthritis or is syn-
chronous with it in 77% of the cases, but in 23% the arthritis comes first. The clue to the
diagnosis of PsA in patients who do not have the psoriasis is a positive family history. Joint
changes were more related to nail lesions than to other skin lesions. Nail lesions occur in
87% of patients with PsA and only 25% of patients with uncomplicated psoriasis.
Subcutaneous nodules do not occur in PsA. Wright concluded that PsA was milder than RA
based on this sample of patients.
Related references (1) Kammer GM, Soter NA, Gibson DJ, Schur PH. Psoriatic arthritis: a
clinical, immunologic and HLA study of 100 patients. Seminars
in Arthritis and Rheumatism 1979; 9:75–97.
(2) Gladman DD, Shuckett R, Russell ML, Thorne JC, Schachter RK.
Psoriatic arthritis (PSA)–an analysis of 220 patients. Quarterly
Journal of Medicine 1987; 62:127–141.
(3) Jones SM, Armas JB, Cohen MG, Lovell CR, Evison G, McHugh NJ.
Psoriatic arthritis: outcome of disease subsets and relationship
of joint disease to nail and skin disease. British Journal of
Key message
There is a specific form of erosive arthritis associated with psoriasis that can be distinguished
from RA.

Although previous authors had noted the association between psoriasis and arthritis, the
description of PsA as a distinct clinical entity with several clinical patterns is attributed to the
late Professor Verna Wright of Leeds. This paper preceded several other contributions from
Wright, and Moll and Wright which further developed the concept of PsA and its clinical pat-
terns, including distal joint disease, oligoarthritis, polyarthritis clinically indistinguishable
from RA, arthritis mutilans and a SpA. Subsequent studies of large populations of patients
with PsA have been published and confirm the existence of these patterns. However, studies
that included patients with long-standing disease demonstrate that many patients go on to
develop a polyarthritis, and that the disease is not as benign as originally thought (1–3).
Strength
Systematic approach to the evaluation of patients with PsA.
Weaknesses
1. Small number of patients.
2. Short duration of disease.
3. Inclusion of patients with erosive arthritis only.
Relevance
The identification and description of PsA as a distinct entity facilitates correct diagnosis for
patients with arthritis and psoriasis and provides a framework for studies of pathogenesis in
this condition.
138
Paper 2
Familial occurrence of psoriatic arthritis

Authors
Moll JM, Wright V
Reference
Annals of Rheumatic Diseases l973; 32:l81–20l
Summary
The authors assessed first and second degree relatives and spouse controls of 88 patients with
established PsA and 20 patients with psoriasis and other forms of arthritis (12 RA, seven
osteoarthritis (OA), one gout) identified from a hospital population. The prevalence of PsA
among the 253 first degree relatives of patients with PsA was 5.5% compared to the calculat-
ed prevalence in the UK population of 0.1%. None of the relatives of patients with psoriasis
with other forms of arthritis had PsA. The degree of familial aggregation was expressed by
the Kellgren factor at 48.8. The prevalence of psoriasis was increased in first degree relatives
of probands with either PsA or with psoriasis and other forms of arthritis compared to the
population control.
Related reference (1) Risch N. Linkage strategies for genetically complex traits. 1.
Multilocus model. American Journal of Human Genetics 1990;
46:222–228.
Key message
This study identified the heritability of PsA.

This is a large family investigation which demonstrates unequivocally the role of hereditary
factors in PsA. It has paved the way for several other investigations, and provided a rationale
to investigate the prevalence of HLA antigens in this disease. It is now clear that there is an
association between markers on chromosome 6 and the susceptibility to PsA. Moreover, this
family investigation provides impetus for genome-wide studies in this disease. Indeed analyz-
ing Moll and Wright’s data with current methodology (1) one finds that the relative risk for
first degree relatives (␭1) of patients with PsA is 55, and for siblings (␭s) it is 27.
Strength
This remains the largest family study of PsA to date.
Weakness
Lack of HLA typing for the individuals included in this study.
Relevance
This study confirms that genetic factors are relevant to the susceptibility to PsA. It remains
the framework for family studies to date. Several genome-wide studies in psoriasis have been
published, and there are current studies looking at families of patients with PsA.
140
Paper 3
Reduced synovial membrane macrophage

numbers, ELAM-1 expression, and lining layer
hyperplasia in psoriatic arthritis as compared
with rheumatoid arthritis
Authors
Veale D, Yanni G, Rogers S, Barnes L, Bresnihan B, Fitzgerald O
Reference
Summary
Veale and co-workers used monoclonal antibodies to surface molecules to identify cells and
vascular endothelium in synovial biopsies from 15 patients with PsA and 15 RA controls
matched by age and disease duration. They found that there was less lining layer hyperplasia,
fewer macrophages and a greater number of blood vessels in the synovium from PsA patients
compared to RA. While ECAM-1 expression was less intense in PsA, there was no difference
in the expression of other adhesion molecules. The number of B cells, and T cells and their
subsets was similar in the two patient groups.
Key message
There are differences in the synovial changes seen in PsA and RA.

In addition to genetic factors, immunological factors are thought to play an important role
in the pathogenesis of PsA. This investigation is the first to compare immunohistological fea-
tures from matched samples of patients with PsA and RA, and notes that for patients with
similar age and disease duration, PsA is associated with an increased vascularity but less
inflammatory response than RA synovia.
Strength
Samples matched by age and disease duration.
Weaknesses
1. This is a small study.
2. While patients were not taking disease modifying anti-rheumatic drugs (DMARDs) at the
time of study, they had taken some within 3 months. Since some DMARDs may remain in
the tissues for prolonged periods the effect of medication on the results is not clear.
3. There were no samples from non-inflammatory arthritis patients.
Relevance
This study supports the role of vascular and immune factors in the pathogenesis of PsA.
142
Paper 4
Fibroblast function in psoriatic arthritis:

I. Alternation of cell kinetics and growth
factor responses
Authors
Espinoza LR, Agular JL, Espinoza CG, Cuellar ML, Scopelitis F, Siveira LH
Reference
Summary
The authors investigated the role of fibroblasts in the pathogenesis of PsA. They obtained
involved and uninvolved skin and synovial fibroblasts from 10 patients with PsA who were
seronegative for rheumatoid factor (RF) and were not on second-line drugs prior to study.
Six controls were also studied. Fibroblast cultures were established and cell cycle analysis was
performed by flow cytometry. A significant increase in S- and G2-M phase values was
observed for psoriatic skin, both involved and uninvolved compared to normal skin, and in
psoriatic synovium compared to normal synovium. The abnormalities did not resolve follow-
ing methotrexate administration despite clinical improvement in the skin lesions. Extracts
from fibroblasts were then used to stimulate NIH-3T3 cell line. A marked proliferative
response was observed in involved psoriatic fibroblasts compared to controls. A significant
increase in kinetic response to growth factors was noted when psoriatic fibroblasts were incu-
bated with growth factors. No correlations with disease duration or severity were noted.
Related reference (1) Espinoza LR, Espinoza CG, Cuellar ML, Scopelitis E, Silveira LH,
Grotendorst GR. Fibroblast function in psoriatic arthritis. II.
Increased expression of beta platelet derived growth factor
receptors and increased production of growth factors and
cytokines. Journal of Rheumatology 1994; 21:1507–1511.
Key message
There is an abnormality in fibroblast activity in both the skin and synovium of patients with
PsA.

These studies demonstrate a significant in vitro alteration in skin and synovium fibroblasts
from patients with PsA. A diffusible factor secreted by the fibroblasts appears to be responsi-
ble for these alterations. Alterations in DNA cell cycle and growth factor response of psoriatic
fibroblasts in both skin and synovium suggest involvement of this cell type in the pathogene-
sis of both skin and joint manifestations of the disease. The presence of these abnormalities
despite therapy supports an intrinsic role for this abnormality. The authors further demon-
strated increased expression of ß- platelet derived growth factor (PDGF) receptor in the pso-
riatic fibroblasts and increased production of interleukin-1ß and PDGF by psoriatic fibrob-
lasts, providing further support for the role of fibroblasts in the pathogenesis of psoriasis and
PsA.
Strengths
1. Clearly demonstrates fibroblast abnormalities from both skin and joint in patients with
PsA.
2. Included healthy controls.
Weaknesses
1. Small number of patients were included.
2. All patients had rheumatoid-like arthritis.
3. Disease severity was not defined thus the relationship between the in vitro abnormalities
and clinical features is difficult to evaluate.
Relevance
The pathogenesis of PsA is unclear. This study provides support for the role of fibroblast
abnormality in the development and perpetuation of both skin and joint lesions of the dis-
ease.
144
Paper 5
The role of HLA antigens as indicators of

progression in psoriatic arthritis (PsA):
multivariate relative risk model
Authors
Gladman DD, Farewell VT
Reference
Summary
An association between HLA antigens and both psoriasis and PsA had been demonstrated. In
this study the authors investigated the role of HLA antigens in disease progression in PsA.
Disease progression was defined as an increase in clinical damage measured by the number
of deformities. The authors developed a model to identify features which would predict dis-
ease progression, and had previously demonstrated that among clinical features at the time
of first visit five or more swollen joints and a high medication level were predictors for pro-
gression, while a low erythrocyte sedimentation rate (ESR) was ‘protective’. Applying a simi-
lar model but including HLA antigens previously reported to be associated with psoriasis or
PsA they identified HLA-B39 as predictive of early progression, HLA-B27 in the presence of
HLA-DR7, and HLA-DQw3 in the absence of HLA-DR7 as predictors for clinical progression.
Including the clinical features outlined previously in the model they showed that the HLA
antigens were stronger than the clinical predictors.
Related references (1) Gladman DD, Farewell VT, Nadeau C. Clinical indicators of pro-
gression in psoriatic arthritis (PsA): multivariate relative risk
model. Journal of Rheumatology 1995; 22:675–679.
(2) Gladman DD, Farewell T, Kopciuk K, Cook RJ. HLA antigens and
progression in psoriatic arthritis. Journal of Rheumatology 1998;
25:730–733.
Key message
Genetic markers in the HLA region are important in the expression and progression of PsA.

These studies were the first to use clinical damage as an outcome in PsA and to look for pre-
dictors for this outcome While the presence of five or more swollen joints and use of a high
level of medication were predictors for progression and a low ESR a ‘protector’ from such
progression, HLA antigens were even stronger predictors. A subsequent study including
more patients and all HLA antigens represented in the patient population confirmed the
above data and also demonstrated HLA-B22 as protective. Thus the HLA region is important
both in the susceptibility and expression of PsA.
Strengths
1. Large number of patients.
2. Good methodology.
Weakness
The study is based on serological typing only.
Relevance
The information gained from this study supports the role for chromosome 6p in the patho-
genesis of PsA. It also provides evidence that genetic markers may influence disease progres-
sion. Moreover, the identification of markers for disease progression is important in the strat-
ification of patients in therapeutic trials.
146
Paper 6
Mortality studies in psoriatic arthritis. Results from

a single centre. II. Prognostic factors for death
Authors
Gladman DD, Farewell VT, Wong K, Husted J
Reference
Arthritis and Rheumatism 1998; 41:1103–11l0
Summary
The authors set out to investigate prognostic factors associated with mortality in patients with
PsA. They had previously demonstrated that patients with PsA were at an increased risk for
death compared to the general population. Fifty-three of 428 patients followed prospectively
at their PsA clinic between 1978 and 1994 died. The four leading causes of death included
diseases of the circulatory system (36.2%), respiratory system (21.3%), malignant neoplasms
(17.0%), and injuries/poisoning (14.9%). The overall standardized mortality ratio (SMR)
was 1.62. Deaths due to respiratory causes were particularly increased in these patients.
Multivariate analysis of potential predictive factors for death present at first visit to clinic
revealed that an ESR >15, prior medications, radiological damage and the absence of nail
lesions were associated with overall mortality. A gender effect was particularly noted in
injuries/poisoning as six of the deaths occurred in males and only one was a female.
Related reference (1) Wong K, Gladman D, Husted J, Long J, Farewell VT. Mortality
studies in psoriatic arthritis. Results from a single centre. I. Risk
and causes of death. Arthritis and Rheumatism 1997;
40:1873–1877.
Key message
People with PsA have an increased risk of death which is related to the severity of the disease.

The study demonstrate that patients with PsA are at an increased risk for death compared to
the general population. Evidence of previous active and severe disease as manifested by the
use of high level of medication and radiological changes as well as an elevated ESR at presen-
tation are prognostic indicators for death. This study further supports the impression that
PsA is not a mild disease as originally proposed by Wright (Paper 1).
Strength
This is the first study to demonstrate that patients with PsA are at an increased risk from
death.
Weakness
The study is based on a population followed in a speciality clinic and although the clinic
includes patients with the full spectrum of the disease, from mild to severe, it may not reflect
the severity of PsA in the community.
Relevance
Recognizing that patients with PsA have a severe disease which leads to premature death, and
that active and severe disease at presentation are predictive factors in this early mortality will
assure that these patients are treated aggressively early in their course to attempt to prevent
such an outcome.
148
CHAPTER 10
Reactive arthritis and

Reiter’s syndrome
Henning K Zeidler
Introduction
Reactive arthritis (ReA) and Reiter’s syndrome are two terms that were used for sometime
more or less interchangeably. The latter goes back early in the century and defines a triad or
more of clinical manifestations, while the former more recent term relates to the pathophysi-
ological concept of an inflammatory joint process related to an initiating infection elsewhere
in the body. Currently ReA is defined as aseptic arthritis triggered by an infectious agent
located outside the joint but with non-culturable organisms and/or bacterial components in
joint materials (1).
For a long time the classical triad of Reiter’s syndrome with conjunctivitis, urethritis and
arthritis following diarrhoea or urogenital infection was the hallmark of the disease entity.
Later in conjunction with the observation of aseptic arthritis following gut infection with
Yersinia enterocolitica, mostly in HLA-B27 positive individuals, the new term ReA was proposed,
which progressively replaced the previous terms like post-dysenteric, post-venereal and
endemic Reiter’s syndrome. Moreover, most recently the genetic, clinical and radiographic
overlap between ReA, ankylosing spondylitis, and other related diseases, were adapted to be
lumped together under the term ‘spondarthritides’ (2). Due to the complexity surrounding
the nomenclature and diagnostic criteria of this group of arthropathies, and due to the lack
of standardized diagnostic criteria, Table 10.1 gives an overview of the most frequent terms
in use to describe these infection-related diseases. There is an urgent need to further devel-
op our present terminology in the direction of a combined aetiological and clinical terminol-
ogy (e.g. chlamydia-induced ReA, chlamydia-induced spondarthritis, chlamydia-induced
enthesopathy, chlamydia-induced Reiter’s syndrome, yersinia-induced ReA, campylobacter-
induced ReA, salmonella-induced ReA).
References
1. Kuipers J, Kohler L, Zeidler H. Etiological agents: the molecular biology and phagocyte-
host interaction. Bailliere’s Clinical Rheumatology 1998; 12:589–609.
2. Calin A, Taurog JD. The Spondylarthritides. Oxford: OxfordUniversity Press, 1998.
Reactive arthritis and Reiter’s syndrome 149
Table 10.1 Terms used synonymously for reactive arthritis and Reiter’s syndrome.
Reactive arthritis (ReA) Reiter’s syndrome
Urogenic: Incomplete/complete Reiter’s syndrome

Urogenic ReA Fiesinger–Leroy–Reiter’s syndrome
Urethritic ReA Reiter’s disease
Post-venereal ReA Conjunctival–urethral–synovial syndrome
Reactive uroarthritis
Sexually acquired reactive arthritis (SARA)
Chlamydia-induced arthritis
Post-enteric:
Post-enteric ReA
Post-dysenteric ReA
Reactive enteroarthritis
Enteric ReA
Yersinia-triggered ReA
Salmonella-triggered ReA
150
Paper 1a
Contribution a l’étude d’une épidemie de

dysenterie dans la Somme
Authors
Fiessinger N, Leroy E
Reference
Bulletin de la Société Médicale des Hopitaux de Paris 1916; 40:2030–2069
Summary
The two French physicians Noel Fiessinger, microbiologist, and Edgar Leroy, military physi-
cian, described four cases with a conjunctivo-urethro-synovial syndrome 10 to 20 days after
diarrhoea. These patients were part of a very extensive microbiological and clinical analysis
of an epidemic of gastroenteritis, which was studied in a military hospital for infectious dis-
ease at the Somme, the French part of the First World War front. Although the causative bac-
teria were not stated, all cases were classified as bacillary dysentery by positive serology, stool
culture and exclusion of other causes of gastroenteritis. The joint puncture revealed an
inflammatory synovial fluid with many polymorphonuclear cells, but cultures were negative.
Nevertheless in the same report in other patients the bacillary dysenteria could be unequivo-
cally attributed by culture to the bacteria shigella, salmonella and yersinia.
Paper 1b
Über eine bisher unerkannte Spirochäteninfektion

(Spirochaetosis arthritica)
Author
Reiter H
Reference
Deutsche Medizinische Wochenschrift 1916; 42:1535–1536
Summary
Hans Reiter reported the case of a young officer serving at the Balkan front who developed,
7 days after acute diarrhoea, an illness characterized by arthritis, urethritis and bilateral con-
junctivitis. From the venous blood culture bacteria were grown and classified as a so far
unknown bacteria which was considered to be a spirochaete, and the disease was called
spirochaetosis arthritica.
Related references (1) Bauer W, Engleman EP. Syndrome of unknown etiology character-
ized by urethritis, conjunctivitis and arthritis (so-called Reiter’s
disease). Transactions of the Association of American Physicians
1942; 57: 307–313.
Related references (2) Wallace DJ. Should a war criminal be rewarded with eponymous
distinction? The double life of Hans Reiter (1881–1969). Journal
of Clinical Rheumatology 2000; 6:49–54.
Key message
Although the clinical association of arthritis and urethritis was described much earlier, both
references for the first time described concordantly the clinical picture of the later so-called
Reiter’s syndrome.

The aetiology of the syndromes were differently attributed as post-infectious following bacil-
lary dysentery and spirochaetosis arthritica, respectively. Both publications appeared at near-
ly the same time – December 8 (1) and 14 (2) 1916, but for language and other reasons the
denomination as Reiter’s disease or Reiter’s syndrome were adopted in the German and
English speaking medical literature and textbooks. Therefore the publication of Reiter is
mostly quoted as the first report and the most frequent cited paper, but the French speaking
medical society more correctly gave credit to all first observers of the ‘Fiessinger-Leroy-Reiter-
Syndrome’.
Strengths
1. Clear and simple observation of a clinical arthritic syndrome in association with a pre-
ceeding enteritic infection.
2. An example of a good clinical observation.
3. Over the whole century stimulated the thinking and research on association between bac-
terial infection and arthritis.
Weakness
The description of a spirochaete by Reiter was never confirmed.
Relevance
Although Reiter’s paper made only a negligible and some what misleading contribution to
the subject, the eponym Reiter’s syndrome has been retained in the literature and textbook
up to recent years because of a wide usage and the preference for the German and English
languages in the medical literature of this century. Moreover, the urethra-conjunctiva-syn-
ovial syndrome was described before the report of Hans Reiter and the French authors. Both
papers for the first time made a distinction between gonorrhoea and the newly described sus-
pected causative bacteria. The French authors clearly described the association with dysen-
tery and added the microbiological identification of enteric bacteria like salmonella and
shigella. Therefore the French description is the first one really to establish the association of
bacterial enteric infection and the syndrome. Hans Reiter wrongly attributed the syndrome
to a spirochaetal organism. Most importantly, Reiter was a fanatic anti-Semite and a vocal
proponent of the eugenics movements of the Nazi regime (2).
Altogether, for historical and ethical reasons it is certainly time to rename the syndrome
and no longer include Reiter’s name. To be in line with the classical clinical triad cited in
many textbooks the name ‘urethra-conjunctiva-synovial syndrome’ should be preferred to
the term ‘reactive cutaneo-arthropathy’, which now is classified as one clinical manifestation
or subtype of the whole spectrum of the group of reactive arthritides.
152
Paper 2
Isolation of bedsoniae from the joints of patients

with Reiter’s syndrome
Authors
Schachter J, Barnes MG, Jones JP, Engleman EP, Meyer KF
Reference
Proceedings of the Society for Experimental Biology (New York) 1966; 122:283–285
Summary
The study investigated 16 patients with Reiter’s syndrome all of whom had arthritis and ure-
thritis and/or conjunctivitis. Specimens (synovial membrane, synovial fluid, urethra, con-
junctiva) for bedsoniae isolation attempts were obtained from eight patients. Bedsoniae were
cultured in four patients from synovial material (two synovial membrane, one synovial fluid,
one from both sources), in two patients from urethra, and in one patient from conjunctiva.
In total five patients were positive for one or more of the investigated samples. Bedsoniae
were not isolated from synovial samples of 15 control patients with osteoarthritis (OA) or
rheumatoid arthritis (RA).
Related references (1) Dunlop EMO, Harper IA, Jones BR. Seronegative polyarthritis: the
bedsonia (chlamydia) group of agents and Reiter’s disease.
(2) Engleman EP, Schachter J, Gilbert RJ, Smith DE, Meyer KF.
Bedsonia and Reiter’s syndrome: a progress report. Arthritis and
Rheumatism 1969; 12:292 (abstract).
(3) Gordon FB, Quan AL, Steinman TI, Philips RN. Chlamydial iso-
lates from Reiter’s syndrome. British Journal of Venereal Diseases
1973; 49:376–380.
(4) Keat AC, Thomas BJ, Dixey J, Osborne MF, Sonnex C, Taylor-
Robinson D. Chlamydia trachomatis and reactive arthritis – the
missing link. The Lancet 1987; 1:72–74.
Key message
First isolation of bedsoniae from synovial material of patients with Reiter’s syndrome, the
psittacosis-lymphogranuloma venerum-trachoma agent later classified as Chlamydia trachoma-
tis. This report lends support to the hypothesis that this agent plays an aetiological role and
can be cultured from the joint, urethra and conjunctiva.

Although the infectious aetiology of post-venereal Reiter’s syndrome had been suggested for
a long time, and two earlier reports described the isolation of large trachoma-like organisms
and of inclusion conjunctivitis agents, respectively, from the urethra of patients with Reiter’s
syndrome and male patients with nonseptic urethritis, the observation of Schachter, et al. for
the first time definitely identified bedsoniae by culture from the joint.
More importantly in three patients the microorganisms were grown from two different
sources (one synovial membrane and synovial fluid, one synovial membrane and urethra,
one urethra and conjunctiva) indicating the multilocular presence of the bacterial agents. In
all cases in which positive results were obtained, the patients were studied during the initial
attack of the diseases, although detailed information on the presentation are missing and the
exact duration of the symptoms in the investigated population were not given. All samples
from 15 control patients with OA and RA were negative for bedsoniae, which indicated a
putative causative role of the agent. Most importantly, the authors also refer to preliminary
studies, in which intraperitoneal infection with yolk-sac suspensions of the isolates in guinea
pigs produced arthritis in some instances, and of greater significance, arthritis has been regu-
larly produced by intra-articular injection of the isolate into subhuman primates.
Surprizingly, only a few studies using the original yolk-sac culture method confirmed the
positive isolation from patients with Reiter’s syndrome (1, 2). In contrast, more recent inves-
tigations using new cell culture techniques sensitive for C. trachomatis were not successful in
culturing live replicating bacteria (3, 4). Therefore, the question of the presence and persis-
tence of live culturable chlamydia in the joint is not finally resolved, although new molecular
biology data demonstrate short-lived chlamydial RNA in the synovial membrane, indicating
the persistence of live organisms.
Strengths
1. Comprehensive search for one putative causative organism in different relevant clinical
samples (joint, urethra, conjunctiva).
2. All controls negative.
3. Stimulated the rheumatological research community to look more intensively for the role
of chlamydia in patients with arthritis.
Weaknesses
1. No exact description of the clinical presentation, symptoms and disease duration.
2. More recent cell culture techniques were not successful in culturing live replicating bacte-
ria.
Relevance
First isolation of chlamydia, formerly termed bedsoniae, from joint material, urethra and
conjunctiva of patients with Reiter’s syndrome. Although later studies with cell culture tech-
niques failed to isolate the organisms, currently the intra-articular presence of chlamydia has
been convincingly shown. Future investigations have definitively to elucidate whether only
inapparent non-culturable but live bacteria can be isolated, or replicating chlamydia only cul-
turable by sophisticated methods.
154
Paper 3
Arthritis associated with Yersinia enterocolitica

infection
Authors
Ahvonen P, Sievers K, Aho K
References
Acta Rheumatologica Scandinavica 1969; 15:232–253
Summary
During a 3 month period in 1968, 3875 unselected sera sent in for rheumatoid factor (RF)
test were screened for agglutination antibodies against Yersinia enterocolitica types 3 and 8.
Nineteen of the 46 patients with a significant titre of 160 or higher had polyarthritis with
acute onset. Of the remaining 27 patients 11 had erythema nodosum and febrile diarrhoea,
usually associated with some symptoms in joints. Most of the patients with acute polyarthritis
had preceding fever and gastrointestinal symptoms, five patients had urinary symptoms or
signs, and three also had conjunctivitis. The joints most often involved were fingers, knees,
and ankles. Three patients had acute sacroiliitis and some others back pain or severe myal-
gia.
Related references (1) Terrti R, Granfors K, Lehtonen O-P, Mertsola J, Makela AL,
Valimaki I, Hanninen P, Toivanen A. An outbreak of yersinia
pseudotuberculosis infection. Journal of Infectious Diseases 1984;
149:245–250.
(2) Winblad S. Arthritis associated with Yersinia enterocolitica infections.
Scandinavian Journal of Infectious Diseases 1975; 7:191–195.
Key message
First serological study convincingly describing the association of Yersinia enterocolitica infec-
tions with arthritis and other musculoskeletal symptoms.

Although these and others authors earlier observed a few cases of arthritis, arthralgia, and
Reiter’s syndrome in patients with positive Yersinia enterocolitica serology or stool cultures, the
present study is the first one definitely describing the association with this infection as well as
a spectrum of rheumatological manifestations including arthritis, sacroiliitis, back pain, gas-
troenteritis, erythema nodosum, fever, urinary symptoms, and conjunctivitis.
Strengths
1. Comprehensive clinical, serological and laboratory investigation.
2. Large population with rheumatic symptoms screened for yersinia antibodies.
Weaknesses
1. Most of the patients had been admitted to hospital because of the severity of the disease,
thus a selective series, comprising only moderate to severe cases.
2. Possibly did not include cases with short duration or mild symptoms.
Relevance
The diagnosis of yersina-associated arthritis was based on serological evidence and was sup-
ported to some extent by epidemiological data and by exclusion of other possible causes. All
cases had negative serology for brucella and salmonella infection (Widal test), RF (Waaler-
Rose and Latex test) and anti-streptolysin 0 titre. Specimens for yersinia stool culture were
obtained relatively late and in all cases after or during antibiotic therapy, the possible reason
why in none of the patients yersinia could be isolated from the stool.
Altogether the study not only described yersinia as an enteric infectious agent triggering
arthritis, but for the first time used the term ReA and also made the point that Y. enterocolitica
arthritis should be differentiated from rheumatic fever, arthritis associated with salmonella,
shigella, brucella and gonococcal infections and finally, if protracted, from RA.
156
Paper 4
Reiter’s disease and HL-A27

Authors
Brewerton DA, Caffrey M, Nicholls A, Walters D, Oates JK, James DCO
Reference
The Lancet 1973 ;ii:996–998
Summary
HL-A27 has been found in 25 out of 33 patients (76%) with Reiter’s disease, but only three
out of 33 patients (9%) with non-specific urethritis and two out of 33 controls (6%). When
all patients with sacroiliitis or spondylitis were excluded, HL-A27 was present in 15 out of 33
patients (65%) with peripheral involvement alone. The eight patients who were HL-A27 neg-
ative did not differ clinically from those who did, except that they did not have sacroiliitis or
spondylitis.
Related references (1) Brewerton D, Caffrey M, Nicholls A, Walter D, James DCO. Acute
anterior uveitis and HL-A27. The Lancet 1973; ii:994–996.
(2) Moll JMH, Haslock J, Macrae JF, Wright V. Associations between
ankylosing spondylitis, psoriatic arthritis, Reiter’s disease, the
intestinal arthropathies and Behçet’s syndrome. Medicine 1974;
53: 343–364.
(3) Aho K, Ahvonen P, Lassus A, Sievers K, Tiilikainen A. HL-A anti-
gen 27 and reactive arthritis. The Lancet 1973; ii:157.
Key message
The high frequency of HLA-B27 (initially termed HL-A27) was reported first in patients with
ankylosing spondylitis and their relatives. The present paper and a related study in acute
anterior uveitis and the diseases associated with it (1) extend this observation and clearly
indicate that a group of rheumatic diseases with common clinical features are also interrelat-
ed by a common background.

Clinically, ReA (Reiter’s syndrome) follows venereally-aquired non-specific urethritis or it fol-
lows dysentery or diarrhoea. Before the discovery of the association with HLA-B27 it was not
known why only a small proportion of those who contact infection develop the disease. The
repeated finding of HLA-B27 in ankylosing spondylitis, Reiter’s syndrome and other related
diseases has put forward the hypothesis that one factor in pathogenesis is inherited, related
to different triggered infections, and closely associated with the histocompatibility system.
Strengths
1. Only those patients were included in the study who had an acute episode of peripheral
arthritis of typical distribution for which no other cause could be found, occurring soon
after an attack of urethritis.
2. Patients selection was agreed before the HLA-B27 results were known.
3. Careful subanalysis of the frequency of HLA-B27 according to clinical presentations and
manifestations.
Relevance
The concordant high frequency of HLA-B27 in Reiter’s syndrome, ankylosing spondylitis and
other related diseases support the concept of seronegative spondarthritis developed by Moll
and Wright from clinical, serological, and radiological evidence (2). Moreover, the relation-
ship of Reiter’s disease with triggering infections by various infective agents opened the view
to suggest a common underlying process contributing to the pathogenesis of a wide group of
clinical disorders. In fact, the earlier reports of yersinia arthritis also indicated the association
between a bacterial infection, HLA-B27, and arthritis (3).
158
Paper 5
Followup study of patients with Reiter’s disease and

reactive arthritis with special reference to HLA-B27
Authors
Leirisalo M, Skylv C, Kousa M, Voipio-Pulkki L-M, Souranta H, Nissila M,
Hvidman L, Nielsen ED, Svejgaard A, Tiilikainen A, Laitinen O
Reference
Summary
The study reports the follow-up of patients with Reiter’s syndrome (n = 160), yersinia arthri-
tis (n = 144), and salmonella arthritis (n = 8). After a follow-up period of a mean time of 65,
52, and 32 months, respectively in the different diagnostic categories, chronic back pain and
joint symptoms were frequent in all patients groups. Patients with yersinia arthritis who were
HLA-B27 positive had more severe acute disease (more frequent back pain, urological symp-
toms, mucocutaneous manifestations, and a longer duration of the disease) and more fre-
quent chronic back pain and sacroiliitis. In contrast, in Reiter’s syndrome the follow-up
revealed no difference in the occurrence of any single chronic symptom between B27 posi-
tive and B27 negative patients.
Related references (1) Csonka GW. The course of Reiter’s syndrome. British Medical
Journal 1958; 1:1088–1090.
(2) Leirisalo-Repo M, Suorenta H. Ten-year follow-up study of patients
with yersinia arthritis. Arthritis and Rheumatism 1988;
31:533–537.
(3) Sairanen E, Paronen I, Mahonen H. Reiter’s syndrome: a follow-up
study. Acta Medica Scandinavica 1969; 185:57–63.
Key message
Chronic and recurrent joint symptoms and back pain, although usually mild, occur in a con-
siderable number of patients, and even radiological sacroiliitis develops in some patients.
Contrary to yersinia arthritis, the presence of B27 in Reiter’s syndrome has no influence on
the clinical picture of the disease.

A large study following the disease course of different subgroups of patients with ReA, which
most importantly also include HLA-B27 tissue typing.
Strengths
1. Comparison of different subgroups of ReA.
2. Extended analysis of the association of HLA-B27 with disease manifestations in clinical
course.
Weaknesses
1. Not a prospective study, therefore, large variability of follow-up.
2. Patients were collected in hospitals, which may restrict the generalizability of the results
because of selection bias.
3. Only a very small number with salmonella arthritis.
Relevance
Although previous studies have addressed the question of prognosis of Reiter’s syndrome (1,
3) and yersinia arthritis, the present study is a very large one, which for the first time directly
compared different subgroups of ReA with different triggering infective agents, and most
importantly, included HLA-typing in the analysis of the acute disease and follow-up. The
main conclusions were that the HLA-B27 in ReA is highly variable, absent in many patients
with Reiter’s syndrome and not obligatory for the disease development, but so far in yersinia
arthritis is of prognostic relevance indicating more severe disease and more frequent late
sequelae (see also 2).
160
Paper 6
Chlamydia trachomatis and reactive arthritis:

the missing link
Authors
Keat A, Thomas B, Dixey J, Osborn M, Sonnex C, Taylor-Robinson D
Reference
The Lancet 1987; i:72–74
Summary
Synovium, synovial fluid cells, or both from eight patients with sexually acquired ReA
(SARA) were examined by means of a fluorescence labelled antibody to Chlamydia trachomatis
(C. tr.). Chlamydial particles were seen in five of the 8 patients with SARA but in none of the
controls (n = 8) with other rheumatic diseases. All five patients had high titres of serum
chlamydial antibodies.
Related references (1) Granfors K, Jalkanen S, von Essen R, Lahesmaa-Rantala R, Isomaki

O, Pekkola-Heino K, Merilahti-Palo R, Saario R, Isomaki H,
Toivanen A. Yersinia antigen in synovial-fluid cells from patients
with reactive arthritis. New England Journal of Medicine 1989;
320:216–221.
(2) Schumacher HR Jr, Magge S, Cherian PV, Sleckman J, Rothfuss S,
Clayburne G, Sieck M. Light and electron microscopic studies
on the synovial membrane in Reiter’s syndrome. Arthritis and
Rheumatism 1988; 31:937–946.
(3) Taylor-Robinson D, Thomas BJ, Dixey J, Osborne MF, Furr PM,
Keat AC. Evidence that Chlamydia trachomatis causes seronega-
tive arthritis in woman. Annals of Rheumatic Diseases 1988;
47:295–299.
Key message
Identification of fluorescent chlamydial particles in joint samples of patients with SARA indi-
cating that the synovitis may result directly from the intra-articular presence of chlamydia.

Although C.tr. has been implicated for long time as a triggering infection in Reiter’s syn-
drome and SARA, earlier isolations of the bacteria from the joint have been questioned and
not replicated (Paper 2). Therefore, the present study was a major step towards the accep-
tance that chlamydial antigen and/or chlamydial particles are present in the joint of patients
with post-urethritic ReA. The authors concluded that renewed efforts to culture the organ-
isms are essential and that early treatment with an appropriate antibiotic therapy might alter
or thwart the progress of disease. Unfortunately, both suggestions although so promising
have not been realized up to today, especially as all efforts to cure ReA by short- or long-term
antibiotic treatment have failed.
Strengths
1. The fluorescence monoclonal antibody method is well established for diagnosis of C.tr. in
urogenital swabs.
2. The method is as least as sensitive as culture.
3. With the same antibody chlamydial particles have been seen in preparations from mouse
joints after experimental intra-articular inoculation of C.tr.
Weaknesses
1. The identified fluorescent particles were classified as elementary body without any mor-
phological confirmation; other possible forms like reticulate body, intermediate body or
aberrant forms were not excluded.
2. Only limited information on the demographics of the patients.
3. No comprehensive descriptions of the clinical signs and symptoms of the patients.
Relevance
The study has been widely accepted as the missing link to confirm that C.tr. is the causative
agent of post-urethritic ReA and that the intra-articular presence of the organism may direct-
ly result in the synovitis, instead of the long-time favoured hypothesis of an extra-articular
infection causing the arthritic immune reaction. The observation of the intra-articular pres-
ence of C.tr. not only changed the concept of pathogenesis of ReA but also stimulated inten-
sive research activities to elucidate the question of whether C.tr. is viable or dead in the joint.
162
Paper 7a
Chlamydial rRNA in the joints of patients

with chlamydia-induced arthritis and
undifferentiated arthritis
Authors
Hammer M, Nettelnbreker E, Hopf S, Schmitz E, Porschke K, Zeidler H
Reference
Clinical and Experimental Rheumatology 1992; 10:63–66
Summary
In one of 11 patients with chlamydia-induced arthritis (CIA) and three of 24 patients with
undifferentiated arthritis (UndA), but none of 16 controls (post-enteric ReA, n = 4; Lyme
arthritis, n = 3; RA, n = 8) chlamydial rRNA was found in the synovial fluid using a commer-
cially available nucleic acid hybridization test based on a isotopic (125J) conjugated single
stranded DNA-probe (GEN-PROBE). In one additional patient with CIA the synovial mem-
branes were positive for chlamydial rRNA.
Paper 7b
Molecular evidence for the presence of chlamydia in

the synovium of patients with Reiter’s syndrome
Authors
Rahman MU, Cheema MA, Schumacher HR, Hudson AP
Reference
Summary
Seven of 9 patients with Reiter’s syndrome were found positive in synovial biopsy samples for
chlamydial rRNA with a molecular hybridization technique using a cloned 466-basepair DNA
fragment encoding the 5’-most portion of Chlamydia trachomatis (serovar L2) 16sRNA gene.
Three of 13 patients with other mostly incompletely explained arthritis were also positive for
chlamydial RNA.
Related references (1) Wordsworth BP, Hughes RA, Allan I, Keat AC, Bell JI. Chlamydial
DNA is absent from the joint of patients with sexually acquired
reactive arthritis. British Journal of Rheumatology 1990;
29:208–210.
Related references (2) Kuipers JG, Jurgens-Saathoff B, Bialowons A, Wollenhaupt J,

Zeidler H. Detection of Chlamydia trachomatis in peripheral
blood leucocytes of reactive arthritis patients by polymerase
chain reaction. Arthritis and Rheumatism 1998; 48:1894–1895.
(3) Gaston JSH, Cox C, Granfors K. Clinical and experimental evi-
dence for persistent yersinia infection in reactive arthritis.
Key message
Both studies for the first time identified by molecular hybridization techniques chlamydial
RNA in joint samples of patients with Reiter’s syndrome, post-urethritic ReA and unclassified
arthritis.

The detection of chlamydial RNA in combination with previous findings of chlamydia-like
particles and/or chlamydial antigens in synovium and synovial fluid of patients with Reiter’s
syndrome and ReA suggest that whole, live bacterial organisms are present in inflamed
joints. In contrast, an earlier search for chlamydial DNA in synovial fluid of patients with
SARA had failed (1).
It is now well accepted that aberrant, live bacteria are persisting in synovial cells especially
in macrophages. Most recently chlamydia has also been identified in the blood in patients
with CIA indicating the bacterial dissemination from the entry site via the bloodstream into
the joint (2). Therefore, the definition of ReA at least for CIA now has to be modified in the
sense that it is no longer a sterile immune-mediated inflammatory process occurring distant
to a primary focus of infection, but rather bacteria driven disease with persistence of the
agent in the joint for a longer time. If this holds true for other forms of ReA (e.g. induced by
yersinia, salmonella or shigella) has to be elucidated in the future. Only recently, yersinia
DNA was found in chronic remittent ReA in the joint, whereas for a long time yersinia anti-
gens (lipopolysaccharide) have been shown in the joint and circulation (3).
Strength
Both studies although varying in the applied techniques, prevailing samples (synovial fluid
versus synovium) and diagnostic categories concordantly identified chlamydial RNA in joint
material.
Weaknesses
1. The data provide no insight into the detailed biological state of chlamydia in the joint, i.e.
whether it is actively, if slowly, dividing, or whether vegetative growth has been suspended.
2. In the study of Hammer, et al. the commercially available nucleic acid hybridization test
GEN-PROBE was of low sensitivity although of good specificity.
Relevance
The identification of chlamydial RNA by simple hybridization techniques indicated the intra-
articular persistence of intact, possibly viable chlamydia because in general RNA is degraded
soon after lysis of bacterial cells. This observation stimulated further studies to demonstrate
chlamydial DNA, short lived RNA and other genetic material proving the presence of whole,
metabolically active chlamydia in joint material.
164
CHAPTER 11
Lupus
Graham RV Hughes and Munther Khamashta
Introduction
The second half of the 20th century saw a revolution in the perception and treatment of
many major diseases, none more so than lupus. In the 1950s, 1960s and even (in some coun-
tries) the 1970s, lupus was considered a rare ‘small print’ disease. Not a part of ‘main-stream’
rheumatology. A fatal disease which required life-long steroids. No pregnancies allowed. A
kidney disease.
Now, we have dedicated lupus centres run by physicians with the proper skills and training
in dermatology, in medical aspects of pregnancy, in the neurology of lupus, and more. All
countries recognize the huge increase in the frequency of new lupus cases, a prevalence fig-
ure in many studies far exceeding that of better known diseases such as multiple sclerosis and
leukaemia.
This chapter selects five articles which have contributed to this change; the early clinical
studies of Edmund Dubois, the introduction of anti-DNA testing, the description of the
antiphospholipid syndrome (APS), and an example of the changing pattern of treatment,
the use of pulse-cyclophosphamide regimes. Finally, a paper, which in pointing out the
increased late coronary risk in lupus, paved the way for a whole new role for lupus, that of a
dIsease model for accelerated atheroma.
In the 1950s there were few large published series of lupus patients. One man changed all
that, Dr E.L. Dubois of Los Angeles. His clinical description of 520 cases provided the base-
line for the majority of later clinical series. They contributed to our understanding of the
breadth of features of systemic lupus erythematosus (SLE), notably, for example, highlight-
ing the importance of central nervous system (CNS) involvement in the disease. Finally, this
series provided the basis for his monumental textbook on lupus, the first ‘Bible’ of the dis-
ease.
In Dubois’ time, steroids were the dominant treatment for lupus, with doses of over 80 mg
daily given by some physicians. The introduction of pulsed intravenous (I.V.) cyclophos-
phamide regimes by the National Institutes of Health (NIH) group in the early 1970s proved
a major advance, largely in allowing the use of lower steroid doses. The NIH regime,
although considered over-aggressive by some workers (including ourselves) has, with modifi-
cation stood the test of time and has undoubtedly improved the prognosis of lupus nephritis
out of all recognition.
Of the multiplicity of antibodies found in lupus, two have achieved over-riding impor-
tance, anti-DNA antibody testing (the benchmark test for lupus, a test remarkable in its speci-
ficity for the disease), and antiphospholipid antibodies. It is our belief that the description of
the APS will have implications far beyond lupus. In cerebral disease, for example cases of
idiopathic stroke, of ‘atypical multiple sclerosis’ and of severe migraine are being attributed
to the pro-coagulant state associated with antiphospholipid antibodies, and being treated
successfully with anticoagulants.
Lupus 165
The syndrome is probably the single most important treatable cause of ‘idiopathic’ recur-
rent foetal loss. One of the results of studies of antiphospholipid antibodies has been an
important link with accelerated arterial disease. This discovery has opened up a new dimen-
sion in the study of disease models of arterial disease.
While 40 years ago, the causes of mortality in lupus were renal disease, steroid overdosage
and infection, the disease pattern has changed. The last of our five classic papers goes back
to the early clinical observation that there might be another ‘late’ threat, arterial disease.
Lupus and the APS are now important models for the study of accelerated atheroma, anoth-
er lateral move for those who treat and think about these diseases.
166
Paper 1
Clinical manifestations of systemic lupus

erythematosus. Computer analysis of 520 cases
Authors
Dubois EL, Tuffanelli DL
Reference
Journal of the American Medical Association 1964; 190:104–111
Summary
Diagnosis of SLE was confirmed by the presence of lupus erythematosus (LE) cells in 75.7%
of the patients, findings of skin biopsies in 6.0% and of renal biopsies in 1.2%, and by the
clinical picture alone in 17.1%. Blacks comprised 34% of the subjects. Spontaneous remis-
sions occurred in 35% of the patients. Proven familial SLE occurred in 2%. Myalgia was pre-
sent in 48.2%. No history of cutaneous involvement at any time was found in 28%. Classic
skin lesions of chronic discoid lupus at the onset of their illness were present in 10.8%.
Urinary abnormalities were noted in only 46.1%. Uremia caused 34% of the 135 deaths and
progressive CNS involvement caused 18.4%. The prognosis has markedly improved. The
mean duration is now 94.8 months for the entire series versus 38.5 months in an untreated
control group.
Related references (1) Wallace DJ, Hahn BH (eds.). Dubois’ Lupus Erythematosus. 5th edi-
tion. Baltimore: Williams & Wilkins, 1997.
(2) Lahita R (ed.). Systemic Lupus Erythematosus. 3rd edition. San Diego:
Academic Press, 1998.
Key message
This paper summarized the clinical diversity of SLE, as studied by an individual clinician.

After the initial clinical descriptions of lupus by Osler and others, a century ago, no single
physician contributed more to the broadening of the clinical concept of lupus. Dubois’ huge
clinical experience, based on his private practice, would probably not now be accepted for
publication, an example if ever one were needed, of ‘eminence-based medicine’ rather than
‘evidence-based medicine’. His Los Angeles observations began with eight patients who all
had a positive lupus erythematosus cell test, and ended with the world’s largest lupus centre,
with over 500 patients.
His clinical observations were numerous, notably the highlighting of the importance of
CNS involvement in lupus, the linking of avascular necrosis of bone with SLE, his pioneering
of treatment protocols, and, especially, his championing of the use of antimalarials in lupus.
He encouraged many, many physicians (including strong personal encouragement for
me, Graham Hughes) to study this disease, at that time widely regarded as ‘rare’ and ‘small
print’. His work spawned the first ‘classic’ textbook on lupus, an almost single-handed tome,
now together with Lahita’s textbook, one of the two ‘Bibles’ of the disease.
Lupus 167
Strength
The clinical observations of one man, heavily and directly involved in the practice of
medicine.
Weakness
Many of his patients would now come under different diagnostic labels, chronic active
hepatitis, Sjögren’s syndrome and so on, but the evolution of these subsets came later with
the development of newer immunoassays.
Relevance
The series which, more than any other, portrayed the broad sweep of clinical features of SLE.
168
Paper 2
Therapy of lupus nephritis: controlled trial of

prednisone and cytotoxic drugs
Authors
Austin HA, Klippel JH, Balow JE, Le Riche NGH, Steinberg AD,
Plotz PH, Decker JL
Reference
New England Journal of Medicine 1986; 314: 614–119
Summary
We evaluated renal function in 107 patients with active lupus nephritis who participated in
long-term randomized therapeutic trials (median follow-up, 7 years). For patients taking oral
prednisone alone, the probability of renal failure began to increase substantially after 5 years
of observation. Renal function was better preserved in patients who received various cytotox-
ic drug therapies, but the difference was statistically significant only for I.V. cyclophos-
phamide plus low-dose prednisone as compared with high-dose prednisone alone
(p = 0.027). The advantage of treatment with I.V. cyclophosphamide over oral prednisone
alone was particularly apparent in the high-risk subgroup of patients who had chronic histo-
logical changes on renal biopsy at study entry. Patients treated with I.V. cyclophosphamide
have not experienced haemorrhagic cystitis, cancer or a disproportionate number of major
infections. We conclude that, as compared with high-dose oral prednisone alone, treatment
of lupus glomerulonephritis with I.V. cyclophosphamide reduces the risk of end-stage renal
failure with few serious complications.
Related references (1) Klippel JH. Indications for and use of, cytotoxic agents in SLE.
Balliere’s Clinical Rheumatology 1998; 12:511–527.
(2) D’Cruz D, Cuadrado MJ, Mujic F, Tungekar MF, Taub M, Lloyd M,
Khamashta MA, Hughes GRV. Immunosuppressive therapy in
lupus nephritis.Clinical and Experimental Rheumatology 1997;
15:275–282.
Key message
Bolus I.V. cyclophosphamide together with corticosteroids, plays a key role in the manage-
ment of severe SLE, notably nephritis.
Lupus 169

The diagnosis of lupus was once a password for routine long-term high-dose steroid therapy.
In the 1970s and early 1980s, immunosuppressives became more widely used as ‘steroid-spar-
ing’ agents, azathioprine for milder cases, cyclophosphamide for more severely ill patients.
The side-effects of long-term oral cyclophosphamide, notably cystitis, amenorrhoea and
infections, severely limited the use of this drug in most young lupus patients. The group at
the NIH deserves the credit not only for the introduction and championing of intermittent
‘pulse’ regimes, but for their dogged perseverance in following the treatment sub-groups of
patients for many years. They have shown that intermittent pulses combined with steroids is
far superior in improving overall prognosis in lupus nephritis than steroids alone (1).
Over the past decade, a number of groups, including ours have modified the somewhat
strong original NIH dose regimes, resulting in similar success in the majority of patients, but
with far fewer adverse effects, notably the almost total eradication of amenorrhoea (2).
Strength
Intermittent ‘pulse’ regimes are well tolerated and improve compliance.
Weakness
I.V. administration requires regular close out-patient monitoring.
Relevance
Intermittent I.V. cyclophosphamide regimes are well tolerated, are beneficial in severe lupus,
and have contributed directly to improved prognosis in SLE.
170
Paper 3
Complement fixation with cell nuclei and DNA in

lupus erythematosus
Authors
Robbins WC, Holman HR, Deicher H, Kunkel HG
Reference
Proceedings of the Society for Experimental Biology and Medicine 1957; 96:575–579
Summary
Sera from patients with active LE fixed complement with a wide variety of nuclei from differ-
ent organs and species, with calf thymus nucleo-protein, and in two instances with histone.
Isolated calf thymus, salmon sperm, human leucocyte and pneumococcal DNA also fixed
complement with many of these sera. Similar reactions were not encountered in a limited
control series including normal individuals and other pathological states.
Most active LE sera fixed complement with both nuclei and DNA in roughly parallel titre.
However, exceptions were encountered and one serum reacted strongly with nuclei but
failed to react with DNA. Cross-absorption experiments with nuclei and DNA suggested the
presence of two distinct serum factors.
The LE factor appeared to be related to the factor responsible for complement fixation
with nuclei but distinct from that responsible for DNA fixation. The significance of these
findings with respect to antibodies against unclear constituents is discussed.
Related references (1) Pincus T, Schur PH, Rose JA, Decker JL,Talal N. Measurement of
serum DNA-binding activity in SLE. New England Journal of
Medicine 1969; 281:701–703.
(2) Hughes GRV. Significance of anti-DNA antibodies in SLE. The
Lancet 1971; 861.
Key message
Four papers, published almost simultaneously in 1957, followed up on earlier observations
that lupus sera contained specific antibodies against nuclear constituents. Lupus serum anti-
bodies could react against DNA.
Lupus 171

The finding of anti-DNA antibodies in serum has proved to be one of the most abidingly spe-
cific in medicine, and proved vital in distinguishing anti-nuclear antibody (ANA) specificity
in lupus from that in other autoimmune diseases. It is exceptionally rare to find antibodies
against double-stranded (ds) DNA in any condition other than lupus.
The clinical studies by Pincus and colleagues in 1969 and by Cohen, Hughes and Christian
in 1970, showed that not only could an immunoassay for anti-dsDNA be diagnostically impor-
tant, but that the level of antibody could provide a rough indicator of disease activity.
Although some clinicians make therapeutic decisions based on antibody titre, we consider
the practice too imprecise. Anti-DNA antibodies measured either by immunoassay, immuno-
fluorescence (Crithidia) or ELISA are now the standard test for lupus, and have directly con-
tributed to the growth in prevalence of lupus world-wide, changing it from a rare, ‘small
print’ disease to a major illness, overtaking RA in some countries.
Strength
Measurement of anti-dsDNA is the specific test for lupus.
Weakness
Some lupus patients (notably some with skin disease) remain persistently anti-dsDNA
negative.
Relevance
The application of anti-DNA testing to connective tissue diseasess has provided both a diag-
nostic test for lupus, and a rough guide to disease severity.
172
Paper 4
Thrombosis, abortion, cerebral disease and the

lupus anticoagulant
Author
Hughes GRV
Reference
British Medical Journal 1983; 287:1088–1089
Summary
Systemic lupus erythematosus, with its broad range of clinical and immunological abnormali-
ties continues to provide lessons relevant to research in a wider variety of disciplines and dis-
eases. In some patients three apparently unrelated clinical features of SLE, recurrent venous
thrombosis, CNS disease (including myelitis), and recurrent abortions, may, it seems, have
common pathogenic mechanisms. Clinicians have suspected as much for some time; those
dealing with many patients with SLE recognize a group of women who have as features of
their disease multiple (even a dozen or more) spontaneous abortions, multiple deep vein
and other thromboses, and neurological abnormalities including either putative cerebral
thrombosis or myelitis or both. Interestingly, some of these patients have negative test results
for ANA. Recent studies from the Hammersmith Hospital, London have confirmed and
extended the association of the lupus anticoagulant with thrombosis.
Harris, et al. have developed a sensitive solid phase radioimmunoassay for anticardiolipin
antibodies some 200–400 times more sensitive than, for example, the precipitation method
used in the Venereal Disease Research Laboratory test. There was a strong correlation
between raised titres of anticardiolipin antibodies and venous and arterial thrombosis. Of the
15 patients with the highest anticardiolipin antibody titres, six had a history of venous throm-
bosis and five had cerebral thrombosis without other predisposing factors. Two each had pul-
monary hypertension and multiple abortions. For those of us hardened into nihilism by years
of study of various autoantibodies in SLE, there is a rare sense of excitement at the implica-
tions of the associations now being reported.
Related references (1) Harris EN, Gharavi AE, Boey ML, Patel BM, Mackworth-Young CG,
Loizou S, Hughes GRV. Anticardiolipin antibodies: detection by
radioimmunoassay and association with thrombosis in systemic
lupus erythematosus. The Lancet 1983; ii:1211–1214.
(2) Hughes GRV. Hughes’ syndrome: the antiphospholipid syndrome.
A historical review. Lupus 1998; 7(suppl.2) S1–S3.
Key message
Antiphospholipid antibodies, detected by sensitive immunoassays, are a strong risk-marker
for a syndrome characterized by venous and arterial thrombosis (especially strokes), and
recurrent abortion.
Lupus 173

Originally described in our lupus clinic patients, it quickly became clear that many patients
with the APS have no evidence of lupus, indeed it is likely that the so-called primary APS, or
Hughes’ syndrome, far exceeds the prevalence of lupus itself. Five examples of the impor-
tance of the syndrome in internal medicine are given:
• It is an important, and potentially preventable cause of stroke, up to 20% of strokes in the
under-40s being due to Hughes’ syndrome.
• It is now recognized as one of the most important causes of repeated pregnancy loss.
• Treatment of pregnant APS women has improved pregnancy success rates from under
20% to over 70%.
• It is a potentially treatable cause of a number of common medical conditions including
migraine, epilepsy, memory loss, multiple sclerosis and deep vein thrombosis.
• In lupus, management has been profoundly altered by the recognition that some features,
notably many aspects of CNS lupus, are due to thrombosis rather than inflammation.
Strength
An easy-to-perform test now picks up a group of individuals susceptible to thrombosis.
Weakness
Many, possibly the majority of antibody-positive individuals do not thrombose. Other factors,
as yet unknown, may be involved.
Relevance
A pro-thrombotic syndrome, identified by the presence of antiphospholipid antibodies,
which has clinical implications not only in rheumatology, but also in fields as diverse as
obstetrics, neurology and cardiology.
174
Paper 5
The bimodal mortality pattern of systemic

lupus erythematosus
Authors
Urowitz MB, Bookman AAM, Koehler BE, Gordon DA, Smythe HA,
Ogryzlo MA
Reference
Summary
The changing pattern of mortality in SLE led to an examination of the deaths in a long-term
systematic analysis of 81 patients followed for 5 years at the University of Toronto Rheumatic
Disease Unit. During the follow-up 11 patients died, six patients died within the first year
after diagnosis (group I) and five patients died an average of 8.6 years (from 2.5 to 19.5
years) after diagnosis (group II). In those who died early the SLE was active clinically and
serologically, and nephritis was present in four. Their mean prednisone dose was
53.3 mg/day. In four patients a major septic episode contributed to their death. In those
who died late in the course of the disease, only one patient had active lupus and none had
active lupus nephritis. Their mean prednisone dose was 10.1 mg/day taken for a mean of 7.2
years. In none was sepsis a contributing factor to their death. All five of these patients had
had a recent myocardial infarction at the time of death; in four, it was the primary cause of
death. Mortality in SLE follows a bimodal pattern. Patients who die early in the course of
their disease, die with active lupus, receive large doses of steroids and have a remarkable inci-
dence of infection. In those who die late in the course of the disease, death is associated with
inactive lupus, long duration of steroid therapy and a striking incidence of myocardial infarc-
tion due to atherosclerotic heart disease.
Related references (1) Vaarala O, Alfthan G, Jauhiainen M, et al. Cross-reaction between

antibodies to oxidised low-density lipoprotein and to cardi-
olipin in systemic lupus erythematosus. The Lancet 1993;
341:923–925.
(2) Shoenfeld Y. The anti-phospholipid (Hughes) syndrome: a cross-
roads of autoimmunity and atherosclerosis. Lupus 1997;
6:559–560.
Key message
Lupus is now a major disease model for the study of accelerated vascular disease.
Lupus 175

Traditionally, most epidemiological studies of lupus had listed renal disease and infection as
the major cause of death. The Toronto publication in 1976 was the first major study to show
a bimodal mortality pattern. While it was widely recognized that the statistical prognosis of
lupus improved with each decade, this study highlighted a second mortality peak from coro-
nary and arterial disease.
The number of studies confirming this observation is now becoming a flood, with esti-
mates of as high as a 50-fold increase in coronary artery disease in lupus patients. The causes
of this phenomenon are still being studied. Hypercholesterolaemia, nephrotic syndrome,
steroid treatment and hypertension are all candidates, but clearly do not provide the whole
solution.
The discovery of the APS opened up a new line of investigation. It was clear that strokes,
coronary artery disease and accelerated atheroma were a feature in many patients with APS,
patients who had never received steroids or suffered renal disease, for example. In 1993,
Vaarala and her colleagues in Finland published an important study suggesting that some
antiphospholipid antibodies were capable of cross-reacting with oxidised low-density lipopro-
teins (oxLDL) (1). A subsequent study suggested that anti-oxLDL were in turn more closely
associated with arterial than with venous thrombosis.
These observations have opened up a major research effort in the immunology of athero-
ma, with a focus on the role of antibodies in the development of oxidation-mediated
endothelial injury. Many workers both in cardiovascular medicine and lupus are now collabo-
rating in this area (reviewed in 2), shifting the emphasis from lipids, through inflammatory
processes, to immunology. Lupus and Hughes syndrome are now important models for the
study of atheroma.
Strength
These observations have already shifted the emphasis of management of lupus patients
towards prophylaxis of risk factors for coronary artery disease. They have also provided a new
disease model for the study of atheroma.
Weaknesses
1. Arterial disease in lupus and Hughes syndrome is almost certainly multifactorial.
2. The importance of antibody-mediated disease remains unknown.
Relevance
Coronary and cerebral vascular diseases are an important causes of death in lupus. New clues
involving immunological mechanisms may lead to a modification of treatment and
prophylaxis.
176
CHAPTER 12
Scleroderma (systemic
sclerosis)
E Carwile LeRoy
Introduction
Reviewing a century of progress in the understanding of a disease as cryptic as scleroderma
(systemic sclerosis, SSc) is daunting. In one sense direct involvement in SSc research for the
last three decades of the century places one perhaps too close to be objective; in another
sense not being around for the first seven decades creates a distance which, while enhancing
objectivity, reduces the awareness gained by direct involvement. Nonetheless, a perspective
in SSc understanding gained in the 20th century can perhaps best be begun by examining
briefly the contributions of earlier times.
In Naples, Italy, in 1753 Carlo Curzio, a physician, described a young woman whose skin
became hardened over the entire body. This has been taken by some as the first reported
case of SSc. Doubt is cast by Curzio’s reported follow-up that complete remission followed
symptomatic management, suggesting as an alternative diagnosis the self-remitting sclerede-
ma of Bushke which follows bacterial infections of the neck and throat. Little else of note is
recorded which can be traced to the 18th century.
In contrast, the 19th century, with major medical centres in London, Edinburgh, Berlin,
Paris, Vienna and both the Middle and Far East, recorded numerous reports of a wasting
condition associated with hard skin. Physicians were wont to derive their own Latin and
Greek based terms for this uncommon disease, some of which are listed in Table 12.1. Many
well-recognized physicians, including Wernike, Addison, Kaposi, Osler, and Raynaud con-
tributed descriptions which introduced the multisystemic nature of SSc. Publications by
Horteloup and Gintrac are recommended (1, 2). This period is reviewed thoroughly by
Rodnan and Benedek (3).
Table 12.1 Alternative terms for scleroderma.
Endurcissement du tissue cellulaire Oedématie concrète

Sclérème en placards Sclérème lardacée
Sclerostenosis cutanea Morphée
Rheumatische Sclerose des Unterhautz-ellgewebes Sclérème simple ou nonoedémateux
Sclerma or pachydermateous disease Cacirnus eburneus
Cutis tensa chronica Elephantiasis sclerosa
Sclerosis corii Sclerosis telae cellularis et adiposae
Trophoneurosis disseminata Cicatrisirendes Hautsclerom
Scleriasis Hautsclerom
Keloid Sclérème des adultes
Textus cellularis duritiens Scleroma
Sclerema Scirrhosarca
Rodnan GP, Benedek TG. Annals of Internal Medicine 1962; 57:305–318.

Scleroderma (systemic sclerosis) 177
References
1. Horteloup P. De la Sclerodermie. Paris: P. Asselin, 1865.
2. Gintrac E. Note sur la sclerodermie. Review Medicin Chirurgie (Paris) 1847 2:263.
3. Rodnan GP, Benedek TG. An historical account of the study of progressive systemic
sclerosis (diffuse scleroderma). Annals of Internal Medicine 1962; 57:305–318.
178
Paper 1
Survival with systemic sclerosis (scleroderma).

A life-table analysis of clinical and demographic
factors in 309 patients
Authors
Medsger TA Jr, Masi AT, Rodnan GP, Benedek TG, Robinson H
Reference
Abstract
A life-table analysis of survivorship with SSc was performed, using 223 patients diagnosed in
Pittsburgh, Pennsylvania, and 86 patients in Memphis, Tennessee. The demographic and clinical
characteristics of the two series were similar, thus allowing for both comparison of the two groups
and analysis of the total 309 patients. A follow-up during 1970 was successful in 94% of all patients.
No difference in survival was found between the two patient groups, the combined 7 year cumula-
tive survivorship being 35%. Significantly decreased survival was found in older patients of both
series after allowance was made for the natural increase of mortality with age. Males had significant-
ly worse survival than females. Negroes had significantly worse survival than whites during the first
year of follow-up of all patients. When no internal organ involvement was detected at entry to
study, the negro prognosis was significantly worse throughout a 7 year follow-up period. Renal, car-
diac, and pulmonary involvement were each independently correlated with decreased survival.
Summary
Two major centres combined their hospital case records to analyze >300 SSc patients. With
an overall 7 year survivorship of 35%, a progressively poorer prognosis was noted with no
internal involvement, lung, heart or kidney involvement, respectively, and survival dropped
below 50% at 4 years for lung, at 2 years for heart, and at less than 1 year for kidney involve-
ment. This report placed the prognosis of SSc on a firm foundation for the first time.
Citation count
225
Related references (1) Goetz RH. Pathology of progressive systemic sclerosis (generalized
scleroderma) with special reference to changes in the viscera.
Clinical Proceedings (South Africa) 1945; 4:337–392.
(2) Tuffanelli DL, Winkelmann RK. Systemic scleroderma. A clinical
study of 727 cases. Archives of Dermatology 1961; 84:359–371.
Key message
Internal organ involvement predicts the survival of SSc patients.

This was the first study in which patients were evaluated uniformly for internal organ involve-
ment.
Strengths
1. Records from multiple hospitals were used.
2. Patient selection techniques were standardized.
3. Statistical analysis was careful and complete.
4. Authors were experienced sclerodermologists.
Weaknesses
1. Survival onset began at diagnosis, not at symptom onset.
2. Retrospective, dependent on physician who completed hospital records.
Relevance
A new era of outcome prediction in SSc began with this article. Figure 12.1 (Figure 5 in origi-
nal paper), showing a progressive worsening of prognosis with lung, heart, and kidney fail-
ure, has become a landmark in SSc thinking and spurred organ-specific and vascular studies
of pathogenesis and management.
Figure 12.1 Observed

1.00 cumulative survival rates
Expected (n=141)
(CSR) for scleroderma
0.75 patients with lung, heart,
No lung, no heart, or kidney involvement
CSR
0.50 no kidney (n=141)

and observed and
Lung, no heart, expected CSR for
0.25 no kidney (n=98) patients with none of
All kidney (n=16) Heart, no these involvements at
kidney (n=43)
entry to study.
0 1 2 3 4 5 6 7
Years of follow up
180
Paper 2
The relationship of hypertension and renal failure

in scleroderma (progressive systemic sclerosis) to
structural and functional abnormalities of the
renal cortical circulation
Authors
Cannon PJ, Hassar M, Case DB, Casarella WJ, Sommers SC, LeRoy EC
Reference
Medicine 1974; 53:1–46
Abstract
A clinicopathological review of 210 patients with SSc seen in one institution with emphasis on renal
(94 patients) involvement, the vascular lesions and the outcome (autopsies in 40 patients), all in the
era prior to the introduction of angiotensin-converting enzyme (ACE) inhibitors, which dramatically
changed the natural history of scleroderma renal crisis (SRC), characterized by oliguria, proteinuria,
hyperreninemia, and renal cortical infarction with renal insufficiency.
Summary
Proteinuria, azotemia and hypertension were used as clinical markers of renal involvement in
a survey of 210 SSc patients who were seen at Columbia-Presbyterian Medical Center (CPMC)
over a 20 year period. One or more markers were observed in 94 patients, an incidence of
renal involvement of 45%. Proteinuria occurred in 36%, azotemia in 19% and the syndrome
of ‘malignant’ hypertension in 7%. The mean onset of renal markers was within 3 years of
onset of the disease. No particular subsets of patients (age, sex, race) showed increased sus-
ceptibility to renal involvement (diffuse and limited subsets of SSc were not evaluated).
The onset of hypertension in SSc was not related to corticosteroid administration. Two
clinical patterns were observed: ‘malignant’ hypertension, and mild chronic hypertension
which terminated abruptly with renal failure. In the latter group, as in the 10 normotensive
SSc patients who developed oliguric renal failure, the onset of azotemia was usually preceded
by episodes of heart failure, pericardial effusion, dehydration or abdominal surgery.
Only 10% of the SSc patients without clinical markers died during the 20 years, whereas
60% of those with renal involvement died during the same period. The relative incidence of
death in patients with different markers was: proteinuria 63%, hypertension 66%, azotemia
77% and ‘malignant’ hypertension 80%. The incidence of markers in the patients that died
(82%) was higher than in the total population (45%). Among the 40 patients autopsied at
CPMC, renal involvement was the leading cause of death (43%). Two-thirds of all deaths and
76% of deaths from renal involvement by SSc occurred in fall and winter. The mean time
from onset of a clinical marker to death in this group was: hypertension, 13 months; protein-
uria, 7 months; ‘malignant’ hypertension and azotemia, 1 month.
Ninety percent of autopsied patients exhibited histological renal abnormalities; in 63% a
lesion was found in the interlobular arteries which appeared different from the vascular
changes seen in hypertensive disease. In the smaller arteries, arterioles and glomeruli the
abnormalities found (‘fibrinoid’ necrosis, arteriosclerosis, capillary thrombi, glomerular

hyperlobulation and endothelial swelling), were similar in normotensive and hypertensive
SSc patients and mimicked those of essential and ‘malignant’ hypertension. Patients with
renal failure exhibited focal areas of cortical necrosis.
The studies indicate the prognostic importance of the kidney vascular lesions in SSc. A
hypothesis is proposed that the structural and functional changes in the renal cortical circu-
lation contribute to the pathogenesis of both hypertension and oliguric renal failure in this
disorder. Possible therapeutic roles for renal vasodilators and for dialysis and renal transplan-
tation in SSc are suggested.
Citation count
228
Related references (1) Klemperer P, Pollack AD, Baehr G. Diffuse collagen disease. Acute
disseminated lupus erythematosus and diffuse scleroderma.
Journal of the American Medical Association 1942; 119:331–332.
(2) Weiss S, Stead EA Jr, Warren JV, Bailey OT. Scleroderma heart dis-
ease with a consideration of certain other visceral manifesta-
tions of scleroderma. Archives of Internal Medicine 1943;
71:749–776.
Key message
Simple clinical surrogate markers (proteinuria, azotemia, microangiopathic haemolytic
anaemia) can identify a population of SSc patients at risk for SRC. Without sensitive detec-
tion and appropriate therapy (ACE inhibitors), SRC is a lethal feature of SSc. All SSc patients
should be screened for SRC.
Strengths
1. Coordinates clinical evaluation with renal cortical blood flow measurements (133Xe), out-
comes, and renal histopathology.
2. Because therapy was ineffective in the era before 1974, these patients represent the natur-
al history of SRC.
3. In general, these patients were thoroughly evaluated.
Weaknesses
1. Retrospective.
2. Experience of one institution.
Relevance
In the 1970s, SRC was the most lethal of the features of SSc. This article brings clinical,
physiological, radiographic, and histopathological studies together to define the renal
involvement of SSc.
182
Paper 3
Scleroderma (systemic sclerosis): classification,

subsets and pathogenesis
Authors
LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA Jr,
Rowell N, Wollheim F
Reference
Abstract
A subclassification of SSc patients is proposed by a group of experienced investigators. The subsets
are defined by skin involvement with limited cutaneous SSc (lcSSc) patients having no taut skin
proximal to the knees, elbows or clavicles and diffuse cutaneous SSc (dcSSc) patients having truncal
skin tautness.
Summary
A two-subset classification was proposed which is justified by survival data from the published
studies of other investigators (Table 12.2).
Table 12.2 Survival in SSc (%) *
1 year 6 years 12 years
lcSSc 98 80 50
dcSSc 80 30 15
*Adapted from Giordano M, Valentini G, Migliaresi S, et al. Different antibody patterns and different
prognoses in patients with scleroderma with various extent of skin sclerosis. Journal of Rheumatology
1986; 13:911–920.
Citation count
384
Related references (1) Masi AT, Rodnan GP, Medsger TA, et al. Subcommittee for
Scleroderma Criteria of the American Rheumatism Association
Diagnostic and Therapeutic Criteria Committee: Preliminary
criteria for the classification of systemic sclerosis (scleroderma).
(2) Littlejohn GO, Barnett A, Miller M. Survival study of patients with
scleroderma diagnosed over 30 years (1953–1983). The value of
a cutaneous classification in the early stage of the disease.
Key message
Subsets defined by skin involvement can be instructive in overall prognosis, implying that
skin and internal organ involvement in SSc are related to the same pathophysiological mech-
anism.

These guidelines have proved helpful to the physician in the evaluation of individual patients
and to investigators in the classification of patient populations.
Strength
The consensus experience of the authors.
Weakness
Not subjected to a rigorous prospective study of new patients from multiple geographic
areas.
Relevance
A consensus definition of two useful subsets of SSc.
184
Paper 4
Identification of a nuclear protein (Scl-70) as a

unique target of human antinuclear antibodies
in scleroderma
Authors
Douvas AS, Achten M, Tan EM
Reference
Journal of Biological Chemistry 1979; 254:10514–10522
Abstract
The initial observations concerning an SSc-specific autoantigen, herein called Scl-70 from its size
(partially degraded), which turned out to be topoisomerase I, a ubiquitous nuclear enzyme.
Summary
This is a careful immunohistochemical analysis of SSc-specific (both false positive and false
negative examples are rare) and diffuse SSc-selective autoantibodies later shown to recognize
the nuclear enzyme topoisomerase I. Other SSc-selective autoantibodies include anti-cen-
tromere, anti-RNA polymerase III, and anti-fibrillin I. Overlap syndromes have different
autoimmune profiles. When careful autoimmune serology was combined with immunogenet-
ic studies (HLA and allotypy), the very strong association between sequence-specific HLA
haplotypes and autoantibodies of a specific type was realized to be the rule in human autoim-
mune disease. Often the HLA haplotype was more closely associated with the autoimmune
pattern than was the clinical disease profile or outcome.
Citation count
207
Related references (1) Rothfield NF, Rodnan GP. Serum antinuclear antibodies in pro-
gressive systemic sclerosis (scleroderma). Arthritis and
Rheumatism 1968; 11:607–617.
(2) Arnett FC. HLA and autoimmunity in scleroderma (systemic scle-
rosis). Intern Rev Immunol 1995; 12:107–128.
Key message
With precise technique, autoantibodies can be useful in the diagnosis, subset staging, and
prognosis in SSc.

This study provides a broad base of information regarding an important autoantibody in the
SSc clinical armamentarium, touching on clinical issues and providing detailed immuno-
chemical evaluation of the nature of the antigen by classical techniques.
Strengths
1. Immunochemical expertise.
2. Experience comparing other autoantibody reactions.
Weakness
Molecular genetic techniques, unavailable at the time of this study, could have provided full
antigen identification.
Relevance
This study represents major progress in the precise definition of the autoimmune response
of patients with SSc.
186
Paper 5
Connective tissue synthesis by scleroderma skin

fibroblasts in cell culture
Author
LeRoy EC
Reference
Journal of Experimental Medicine 1972; 135:1351–1362
Abstract
Skin fibroblasts are propagated by monolayer-explant tissue culture techniques from wrist and fore-
arm biopsies of clinically and histologically involved skin from nine patients with SSc and grown as
pairs with controls from skin biopsies from healthy donors matched for biopsy site, age, gender,
and ethnicity. Hydroxyproline measurements showed 3-fold greater soluble collagen in media and
2.5-fold greater insoluble collagen in cell pellets from the SSc fibroblasts in the paired mass cell cul-
tures. Glycoprotein production was also increased, based on sialic acid measurements.
Summary
Skin fibroblasts from subjects with SSc and control subjects were grown in tissue culture to
compare the characteristics of connective tissue metabolism. A striking increase in soluble
collagen (media hydroxyproline) was observed in eight of nine SSc cultures when they were
compared with identically handled control cultures matched for age, sex, and the anatomic
site of donor skin. Glycoprotein content as estimated by hexosamine and sialic acid was also
significantly increased in the SSc cultures. Estimations of protein-polysaccharide content by
uronic acid determinations were low in all cultures and not significantly increased in SSc
cultures.
This report demonstrated the feasibility of using fibroblast cell cultures to study chronic
rheumatic and connective tissue disorders (CTD). The initial results suggest a net increase in
collagen and glycoprotein synthesis in SSc fibroblast cultures. The implications of an abnor-
mality of connective tissue metabolism by skin fibroblasts propagated in vitro in the acquired
disorder SSc are discussed.
Citation count
150
Related references (1) Uitto J, Bauer EA, Eisen AZ. Scleroderma: increased biosynthesis
of triple-helical type I and type III procollagens associated with
unaltered expression of collagenase by skin fibroblasts in cul-
ture. Journal of Clinical Investigation 1979; 64:921–930.
(2) Jimenez SA, Feldman G, Bashey RI, Bienkowski R, Rosenbloom J.
Co-ordinate increase in the expression of type I and type III col-
lagen genes in progressive systemic sclerosis fibroblasts.
Biochemical Journal 1986; 237:837–843.

The first demonstration of increased collagen production by cells from an involved organ of
patients with SSc (or any other diffuse CTD), providing solid biochemical evidence to sup-
port the designation 30 years before that SSc (and others) was a ‘collagen’ disease.
Subsequently, many other characteristics of the activated state of the SSc fibroblast have been
determined.
Strengths
1. The first demonstration of an abnormality of collagen production in the fibrotic disease
SSc.
2. An example of the value of direct cellular studies from lesional biopsies of patients with
autoimmune disease.
Weaknesses
1. Collagen was not characterized as to type.
2. Mechanism of increased production was not delineated as to transcription, translation, or
post-translational regulation.
Relevance
Increased levels of collagen production by fibroblasts from skin biopsies of involved skin of
SSc patients early in their course of skin involvement was first observed here. This observa-
tion spawned an intense and continuing investigation by many laboratories of the character-
istics of the activated SSc fibroblast and the regulation of the extracellular matrix in SSc and
other fibrotic diseases.
188
Paper 6
Variable response to oral angiotensin-converting

enzyme blockade in hypertensive scleroderma
patients
Authors
Whitman HH III, Case DB, Laragh JH, Christian CL, Botstein G,
Maricq H, LeRoy EC
Reference
Abstract
Twelve SSc patients with hypertension, seven of whom had malignant hypertension and renal failure
or SRC, were treated with captopril. The first dose lowered mean pressure in all patients by
2.84 kPa (21.3 mmHg); in six patients it relieved encephalopathy. Blood pressure was controlled in
all patients. Two of seven patients with SRC had improvement in renal function; the five patients
who did not have malignant hypertension improved or stabilized. Despite good pressure control,
however, renal failure developed in five patients with SRC. The data indicated that captopril is
effective antihypertensive therapy in SSc and, when given early, may prevent renal failure and death.
Citation count
51
Related references (1) D’Angelo WA, Fries JF, Masi AT, Shulman LE. Pathologic observa-
tions in systemic sclerosis (scleroderma). American Journal of
Medicine 1969; 46:428–440.
(2) Lopez-Overjero JA, Saal SD, D’Angelo WA, Cheigh JS, Stenzel KH,
Laragh JH. Reversal of vascular and renal crisis of scleroderma
by oral angiotensin-converting-enzyme blockade. New England

This study of twelve SSc patients with SRC treated with the first ACE inhibitor (captopril)
demonstrated the efficacy of early detection and treatment, probably before renal cortical
infarction had become complete.
Strength
An effective therapy for the often fatal SRC.
Weaknesses
1. Uncontrolled, not blinded, open study.
2. Relatively short follow up.
Relevance
Captopril and other ACE inhibitors have significantly prolonged life in SSc patients by pre-
venting the sudden, severe renal insufficiency which half or so of SSc patients were at risk to
develop. On ACE inhibitors, renal failure occurs rarely, and it is much more chronic and
insidious in its course when compared with the pre-ACE inhibitor era.
190
Paper 7
Diagnostic potential of in vivo capillary microscopy

in scleroderma and related disorders
Authors
Maricq HR, LeRoy EC, D’Angelo WA, Medsger TA Jr, Rodnan GP,
Sharp GC, Wolfe JF
Reference
Abstract
The prevalence of scleroderma-type capillary abnormalities, as observed by in vivo microscopy, was
determined in 173 patients from three rheumatic disease centres. The patients had a variety of
CTDs: SSc 50; systemic lupus erythematosus (SLE) 60; mixed connective tissue disease (MCTD) 26;
Raynaud’s disease 11; other rheumatic disorders 26. Enlarged and deformed capillary loops sur-
rounded by relatively avascular areas, most prominently in the nailfolds, were found in 82% of
patients with SSc and in 54% with MCTD. The rarity of these abnormalities in SLE (2%) despite the
presence of Raynaud’s phenomenon suggests that they are not an expression of the Raynaud’s phe-
nomenon frequently associated with SSc and MCTD. The single patient with Raynaud’s disease and
scleroderma-type capillary changes subsequently developed SSc.
Summary
Maricq, et al. adapted widefield microscopy to study the capillaries of the nailfold. These
studies, over 25 years, have helped to distinguish SSc subsets, have served as a guide to dis-
ease activity, and to separate the varieties of fasciitis from the major groups of limited and
diffuse SSc (diffuse fasciitis with eosinophilia, eosinophilia myalgia syndrome, toxic oil syn-
drome), as well as MCTD patients at risk for pulmonary hypertension and localized SSc, by
either the presence or absence of nailfold capillary changes of SSc type (dilatation and avas-
cular areas).
Citation count
163
Related references (1) Brown GE, O’Leary PA. Skin capillaries in scleroderma. Archives of
Internal Medicine 1925; 36:73–88.
(2) Lewis T. Experiments relating to the peripheral mechanism
involved in spasmodic arrest of the circulation in the fingers, a
variety of Raynaud’s Disease. Heart 1929; 15:7–101.
(3) Banks BM. Is there a common denominator in scleroderma, der-
matomyositis, disseminated lupus erythematosus, The Libman-
Sacks Syndrome and polyarteritis nodosa. New England Journal of
Medicine 1941; 225:433–444.
Key message
A safe and user friendly technique provides significant clinical information early in the
course of CTD syndromes, with implications as to diagnosis, subset delineation, activity
assessment and prognosis.

Both generalist and specialist physicians have access in their office or clinic to the few tools
necessary (low power microscope, epillumination, immersion oil, camera for reimbursement
documentation) to evaluate all patients with Raynaud’s phenomenon (5–20% of adult popu-
lation) for early features of CTD.
Strengths
1. Adaptable to office practice.
2. Non-invasive, thus user friendly.
3. Inexpensive.
4. In SSc, dilated capillaries remain constant; thus test is reproducible.
Weaknesses
1. Requires modest training and experience on part of examiner.
2. Photography adds significant complexity.
Relevance
Widefield nailfold capillaroscopy for CTD has stood the test of time and widespread physi-
cian use. Hand in hand with SSc-selective serology (anti-centromere, topoisomerase I, RNA
polymerase III and Th/To) it is the best screening procedure available. In fact, nailfold capil-
lary patterns with serology constitute and define the most limited form of lcSSc (formerly
CREST).
192
CHAPTER 13
Systemic vasculitis
David GI Scott
Introduction
The systemic vasculitides are a heterogeneous group of diseases that appear to be becoming
commoner (or are recognized more frequently) in the last 5–10 years. Vasculitis can occur de
novo (primary vasculitis) including diseases such as Wegener’s granulomatosis, polyarteritis
nodosa and Churg–Strauss syndrome, or can also be due to infections (especially viral),
malignancy and chronic connective tissue diseases, such as rheumatoid arthritis (RA) and
systemic lupus erythematosus (SLE).
The first description of systemic vasculitis is usually attributed to Kussmaul and Maier
(1866) (1) though there were earlier descriptions by Rokitansky (2) and others. From this
time until the early 1950s almost all vasculitides were called peri- (poly-) arteritis nodosa. The
first author to rationally subdivide the many different clinical syndromes encompassing sys-
temic vasculitis was Pearl Zeek who in a seminal review in the American Journal of Clinical
Pathology in 1952 (Paper 1) classified vasculitis into five distinct subgroups. Classification sys-
tems have evolved over the ensuing 50 years but most are based on Zeek’s original system.
Classification criteria were first published in 1990 by the American College of Rheumatology
(ACR) (3) who compared clinical features of eight different vasculitic syndromes and devel-
oped criteria which were useful only in distinguishing one type of vasculitis from another. A
consensus was reached on the nomenclature to be used for vasculitis and definitions of spe-
cific vasculitides at a consensus conference in Chapel Hill in 1992 (published in 1994, Paper
2). These definitions and classifications have allowed more accurate recording on the epi-
demiology of vasculitis. The primary systemic vasculitides, though rare, probably occur in
approximately 20/million/year, but when all primary and secondary vasculitides are includ-
ed the incidence is probably in excess of 200/million/year.
The aetiology of vasculitis is poorly understood. Early studies in animal models examined
the role of allergy and serum sickness. More recently there have been links with infections
(hepatitis B and hepatitis C) and the link with autoantibodies has been established following
the description of antibodies to anti-neutrophil cytoplasmic antigens (ANCA) in 1984 by van
der Woude and colleagues (Paper 3). ANCA are now established in clinical practice to aid
diagnosis and prognosis of systemic vasculitides, particularly Wegener’s granulomatosis.
Corticosteroids improved the survival of patients with primary systemic vasculitis from
approximately 10 to 50% at 2 years, though had less of an effect on Wegener’s granulomato-
sis. The introduction of cyclophosphamide in the 1970s to treat systemic vasculitis has dra-
matically improved outcome significantly further (Paper 4). Cyclophosphamide is now the
gold standard treatment for systemic vasculitis and current research is examining different
ways of delivering cyclophosphamide or similar immunosuppressive drugs to reduce toxicity.
This improved outcome has resulted in vasculitides now being thought of as chronic autoim-
mune diseases with many of the physical and psychological consequences seen in other
rheumatic conditions such as RA and SLE.
Systemic vasculitis 193
References
1. Kaussmaul A, Maier R. Über eine bisher nicht beschreibene eigenthümliche
Arterienerkrankung (Periarteritis nodosa), die mit Morbus Bright und rapid
fortschreitender allgemeiner Muskellhmung einhergeht. Deutsche Archive Klinical Medizin
1886; 1:484–514.
2. Rokitansky K. Ueber einige der wichtigsten Krankheiten der Arterien. Deukschriften der
Kais. Akademie der Wissenschaften Besonders Abgedrucket 1852; 4:49.
3. Fries JF, Hunder GG, Block DA, et al. The American College of Rheumatology 1990
criteria for the classification of vasculitis: summary. Arthritis and Rheumatism 1990;
33:1135–1136.
194
Paper 1
Periarteritis nodosa: a critical review

Author
Zeek PM
Reference
American Journal of Clinical Pathology 1952; 22:777–790
Summary
This paper reviews literature from 1866–1952 in three eras, 1866–1900, 1900–1925 and
1925–1952. It covers clinical descriptions of vasculitis concentrating on putative aetiologies
and the descriptions of vasculitis in animal models. In the early periods allergy and immu-
nization, including serum sickness, were thought to be important factors. Clinical and histo-
logical descriptions appeared to include a number of different types of vasculitis which Zeek
summarized and classified into five different types: (a) hypersensitivity angiitis, (b) allergic
granulomatous angiitis, (c) rheumatic arteritis, (d) periarteritis nodosa, and (e) temporal
arteritis.
Related references (1) Lie JT. Classification and immunodiagnosis of vasculitis: a new
solution or promises unfulfilled? (Editorial). Journal of
Rheumatology 1988; 15:5.
(2) Scott DGI, Watts RA. Classification and epidemiology of systemic
vasculitis. British Journal of Rheumatology 1994; 33:897–899.
Key message
The first description/classification of vasculitis into five different types recognizing differ-
ences in clinical and histological features and outcome.

Prior to this review the term peri- (poly-) arteritis nodosa was used to describe a wide range
of different diseases ranging from relatively benign small vessel vasculitis to potentially fatal
systemic vasculitis (polyarteritis) and temporal arteritis, diseases which are now recognized to
have very different aetiologies and outcomes. Zeek summarized the different putative aeti-
ologies in some detail recognizing previous work on the potential role of allergy and serum
sickness.
This important careful review of the literature on vasculitis up to 1952 explained not only
the different clinical patterns but grouped together different types of vasculitis into five dis-
tinct patterns which has formed the basis for all current classification systems. The basis of
Zeek’s diseases can, for example, be seen in the classification systems described by Fauci and
colleagues in the 1970s (Paper 4), Lie in the 1980s (1), the Chapel Hill consensus definitions
(Paper 2) and Scott and Watts in the 1990s (2). It is interesting to note that only in the last 6
years has polyarteritis nodosa been clearly separated into microscopic polyangiitis and ‘clas-
sic’ polyarteritis nodosa in a way that has now been accepted internationally. Zeek in fact
undertook this separation to some degree in her review.
Strengths
1. A comprehensive view of the literature.
2. A logical classification which formed the basis of future classifications.
Weaknesses
1. A review article and, therefore, contains no original research data.
2. Takayasu arteritis and Wegener’s granulomatosis, though previously described, were not
included.
Relevance
Zeek’s classification of vasculitis has formed the basis of our understanding of the different
vasculitic diseases. This paper was the first to acknowledge properly the different clinical and
histological expressions of vasculitis and their prognosis, which was to become particularly
relevant with concurrent introduction of corticosteroid treatment.
196
Paper 2
Nomenclature of systemic vasculitides: proposal

of an international consensus conference
Authors
Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, Gross WL, Hagen EC,
Hoffman GS, Hunder GG, Kallenberg CGM, McCluskey RT, Sinico RA,
Rees AJ, van Es LA, Waldherr R, Wiik A
Reference
Summary
An ad hoc committee comprising clinicians and pathologists from six countries and multiple
medical disciplines convened in Chapel Hill, North Carolina, with the aim of reaching a con-
sensus on the names of the most common forms of non-infectious systemic vasculitis and to
construct root definitions of the vasculitides so named. This publication includes a classifica-
tion of vasculitis as well as definitions based on classical, clinical and pathological features of
the commoner vasculitides linking them to vessel size.
Related references (1) Fries JF, Hunder GG, Block DA, et al. The American College of
Rheumatology 1990 criteria for the classification of vasculitis:
summary. Arthritis and Rheumatism 1990; 33:1135–1136.
(2) Scott DGI, Watts RA. Classification and epidemiology of systemic
vasculitis. British Journal of Rheumatology 1994; 33:897–899.
Key message
The current nomenclature of vasculitis requires standardization. Definitions, if adopted
internationally, will improve our understanding of these diseases.

In 1990 the ACR (1) published data describing classification criteria of eight different vas-
culitides based on an analysis collected prospectively from approximately 1000 patients with
clinically well-defined vasculitis. Criteria were developed to separate one type of vasculitis
from another. There were, however, no standard definitions for making the initial diagnosis.
The ACR data did not, therefore, resolve many of the basic nomenclature problems sur-
rounding systemic vasculitis and the classification criteria are only helpful in differentiating
one type of vasculitis from another.
This paper provided the basic framework in terms of nomenclature for systemic vasculitis
in that:
• It was the first to recognize microscopic polyangiitis within the overall classification system
of vasculitis.
• It was the first to link putative aetiological factors (e.g. ANCA) with definitions.
• It redefined classical polyarteritis nodosa.
The distinction between polyarteritis nodosa and microscopic polyangiitis initially caused
controversy but these descriptions have stood the test of time (so far).
Strengths
1. The definitions are clear and succinct and easy to apply in the clinical setting.
2. The consensus came from a wide spectrum of specialities.
3. These definitions are now widely used internationally.
Weaknesses
1. The definitions are subjective but have been (mis)used as classification criteria.
2. The international experts were to some degree self-selected.
3. The classification grouped together a disparate group of small vessel vasculitides ranging
from potentially fatal Wegener’s granulomatosis to benign cutaneous vasculitis. A more
logical classification would separate Wegener’s granulomatosis, Churg–Strauss syndrome
and microscopic polyangiitis as being not only associated with ANCA, but also strongly
associated with glomerulonephritis and more responsive to cyclophosphamide treatment,
whereas Henoch Schonlein purpura, essential cryoglobulinaemia and cutaneous leucocy-
toclastic angiitis are associated with immune complex deposition rather than ANCA, cause
less, if any, renal disease and rarely require aggressive immunosuppressive treatment (see
Scott and Watts (2)).
Relevance
The definition of the different vasculitic diseases allows the international community to use a
common language. This paper changed our understanding of polyarteritis nodosa and clear-
ly identified differences between classic polyarteritis nodosa and microscopic polyangiitis
(which had previously been included under the group hypersensitivity vasculitis). The ACR
review found this group to have the poorest specificity and sensitivity in terms of classifica-
tion criteria.
198
Paper 3
Autoantibodies against neutrophils and monocytes:

tool for diagnosis and marker of disease activity
and marker of disease activity in Wegener’s
granulomatosis
Authors
Van der Woude FJ, Lobatto S, Permin H, van der Giessen M, Rasmussen N,
Wiik A, van Es LA, van der Hem GK, The TH
Reference
The Lancet 1985; i(February 23):425–429
Summary
Immunoglobulin G (IgG) autoantibodies against extra-nuclear components of polymor-
phonuclear granulocytes were detected in 25 of 27 serum samples from patients with active
Wegener’s granulomatosis and in only four of 32 samples from patients without signs of dis-
ease activity (Figure 13.1). In a prospective study these antibodies proved to be better mark-
ers of disease activity than several other laboratory measurements used previously. The
autoantibodies were disease specific. This autoantibody may have a pathogenetic role in
Wegener’s granulomatosis. The detection of this antibody is valuable for diagnosis and esti-
mation of disease activity.
Figure 13.1 Relation of IgG

p<0.01 p<0.01
512
ACPA titres and IGPT
scores with disease activity.
*1000 256
800 128
IGPT score
ACPA titre
600 64
400 32
200 16
0 Neg
Controls WG WG clinically WG WG clinically
not active active not active active
WG = Wegener's granulomatosis
Key message
This is the first description of a specific antibody associated with primary systemic vasculitis
particularly Wegener’ s granulomatosis.

Prior to this paper the aetiology of vasculitis was poorly understood. Vasculitis had often
been considered amongst the connective tissue diseases but was notable for the absence of
any obvious immune response. Indeed, the differentiation by histology between inflamma-
tion of the kidney due to connective tissue diseases such as lupus and the primary vasculitides
was the absence of specific immune deposition in the primary vasculitides.
This paper was the first to recognize the link between these antibodies (which had only
recently been described) and systemic vasculitis. The characterization of the antibody which
evolved from this paper has resulted in an industry of research linking autoantibodies, neu-
trophils and endothelial cells in the pathogenesis of vasculitis. The first ANCA workshop
meet in Copenhagen with 40 interested clinicians and pathologists. This has resulted in regu-
lar meetings, the last being the 9th ANCA workshop in Groningen including over 275 dele-
gates. These antibodies are currently used to aid diagnosis and assess prognosis of primary
systemic vasculitis, especially Wegener’s granulomatosis. Subsequent studies have shown that
in Wegener’s the antibody has specificity against proteinase 3 (PR3). ANCA are also now
used in the classification of vasculitis and link three diseases, Wegener’s granulomatosis,
microscopic polyangiitis and Churg–Strauss syndrome. The latter two are usually associated
with a perinuclear staining with antibodies directed against myeloperoxidase, whereas
Wegener’s granulomatosis is associated with antibodies against PR3.
Strengths
1. Relatively large numbers of patients with classical Wegener’s granulomatosis.
2. Wide range of normal and disease controls.
3. Clear message linking presence of the antibody to the disease.
Weaknesses
1. Disease activity criteria were only applied to the patient group.
2. Disease activity scores were applied retrospectively in many of the patients.
Relevance
Anti-nuclear cytoplasmic antibodies are now recognized as the most useful laboratory test in
the diagnosis of systemic vasculitis. The link between specific antibodies against PR3 and
Wegener’s granulomatosis has been confirmed in many subsequent studies. The description
of this antibody has led to huge expansion in the understanding of the relationship between
autoimmunity, neutrophil and endothelial cell interactions leading to potentially new and
exciting therapeutic targets.
200
Paper 4
Cyclophosphamide therapy of severe systemic

necrotizing vasculitis
Authors
Fauci AS, Katz P, Barton MD, Haynes BF, Wolff SM
Reference
New England Journal of Medicine 1979; 301(5):235–238
Summary
Seventeen patients with severe systemic vasculitis were studied over an 11 year period.
Sixteen were treated daily with cyclophosphamide (2 mg/kg/day) and one with azathio-
prine. Before entering the study all patients had active and progressive disease even though
16 patients had been receiving corticosteroids that had caused severe and often incapacitat-
ing toxic effects. Three patients died during the study. Complete and often dramatic remis-
sions occurred in the surviving 14 patients who were then placed on alternate-day corticos-
teroid treatment with continuation of cyclophosphamide. The mean duration of remission
was 22 months (range 2–61).
Key message
First description of cyclophosphamide treatment for a wide range of systemic vasculitides in
significant numbers showing a dramatic improvement in outcome.

Prior to the use of cyclophosphamide, the prognosis of systemic vasculitis, particularly
Wegener’s granulomatosis, was poor with a median survival in Wegener’s with renal involve-
ment of only 5 months. The overall mortality at 2 years for all diseases was less than 15% and
even with corticosteroids this had only improved to 48%. The use of corticosteroids was also
associated with significant toxicity. Previous descriptions of the use of immunosuppressive
treatments was restricted to case reports of one or two patients only.
This paper established cyclophosphamide as the standard treatment for severe systemic
vasculitis. The regime described by this group became known as the Fauci regime and was
until only a few years ago the gold standard against which all other treatments were judged.
Current international studies looking at new treatments of systemic vasculitis still use oral
cyclophosphamide in most cases, though the length of treatment has been shortened owing
to the discovery of significant long-term bladder toxicity with prolonged oral treatment.
Strengths
1. Large numbers of patients (when compared to previous studies).
2. Clear message showing clinical benefit in patients with severe disease.
Weaknesses
1. Patients studied over quite a long period of time in an uncontrolled fashion.
2. Varied clinical diagnoses included rheumatoid vasculitis, polyarteritis nodosa, microscopic
polyangiitis and Wegener’s granulomatosis.
Relevance
Vasculitis was transformed from a fatal disease to a treatable, though not always curable dis-
ease, with the use of cyclophosphamide which was introduced by Fauci and his colleagues
from the National Institutes of Health. The outcome of these serious diseases was dramatical-
ly altered by a treatment which over the ensuing 20 years was consider the ‘gold standard’.
202
CHAPTER 14
Sjögren’s syndrome
Roland Jonsson
Introduction
Sjögren’s syndrome (SS) is a chronic inflammatory and lymphoproliferative disease affecting
~0.5% of the population with autoimmune features and characterized by a progressive
mononuclear cell infiltration of exocrine glands, notably the lacrimal and salivary glands
(autoimmune exocrinopathy). These lymphoid infiltrations leads to dryness of the eyes (ker-
atoconjunctivitis sicca), dryness of the mouth (xerostomia), and very frequently dryness of
the nose, throat, vagina and skin. SS is associated with the production of autoantibodies since
B cell activation is a consistent immue-regulatory abnormality. The spectrum of the disease
extends from an organ-specific autoimmune disorder to a systemic process (musculoskeletal,
pulmonary, gastric, haematological, dermatological, renal, and nervous system involvement).
SS may develop alone (primary) or in association with almost any of the autoimmune
rheumatic diseases (secondary), the most frequent being rheumatoid arthritis (RA) and sys-
temic lupus erythematosus (SLE). SS is also associated with an increased risk of B cell lym-
phoma development. Current therapy provides only marginal symptomatic relief.
Sjögren’s syndrome was described during the 19th century in a number of case reports
with various combinations of dry mouth, dry eyes and chronic arthritis between the years of
1882 and 1924 (in 1). In 1892, Mikulicz reported a man with bilateral parotid and lachrymal
gland enlargement associated with massive round cell infiltration (2). Gourgerot, in 1926,
described three patients with salivary and mucous gland atrophy and insufficiency (3). In
1927 Mulock Houwer reported the association of filamentary keratitis, the major ocular man-
ifestation of the syndrome, with chronic arthritis (in 1). In 1933 Henrik Sjögren, a Swedish
opthalmologist, reported in his classical doctoral dissertation detailed clinical and histologi-
cal findings in 19 women with xerostomia and keratoconjunctivitis sicca, of whom 13 had
chronic arthritis (4). Later, in 1953, Morgan and Castleman established that SS and Mikulicz
disease were the same entity (5). The link between SS and malignant lymphoma was
described in a classical paper in 1964 (6). The distinction between primary and secondary SS
was suggested in 1965 (7) and later verified (8, 9). From the diagnostic point of view the first
histological grading assessing the infiltration of labial glands was described in 1968 (10). The
SS associated (Ro/SSA) autoantibodies in the sera were described in 1969 (11). SS in fami-
lies was described in a subsequent study (12). Gene interaction and complementation of the
immune response was shown for human leucocyte antigen (HLA) and the RNA proteins
Ro/SSA and La/SSB (13). A set of preliminary classification criteria was identified by a
European Concerted Action group in 1993, which has been widely accepted (14).
The future challenge will be to further clarify the disease process including genetic and
environmental influences. For this purpose, SS is a useful disease model to study the mecha-
nisms of autoimmunity perhaps applicable to other autoimmune diseases.
Sjögren’s syndrome 203
References
1. Jonsson R, Haga H-J, Gordon T. Sjögren’s syndrome. In: Arthritis and Allied Conditions - A
Textbook of Rheumatology (Koopman WJ, ed.), 4th edition. Philadelphia: Williams & Wilkins,
2001: pp 1736–1759.
2. Mikulicz J. Über eine eigenartige symmetrische erkrankung der tränen- und mundspe-
icheldrüsen. Beitr z Chir Festscr f Theodor Billrodt, Stuttgart, 1892: 610–630.
3. Gourgerot H. Insufficance progresive et atrophie des glands salivaires et muqueuses de la
bouche, des conjonctives (et parfois des muqueuses nasale, laryngée, vulvaire) sécheresse
de la bouche, des conjonctives. Bull Méd (Paris) 1926; 40:360–368.
4. Sjögren H. Zur kenntnis der keratoconjunctivis sicca. Acta Opthalmologica 1933; 11(suppl.
II):1–151.
5. Morgan WS, Castleman B. A clinicopathologic study of ‘Mikulicz’s disease’. American
Journal of Pathology 1953; 29:471–503.
6. Talal N, Bunim JJ. Development of malignant lymphoma in the course of Sjögren’s syn-
drome. American Journal of Medicine 1964; 36:529–540.
7. Bloch KJ, Buchanan WW, Wohl MJ, et al. Sjögren’s syndrome: a clinical, pathological and
serological study of sixty-two cases. Medicine 1965; 44:187–231.
8. Moutsopoulos HM, Webber BL, Vlagopoulos TP, et al. Differences in the clinical manifes-
tations of sicca syndrome in the presence and absence of rheumatoid arthritis. American
9. Moutsopoulos HM, Mann DL, Johnson AH, et al. Genetic differences between primary
and secondary sicca syndrome. New England Journal of Medicine 1979; 301:761–763.
10. Chisholm DM, Mason DK. Labial salivary gland biopsy in Sjögren’s disease. Journal of
Clinical Pathology 1968; 21:656–660.
11. Clark G, Reichlin M, Tomasi TB. Characterization of a soluble cytoplasmic antigen reac-
tive with sera from patients with systemic lupus erythematosus. Journal of Immunology 1969;
102:117–122.
12. Reveille JD, Wilson RW, Provost TT, et al. Primary Sjögren’s syndrome and other autoim-
mune diseases in families. Prevalence and immunogenetic studies in six kindreds. Annals
of Internal Medicine 1984; 101:748–756.
13. Harley JB, Reichlin M, Arnett FC, et al. Gene interaction at HLA-DQ enhances autoanti-
body production in primary Sjögren’s syndrome. Science 1986; 232:1145–1147.
14. Vitali C, Bombardieri S, Moutsopoulos H, et al. Preliminary criteria for the classification
of Sjögren’s syndrome: results of a prospective concerted action supported by the
European Community. Arthritis and Rheumatism 1993; 36:340–347.
204
Paper 1
Zur kenntnis der keratoconjunctivis sicca: Keratitis

filiformis bei hypofunktion der tränendrüsen
Author
Sjögren H
Reference
Acta Opthalmologica 1933; 11(suppl.II):1–151
Summary
The first detailed description of 19 female patients with xerostomia and keratoconjunctivitis
sicca, of whom 13 had chronic arthritis. Careful clinical and ophthalmological examinations
are presented, which included microscopic examination of the lachrymal glands in 10 and
parts of conjunctivae/corneae in 12 patients. In one patient who died, an autopsy was per-
formed, which also included an examination of the salivary glands. Infiltration of round cells
(mononuclear cells) in glandular parenchyma was described and found to be
accumulated/focal in nature. The study laid the ground for the designation of a syndrome.
Related references (1) Mikulicz J. Über eine eigenartige symmetrische erkrankung der
tränen–und mundspeicheldrüsen. Beitr z Chir Festscr f Theodor
Billrodt, Stuttgart, 1892: 610–630.
(2) Gourgerot H. Insufficance progresive et atrophie des glands sali-
vaires et muqueuses de la bouche, des conjonctives (et parfois
des muqueuses nasale, laryngée, vulvaire) sécheresse de la
bouche, des conjonctives. Bull Méd (Paris) 1926; 40:360–368.
(3) Morgan WS, Castleman B. A clinicopathologic study of ‘Mikulicz’s
disease’. American Journal of Pathology 1953; 29:471–503.
Key message
The first detailed description of a fairly large number of patients, all women with xerostomia
and keratoconjunctivitis sicca, of whom 13 had chronic arthritis, thereby establishing the
name of the syndrome, and its association with RA.

At the time when Henrik Sjögren presented his dissertation he was well aware that each sepa-
rate condition had been described before (1), i.e. keratitis filiformis by Leber already in 1882
(in reference1, Introduction), dry mouth (xerostomia) by Hadden in 1888 (in reference 1,
Introduction) and the combination by the French dermatologist Henri Gourgerot in 1926
(2). In 1927, Muloch Houwer described joint symptoms in connection with keratitis fili-
formis, and Houwer in fact found that one case report had been published by the German
Erich Fischer already in 1889 (in reference 1, Introduction).
In 1933, Henrik Sjögren had collected a total of 19 cases, all female aged 29–72 years. He
had conducted careful clinical and ophthalmological studies, which included microscopic
examination of the lachrymal glands and parts of conjunctivae/corneae and in one patient
examination of the salivary glands. This made up the basis of Henrik Sjögren’s classical doc-
toral dissertation. Further, Sjögren coined the expression ‘keratoconjunctivitis sicca’ and he
had previously carefully described a method to stain the damaged cells in the conjunctiva
and cornea by using 1% Bengal rose, a staining similar to that seen in ‘keratitis filiformis’.
Strengths
1. The first detailed description of a fairly large number of patients with xerostomia and ker-
atoconjunctivitis sicca.
2. Definition of keratoconjunctivitis sicca.
3. The suggestion that SS was part of a systemic disease.
Weaknesses
1. The thesis was published in German and thus gained limited initial recognition.
2. Salivary gland tissue was examined in only one patient.
Relevance
The thesis increased the awareness of SS as a systemic disease in particular after it was trans-
lated to English by Dr Bruce Hamilton in 1943. It also laid the foundation for future studies
in the field (3), which started predominantly in the 1960s.
206
Paper 2
Development of malignant lymphoma in the course

of Sjögren’s syndrome
Authors
Talal N, Bunim JJ
Reference
Summary
Of 58 patients with SS followed at the National Institute of Arthritis and Metabolic Diseases,
reticulum cell sarcomas developed in three, and lesions resembling Waldenström’s
macroglobulinaemia in a fourth. These changes were present in lymph nodes and organs
exclusive of the salivary and lacrimal glands. These patients have certain clinical and labora-
tory features in common which distinguish them from patients with the usual benign cases of
SS. These features include a relatively high incidence of splenomegaly, purpura, vasculitis,
leukopenia, lymphopenia and hypogammaglobulinaemia. The gamma globulin abnormali-
ties have been investigated by immunoelectrophoretic analysis, utracentrifugation and fluo-
rescent antibody techniques. In one patient, gamma globulin decreased from elevated to
markedly low levels; the rheumatoid factor (RF) and tissue antibodies disappeared as reticu-
lum cell sarcoma developed. In another patient, lymph nodes were populated with an abnor-
mally large proportion of cells containing 19S macroglobulin, and several cells exhibited
unusual intranuclear inclusions, which stained positive with periodic acid-Schiff stain. The
relationship between connective tissue diseases, gamma globulin abnormalities, thymomas
and malignant lymphomas is discussed. The hypothesis is presented that in SS the chronic
state of immunological hyperactivity and the proliferation of immunologically competent
cells producing abnormal tissue antibodies predispose to the relatively frequent development
of malignant lymphoma.
Related references (1) Bunim JJ, Talal N. The association of malignant lymphoma with
Sjögren’s syndrome. Transactions of the Association of American
Physicians 1963; 76:45–56.
(2) Kassan S, Thomas T, Moutsopoulos HM. Increased risk of lym-
phoma in sicca syndrome. Annals of Internal Medicine
1978; 89:888–892.
(3) Voulgarelis M, Dafni UG, Isenberg DA, Moutsopoulos HM and
Members of the European Concerted Action on SS: Jonsson R,
Haga H-J, et al. Malignant lymphoma in primary Sjögren’s syn-
drome – A multicenter, retrospective, clinical study by the
European Concerted Action on Sjögren’s syndrome. Arthritis
and Rheumatism 1999; 42:1765–1772.
Key message
The paper illustrates development of malignant lymphoma, the most severe outcome/com-
plication of SS.

Sjögren’s syndrome is considered a benign but systemic condition characterized by infiltra-
tion of the salivary and lacrimal glands with lymphoid cells, by hypergammaglobulinaemia
and by the presence of multiple circulating antibodies directed against various tissue compo-
nents. It is clear from the sequence of the clinical pathological events described in the paper
that the occurrence of SS and the enhanced immunological reactivity preceded by several
years the development of malignant lymphomas. It seems reasonable to speculate that the
chronic state of immunological hyperactivity and presumably, the associated proliferation of
lymphoid cells predispose to the development of malignant lymphoma. It is possible,
although unlikely, that SS is actually an early manifestation of an occult lymphoma. However,
this appears unlikely due to the following: (a) the benign lesions present in the biopsy speci-
mens of salivary glands; (b) the long course of illness, and (c) the frequency of purpura, vas-
culitis and abnormal tissue antibodies.
Strengths
1. Documentation of a severe complication in SS.
2. The study comprises well characterized patients.
Weaknesses
1. The report is only a collection of cases and without a control/comparative group.
2. The prevalence and relative risk of lymphoma development is not presented.
Relevance
The awareness that malignant lymphoma can develop in the course of SS is of fundamental
clinical importantance (1, 2). One of the risk factors is parotid swelling. Other frequent
accompanying clinical presentations are skin vasculitis, peripheral nerve involvement,
anaemia, and lymphopenia (3). The clinical and laboratory data suggest that SS is a disease
of unusually aggressive B cell activation which begins in the salivary glands and later becomes
extra-glandular with a strong propensity to malignant transformation at any time in its
course. The finding of monoclonal immunoglobulins or light chains in many SS patients sug-
gests that a monoclonal process is present very early and coexists with the polyclonal B cell
disorder.
208
Paper 3
Sjögren’s syndrome: a clinical, pathological and

serological study of sixty-two cases
Authors
Bloch KJ, Buchanan WW, Wohl MJ, Bunim JJ
Reference
Medicine 1965; 44:187–231
Summary
The clinical, serological, and pathological characteristics of five different subgroups of
patients with SS are reported. These groups consisted of 30 patients in whom the sicca com-
plex was associated with definite or classical RA (Group A), two with ‘probable’ RA (Group
B), three with progressive systemic sclerosis (Group C), and four patients with polymyositis
(Group D). In addition, 23 patients had the sicca complex in the absence of an associated
connective tissue disease (Group E).
Histological examinations of salivary glands from 20 patients in this series (16 biopsy and
five post-mortem specimens) presented a fairly wide spectrum of changes from acinar atro-
phy with adipose replacement and relatively sparse lymphocytic infiltration, to massive lym-
phocytic infiltration and replacement of acinar by lymphoid tissue. Proliferative changes in
the duct-lining cells were noted in 70% of specimens, but in only 45% had these progressed
to epi-myo-epithelial islands. Germinal centres were found in six biopsies. The integrity of
lobular architecture was preserved in all cases. Patients with SS, particularly those in Group
E, frequently had hypergammaglobulinemia, and RFs were demonstrated in the serum of
nearly all patients in this series regardless of the absence of rheumatoid joint disease.
Rheumatoid factors occurred in Group E with a frequency found in only one other condi-
tion, RA accompanied by subcutaneous nodules. Antibodies directed against several nuclear
and cytoplasmic antigens were detected, especially in the serum of patients in Group E. It
was not possible to demonstrate antibodies directed specifically against lacrimal or salivary
gland constituents. Reticulum-cell sarcoma developed in three patients several years after the
onset of SS and one patient developed extensive abnormal but not malignant lymphoid infil-
trates in several organs. The relationship between radiation therapy directed at the salivary
and lacrimal glands and the development of these lesions was discussed.
Methylcellulose is the drug of choice for keratoconjunctivitis sicca. Although systemic cor-
ticosteroid administration does cause a reduction in size of enlarged lacrimal and salivary
glands, it does not increase secretion and is therefore not indicated as treatment of SS. Anti-
malarial drugs have not been effective. Radiation therapy to enlarged glands cannot be rec-
ommended. Neuropathy in one patient associated with arteritis but not with RA, responded
well to corticosteroid therapy.
Related references (1) Moutsopoulos HM, Webber BL, Vlagopoulos TP, Chused TM,
Decker JL. Differences in the clinical manifestations of sicca
syndrome in the presence and absence of rheumatoid arthritis.
American Journal of Medicine 1979; 66:733–736.
(2) Moutsopoulos HM, Mann DL, Johnson AH, Chused TM. Genetic
differences between primary and secondary sicca syndrome.
New England Journal of Medicine 1979; 301:761–763.
Key message
The study offered the first and current definition of SS as a triad of keratoconjunctivitis sicca,
xerostomia and another connective tissue disease with two of three components considered
sufficient for diagnosis.

Much interest has been focussed on the clinical and immunological derangements in the
connective tissue diseases, which have in common the presence of circulating autoantibodies
against one or several tissues or proteins. The present study delineated and characterized sev-
eral clinical subgroups of SS. About one-half of the patients had RA associated with the sicca
syndrome. In the majority of these, widespread and destructive joint disease was evident. The
coexistence of the sicca syndrome in such patients suggests an association with more fully
expressed RA. From a wide spectrum of clinical manifestations and course an ordered defini-
tion is presented. Such clinical information is highly appreciated. Altogether, this was an
early paper which was the first comprehensive study of SS.
Strengths
1. A clear and concise definition of SS is presented.
2. Description of diseases to exclude.
Weaknesses
1. Imperfect definition of xerostomia.
2. The study lacks analysis of anti-Ro/SSA and anti-La/SSB.
Relevance
The definition of SS together with the distinction of one primary and one secondary form to
a great extent increased the attention of this enigmatic and distressing syndrome.
Subsequent studies delineated further both the clinical manifestations (1) as well as the
genetic differences (2) between primary and secondary SS. These early clinical studies laid
the ground for subsequent attempts to further define the genetics behind the disease as well
as defining diagnostic and classification criteria for SS.
210
Paper 4
Labial salivary gland biopsy in Sjögren’s disease

Authors
Chisholm DM, Mason DK
Reference
Journal of Clinical Pathology 1968; 21:656–660
Summary
A labial biopsy technique is described and was used to study 40 patients with connective tis-
sue disease and 60 post-mortem subjects. More than one focus of lymphocytes per 4 mm2
(0.08 in2) of minor salivary tissue (grade 4) was found to be a consistent finding in patients
with SS. The labial biopsy is shown to be a further valuable investigative procedure in such
patients.
Related references (1) Greenspan JS, Daniels TM, Talal N, Sylvester RA. The histopathol-
ogy of Sjögren’s syndrome in labial salivary gland biopsies. Oral
Surgery, Oral Medicine and Oral Pathology 1974; 37:217–229.
(2) Daniels TE. Labial salivary gland biopsy in Sjögren’s syndrome.
Assessment as a diagnostic criterion in 362 suspected cases.
(3) Daniels TE, Whitcher JP. Association of patterns of labial salivary
gland inflammation with keratoconjunctivitis sicca. Analysis of
618 patients with suspected Sjögren’s syndrome. Arthritis and
Rheumatism 1994; 37:869–877.
Key message
The labial salivary gland biopsy is shown to be a valuable diagnostic tool in SS.

A useful and internationally agreed definition of the disease exists (reference 7,
Introduction) but there are no fully agreed criteria for diagnosing SS. The oral component
of SS begins with lymphocytic infiltration of the major and minor salivary glands, causing var-
ious oral symptoms, insidious decrease in resting and stimulated salivary flow, and changes in
the composition and physical characteristics of saliva. Occasional enlargements of major
glands in SS usually represent the histopathological entity called ‘(benign) lymphoepithelial
lesion’, which is a firm, non-tender swelling that usually affects the entire gland. The pres-
ence of a benign lymphoepithelial lesion in a major salivary gland or sufficient focal lympho-
cytic sialadenitis (Figure 14.1) in a minor salivary gland biopsy specimen provides objective,
disease-associated evidence of the salivary component of SS. This paper was the first to
describe that the severity of infiltration in minor glands can be graded. Further, the subse-
quent suggestion of a focus scoring (1–3) of sialadenitis can provide a threshhold for individ-
ual diagnosis representing a semi-quantitative assessment.
Figure 14.1 Focal sialadenitis in a

labial salivary gland biopsy.
Strengths
1. Introduced focal lymphocytic sialadenitis in the minor salivary glands as an important
diagnostic tool.
2. Established a semi-quantitative significance threshold for patients at >1 lymphocytic
focus/4 mm2 (0.08 in2) glandular tissue.
3. Correlation with post-mortem specimens.
Weaknesses
1. Related focal sialadenitis to only a few other clinical manifestations of the syndrome.
2. Rather small sample material.
Relevance
There is great need for internationally accepted diagnostic criteria for both primary and sec-
ondary SS, but such criteria need to be as disease-specific as possible. There is no current
diagnostic gold standard against which to calculate accurately the sensitivity and specificity
values for diagnostic tests in SS, or for any other syndrome. Based on the strong association
between focal sialadenitis in an adequate labial salivary gland biopsy specimen and consistent
and rigorous tests for keratoconjunctivitis sicca (2, 3), proposed tests can be assessed against
the histological criterion. Alternative tests in the diagnosis of SS should be proposed to be
included in the criteria only when it is clear that all such tests are diagnostically equivalent.
212
Paper 5
Characterization of a soluble cytoplasmic antigen

reactive with sera from patients with systemic
lupus erythematosus
Authors
Clark G, Reichlin M, Tomasi TB
Reference
Summary
A tissue antigen reactive with sera from SLE patients has been partially characterized. The
antigen is a soluble cytoplasmic component which is present in a variety of human tissues
and is distinct from known nuclear antigens. It is an acidic macromolecule which has the
electrophoretic mobility of an alpha1 globulin and is resistant to most proteolytic enzymes
including trypsin, pepsin and chymotrypsin. Antigenicity is destroyed by 0.02 M periodate
and by 0.001 M parahydroxy mercuribenzoate. Antibodies to this antigen have been found in
40% of unselected SLE sera and were absent from a large number of sera from normal per-
sons and from the sera of patients with other connective tissue disorders.
Related references (1) Anderson JR, Gray KG, Beck JS, Kinnear WF. Precipitating autoan-
tibodies in Sjögren’s syndrome. The Lancet 1961; ii:456–460.
(2) Mattioloi M, Reichlin M. Heterogeneity of RNA protein antigens
reactive with the sera in systemic lupus erythematosus.
Description of a cytoplasmic non-ribosomal antigen. Arthritis
and Rheumatism 1974; 17:421–429.
(3) Alspaugh MA, Tan EM. Antibodies to cellular antigens in Sjögren’s
syndrome. Journal of Clinical Investigation 1975; 55:1067–1073.
(4) Tengnér P, Halse A-K, Haga H-J, Jonsson R, Wahren-Herlenius M.
Detection of anti-Ro/SSA and anti-La/SSB autoantibody pro-
ducing cells in salivary glands from patients with Sjögren’s syn-
drome. Arthritis and Rheumatism 1998; 41:2238–2248.
Key message
Present the first characterization of a tissue antigen, Ro, which reacts with many Sjögren’s
syndrome sera.

The results from this study show the Ro antigen to be distinct from the nuclear antigens thus
far described (1) (Figure 14.2). The antigen is derived primarily from the soluble cytoplas-
mic fraction of a human tissue homogenate (2). It is present in many human organs as well
as tissues of some other species. The Ro peptide is a 60 kD single chain protein. Peptides
derived from Ro retain a portion of immune reactivity.
Antibodies to the Ro antigen were found in SS and SLE (3) and have not been detected in
normal sera or in significant numbers in sera of patients with scleroderma, polymyositis, RA,
myeloma and other diseases. Together with La, these autoantibodies established SS as an auto-
immune disease (3). The clinical relationships of autoantibodies to forms of disease expres-
sion are the most powerful evidence that they are involved in the immunopathogenesis of SS.
Figure 14.2 Ouchterlony

agar diffusion demonstrating
the Ro antigen (from refer-
ence 11, Introduction).
Strengths
1. First characterization of the Ro antigen.
2. Established SS as an autoimmune disease.
Weakness
No molecular forms (e.g. Ro52) were analysed separately.
Relevance
Like RF and anti-nuclear antibody (ANA), anti-Ro but also anti-La cannot be considered spe-
cific for SS. There is also some variation in presented figures for positivity depending on lab-
oratory techniques for detection as well as patient cohort sampling and referrals (in refer-
ence 1, Introduction). Over 60% of patients with subacute cutaneous lupus have Ro and half
of whom also have La precipitins. Approximately 30 and 10% of SLE have Ro and La precip-
itins, respectively. Anti-Ro is also common in neonatal lupus. The figures for SS are 60–75%
positive for Ro and about 40% positive for La.
Though autoantibodies are clearly at least associated with extraglandular manifestations
such as vasculitis, purpura and leukopenia there is less convincing evidence that they are
directly involved in the destruction of the salivary (4) and lacrimal glands. However, this
paper started the autoantibody/autoimmune parade for SS and served to illustrate the close
relationship between lupus and SS.
214
Paper 6
Primary Sjögren’s syndrome and other autoimmune

diseases in families. Prevalence and immunogenetic
studies in six kindreds
Authors
Reveille JD, Wilson RW, Provost TT, Bias WB, Arnett FC
Reference
Annals of Internal Medicine1984; 101:748–756
Summary
The relationship of HLA and heavy chain immunoglobulin (Gm) haplotypes to disease and
autoantibody expression were examined in six large kindreds, each having one or more
members with primary SS. Various other autoimmune diseases and autoantibodies occurred
among the 117 relatives in these families. The HLA and Gm haplotypes did not necessarily
segregate persons into those with SS, other autoimmune disorders, or serological abnormali-
ties, but HLA alleles DR3 and DR2 occurred in significant excess in relatives with SS, irre-
spective of HLA haplotype. Segregation analysis suggested a Mendelian dominant genetic
defect common to the many autoimmune diseases and serological reactions that was not
linked to HLA or Gm. A significant effect of female sex was also documented. These studies
suggest that SS results from the interaction of several HLA-linked and non-HLA genes.
Related references (1) Provost TT, Talal N, Harley JB, Reichlin M, Alexander E. The rela-
tionship between anti-Ro (SS-A) antibody positive Sjögren’s syn-
drome and anti-Ro (SS-A) antibody-positive lupus erythemato-
sus. Archives of Dermatology 1988; 124:63–71.
(2) Provost TT, Talal N, Bias W, Harley JB, Reichlin M, Alexander E.
Ro (SS-A) positive Sjögren’s/lupus erythematosus (SS/LE)
overlap patients are associated with the HLA-DR3 and/or DRw6
phenotypes. Journal of Investigative Dermatology 1988; 91:369–371.
(3) Rischmueller M, Lester S, Chen Z, Champion G, Van Den Berg R,
Beer R, Coates T, McCluskey J, Gordon T. HLA class II pheno-
type controls diversification of the autoantibody response in pri-
mary Sjögren’s syndrome. Clinical and Experimental Immunology
1998; 111:365–371.
Key message
The study suggests that SS results from the interaction of several HLA-linked but also genes
outside the HLA system.

The occurrence of SS as well as other autoimmune diseases in family members of patients
with this syndrome has previously been well described but most often as case reports. This
paper was one of the first indications that autoimmunity in general tends to aggregate in
families. A survey of family histories in 98 unrelated patients with SS has shown similar high
frequencies of sicca syndrome as well as other diseases of autoimmunity. Twelve percent of
51 patients with primary SS had at least one other relative with the same disorder.
Interestingly, this 12% familiar clustering of primary is comparable to that reported for SLE.
In fact, this study suggests intimate relationships between primary and SLE in that these two
disorders tended to cluster in the same families. The study presented a future-looking idea,
since the studies to check this out are only now just underway.
Strengths
1. The relationships of HLA to disease and autoantibody expression is presented.
2. A significant effect of female sex was documented.
3. Clustering of SS and SLE in families.
Weaknesses
1. Twin studies are lacking.
2. The study is fairly descriptive.
3. Full genome scanning is lacking (but not available at that time).
Relevance
The apparent excess of SS and other autoimmune diseases in relatives of patients with SS
suggests an heriditary predisposition. Relatives with SS, other disorders, or serological abnor-
malities (1) do not necessarily need to share HLA haplotypes. Thus, a simple model of
Mendelian inheritance linked to HLA seems unlikely. However, the inheritance of a certain
HLA allele, HLA-DR3, appears to occur independently of HLA haplotypes shared with other
affected relatives (2). Mechanisms underlying certain HLA effects are also unknown,
although some studies have suggested that HLA alleles are more strongly associated with the
production of a particular antibody response in these disorders (3). Altogether, these studies
suggest that SS and the other autoimmune diseases seen in families result from many genet-
ics effects and are multifactorial.
216
Paper 7
Gene interaction at HLA-DQ enhances

autoantibody production in primary
Sjögren’s syndrome
Authors
Harley JB, Reichlin M, Arnett FC, Alexander E, Bias WB, Provost TT
Reference
Science 1986; 232:1145–1147
Summary
Primary SS is an autoimmune disorder characterized by dryness of the mouth and eyes. The
HLA locus DQ is related to the primary SS autoantibodies that bind the RNA proteins
Ro/SSA and La/SSB. Both DQ1 and DQ2 alleles are associated with high concentrations of
these autoantibodies. An analysis of all possible combinations at DQ has shown that the
entire effect was due to heterozygotes expressing the DQ1 and DQ2 alleles. These data sug-
gest that gene interaction between DQ1 and DQ2 (or alleles at associated loci), possibly from
gene complementation of trans-associated surface molecules, influences the autoimmune
response in primary SS.
Related references (1) Hamilton RG, Harley JB, Bias WB, Roebber M, Reichlin M,
Hochberg MC, Arnett FC. Two Ro (SS-A) autoantibody respons-
es in systemic lupus erythematosus. Correlation of HLA-DR/DQ
specificities with quantitative expression of Ro (SS-A) autoanti-
body. Arthritis and Rheumatism 1988; 31:496–505.
(2) Scofield RH, Frank MB, Neas BR, Horowitz RM, Hardgrave KL,
Fujisaku A, McArthur R, Harley JB. Cooperative association of T
cell beta receptor and HLA-DQ alleles in the production of
anti-Ro in systemic lupus erythematosus. Clinical Immunology and
Immunopathology 1994; 72:335–341.
(3) Fujisaku A, Frank MB, Neas B, Reichlin M, Harley JB. HLA-DQ
gene complementation and other histocompatibility relation-
ships in man with the anti-Ro/SSA autoantibody response of
systemic lupus erythematosus. Journal of Clinical Investigation
1990; 86:606–611.
Key message
Gene interaction between two HLA loci suggest gene complementation.

The genetic composition at the HLA locus DQ has been shown to have an important effect
upon the relative quantitative concentrations of autoantibodies present in sera from patients
with primary SS. Patients who are heterozygous at the HLA-DQ for the alleles DQ1 and DQ2
have much higher levels of anti-Ro/SSA, anti-La/SSB, RF and ANA. Gene interaction at
HLA-DQ then contributes to higher levels of autoantibodies. The most intriguing possibility
is that this data does indeed reflect gene complementation by the physical association of
multiple polymorphic gene products. This paper shows that the HLA relationships are com-
plicated and that they operate at the level of risk for disease as well as at the level of probabil-
ity of anti-Ro and anti-La antibodies in SS. Later work confirmed these relationships for anti-
Ro in SLE (1) and also implicated the T cell receptor (2).
Strengths
1. Illustrates clearly the HLA linkage in SS.
2. Gene complementation gives raised autoantibody levels.
Weakness
Used serology for HLA-typing (the only available method in 1986).
Relevance
In a subsequent study it was shown that the anti-Ro response had gene complementation in
SLE (3) consistent with what had been observed in SS. However, this time it was done at the
DNA level using an approach which revealed more of the polymorphic richness of the HLA.
The pathogenic mechanisms and disease perpetuation/chronicity in SS remains unknown.
The most urgent issues for understanding SS are consequently studies of aetiology and
detailed pathogenic processes. The roles that genetic factors and autoantibodies play will be
of utmost importance for progress in this field.
218
Paper 8
Preliminary criteria for the classification of Sjögren’s

syndrome: results of a prospective concerted action
supported by the European Community
Authors
Vitali C, Bombardieri S, Moutsopoulos H, et al.
Reference
Summary
Objective. Different sets of diagnostic criteria have been proposed for SS, but none have been
validated with a large series of patients or in a multicentre study. We conducted the present
study involving 26 centres from 12 countries (11 in Europe, plus Israel), with the goals of
reaching a consensus on the diagnostic procedures for SS and defining classification criteria
to be used in epidemiological surveys and adopted by the scientific community.
Methods: The study protocol was subdivided into two parts. For part 1, questionnaires
regarding both ocular and oral involvement were developed; they included 13 questions and
seven questions, respectively. For part II a limited set of diagnostic tests was selected, and the
exact procedure to be followed in performing these tests was defined. Part 1 of the study
included 240 patients with primary SS and 240 age- and sex-matched controls. Two hundred
and forty-six patients with primary SS, 201 with secondary SS, 113 with connective tissue dis-
eases but without associated SS, and 133 control patients were studied in part II.
Results: The study resulted in (a) the validation of a simple six-item questionnaire for
determination of dry eyes and dry mouth, which showed good discriminant power between
patients and controls, to be used in the initial screening for sicca syndrome, and (b) the defi-
nition of a new set of criteria for the classification of SS. The sensitivity and specificity of the
criteria in correctly identifying patients with either the primary or the secondary variant of SS
were also determined.
Conclusion: Using the findings of this prospective multicentre European study, general
agreement can be reached on the diagnostic procedures to be used for patient with SS. Final
validation of the preliminary classification criteria for SS is underway.
Related references (1) Daniels TE, Silverman S, Michalski JP, Greenspan JS, Sylvester RA,
Talal N. The oral component of Sjögren’s syndrome. Oral
Surgery, Oral Medicine and Oral Pathology 1975; 39:875–885.
(2) Manthorpe R, Frost-Larsen K, Isager H, Prause JU. Sjögren’s syn-
drome. A review with emphasis on immunlogical features.
Allergy 1981; 36:139–153.
(3) Fox RI, Robinson CA, Curd JG, Kozin F, Howell FV. Sjögren’s syn-
drome: proposed criteria for classification. Arthritis and
Rheumatism 1986; 29:577–585.
Related references (4) Vitali C, Moutsopoulos HM, Bombardieri S, et al. The European
community study group on diagnostic criteria for Sjögren’s syn-
drome. Sensitivity and specificity of tests for ocular and oral
involvement in Sjögren’s syndrome. Annals of Rheumatic Diseases
1994; 53:637–647.
Key message
Definition and validation of simple tools that may be used to measure the prevalence of SS in
the general population, and to reach agreement on classification criteria for this disease.

Although there has been a useful and internationally agreed-upon definition of SS available
for more than 30 years (Paper 3), there has been shortage of fully agreed criteria for diag-
nosing it. Understanding the clinical spectrum for SS and the various tests used to diagnose
its components (1) provides a basis for further refinement of potential criteria (2, 3). The
spectrum of the presentation of the disorder is very broad, ranging from the local conse-
quences of exocrine gland dysfunction to major, life-threatening systemic complications.
Although SS was once thought to be rare, recent data suggest that the prevalence of this con-
dition may approach or even exceed that of RA. Up to now, patients with SS have been
missed at diagnosis or uncorrectly classified, due to both great variability at disease presenta-
tion and to the lack of well-defined and commonly accepted diagnostic criteria. Thus, this
study report the results of a multicentre study performed in Europe and supported by the
European Commission.
Strengths
1. Use of diagnostic sensitivity and specificity calculations for each symptom and test.
2. The study was a multicentre procedure.
Weaknesses
1. The criteria are too inclusive by including subjective symptoms and having few/no exclu-
sions.
2. Each test was performed on only 26–90% of the patients studied.
3. Thus, the sensitivity and specificity calculations might not be methodologically sound.
Relevance
The main accomplishments of the study were: (a) the validation of a simple questionnaire
for dry eyes and dry mouth, to be used as the first step in the selection of potential patients
with SS, and (b) the definition of a set of classification criteria that is both highly specific and
highly sensitive (4) in discriminating patients with primary or secondary SS from controls
and from patients with connective tissue disease but without SS (Figure 14.3, overleaf).
There is clearly a need for internationally accepted diagnostic criteria for this disorder, but
those criteria must be as disease-specific as possible. The diagnosis of SS is significant for at
least two reasons: (a) clinically, a diagnosis of SS commits the patient to living with the spec-
tre of an incurable and potentially progressive disease; (b) scientifically, if patients are
included in studies of SS on the basis of subjective or less-specific criteria, some will be
included who do not have a systemic and immunological mediated disease. Our ability to
increase knowledge on the genetic background, the epidemiology and the pathogenesis of
this clinically and scientifically important entity will thus be hampered.
220
Ocular symptoms: Oral symptoms:

Dry eyes for more than 3 months, Daily feeling of dry mouth for more than
recurrent sensation of sand or gravel 3 months, recurrent or persistent swollen
Or
in the eyes or use tear substitutes salivary glands or use liquids to aid in
more than 3 times per day swallowing dry food
– +
No SS Ocular signs:
Positive Shrimer's test ()5 mm in 5 min),
or a Rose Bengal score of *4 according
– to van Bijsterveld scoring system
Histopathology: Histopathology:
Focus score *1 in a minor salivary Focus score *1 in a minor salivary
gland biopsy gland biopsy
– + – +
No SS SS
Objective salivary gland test: Autoantibodies:

Positive result in one of the following Antibodies to Ro(SS-A) or La(SS-B),
tests: salivary scintigraphy, parotid antinuclear antibodies, or rheumatoid
sialography or unstimulated salivary flow factors
()1.5 in 15 min)
– + – +
No SS SS SS
Objective salivary gland test:

Positive result in one of the following tests:
salivary scintigraphy, parotoid sialography
or unstimulated salivary flow ()1.5 ml in
15 min)
– +
No SS SS
Figure 14.3 An algorithm for the diagnosis of Sjögren’s syndrome based on the European criteria
(reference 14, Introduction).
Acknowledgements
The following experts on Sjögren’s syndrome and rheumatic disease selection of papers and
the content and are hereby acknowledged: Anne Isine Bolstad, Troy Daniels, Tom P Gordon,
Hans-Jacob Haga, John B Harley, Haralampos Moutsopoulos, Norman Talal, Claudio Vitali,
Kate Frøland and Karl Brokstad are acknowledged for editorial assistance. R.J. is supported
by EU grants BMH4-CT96-0595, BMH4-CT98-3489, The NorFA grant # 98.15.022-0 and the
Research Council of Norway grant # 115563/320.
222
CHAPTER 15
Myositis
Frederick W Miller and Paul H Plotz
Introduction
We have selected six key publications for this review which have helped to mould our current
understanding of a rare but increasingly recognized group of syndromes characterized by
chronic idiopathic muscle inflammation. These classic papers include: the first scholarly
review of the then known myositis cases; the first use of methotrexate in steroid-resistant
patients; the first careful delineation of diagnostic criteria and clinical groups; the best
descriptions of immunopathological differences among dermatomyositis (DM), polymyositis
(PM), and inclusion body myositis (IBM); the clearest description of the phenotypic and
prognostic distinctiveness of IBM, and finally a large study documenting the recent discovery
that myositis can be divided into clinically useful groups by the use of the myositis-specific
autoantibodies.
Although cases of myositis were described in Europe in the latter part of the 19th century,
Steiner’s detailed and scholarly report of a case of DM in 1903 (Paper 1) introduced the dis-
ease to the English-speaking world. Just as the concept of rheumatoid arthritis (RA) as a con-
dition separate from its phenocopies did not become clear until several decades into this
century, so too did PM, the subtler cousin of DM, not crystallize as a distinct entity until the
publication of Walton and Adams’ wonderful monograph in the 1950s (1). In our view, the
separation of the inflammatory myopathies from the genetic dystrophies remains incomplete
today. Some cases of ‘limb girdle dystrophy’ do respond well to anti-inflammatory therapy,
and some cases of ‘polymyositis’ are highly resistant to it. Thus, nosology is still a living issue
in the inflammatory myopathies.
The rarity and heterogeneity of the myositis syndromes has resulted in little information
to guide physicians in the treatment of myositis. The beneficial effects of corticosteroids in
myositis were recognized early, but the discovery of the benefits of methotrexate by Malaviya
and his associates in 1968 (Paper 2) antedates recognition of its usefulness in other inflam-
matory conditions. Subsequently the roles of azathioprine, high dose intravenous gamma
globulin, and combination chemotherapy in the treatment of myositis have been demon-
strated, but a real therapeutic breakthrough has not yet occurred.
Diagnostic criteria and the categorization of myositis into five more homogenous groups
were two major legacies of Bohan and coworkers (Paper 3). Among the important areas in
which recent progress has been made in extensions of this work are juvenile and cancer-asso-
ciated myositis. A large, careful study of juvenile myositis is just now being undertaken by a
consortium of clinics. The relationship of cancer to myositis is real, but biologically impene-
trable. Several excellent studies, most especially the large controlled investigation by
Sigurgeirsson and colleagues (2), however, have at least quantitated the risk and pointed the
way for practitioners who must decide how deeply to search for cancer when they first
encounter a new patient with myositis.
The meticulous immunopathological observations of A. Engel and his collaborators
(Paper 4) set the stage for understanding the mechanisms of muscle damage. When the new
biological agents are applied to myositis, the therapeutic rationale will have been derived
Myositis 223
from those beautiful studies. The recognition of IBM represents a milestone on a still unfin-
ished road. The description of IBM did not come abruptly, and we have included here not
the first hints of the illness (3, 4), but the landmark clinical and laboratory description
(Paper 5). The exact bounds of this entity remain uncertain for several reasons. Sporadic
IBM does not always show inflammation, and anti-inflammatory therapy has been only mar-
ginally effective when inflammation is present. Thus, here too nosology remains alive, and
the role of interacting genetic predispositions is likely to be increasingly recognized.
The pathological as well as clinical distinction between DM and PM has been supplement-
ed recently by the discovery that certain autoantibodies, which are wholly disease-specific for
myositis, delineate distinct clinical entities. Mathews and Bernstein (5) identified the
enzymes that join an amino acid to the correct tRNA as myositis-specific targets of autoimmu-
nity, and Love and her colleagues (Paper 6) expanded the observations of others to show
that myositis is really a complex family of diseases. This latter work was an important and nec-
essary step in understanding pathogenesis and in grouping patients into appropriate subsets
for therapeutic and genetic studies.
We are only beginning our exploration of the myositis syndromes, and the millennium
ahead holds unimaginable milestones on the path toward discovery of how these sign-symp-
tom-laboratory complexes we so poorly understand today can be defined, cured, or even pre-
vented.
References
1. Walton JN, Adams RD. Polymyositis. Edinburgh and London: E. & S. Livingstone Ltd, 1958.
2. Sigurgeirsson B, Lindelof B, Edhag O, Allander E. Risk of cancer in patients with der-
matomyositis or polymyositis. A population-based study. New England Journal of Medicine
1992; 326(6):363–367.
3. Chou SM. Myxovirus-like structures and accompanying nuclear changes in chronic
polymyositis. Archives of Pathology 1968; 86:649–658.
4. Yunis EJ, Samaha FJ. Inclusion body myositis. Laboratory Investigations 1971; 25:240–248.
5. Mathews MB, Bernstein RM. Myositis autoantibody inhibits histidyl-tRNA synthetase: a
model for autoimmunity. Nature 1983 Jul 14–20; 304(5922):177–179.
224
Paper 1
Dermatomyositis with report of a case which

presented a rare muscle anomaly but once
described in man
Author
Steiner WJ
Reference
Summary
After a thorough analytical summary of all previously described cases of similar illnesses,
Steiner here describes a patient with myositis whom he examined and whose biopsy he
describes. He notes many features of the clinical illness and of the histologic changes still
considered central to the diagnosis, and he separates it from trichinosis, infective myositis,
syphilitic myositis, and, most importantly, from ‘neuromyositis’–that is myopathy accompany-
ing neurologic disease. ‘Dermatomyositis’, he wrote, ‘may be defined as an acute, subacute,
or chronic disease of unknown origin, characterized by a gradual onset with vague and indef-
inite prodromata, followed by oedema, dermatitis, and a multiple muscle inflammation.’
Key message
This is a convenient starting point for the modern conception of myositis as an idiopathic,
non-infectious inflammation primarily limited to the skeletal muscles and skin with an acute,
sub-acute, or chronic course. Although rash was not a significant feature of this case, Steiner
recognized the resemblance to cases with dermatitis. The delineation of PM and its separa-
tion from the dystrophies only became clear (and is still occasionally obscure) with the publi-
cation of Walton and Adams’ beautiful monograph in 1958 (reference 1, Introduction).

When Steiner was a Medical House Officer under Osler at Johns Hopkins in 1899, he cared
for a 31-year-old labourer from the stone quarries who presented with sore, swollen, and
weak muscles. Despite a diet that included raw sausage, the patient did not have trichinosis
(no worms in the teased biopsy and a normal eosinophil count). He also had dyspnoea, but
no dermatitis. Distal as well as proximal muscles were clinically involved, and the biopsy
described in detail (five pages of text) was from a gastrocnemius muscle. Neurological exami-
nation was normal, except of the muscles. An extensive electrical examination consisted
entirely of a description of the gross response to electrical stimulation of the muscles rather
than the spontaneous discharge pattern, which is central now.
In his careful summary of earlier descriptions, combined with his own observations,
Steiner recognized the relationship to the disease now called PM, to scleroderma, and to vas-
culitis and a number of epidemiological features which still hold. His summary of the previ-
ously described pathological features, even more than the description of his own case, closely
matches what are still recognized as central features: ‘parenchymous and interstitial inflam-
mation...being either focal and diffuse;’ muscle fibres separated by oedema and inflammato-
Myositis 225
ry cells; perivascular inflammation; degeneration including the loss of striations, and regen-
eration. Of the clinical findings, he particularly noted the course of spontaneous remission
and relapse; the variable occurrence of dermatitis which may be succeeded by hyperpigmen-
tation; the involvement of the respiratory muscles, and the ominous implication of dyspha-
gia. He also discussed the possibility that infection and toxins have an aetiological role, and
comes to the same uncertain conclusion we would today.
Strengths, weaknesses, and relevance

In assessing the earliest descriptions of familiar diseases, one must always confront the fact
that the first cases are almost invariably the severest. Nevertheless, they offer a glimpse of the
natural history, and consequently a view of the aetiology and pathogenesis, which early
recognition and aggressive intervention obscure. Like most cases of all its autoinflammatory
relatives, myositis has a natural cycle of relapse and remission – a fact which should not be
forgotten. With the publication of this classic, Steiner’s paper defined the field.
226
Paper 2
Treatment of dermatomyositis with methotrexate

Authors
Malaviya AN, Many A, Schwartz RS
Reference
The Lancet 1968; 2(7566):485–488
Summary
Four patients with DM were treated with intravenous methotrexate. Three of them were
refractory to corticosteroids, and one had received no other treatment. Each patient was bed-
ridden by severe muscular weakness before treatment, and one was in the terminal phases of
the disease. All patients responded to methotrexate with improvement of muscular strength
to normal or near normal and disappearance of the rash. Concomitantly, laboratory abnor-
malities indicative of muscle disease disappeared.
Key message
Prior to the publication of this paper, the effective management of the debilitating weakness
and troublesome rash of DM had been limited to corticosteroids. Malaviya, et al. realized that
dermatomyositis was closely enough associated with other diseases recognized as immunolog-
ical in origin and was occasionally accompanied by serological abnormalities so that it might
respond to the anti-metabolite methotrexate. The remarkable success in their first four
patients reported in this paper not only introduced what is now known to be one of the most
efficacious treatments, but also strengthened the place of this illness in the family of autoim-
mune rheumatological diseases.
Related references (1) Sokoloff MC, Goldberg LS, Pearson CM. Treatment of corticos-
teroid-resistant polymyositis with methotrexate, The Lancet 1971;
1:14–16.
(2) Villalba L, Hicks JE, Adams EH, Sherman JB, Gourley MF, Leff RL,
Thornton BC, Burgess SH, Plotz PH, Miller FW. Treatment of
refractory myositis: A randomized crossover study of two new
cytotoxic regimens. Arthritis and Rheumatism 1998; 41:392–399.
Myositis 227

The introduction of corticosteroids for the treatment of RA eclipsed the slightly earlier
observations on the efficacy of the anti-metabolite aminopterin in refractory arthritis and it
took decades for rheumatologists to re-discover the benefits of methotrexate in RA. Well
before that re-discovery, however, Malaviya, et al., faced with the hopeless case of a boy dying
of DM despite corticosteroids, drew the analogy between DM and other diseases which had
responded to methotrexate, supposedly because immunological events were involved in
pathogenesis, and tried methotrexate. The boy recovered dramatically. It was a year before
they tried it again, and again they achieved an apparently durable remission in a patient with
cancer-associated DM who was failing on corticosteroids. Two years later, faced with a fresh
case of DM, they moved first to methotrexate and induced a remission without ever exposing
the patient to corticosteroids.
A confirmatory paper by Sokoloff, Goldberg, and Pearson in 1971 (1) extended the obser-
vations to patients with PM and reported two failures in seven patients treated, bringing the
state of knowledge of the place of methotrexate in the treatment of the idiopathic inflamma-
tory myopathies almost to the point where it remains today. The recognition that methotrex-
ate by mouth could be as effective was the most important remaining step, along with the
proof that methotrexate could achieve substantial improvement when given with azathio-
prine and corticosteroids in patients who had failed to respond to steroids and either of the
cytotoxics alone (2). It is now becoming common to use methotrexate or azathioprine along
with corticosteroids as the first-line of therapy in patients with severe disease or poor prog-
nostic features to spare the toxicities of prolonged, high dose steroids.

We are no longer so sure that we understand the mode of action of this useful drug. The fail-
ure of high intravenous doses in the range commonly used to treat malignancy to improve
upon low oral doses certainly suggests that the pharmacological mechanisms in the two fami-
lies of diseases are different. It is not even clear that an effect on immune mechanisms is the
dominant one at play in the therapy of myositis or arthritis, but the bold leap of the Tufts
group was a great advance that has stood the test of time.
228
Paper 3
A computer-assisted analysis of 153 patients with

polymyositis and dermatomyositis
Authors
Bohan A, Peter JB, Bowman RL, Pearson CM
Reference
Medicine 1977; 56(4):255–286
Summary
Diagnostic criteria and a classification scheme were developed by an empirical analysis of a
large number of well-studied patients with PM-DM. Diagnosis was based upon a combined
clinical-laboratory-pathological evaluation. The most useful criteria for diagnosis were found
to be: ‘a) proximal muscle weakness; b) elevation of serum enzymes; c) the characteristic
electromyographic triad; d) typical muscle biopsy histopathology, and e) the classical skin
rash of dermatomyositis’. The most sensitive diagnostic criteria were CPK, aldolase, and prox-
imal muscle weakness. Definitions were developed for possible, probable and definite dis-
ease. The myositis syndromes were further classified into five distinct groups of patients: PM,
DM, cancer-associated myositis, childhood myositis, and myositis syndromes in association
with other connective tissue diseases (overlap myositis).
Key message
Five criteria were found to diagnose probable or definite disease in nearly 99% of clinically-
defined PM-DM patients. Furthermore, the myositis syndromes defined by these criteria
could be divided into five homogeneous groups based upon a number of clinical features.
Related references (1) Wagner E. Fall einer seltenen Muskelkranheit. Arch Heilkd 1863;
IV:282.
(2) Unverricht H. Dermatomyositis acuta. Deutsche Medizinische
Wochenschrift 1891; 17:41–49.
(3) Hoffman GS, Franck WA, Raddatz DA, et al, Presentation, treat-
ment, and prognosis of idiopathic inflammatory muscle disease
in a rural hospital, American Journal of Medicine 1983;
75:433–438.
Myositis 229

Although PM and DM had been recognized for a century before the publication of this
paper (1, 2), no well-accepted criteria for the diagnosis or classification of these syndromes
existed before this work. This study has stood the test of time, in that decades later, these
approaches defined by Bohan, et al. are still the most frequently utilized for the diagnosis and
classification of myositis. The generalized acceptance of these criteria resulted in much need-
ed consistency among the clinical, pathogenic and therapeutic studies that followed and
allowed for more reliable meta-analyses of studies so necessary for rare diseases.
In addition to defining practical criteria and classifications, this study included detailed
descriptions of the clinical, electromyographic, pathological, serological, and prognostic fea-
tures of the largest cohort of patients published up to that time. It condensed 15 years of
experience by this major centre for the study of myopathies into a clinically useful summary
that remains one of the best descriptions of disease in one of the largest populations of
myositis patients ever studied. A number of aspects of the inflammatory myopathies, later
confirmed by other investigations, first emerged from this study. These include the findings
that: persistently normal serum enzymes, normal EMGs and normal muscle biopsies are seen
in some patients with clinically obvious PM-DM; the paraspinal muscles are likely to be the
most sensitive for EMG assessments; malignancy is associated with both DM and PM and is
the leading cause of death (71%) in this group, and responses to treatment cannot be
defined by muscle enzymes alone.
While we now know that additional clinical groups, including IBM, are important compo-
nents of the idiopathic inflammatory myopathies (Paper 5), and that the serological classifi-
cation of these diseases also adds useful information (Paper 6), the comprehensive and care-
ful description and analysis of this large group of patients served as the classic foundation for
defining and classifying these diseases.

This is the best-documented description of myositis in almost all of its clinically relevant
aspects that we have, but, like all but a handful (3) of other descriptions of myositis, it por-
trays the most seriously affected patients and probably therefore must be considered still
incomplete. Knowledge of the actual spectrum of the idiopathic inflammatory myopathies
awaits the elucidation of the pathogeneses of the myositis syndromes and the many other
myopathies that mimic these disorders.
230
Paper 4
Mononuclear cells in myopathies

Authors
Engel AG, Arahata K
Reference
Human Pathology 1986; 17:704–721
Summary
Detailed immuno-phenotyping of all the inflammatory cells in inflammatory myopathies
demonstrated that gradients of B cells and CD4+ T cells declined and the gradient of CD8+
T cells rose from perivascular to perimysial to endomysial areas of inflamed muscle in biop-
sies from DM, PM, and IBM patients, but there were marked differences in the proportion of
cells in the various locations. In DM, B cells and CD4+ T cells were generally found together,
there was a high CD4+/CD8+ ratio, and there was a relatively greater perivascular infiltrate.
Few T cells were found abutting or invading non-necrotic muscle fibres. In both PM and
IBM, activated CD8+ T cells were commonly found invading non-necrotic muscle cells at
endomysial sites. Macrophages were common and NK cells rare in all three diseases. Two
major processes are proposed. In DM, the predominant process is humoral immunity, where-
as in PM and IBM, the predominant process is a cytotoxic T cell attack on muscle cells which
appears to be antigen driven and major histocompatibility complex (MHC) restricted.
Key message
At least two major immunopathogenic processes are evident in the inflammatory myopathies:
a humoral and vascular-based process which dominates DM and a cytotoxic cellular process
which dominates PM and IBM, although both processes may be at play in a given patient.
The close resemblance of immunopathogenic findings in PM and IBM is striking given their
clinical and serological differences, and in both, non-necrotic muscle cells are invaded by
activated cytotoxic T cells.
Related reference (1) Banker BO, Victor M. Dermatomyositis (systemic angiopathy) of

childhood. Medicine 1966; 45:261.

Over many years, scattered observations on the immunopathogenesis of the inflammatory
myopathies had been gathered by a number of groups, but no clear picture had emerged
(1). In part, of course, this was due to the only gradual emergence of the now canonical sep-
aration of lymphocyte cell types using immunological cell surface markers. At the time of this
study, the main cell types and some of the cytokines that regulate immune function were
known and good reagents were available. Engel and Arahata applied exquisitely careful and
comprehensive studies using combinations of markers to identify the location, the cell type,
and the state of activation of all the inflammatory cells in many muscle biopsies from all the
major inflammatory myopathies.
They could document the vasculocentric and humoral process most commonly found in
both adults and children with DM by demonstrating the dominance of B and CD4+ T cells in
the inflammation at perivascular and perimysial areas. In PM, although some of these
Myositis 231
features were present, a major process was the presence of activated CD8+ T cells at
endomysial locations and within non-necrotic muscle cells (Figure 15.1). NK cells were virtu-
ally absent and macrophages were widely dispersed in both types of inflammation. IBM strik-
ingly resembled PM, echoing the imprecise clinical boundary between treatment-resistant
PM and IBM in older adults. The significant up-regulation of MHC Class I on non-necrotic
muscle cells and the paucity of pro-inflammatory cytokines have also emerged as characteris-
tic parts of the pathogenic picture.
Additional observations on the inflammatory infiltrate of Duchenne dystrophy document
the qualitative resemblance to, but wide quantitative difference from, PM and reinforce the
role that the reversal of inflammation may play in the partial response dystrophy patients
may experience from anti-inflammatory treatment. Just as the autoantibodies and the rashes
mark different groups of myositis patients, so do the basic immunopathological processes.
The kind of understanding of this family of diseases that will one day allow a rational and
sharply targeted approach to therapy was substantially advanced by the painstaking observa-
tions in this paper, and it set an extremely high standard for future quantitative immunohis-
tological studies.
A B C
Figure 15.1 Non-necrotic muscle fibres surrounded and invaded by mononuclear cells in polymyositis:
(A) CD8+ cytotoxic T cells are visualized by green fluorescence. (B) HLA DR+ activated cells are
demonstrated by red fluorescence. (C) Double exposure of A and B reveals that many of the CD8+
T cells are also activated as shown by their yellow colour.

This study and others by Engel, Arahata, and their colleagues defined the anatomy and cell
surface markers in the inflammatory myopathies, but those markers alone do not define the
immunopathogenesis. Later studies, building upon this work, are continuing to refine and
deepen understanding of the pathogenesis of myositis.
232
Paper 5
Inclusion body myositis. Observations in

40 patients
Authors
Lotz BP, Engel AG, Nishino H, Steven JC, Litchy WJ
Reference
Brain 1989; 112(3):727–747
Summary
The histopathological, ultrastructural, and clinical features of IBM were defined in 48 of 170
consecutive myositis patients in whom the diagnosis was suspected on light microscopic
grounds. ‘One or more vacuoles containing membranous material, groups of atrophic fibres,
and an autoaggressive endomysial inflammatory exudate occurred in 100%, 96% and 92% of
the muscle biopsy specimens.’ Filamentous inclusions, typically near vacuoles, were seen by
electron microscopy in 40 of 48 patients and distinguished IBM from other inflammatory
myopathies. ‘The typical clinical features in patients diagnosed by histological criteria as IBM
were: insidious onset after age 50 years with painless, proximal lower extremity weakness;
slow but relentless progression with selectively severe involvement of the quadriceps, iliop-
soas, tibialis anterior, biceps and triceps muscles; relatively early depression of the knee
reflexes, and a normal or mildly elevated serum creatine kinase level. The male: female ratio
was 3:1. Distal weakness occurred in about 50%, but only in 35% was it as great or greater
than proximal weakness.’ Prednisone failed to prevent disease progression in patients
observed for 2 or more years. IBM is a clinically and pathologically distinct form of idiopathic
inflammatory myopathy.
Key message
IBM is a unique form of myositis, defined by a combination of clinical, histological and ultra-
structural features, with poor responses to prednisone.
Related reference (1) Yunis EJ, Samaha FJ. Inclusion body myositis. Laboratory
Investigations 1971; 25:240–248.

In 1968, Chou described a 66-year old man with progressive weakness of all muscles, dyspha-
gia, pronounced atrophy of the shoulder girdle and quadriceps muscles, myopathic elec-
tromyograms, myonecrosis and mononuclear cell infiltrates in muscle biopsies, and unusual
muscle ultrastructural features, including myxovirus-like intranuclear and intracytoplasmic
filaments (reference 3, Introduction). In 1971, Yunis and Samaha (1) coined the term ‘inclu-
sion body myositis’ to describe a 26 year old white woman with an 8 year history of weakness,
normal creatine kinase levels, wasting of the abductors and adductors of the thigh but nor-
mal quadriceps muscles, inflammatory infiltrates on muscle biopsy, and similar fibrillar
nuclear and cytoplasmic inclusions on electron microscopic examination. Other case reports
and case series followed these initial publications.
Myositis 233
Lotz and his colleagues, however, were the first to assemble what is still the largest pub-
lished cohort of IBM patients, and define the extended clinical and pathological spectrum
of this illness. A number of distinguishing and clinically important aspects of IBM, which
have been subsequently confirmed by others, were emphasized in this paper. These include:
the elderly male predominance; the insidious onset; the slowly progressive, painless weak-
ness; quadriceps atrophy and the frequent involvement of distal muscles; the presence of
certain neurogenic features and dysphagia in some patients but few other extramuscular
symptoms; normal or relatively low serum levels of creatine kinase; the lack of a distinguish-
ing electromyographic pattern; occasional associations with other autoimmune disorders
and malignancy, and poor responses to corticosteroids.

We now know that the IBMs are a heterogeneous group of inflammatory and non-inflamma-
tory, and hereditary and sporadic conditions which share the common feature of characteris-
tic purplish rimmed vacuoles on trichrome staining and paired helical filamentous inclusions
on ultrastructural evaluation of muscle biopsies. Much of this progress in understanding
IBM, as well as the increased awareness by physicians that has resulted in the recognition that
IBM is the most common myopathy in persons greater than 50 years of age, can be traced to
the careful delineation of this fascinating entity by the authors of this paper.
234
Paper 6
A new approach to the classification of idiopathic

inflammatory mypathy: myositis-specific autoanti-
bodies define useful homogeneous patient groups
Authors
Love LA, Leff RL, Fraser DD, Targoff IN, Dalakas M, Plotz PH, Miller FW
Reference
Medicine (Baltimore) 1991; 70;360–374
Summary
Idiopathic inflammatory myopathies (IIM) are a heterogeneous group of systemic rheumatic
diseases which share common features. A number of classification schemes have been pro-
posed for them. We compared the usefulness of myositis-specific autoantibodies to the stan-
dard clinical categories in predicting clinical signs and symptoms, human leucocyte antigen
(HLA) types and prognosis in 212 adult IIM patients. Although patients with IBM (n = 26)
differed, there were few other significant differences among other clinical groups. In con-
trast, autoantibody status defined distinct sets of patients and each patient had only one
myositis-specific autoantibody. Patients with anti-aminoacyl-tRNA synthetase autoantibodies
(n = 47) had significantly more arthritis, fever, interstitial lung disease and ‘mechanic’s
hands’, were taking more prednisone and cytotoxic drugs, had higher death rates, and fre-
quently had HLA-DRw52. Those with anti-signal recognition particle antibodies (n = 7) had
increased palpitations, myalgias, severe refractory disease with high death rates, and were fre-
quently DR5, DRw52 +ve. Patients with anti-Mi-2 autoantibodies (n = 10) had increased ‘v-
sign’ and ‘shawl-sign’ rashes, cuticular overgrowth, and a good response to therapy, and were
often DR7 and DRw53+ve. The two patients with anti-MAS antibodies were the only ones to
have rhabdomyolysis preceding myositis; both had insulin dependent diabetes, and both had
HLA-B60, -C3, -DR4 and -DRw53.
Key message
The presence of myositis-specific antibodies help with the interpretation of symptoms and
signs in myositis patients, and in the prediction of their clinical course and prognosis.
Related reference (1) Plotz PH, Dalakas M, Leff RL, Love LA, Miller FW, Cronin ME.
Current concepts in idiopathic inflammatory myopathies:
polymyositis, dermatomyositis and related disorders. Annals of
Myositis 235

This work, amongst others, has helped to make it abundantly clear that myositis includes a
complex family of different diseases. The standard clinical grouping is into primary PM, pri-
mary DM, myositis with another connective tissue diseae, myositis associated with a diagnosis
of cancer, and IBM. However, the authors of this paper showed that of these five categories,
only IBM stands out as being different in its clinical manifestations. In contrast, they showed
that the single, specific type of autoantibody that each patient has is a good predictor of the
clinical features, course and outcome of the disease (Table 15.1). This, along with the HLA
associations, suggests that there are specific immunogenetics and immune responses linked
to specific clinical features in this family of diseases. If this is the case, then these findings
should lead onto our being able to distinguish and treat specific pathological pathways
responsible for disease in different groups of patients. The authors of this article therefore
proposed an adjunct classification of IIM, based on antibody status, which they recommend
should be incorporated into all future studies of epidemiology, aetiology and therapy of IIM.
Table 15.1 Disease profiles of patients with idiopathic inflammatory myositis.
IIM group Synthetase+ SRP+ Mi-2 MAS+ IBM
Gender (M:F) 2.7 6 1.2 1 0.5

Age at onset 41.3 36.5 46.3 36.5 51.8
Onset Acute Very acute Acute Acute Insidious
Major symptoms Interstitial lung Palpitations, cardiac ‘V sign’ rash, ‘shawl Alcoholic Falling, atrophy
and signs disease, dyspnoea disease, severe sign’ rash, cuticular rhabdomyolitis distal and
on exertion, fevers, weakness overgrowth asymmetric
arthritis, ‘mechanic’s weakness
hand’
HLA associations DR3 DR5 DR7 DR4 DR3
DRw52 DRw52 DRw53 DRw53 DRw52
Clinical group Jo-1, PM > DM, PM DM PM Not DM
others DM >> PM
Prognosis:
Steroid response Moderate Poor Good Good Poor
Response to taper Flare Flare Good Good Little
Mortality (%) 21 43 11 0 0
Strengths, weaknesses and relevance

This excellent study is based on a relatively large sample of patients (212). However, they all
came from the National Institutes for Health, which is a tertiary referral centre, and may rep-
resent a more severe group of IIM patients than those seen in other practices or in the com-
munity. Furthermore, some of the rarer clinical groups, such as juvenile DM, were not
included. However, it seems likely that the proposed new classification will stand the test of
time, and the fact that each patient with IIM has a single autoantibody, which acts as a mark-
er of a specific clinical expression, course and outcome, is a finding of huge potential impor-
tance–not only to our understanding of IIM, but to all autoimmune disease.
236
CHAPTER 16
Arthritis and infection

Andrew Keat and Ralph C Williams, Jr
Introduction
Bone and joint sepsis has been described in the medical literature from the times of ancient
Greece and Rome, and generalized infectious diseases which can affect joints, such as gonor-
rhoea and tuberculosis, have been common for centuries. Infections of bones and joints
were probably relatively common and frequently fatal prior to the introduction of antibiotics.
However, it is of interest to note that there is paleopathological evidence for long survival of
some people with severe osteomyelitis long before the germ theory of disease or antibiotics
were thought of (1). Chemotherapy was first used for osteomyelitis in 1936 (2, 3), and the
subsequent use of antibiotics for joint and other infections has reduced the severity and fre-
quency of septic arthritis as well as osteomyelitis.
However, in the second half of the 20th century arthritis and infection became very impor-
tant again, for two different reasons. First it became apparent that infections could trigger
certain forms of rheumatic disease, such as reactive arthritis (see chapter 10) and the search
for the ‘immaculate infection’ which might be the cause of rheumatoid arthritis (RA) contin-
ues today. Secondly, infection became a major issue in relation to the development of joint
replacement surgery.
There are many different types of joint infection, and the choice of just eight classic
papers in this field proved very difficult. One of the early classical descriptions of joint
pathology in septic arthritis has been included (Paper 1). The author and editors agreed on
one other paper that dates from the first half of the 20th century, which is a classical descrip-
tion of gonoccocal arthritis which includes a very early description of the dramatic response
of this condition to sulphanilimide (Paper 2). Septic arthritis is more likely to occur in
patients who are immunosuppressed or in those with other rheumatic diseases, including RA
and systemic lupus erythematosus (SLE). The third paper describes the condition in RA
patients, pointing out how features of sepsis may be masked in RA. The importance of John
Chanley’s work on joint prostheses is mentioned in the chapter on osteoarthritis (chapter
17); here we celebrate his contribution to the understanding of how post-operative and pros-
thetic infection rates could be reduced (Paper 4).
The last three papers are concerned with different sorts of relationships between infec-
tions and joint disease. Paper 5 is the first description of an arthritis being caused by immune
complex disease triggered by an infection (meningococcal). Paper 6 describes the arthritis
seen in adults in response to parvovirus infection, and Paper 7 relates to the recent but classi-
cal work of Alan Steere and his colleagues on Lyme disease. Finally we have included two
classical contributions on rheumatic fever (Paper 8, reviewed by Ralph Williams). There are
several other conditions that we would like to have included, such as mycoplasmal septic
arthritis (4), hepatitis B (5, 6), tuberculosis (7) and syphilis.
Arthritis and infection 237
References
1. Rogers J, Dieppe P. Lessons from paleopathology. The Practitioner 1983; 227: 1191–1199.
2. Le Cocq JF, Le Cocq E. Use of neoarsphenamine in treatment of acute S. aureus septi-
caemia and osteomyelitis. Journal of Bone and Joint Surgery 1941; 23:596–597.
3. Hedstrom SA, Lidgren L. Septic arthritis and osteomyelitis. In: Rheumatology (Klippel J,
Dieppe P. eds.). 2nd edition, London: Mosby, 6.2.1–6.2.10.
4. Furr PM, Taylor-Robinson D, Webster ADB. Mycoplasmas and ureaplasmas in patients
with hypogammaglobulinemia and their role in arthritis: microbiological observations
over twenty years. Annals of Rheumatic Diseases 1994; 53;183–187.
5. Duffy J, Lidsky MD, Sharp JT, Dans JS, Pearson DA, Hollinger FB, Min KW. Polyarthritis,
polyarteritis and hepatitis B. Medicine (Baltimore) 1976; 55:19–37.
6. Schumacher HR, Gall EP. Arthritis in acute hepatitis and chronic active hepatitis.
Pathology of the synovial membrane with evidence for the presence of Australia antigen
in synovial membranes. American Journal of Medicine 1974; 57:655–64.
7. Chapman M, Murray RO, Stoker DJ. Tuberculosis of bones and joints. Seminars in
Roentgenology 1979; 14:266–282.
238
Paper 1
Histological changes in the knee joint in

various infections
Authors
Keefer CS, Parker F, Myers WK
Reference
Archives of Pathology 1934; 18:199–215
Summary
The pathological lesions of eight cases of infective arthritis due to streptococcic, gonococcic,
meningococcic and pneumococcic infections and to an unidentified gram-negative coccus
are reported. The character of the change varied with the mode of infection which occurred
in one of three ways: (a) by the bloodstream; (b) by direct extension from osteomyelitis; (c)
by direct extension from the skin overlying the joint.
In the cases in which the infection of the joints occurred as a result of haematogenous
infection, the process began in the synovial connective tissue, with infiltration of polymor-
phonuclear, lymphoid and plasma cells about the blood vessels between the strands of con-
nective tissues. As the infection progressed, the synovial lining was destroyed and completely
replaced by granulomatous tissue. Later the cartilage and bone were involved in the process
and destroyed. When the bone was involved primarily, the outstanding lesions were a
destruction of bone and overlying cartilage. The inflammation of the synovium showed a
progression from the superficial to the deeper layers. The changes were characteristic of an
inflammatory process that could be readily distinguished from degeneration.
Related references (1) Parker F, Keefer CS, Myers WK, Irwin C. Histologic changes in the
knee joint with advancing age. Relation to degenerative arthri-
tis. Archives of Pathology 1934; 17: 516–532.
(2) Keefer CS, Parker F, Myers WK, Irwin C. Relationship between
anatomic changes in knee joints with advancing age and degen-
erative arthritis. Archives of Internal Medicine 1934; 53:325–336.
Key message
Infections get into joints by one of three routes: the blood, bone infections or from the over-
lying skin. Once the joint becomes infected it is quickly destroyed.

This paper is one of a series of contributions on joint pathology that came from the Boston
City Hospital Department of Pathology in the 1930s. At that time there was little knowledge
of joint pathology, and it was not easy to distinguish between the different causes of joint
destruction.
This was one of the first comprehensive descriptions of what happens to joints in septic
arthritis. The excellent descriptions of both gross pathology (Figure 16.1) and microscopic
changes clearly show the extensive destruction of joints that can result from these infections.
In addition to illustrating the degree of joint damage, and distinguishing it from the types of
damage done by ageing and osteoarthritis (1, 2), the other main contribution was to docu-
ment the three main causes of septic arthritis, and to illustrate that in many cases it was due
to haematogenous spread from other sites.
Figure 16.1 Post-mortem knee joint showing

severe joint destruction due to septic arthritis,
with deep ulceration through the bone and
cartilage of the lateral femoral condyle. This
contemporary specimen shows changes very
similar to those described by Keefer, et al. in
this paper. (From: Slide Atlas of Rheumatology
(Dieppe P, Bacon P, Bamji A, Watt I. eds.).
London: Gower Medical Publishing, 1984.)
Strengths
1. Excellent clinical and pathological descriptions of the eight cases.
2. Clear evidence of the serious destruction of joints that can result from a septic arthritis.
3. Differentiation of the three main sources of infections in joints.
Weaknesses
1. Only eight cases are described, in seven of which the infection was in the knee joint, and
seven of whom died as a result of the sepsis (in one case the leg was amputated, so that a
specimen was obtained without autopsy).
2. There was only one case of local spread from overlying skin and one of spread from a
bone focus; so six of the eight were haematogenous in origin. In spite of this they try to
differentiate pathology in the three types.
3. They infer the time course of the disease in spite of there only being pathological evi-
dence at one time point in a small number of cases.
Relevance
Distinguished septic arthritis as a serious cause of rapid joint damage, usually resulting from
haematological spread from another site.
240
Paper 2
Gonococcic arthritis: pathogenesis, mechanism of

recovery and treatment
Authors
Keefer CS, Spink WW
Reference
Summary
One hundred and forty cases of gonococcal arthritis seen over 5 years are documented. The
diagnosis was based on a history of recent gonorrhoea, evidence of specific gonococcal geni-
tal-tract infection, a positive gonococcal complement fixation test in blood or synovial fluid,
the presence of gonococci in synovial fluid. The course and clinical features of this condition
were described. Particular attention was drawn to the polyarticular nature of the disease, the
common involvement of tendon sheaths, the danger of exacerbating the disease by ‘vigor-
ous’ prostatic massage and the presence of maculopapular or haemorrhagic skin lesions in
patients with bacteraemia. Three patients were treated with sulphanilamide with dramatic
fever resolution and sterilization of the joints. Seven of the 140 patients died.
Related reference (1) Keefer CS, Parker F, Myers WK. Histological changes in the knee
joint in various infections. Archives of Pathology 1934; 18:199–215.
Key message
This was a common form of arthritis which needed to be distinguished from other forms of
polyarthritis including rheumatic fever. The demonstration of gonococci in synovial fluid was
an important diagnostic finding, heralding rapid articular cartilage destruction. Treatment
with sulphanilamide appeared dramatically effective.

Since the very first descriptions of septic gonococcal arthritis 50 years earlier, this condition
has been recognized as a common destructive condition. In laying down clear diagnostic cri-
teria, Keefer and Spink helped to establish more clearly one more distinct rheumatic disease.
They still did not differentiate fully between septic gonococcal arthritis and reactive arthritis
and the use of the gonococcal complement fixation test has not stood the test of time.
Nonetheless, through meticulous scientific method and analysis, in this and an earlier paper
(1), they demonstrated synovial changes and the importance of early articular cartilage
destruction.
Most importantly, however, they demonstrated that treatment with the anti-microbial
agent, sulphanilamide, induced rapid sterilization of the joint and defervescence of fever. By
so doing, they enabled both the specialist and the generalist to make an accurate diagnosis of
gonococcal arthritis and ushered in the era of cure of the condition through effective anti-
microbial chemotherapy.
Strengths
1. Clinical descriptions of arthritis are clear with data based on a large sample.
2. The effects of treatment with sulphanilamide are demonstrated, albeit on a very small
number of patients, with dramatic clarity.
Weaknesses
1. No differentiation is made between septic gonococcal arthritis and reactive arthritis.
2. The value of the gonococcal complement fixation test is over-estimated.
3. Anti-microbial treatment was only used in three patients.
Relevance
Scientific analysis of both the clinical and immunological features of gonococcal arthritis
allowed ready diagnosis for both the generalist and the specialist. Anti-microbial chemother-
apy was shown to be rapidly curative.
242
Paper 3
Suppurative arthritis complicating

Authors
Kellgren, JH, Ball J, Fairbrother RW, Barnes KL
Reference
British Medical Journal 1958; 1:1193-1200
Summary
Thirteen patients (one was added as a post-script) were described in which severe bacterial
infection complicated pre-existing RA. Most patients had severe disease of up to 20 years’
duration with much joint destruction. In 10 of the patients, septic arthritis occurred though
in the remainder abscess formation at other sites, including osteomyelitis, was described. A
pre-disposing source of infection was identified in a few patients but not in most. Six patients
died, sepsis being discovered in most only at necropsy. Radiological investigation was in the
main unhelpful and some patients did not have classical features of infection, particularly
fever and leucocytosis. Diagnosis was complicated by the pre-existing features of active RA. In
earlier patients in the series, the diagnosis of sepsis was unsuspected and joint aspiration and
blood cultures were not carried out. Several of the later patients underwent aspiration of sus-
picious joints with early diagnosis and successful treatment.
Key message
Patients with severe rheumatoid joint destruction are especially susceptible to bacterial sepsis
though the clinical features of infection may be atypical.

Although several earlier authors had described patients with RA complicated by septic arthri-
tis, Kellgren, et al. demonstrated that 50% of cases of joint sepsis seen at their hospitals over
the period of study were in patients with RA. Although an uncommon complication, this
paper raised awareness of this lethal but potentially treatable complication. Having raised the
index of suspicion, this paper also pointed out for the first time that the diagnosis is especial-
ly difficult for two reasons. Firstly, the features of active RA closely mimic those of infection.
Secondly, several patients even with widespread sepsis failed to respond with either fever or
leucocytosis. The paper indicated for the first time that the absence of classical indicators of
sepsis does not exclude the possibility of joint sepsis.
The authors also pointed out the importance of aspirating potentially infected joints.
Their statement ‘in the light of our experience with these patients, aspiration and culture of
fluid from the more painful joints are indicated in every patient with RA in whom there is a
sudden deterioration ....’ has become one of the central tenets of good clinical rheumatolog-
ical practice. Moreover, they were able to show that diligent search for infection and rigorous
energetic treatment, even with the limited antibiotic repertoire of the day, was able to pro-
duce remarkable recoveries. Four of the patients had received either steroid or adrenocorti-
cotropic hormone (ACTH). It is not clear whether this predisposed the patients to sepsis
though Kellgren and others cite evidence that this is likely with the implication that this com-
plication should be particularly sought in patients receiving steroid therapy.
Strengths
1. The key points are clearly and powerfully made.
2. The index of suspicion of septic arthritis in this group of patients was clearly raised.
3. The value of joint aspiration in cases where any doubt occurred was emphasized
unequivocally.
Weaknesses
1. This is a small series of patients.
2. The death of earlier patients and the survival of later patients could have occurred by
chance.
Relevance
The detection of joint sepsis in patients with pre-existing rheumatoid disease remains diffi-
cult though a high index of suspicion and routine joint aspiration to search for sepsis has
become normal good clinical practice.
244
Paper 4
Post-operative infection after total hip replacement

with special reference to air contamination in the
operating room
Author
Charnley J
Reference
Clinical Orthopaedics and Related Research 1972; 87:167–187
Summary
Reviewing deep infections in 5800 total hip replacements performed between 1960 and
1970, the incidence of infection fell from 7% to 0.5%. This was attributed to stringent mea-
sures to prevent exogenous infection in the operating theatre. The performance of surgery
within a field of ultra-clean filtered air was the major factor in reducing infection rates. The
use of impermeable surgeons’ gowns to prevent transmission of bacteria from those within
the operating enclosure and improved wound closure techniques, also contributed to the
reduced infection rate.
Key message
Most deep post-operative prosthetic hip infections are acquired in the operating theatre. The
risk can be minimized by exposing the patient to ultra-clean filtered air and by preventing
shedding of bacteria from surgeons and nurses in the operating theatre by use of imperme-
able theatre clothing.

When joint replacement surgery was in its infancy, it became clear that the risk of prosthetic
joint infection was substantial with significant morbidity (Figure 16.2) and mortality. Several
factors were incriminated including use of acrylic cement and subsequent blood-borne infec-
tion from superficial sites of sepsis.
By a series of painstaking studies using operating enclosures within conventional operat-
ing theatres, air filtration systems capable of changing the air 130–300 times and bacteriolog-
ical assessment using settle plates, the exposure of the wound to bacteria was carefully
assessed. Over a decade, Charnley was able to show that exposing the wound to carefully
directed ultra-clean air could reduce deep infection rates from around 7% at the outset to
1.4%. Further improvements in the infection rate could then be made by the use of imper-
meable theatre clothing and newer wound closure techniques.
Charnley did note that patients with

RA were constitutionally subject to slightly
higher infection risk than others, and he
concluded that it is unlikely that clean air
alone can reduce the infection rate below
1.5%. Subsequently, the introduction of
antibiotic-impregnated cement and pro-
phylactic antibiotic therapy also made its
contribution, but in these days of multire-
sistant micro-organisms, the basic tenet of
protecting the patient from per-operative
infection remains of crucial importance.
Figure 16.2 An infected prosthesis of the

knee joint resulting in ulceration through the
skin. (From Slide Atlas of Rheumatology
(Dieppe P, Bamji A, Watt I. eds.). London:
Gower Medical Publishing, 1984.)
Strengths
1. Data are drawn from very large numbers of patients operated on at a single centre.
2. Clear end-points are used in terms of measured bacterial colony counts in filtered and
unfiltered air and infection rates.
Weakness
The use of operating enclosures was not widely taken up elsewhere due to practical
considerations.
Relevance
Systematic and painstaking methods to prevent exposure to infection at the time of surgery is
of paramount importance particularly in implant surgery. The clear demonstration of this by
an acknowledged pioneer of implant surgery has influenced the practice of joint replace-
ment surgery world-wide.
246
Paper 5
Allergic complications of meningococcal disease.

II. Immunological investigations
Authors
Greenwood BM, Whittle HC, Bryceson ADM
Reference
Summary
Four patients (of 47 described in a related paper (1)) with arthritis-complicating meningo-
coccal infection were investigated immunologically. Complement-depletion in serum and
deposits of meningococcal antigen, together with immunoglobulin and C3, were demonstrat-
ed in synovial fluid white cells of the two patients investigated and in one skin biopsy. These
findings indicate that arthritis and skin lesions associated with a meningococcal infection are
due to immune complex formation rather to actual sepsis.
Related references (1) Whittle HC, Abdullahi MT, Fakunle FA, Greenwood BM, Bryceson
ADM, Parry EHO, Turk JL. Allergic complications of meningo-
coccal disease. I. Clinical aspects. British Medical Journal 1973;
2:733–737.
(2) Henrick WW, Parkhurst GM Meningococcus arthritis. American
Journal of the Medical Sciences 1919; 158:473.
(3) Rolleston H. Lumleian lecture on cerebrospinal fever. The Lancet
1919; 1:645.
Key message
This is the first clear description of arthritis caused by microbial infection but mediated by an
immune complex disease.

Arthritis as a complication of meningococcal infection had been well described more than
50 years before this report (2, 3) and allusion had been made to the possibility that arthritis
occurring late during convalescence from the acute infection might be due to serum sickness
or a related process. These authors were able to demonstrate, via conceptually simple and
clear means, that arthritis in this circumstance is likely to be due to immune complex forma-
tion rather than direct tissue invasion by viable micro-organisms.
By this means the authors were able to demonstrate that acute infection may lead to syn-
ovitis by mechanisms other than persistence of viable micro-organisms. Thus, a half-way
house was built between overtly septic disease and disorders such as RA in which no direct
evidence of an infective cause is apparent.
Strengths
1. Clear and thorough methodology.
2. This work occurs in the context of other clinical and immunological investigations of a
large population of patients with meningococcal disease.
Weakness
Actual immunological data are available for a very small number of patients.
Relevance
The demonstration of an immune mechanism for arthritis in this group of patients has stim-
ulated searches for other micro-organism-driven mechanisms in other forms of inflammatory
arthritis.
248
Paper 6
Human parvovirus-associated arthritis: a clinical

and laboratory description
Authors
Reid DM, Brown T, Reid TMS, Rennie JAM, Eastmond CJ
Reference
The Lancet 1985; 1:422–425
Summary
During an outbreak of human parvovirus (HPV) infection in Scotland, 42 patients were iden-
tified with both serological evidence of infection and joint pain. Details were available on 30
patients (27 adults and three children). Each of the three children had a typical rash but
seven of the 27 adults had no prodromata nor rash. The intervals between the onset of the
rash and the onset of joint pain ranged from 1–16 days. Affected adults were predominantly
young women (mean age 30.3 years).
Arthritis consisted mainly of symmetrical small joint involvement of the proximal inter-
phalangeal and metacarpophalangeal joints, though less commonly knees, wrists and ankles
were involved. All patients had immunoglobulin (Ig) M and IgG human parvovirus antibod-
ies at presentation. IgM antibodies reached a peak at 8 days and declined to near zero at 60
days. IgG antibodies reached a peak at 40 days and declined more slowly thereafter. In 17 of
24 patients, arthritis resolved within 2 weeks. All but two resolved within 4 weeks though
arthralgia persisted in two patients.
Related reference (1) White DG, Mortimer PP, Blake DR, Woolf AD, Cohen BJ, Bacon
PA. Human parvovirus arthropathy. The Lancet 1985; 1:419–421.
Key message
This paper, published simultaneously with (1) clearly describes a clinical arthritis syndrome
associated with a common viral infection. Children develop a trivial exanthem but young
adults, usually women, develop transient polyarthritis, often without features of the exan-
them.

Since the discovery of HPV in 1975, several groups had noted the presence of joint pain
amongst a range of other more prominent disease features. For the first time, Reid and col-
leagues, along with White, et al., clearly delineated one of the most common forms of inflam-
matory arthritis in the western world. They provided a clear picture of its natural history and
likely epidemiology. White and colleagues also made the same observations but under rather
different circumstances. By identifying HPV-infected patients attending an early synovitis
clinic, they described a more selected population, incidentally demonstrating the value of
the relatively new innovation, the ‘early synovitis clinic’. Reid’s description of a short-lived,
spontaneously-resolving disease has been borne out by subsequent observation. White, how-
ever, described a more persistent illness in several patients, symptoms persisting for more
than 4 years in three of the 19 patients described. In spite of several tantalizing subsequent
reports, a clear link between HPV and RA has not been revealed although the possibility that
some cases may persist and lead to joint damage remains unresolved.
Strengths
1. Simple classic description of a common disease.
2. Diagnostic serology is clearly described for clinical usage.
3. This description provides a clear basis for diagnosis and treatment by both specialists and
generalists.
Weaknesses
1. Epidemiology of this condition is hinted at rather than assessed.
2. Small numbers of patients were documented, not all of these after direct clinical examina-
tion by the authors.
Relevance
This and other subsequent papers allowed a common disorder to be diagnosed readily, espe-
cially by primary care physicians. Rapid diagnosis thus prevents unnecessary suffering by the
patients and inappropriate use of resources.
250
Paper 7
Lyme arthritis: An epidemic of oligoarticular

arthritis in children and adults in three
Connecticut communities
Authors
Steere AC, Malawista SE, Snydman DR, Shope RE, Andiman WA,
Ross MR, Steele FM
Reference
Summary
An epidemic form of arthritis has been occurring in eastern Connecticut at least since 1972,
with the peak incidence of new cases in the summer and early autumn. Its identification has
been possible because of tight geographical clustering in some areas, and because of a char-
acteristic preceding skin lesion in some patients. The authors studied 51 residents of three
contiguous Connecticut communities (39 children and 12 adults) who developed an illness
characterized by recurrent attacks of asymmetric swelling and pain in a few large joints, espe-
cially in the knee. Attacks were usually short (median 1 week) with much longer intervening
periods of complete remission (median 2.5 months), but some attacks lasted for months. To
date the typical patient has had three recurrences, but 16 patients have had none. A median
of 4 weeks (range 1–24) before the onset of arthritis, 13 patients (25%) noted an erythema-
tous papule that developed into an expanding, red annular lesion, as much as 50 cm
(19.5 in) in diameter. Only two of 159 family members of patients had such a lesion and did
not develop arthritis (p<0.000001). The overall prevalence of the arthritis was 4.3 cases per
1000 residents, but the prevalence among children living in four roads was 1 in 10. Six fami-
lies had more than one affected member. Nine of 20 symptomatic patients had low serum C3
levels, compared to none of 31 asymptomatic patients (p<0.005); no patient had iridocyclitis
or a positive test for antinuclear antibodies (ANA). Neither cultures of synovium and synovial
fluid nor serological tests were positive for agents known to cause arthritis. ‘Lyme arthritis’ is
thought to be a previously unrecognized clinical entity, the epidemiology of which suggests
transmission by an arthropod vector.
Related references (1) Putkonen T, Muslakallo KK, Salminen P. Erythema chronicum

migrans with meningitis: a rare coincidence of two tick-born
diseases. Dermatologica 1962; 125: 184–188.
(2) Steere AC, Malawista SE, Hardin JA, Ruddy S, Askenease PW,
Andiman WA. Erythema chronicum migrans and Lyme arthri-
tis. The enlarging clinical spectrum. Annals of Internal Medicine
1977; 86:685–698.
(3) Steere AC, Malawista SE. Cases of Lyme disease in the United
States; locations correlated with distribution of Ixodes dammini.
Annals of Internal Medicine 1979; 91:730–733.
(4) Steere AC, Malawista SE, Newman JH, Spieler PN, Bartenhagen
NH. Antibiotic therapy in Lyme disease. Annals of Internal
Medicine 1980; 93:1–8.
Key message
A new and distinctive form of arthritis has been discovered in Connecticut; it is probably
caused by an arthropod-born infection.

The opening paragraph of this paper is a true ‘classic’: ‘In November 1975 a mother from
Old Lyme, Connecticut, informed the State Health Department that 12 children from that
small community of 5000, four of whom lived close together on the same road, had a disease
diagnosed as JRA’. The authors thereby paid tribute to the observations and actions of the
community right at the beginning of this first paper on Lyme disease. The Yale group, led by
Alan Steere and Steven Malawista went on to investigate the phenomenon fully, leading to a
comprehensive description of the disease, the vector, the causative organism and the treat-
ment (see related references).
The importance of the discovery of Lyme disease is hard to overestimate. Not only is this
an important disease, which can be treated effectively, but it has also given us great insights
into infections and arthritis. This first paper provided many of these insights and is a model
of combined clinical and epidemiological work to sort out a clinical syndrome. In this paper,
the authors fully describe their efforts to make sure that the condition is distinctive, and that
it is not a previously recognized form of juvenile arthritis or infectious arthritis.
Strengths
1. Superb clinical and epidemiological investigation of a phenomenon that lay people in the
community discovered.
2. Clear evidence that Lyme arthritis is a new, distinctive disease.
3. Pre-empts the subsequent association with Ixodes dammini, the spirochaete and antibiotic
treatment.
Weakness
It would be churlish to criticize this excellent first paper on Lyme arthritis, although it
should perhaps be noted that the authors did, at this early stage, call it Lyme ‘arthritis’ rather
than Lyme ‘disease’.
Relevance
The original, ground-breaking description of an important, treatable disease and of an
important model for the pathogenesis of other forms of arthritis.
252
Paper 8a
Immunologic relation of streptococcal and tissue

antigens. I. Properties of an antigen in certain
strains of Group A streptococci exhibiting an
immunologic cross-reaction with human heart tissue
Author
Kaplan MH
Reference
Paper 8b
A immunological relationship between the Group A

streptococcus and mammalian muscle
Authors
Zabriskie JB, Freimer EH
Reference
Summary
The two reports by Kaplan and later by Zabriskie and Freimer first established that there was
indeed immunological cross-reactivity or molecular mimicry between Group A streptococcal
bacterial antigens and human heart tissues. The evidence for this cross reactivity included
production of rabbit antisera against streptococcal cell wall antigens which reacted in
immunofluorescence with human cardiac muscle, and complete abolition of immunofluores-
cence staining of heart muscle tissues after absorption of rabbit antisera to Group A strepto-
coccal antigens by streptococcal cell wall extracts. Moreover, the cross-reactive properties of
Group A streptococci were found to be most prominent in streptococcal strains most often
associated with acute rheumatic fever. Initially Kaplan’s work seemed to implicate cross-reac-
tivity with human heart muscle and streptococcal M proteins, whereas the report by Zabriskie
and Freimer localized the cross-reactive antigens to streptococcal cell membranes. Later
work by both groups of authors demonstrated that patients with acute rheumatic fever pro-
duced antibodies which reacted with heart muscle antigens. These observations set the stage
for many other workers who expanded this concept as a probable mechanism of disease
pathogenesis.
Related references (1) Kaplan MH, Svec KH. Immunologic relation of streptococcal and
tissue antigens. III. Presence in human sera of streptococcal
antibody cross-reactive with heart tissue. Association with strep-
tococcal infection, rheumatic fever, and glomerulonephritis.
Journal of Experimental Medicine 1964: 119–151.
(2) Van de Rijn I, Zabriskie JR, McCarty M. Group A streptococcal
antigens cross-reactive with myocardium. Purification of heart-
reactive antibody and isolation and characterization of the
streptococcal antigen. Journal of Experimental Medicine 1977;
146:579–599.
(3) Dale JR, Beachey EH. Epitopes of streptococcal M Proteins shared
with cardiac myosin. Journal of Experimental Medicine 1985;
162:583–591.
Key message
The observations by Kaplan and shortly thereafter by Zabriskie and Freimer were the first to
establish a clear antigenic cross-reactivity between common bacterial antigenic moieties and
human tissues–particularly heart muscle and membranes of heart muscle cells. These find-
ings seemed of particular importance since it was already known that Group A streptococcal
pharyngeal infection always preceded the clinical onset of acute rheumatic fever by
2–3 weeks, and that marked elevations of serum antibodies to Group A streptococcal anti-
gens were a reliable sign of rheumatic fever. The clear demonstration that Group A strepto-
coccal throat infection might therefore induce a cross-reacting natural immune response to
self or autologous heart muscle tissue was the first example of how such molecular mimicry
might be implicated in an acute and subsequent chronic debilitating human disease, namely,
rheumatic fever and rheumatic heart disease. This concept of molecular mimicry was subse-
quently expanded by many other investigators to attempt to explain other manifestations of
rheumatic fever as well as other entirely different disease states.

These initial observations by two independent groups were the first clear-cut scientific
demonstrations that an organism which caused serious infections in humans, namely the
Group A beta-haemolytic streptococcus, could also induce serious illness and consequent
lifelong disability and impairment by another mechanism not previously recognized or
described. This mechanism was molecular mimicry, causing cross-reactivity between antigens
present in the infecting organism and vulnerable human tissues such as human heart muscle
cells, heart valves, endocardium lining heart chambers and heart valves, and virtually all
heart structures including pericardium, myocardium and cardiac conduction system.
Subsequent investigations by these two groups as well as a large number of other investiga-
tors demonstrated that initial exposure to Group A streptococcal antigens during the early
acute pharyngeal infection could initiate a broad anti-streptococcal immune response which
could then damage autologous heart muscle, heart valve structures, and endocardium, as
well as probably being responsible for the acute arthritis and even the strange central ner-
vous system disturbance of Sydenham’s chorea and athetoid movement disoders, known to
be sometimes the only clinical manifestation of acute rheumatic fever. These observations
completely changed some of the mechanics of thinking about how diseases happen and what
actually activates the vast linkage of physiological messages and communications within the
living human organism that could initiate and then perpetuate a self-damaging disease
process such as chronic rheumatic heart disease. This disease, with its slow evolution of heart
254
valve inflammation, scarring and deformity or its progressive weakening of left ventricular
muscle contraction and disability which we now refer to as cardiomyopathy, starts with an ini-
tial Group A streptococcal infection in the throat of the subject who then mounts an
immune response to a wide variety of Group A streptococcal antigens which then cross-reacts
with autologous tissues and induces first an acute and later a chronic inflammatory response.
Strengths
1. Straightforward scientific evidence presented.
2. First to suggest that bacterial or other external antigens in our environment could start a
pathological injurious process in the human host which would by molecular mimicry initi-
ate an acute and then a chronic inflammatory process.
Weakness
Most of the early observations by both groups were directly supported by demonstrations of
various cross-reacting humoral antibodies,with little mention of cell-mediated or T cell
involved immune response. This is understandable since in the early 1960’s not much was
understood about cell-mediated or T cell induced immune responsiveness. It was only much
later that attention to cell-mediated immunity was directed to acute rheumatic fever and con-
sequent tissue damage in rheumatic fever or chronic rheumatic heart disease.
Relevance
The whole concept of molecular mimicry and its basic influence as an important mechanism
of human disease has had its ups and downs in being accepted by the medical and scientific
community at large. Perhaps the strongest set of facts that such a mechanism is really of
major importance is the whole scenario of acute rheumatic fever and the Group A strepto-
coccus.
Figure 16.3 Photomicrograph

of chronic inflammatory cell
infiltrate within a mitral valve
of a patient with chronic
rheumatic heart disease. Most
of the inflammatory cells are
small lymphocytes and plasma
cells indicating principal
involvement of the immune
system in this inflammatory
process. Haematoxylin and
eosin (x380).
256
CHAPTER 17
Osteoarthritis
Paul Dieppe
Introduction
The story of our developing understanding of osteoarthritis (OA) is far from complete. The
papers selected in this section take us through some of the major landmarks of the 20th cen-
tury, at which point we are still left with much to understand, and a somewhat enigmatic dis-
order to grapple with.
The different forms of arthritis could only be distinguished when pathology and radiology
could be combined with clinical pattern recognition. This process began in the early part of
the 20th century with the differentiation of hypertrophic and atrophic forms of arthritis by
Goldthwaite (Related reference 1, Paper 1), followed by the superb contribution of Nichols
and Richardson (Paper 1), who provided us with the first clear descriptions of OA, but also
saddled us with concepts of cartilage degeneration as the basic cause. The 19th century had
seen superb clinical descriptions of OA of the finger joints by Heberden and Haygarth. In
the 1930s and 40s this form of OA was studied extensively by Stecher, who showed that it had
two forms, one idiopathic (and genetic), the other post-traumatic (Paper 2). Until the 1950s,
Heberden’s nodes had been considered as a separate entity, but this was changed by
Kellgren and Moore, who showed the association with large joint OA, and provided us with
the definitive description of generalized OA (Paper 3). One of the main and continuing
problems in studying OA is the difficulty that we have with definitions, grading and measure-
ment. Jonas Kellgren and John Lawrence, whose contributions to the field were immense,
provided us with an x-ray grading system in the 1950s, which remains the gold standard in
the 21st century. The application of this system to population studies was pioneered by
Lawrence, and led to the seminal observation of the discordance between x-ray changes and
symptoms, which changed thinking about this disease for all time (Paper 4). The application
of radiographs as an epidemiological tool for the study of OA also allowed this group to
make the first associations between occupation and OA (1).
There was only been one major development in the treatment of OA throughout the 20th
century, namely the introduction of effective, long lasting prostheses for the hip and knee
joints. John Charnley is rightly credited with leading the way here, and his crucial papers of
1960 and of 1961 (Paper 5) would rate as all time classic medical contributions of the last
century.
The final three chosen contributions concern more recent attempts to try to understand
the aetiopathogenesis of this condition. Only here is the choice difficult and potentially con-
tentious. I could have chosen the work of Henry Mankin on cartilage (2), the work of Eric
Radin on biomechanics (3), Paul Byers’ contributions to progression (4), or the much more
recent innovations in genetics (5). However, it seems to me that biochemistry, pathology and
imaging, when combined with clinical observations, remain the key techniques able to help
us understand clinical aspects of OA. I have therefore chosen one contribution from the
pathologist Peter Bullough who in my view has made the most important contribution to the
field (Paper 6), and one paper from Helen Muir and colleagues (Paper 7), in recognition of
the huge contribution that Dr Muir made to the investigation of cartilage biochemistry as a
Osteoarthritis 257
window on the pathogenesis of OA, and in the understanding that animal models are an
important tool with which to investigate the disease. Finally, I have chosen one paper from
my own group, which, with Iain Watt’s help, has pioneered the application of new imaging
techniques (scintigraphy and MRI) in OA (Paper 8). A further justification for this choice is
that its message seems to fit well with the ideas put forward by the four earlier quoted classic
contributions, and provides us with a neat circle–from correlations of pathology and radiolo-
gy with clinical findings in 65 patient studied cross-sectionally in 1909 (Paper 1) to correla-
tions of imaging with clinical findings in 94 patients studied prospectively in the 1990s
(Paper 8).
References
1. Kellgren JH, Lawrence JS. Rheumatism in Miners. British Journal of Industrial Medicine
1952; 9:197–207.
2. Mankin HJ. The reaction of articular cartilage to injury and osteoarthritis. New England
3. Radin EL, Paul IL. Response of joints to impact loading. Arthritis and Rheumatism 1971;
14:356–362.
4. Byers P, Contepomi C, Farkas T. A post mortem study of the hip joint. Annals of Rheumatic
Diseases 1970; 29;15–31. Prevalence of cartilage lesions in foot joints: a test of the concept
of limited and progressive lesions. Annals of Rheumatic Diseases, 1970; 29:15–31.
5. Alo-Kokko L, Baldwin CT, Moskowitz RW, et al. Single base mutation in the type II procol-
lagen gene (COL2A1) as a cause of primary osteoarthritis associated with a mild chon-
drodysplasia. Proceedings of the National Academy of Sciences USA 1990;87:6565–6568.
258
Paper 1
‘Arthritis deformans’
Authors
Nichols EH, Richardson FL
Reference
Journal of Medical Research 1909; 21:149–221
Summary
In non-tubercular deforming arthritis there are two pathological types of joint change:
(a) the proliferative type, which tends to destroy joint cartilage and lead to ankylosis of the
adjacent joint surfaces, and (b) the degenerative type, which tends to destroy the joint carti-
lage, and produce deformity without ankylosis. These two types do not correspond to two
definite diseases, but each type represents reaction of the joint tissues to a considerable vari-
ety of causes. In neither type if the original injury is sufficiently severe, or if the causative fac-
tor continues to act, is there likelihood of the regeneration of a perfect joint.The nomencla-
ture used in this article is suggested because it describes the pathological process, without
any assumption that the aetiology is known in any given case.
Related references (1) Goldthwaite JE. The treatment of disabled joints resulting from
the so-called rheumatoid diseases. Boston Med Surg J 1897;
136:79–84.
(2) Collins DH. Osteoarthritis. Journal of Bone and Joint Surgery 1953;
358:518–520.
Key message
That there are two main types of arthritis, each of which represents the reaction of a synovial
joint to certain types of injury. The proliferative type is characterized by synovial inflamma-
tion, the degenerative type by focal areas of loss of articular cartilage.

This article represents the definitive classification of arthritis into the two main types still rec-
ognized today, inflammatory and degenerative. This was made possible by a combination of
clinical, radiological and pathological expertise; one of the remarkable things about this arti-
cle, is that its two authors were able to provide clear descriptions of the findings of what
would be regarded as three quite different disciplines today: clinical medicine, radiology and
pathology. The article is also important for its thesis that each of these two forms of arthritis
represent the reaction of joint tissues to a variety of different types of injury, rather than
being disease entities.
Subsequently the inflammatory form of arthritis has been split into several diseases, and in
some cases the cause is known (gout and septic arthritis for example). In the case of the
degenerative type (i.e. OA) it remains appropriate and topical to regard it as the response of
a joint to injury. The article is particularly important in the history of OA as it describes and
illustrates several patients with different types of this condition, including a woman with
Osteoarthritis 259
Heberden’s nodes, and a man with hip disease (‘malum coxae senelis’). This pre-empts the
subsequent descriptions of different expressions of OA, featured in other papers in this
chapter. In addition, Paper 8, in one sense, brings us full circle to this paper.
Strengths
1. Beautiful, full descriptions of clinical, radiographic and pathological findings.
2. Classical deductive reasoning based on sound observation.
3. Extensively illustrated with line drawings, clinical photographs, photomicrographs and
radiographs.
Weaknesses
1. Based on 65 selected cases from one Boston Hospital, only 26 of which came to autopsy.
2. Anchored in the medical science of 1909, which includes several concepts that have been
discredited subsequently.
3. Introduction of the term degenerative arthritis, which is one that may have held back
research into this form of joint disease.
Relevance
The first clear, comprehensive description of OA, the introduction of the concept of two
forms of arthritis, each of which was a reaction of the joints to different forms of insult, and
the introduction of the nomenclature ‘degenerative arthritis’.
260
Paper 2
Heberden’s nodes: heredity in hypertrophic

arthritis of the finger joints
Author
Stecher RM
Reference
American Journal of Medical Science 1941: 201:801–809
Summary
Heberden’s nodes or hypertrophic arthritis of the finger joints may be regarded as a distinct
clinical entity. Two types are recognized and distinguishable, one arising as a result of trau-
ma, the other arising spontaneously or idiopathically.
The occurrence of this disease has been shown to be profoundly influenced by race, sex
and age differences. A previous study has revealed accurately the incidences of these various
classifications. A study of 68 families of patients with idiopathic Heberden’s nodes showed
that the mothers of such subjects are affected twice as frequently and the sisters three times
as frequently as the population in general. These figures, if in error at all, are more likely to
be low rather than high. A control series of 43 families selected and studied in the same way
showed the sisters of unaffected women to have Heberden’s nodes as frequently as the popu-
lation in general.
The occurrence of multiple cases in the same family cannot be accounted for on the basis of
chance alone.
Heredity factors seem best to explain the recorded observations, although the exact mech-
anism of transmission has not been determined.The conclusions concerning Heberden’s
nodes are not directly applicable to other forms of hypertrophic arthritis at the present time.
Related references (1) Heberden W. Commentaries on the history and causes of disease.
London: Payne, 1802.
(2) Haygarth J. A clinical history of diseases. II. Nodosity of the joints.
London: Gadell and Davies, 1805.
Key message
There are two forms of finger joint OA: one caused by trauma, and the other inherited
(Table 17.1).
Table 17.1 Expected-versus-reported prevalence of Heberden’s nodes in the sisters and mothers of
68 cases and 43 controls.
Cases (68) Controls (43)

Mothers (67) Sisters (129) Mothers (43) Sisters (109)
Expected 11 11.4 7 6.3

Observed 21 33 0 5
Osteoarthritis 261

The early descriptions of OA, such as that of Nichols and Richardson (Paper 1), called the
condition a degenerative one, and concentrated on trauma and age as the main causes.
Robert Stecher spent a long time studying OA of the finger joints, and studied all the previ-
ous literature. He wrote extensively on the subject, and was rewarded with recognition,
awards and invitations to speak at prestigious meetings, including (fittingly) giving the
‘Heberden Oration’. His particular contributions include the recognition that there are at
least two separate forms of finger joint OA: a post-traumatic type and an idiopathic form, and
the outcome of this particular paper–the suggestion that the idiopathic form is largely
caused by inherited traits. Stecher also uses the nomenclature hypertrophic arthritis, rather
than degenerative arthritis, and did much to rekindle interest in OA, and this and his related
articles had a high impact at the time and subsequently. Stecher’s conclusion that the mecha-
nism of transmission cannot be determined, and that conclusions made from observations of
finger joint OA should not be applied to other forms of the disease were very perceptive and
remain true today.
Strengths
1. Well illustrated descriptions and illustrations of family trees of three large families with
Heberden’s nodes.
2. Excellent clinical epidemiology for the time, with a study of 68 index case families and 43
controls, resulting in the clean data, and well calculated ratios of expected-versus-observed
rates of Heberden’s nodes (see Table 17.1).
Weaknesses
1. His expected rates were based on observations made previously from a different popula-
tion base, and therefore represent historical controls.
2. The presence or absence of Heberden’s nodes in relatives is based on reported history,
not examination of the subjects, and may therefore, be erroneous. Stecher recognizes this
problem and claims that it is more likely to lead to negative rather than positive bias to his
data. He may have been right.
Relevance
Recognition that genetic factors are important in OA, and the differentiation of idiopathic
(genetic) finger joint OA from post-traumatic OA.
262
Paper 3
Generalized osteoarthritis and Heberden’s nodes

Authors
Kellgren JH, Moore R
Reference
Summary
From a study of 391 cases of OA attending a rheumatic clinic we have been able to define a
distinct clinical entity for which we suggest the name of primary generalized OA. The condi-
tion occurs most often in middle-aged women, and is characterized by a distinct pattern of
joint involvement, by a course in which each affected joint passes through an initial painful
phase and a more or less acute arthritic phase, and by other distinctive clinical and radiologi-
cal features.
Generalized OA is a constitutional disorder affecting the diarthrodial joints and probably
represents a severe form of idiopathic Heberden’s nodes. A therapeutic trial of oestrogen
therapy gave negative results, but patients suffering from this disease may be helped consid-
erably by the proper management of their complaints.
Related references (1) Cecil RL, Archer BH. Classification and treatment of chronic
arthritis. Journal of the American Medical Association 1926;
87:741–746.
(2) Ehrlich GE. Inflammatory osteoarthritis. Journal of Chronic Diseases
1972; 25:317–328.
Key message
There is a special form of OA (mainly affecting women) that affects several different joint
sites in the same individual.

This is one of the most widely quoted papers on OA, being regarded as the original and
definitive clinical description of the now generally recognized entity of generalized OA.
Kellgren and Moore begin their article by pointing out that hitherto OA of the large joints
had been believed to be an age-related degenerative disorder (the legacy of the work of
Nichols and Richardson and others). However, they also recognized that the disease called
Heberden’s nodes (as described by Stecher and others) also seemed to have something to do
with OA. So they posed the question ‘are Heberden’s nodes associated with OA of the large
joints?’.
Osteoarthritis 263
In order to address the question they examined their records from 1,813 cases of all types
of arthritis seen in their Department of Rheumatology in Manchester over the last 3 years,
and found 120 cases of a generalized form of OA, 110 of which were women. They provide us
with a full clinical description of the group, including the joints affected; 103 had
Heberden’s nodes (most of whom also had thumb base disease), the knees were involved in
64, spine in 57 and hips in 36. Three individual illustrative cases are described in detail, with
radiographs and clinical photographs included (Figure 17.1). They describe, for the first
time, the tendency of the joints to go through different stages in the evolution of the disease.
Because of the high prevalence of this condition in middle-aged women they speculated that
it had something to do with the menopause, and took the next, bold step of trying a placebo-
controlled trial of oestrogen therapy, which they describe as ineffective. This was one of the
first clinical trials of drug therapy for OA.
Figure 17.1 The hands of a

patient with nodal generalized
osteoarthritis, of the sort
described in this paper.
Strengths
1. The testing of two hypotheses through clinical research; that Heberden’s nodes might be
associated with large joint OA, and that oestrogen deficiency might be a factor.
2. Excellent descriptions of the condition, that have never been bettered.
3. The first time that anyone had suggested that there are different stages and phases of
activity in the evolution of OA.
4. One of the first clinical trials of drug therapy in OA.
Weaknesses
1. The case selection is not described and is a potential source of bias.
2. No statistical analysis of the expected-versus-observed frequency of the association, or of
the clinical trial results.
3. The clinical trial was not randomized and only involved 10 patients in each group, so did
not have the power to show whether oestrogens were effective.
Relevance
The definitive description of generalized OA.
264
Paper 4
Osteo-arthrosis: prevalence in the population and

relationships between symptoms and x-ray changes
Authors
Lawrence JS, Bremner JM, Bier F
Reference
Summary
A combined urban-rural population in the North of England was questioned about musculo-
skeletal pain and loss of work, and was submitted to a clinical examination of the joints and to a
routine series of joint x-rays. A sample of blood was tested for rheumatoid factor by the sheep-
cell agglutination test. The x-rays were read without knowledge of the history or clinical state.
After standardization of the adult population for age and sex by the unweighted mean,
the prevalence of osteo-arthrosis was found to be 52% in males and 51% in females. Five or
more joint groups were affected in 9% of males and 12% of females, the proportion increas-
ing rapidly with age to a maximum of 37% in males and 49% in females aged 65 years and
over. When minimal disease was excluded, the prevalence of osteo-arthrosis was 19% in
males and 22% in females, and five or more joint groups were affected in 0.5% of males and
1.8% of females. Standardization to the 1961 population of England and Wales gave slightly
lower prevalences in males.
On the basis of the clinical, radiological and serological findings, the population was
divided into a rheumatoid and a non-rheumatoid group. In the non-rheumatoid group there
was a significant correlation between x-ray changes of osteo-arthrosis and the frequency of
symptoms in the corresponding distribution.This was present in all joints except those of the
lumbar spine and was only marginal in the cervical spine. In the rheumatoid groups there
was a higher rate of symptoms, but the increase of complaints with the severity of osteo-
arthrosis was no greater than that in the non-rheumatoid group. Morning stiffness was associ-
ated with an increase in symptoms in the hands and knees in the non-rheumatoid group
regardless of whether osteo-arthrosis was present or not.
Obesity was associated with an increase of pain in the knees in persons with osteo-arthrosis
but had little influence on those without. Persons with osteo-arthrosis who had been classi-
fied as neurotic did not have a higher complaint compared with those not so classified, nor
did those with a history of injury. Persons with osteo-arthrosis living in damp houses did not
have significantly more than the expected proportion of symptoms, but the numbers in very
damp houses were small. Symptoms were only slightly more when bone cysts were present in
the metacarpophalangeal joints, but showed no relationship to symptoms in the hip joints.
Related reference (1) Kellgren J, Lawrence JS. Radiological assessment of osteoarthritis.

Key message
Radiographic changes of OA are very common in the community, and predispose to joint
pain. However, severe radiographic OA is often asymptomatic (Table 17.2).
Osteoarthritis 265
Table 17.2 Data on the association between x-ray grades and symptoms in people without
rheumatoid arthritis, taken from the data in this paper (% of people with symptoms).
Joint Sex Radiographic grading

0–1 2 3–4
DIP M 1.4 8.1 9.4
F 1.3 6.2 25.0
Hip M 5.8 0 56.3
F 9.6 0 100
Knee M 22.9 41.7 55.5
F 20.1 43.3 79.5

This paper reports the prevalence of symptoms and x-ray changes of OA of the spine, hips, knees,
hands and feet in a large community-based cohort of people in the North of England (1098 men
and 1198 women). It illustrates that radiographic changes of OA are extremely common. It also
showed up the discrepancy between x-ray changes, which are graded on the Kellgren and
Lawrence 0–4 system, and the presence of symptoms, such that many of those with the most
severe changes of OA on x-ray are asymptomatic. It is clear that in the spine radiographic changes
are practically meaningless and that the discrepancy between pain and x-ray changes in peripher-
al joints is more obvious in the hand, and least marked in the first metatarsophalangeal joint.
Although the authors are most concerned with prevalence rates, the seminal observation
made to our understanding of OA was this discordance between pathology and clinical fea-
tures. They show clearly that the radiographic surrogate for joint pathology cannot be equat-
ed with the clinical syndrome. The observation, which has since been confirmed by many
other surveys, has been seen as a problem by many, undermining their concept of a disease.
It certainly makes us have to think differently about the nature of OA, but it provided the
freedom to ask questions about the causes of pain, which still remain unanswered today.
Many of their other observations about associations and features of radiographic OA are
unparalleled, and this is a huge paper with a mass of good data.
Strengths
1. Comprehensive clinical and radiographic data on large numbers from a community.
2. A huge amount of data, carefully analysed and reported.
3. Definitive data on the prevalence of radiographic OA in the population.
Weaknesses
1. No concern for potential errors in x-ray readings.
2. Dependence on the Kellgren and Lawrence grading system, which, while useful, has a
number of inherent problems, such as the assumptions of linearity and the assumption
that bone and cartilage changes occur together.
3. A very complicated paper, muddled in places.
4. No hypothesis testing.
Relevance
The baseline data for prevalence of radiographic OA in the community, and the first and
major contribution to show that x-ray changes, while predisposing to symptoms, cannot be
equated with them or seen as their cause.
266
Paper 5
Arthroplasty of the hip: a new operation

Author
Charnley J
Reference
The Lancet 1961; 1:1129–1132
Summary
There is no summary or abstract to this paper, which reports Charnley’s early work on and
experience with his new design of hip replacement. In the introduction to this key piece of
work he writes:
‘In considering how arthroplasty of the hip can be improved, two facts stand out:
(1) After replacement of the head of the femur by a spherical surface of inert material, the
failures are essentially long-term. At first the patient may notice no difference between
the artificial head and the living one which preceded it. Our problem is to make this
temporary success permanent.
(2) Objectives must be reasonable. Neither surgeons nor engineers will ever make an arti-
ficial hip joint which will last 30 years and at the same time in this period enable the
patient to play football.’
This remains as true in 2001 as it was in 1961.

He then describes the principles and laboratory research which led to the design of his
revolutionary low friction arthroplasty, through the use of a polytetrafluorethylene socket
and a metallic femoral head, and the cementing of the prosthesis into the femur. He notes
that only 97 of his new designed hips have been inserted, and long- term results are awaited,
there had only been one death (due to a post-operative myocardial infarction) and the early
results were good.
Related references (1) Charnley J. Anchorage of the femoral head prosthesis to the shaft
of the femur. Journal of Bone and Joint Surgery 1960; 42(B):28–30.
(2) Charnley J. The long term results of low-friction arthroplasty of the
hip performed as a primary intervention. Journal of Bone and
Joint Surgery 1972; 54(B):61–76.
(3) Wroblewski BM. Fifteen to 21-year results of the Charnley low-fric-
tion arthroplasty. Clinical Orthopaedics and Related Research 1986;
211:30–35.
Key message
A low friction arthroplasty of the hip, with a cemented femoral component, provides a good
prosthesis in bioengineering terms, and early clinical results are encouraging.
Osteoarthritis 267

John Charnely is rightly regarded as the father of modern joint replacement. Of all the papers
quoted in this section, this is the only one which heralded a breakthrough of major importance
to those who suffer from OA. The introduction and subsequent wide usage of Charnley’s hip
prosthesis (and subsequent designs and developments in the hip and knee) has brought relief
to literally millions of people with OA resulting in severe pain and disability.
Prior to Charnley’s work, there had been a number of attempts at replacing the hip with
artificial materials, but they had been largely unsuccessful (Charnley describes the problems
with the Smith Peterson and Judet hips in the first part of this article). The innovation
reported in this paper was the design of a low friction prosthesis through the use of new
materials, and the design, which included a small femoral head, cemented into the femur.
This design has proved as good as any of those subsequently developed, most of which have
used the same principles as Charnely. For example, the related paper of Wrobeleski (3)
demonstrates excellent long-term outcomes.
Strengths
1. Charnley was ahead of his time, his insight into the need for a low friction bearing, and
the use of long-lasting high molecular weight plastics and bone cement revolutionized the
outcome of joint replacement, and changed this from a poor option to the preferred
option for severe joint disease.
2. A wonderful, concise description of the scientific rationale for his innovations.
Weaknesses
1. Relatively small numbers of cases studied over a short period of time, and no convincing
outcome measures reported.
2. This is Charnley saying ‘trust me, I’m a surgeon’, but, he was right.
Relevance
An innovation that changed medicine, surgery and the prospects for those with severe OA
for ever and for the better. One of the great contributions of all time.
268
Paper 6
The geometry of diarthrodial joints, its physiological

maintenance and the possible signficicance of age-
related changes in geometry to load distribution
and the development of osteoarthritis
Author
Bullough PG
Reference
Summary
Diarthrodial joints are governed by physiological mechanisms that maintain stability and an
equitable distribution of load. Modelling continues throughout life to maintain the necessary
physiological incongruity. However, in old age the system seems to go awry, and the result is
an increasing congruity yielding possibly increased stability but interfering with cartilage
nutrition and altering the distribution of load. The increasing maldistribution of load with
age, it is proposed, mechanically overtaxes the previously underloaded and, presumably,
atrophic cartilage. Overtaxing the cartilage in turn leads to further depletion of proteogly-
cans, collagen fraying and eventually OA. Thus arthritis, at least in one of its forms, appear to
be inevitable because of the maldistribution of load that results from age-related changes in
joint shape, possibly dictated by the joint’s requirement for stability.
Related references (1) Bullough PG, Goodfellow J, O’Connor J. The relationship between
degenerative changes and load bearing in the human hip.
Journal of Bone and Joint Surgery 1973; 55(B):746.
(2) Lane LB, Bullough PG. Age-related changes in the thickness of the
calcified zone and the number of tidemarks in adult human
articular cartilage. Journal of Bone and Joint Surgery 1980;
62(B):372.
Key message
Age-related changes in the congruity of joints leads to OA, through alterations in the load-
bearing surface.
Osteoarthritis 269

Fittingly, this paper is reported in an edition of the journal dedicated to Truetta, who was
one of the first observers to stress the importance of vascularity and bone turnover in the
aetiopathogenesis of OA. This essay summarizes much of the very important work done by
Peter Bullough and his colleagues in the 1970s (see also related references). He (re)-
described the changes in bone vascularity and in the tidemark seen in normal joints with age-
ing and in OA, suggesting that bone turnover and changes in bone shape are crucial, and he
also showed that cartilage damage develops at sights of maximal loading of joints. He put
these observations together with his very extensive experience of the morbid anatomy and
pathology of osteoarthritic joints to produce a clear story as to how OA develops.
Strengths
1. A clear, concise, plausible story on the aetiopathogenesis of OA.
2. Paper based on a combination of careful pathological observations and descriptions and
laboratory-based tests of hypotheses on joint congruity and loading.
3. Well illustrated.
Weaknesses
1. This is an essay, it contains very little data.
2. Bullough asks you to believe him, but, like Charnley, he is probably right.
Relevance
This work emphasized the dynamic nature of the response of the joint to injury, and was
therefore the forerunner of much of the recent work on turnover of cartilage and bone.
270
Paper 7
An experimental model of osteoarthritis: early

morphological and biochemical changes
Authors
McDevitt C, Gilbertson E, Muir H
Reference
Journal of Bone and Joint Surgery 1977; 59:24–35
Summary
An experimental model of OA resulting from laxity of the joint was induced in 18 mature
dogs (at least 2 years old) by sectioning the anterior cruciate ligament of the right knee (sti-
fle) with a stab incision, the left knee providing a control. A sham operation was also per-
formed in three other dogs, in which a stab incision was made but the ligament left intact.
The dogs were killed at various intervals from 1 to 48 weeks later. Morphological changes in
bone, cartilage, synovial membrane and joint capsule were examined in all joints and bio-
chemical changes in the cartilage of three dogs killed after 2, 8, and 16 weeks. All the
changes resulting from the operation progressed with time and became indistinguishable
from those found in three dogs with natural OA of the knee. There were no changes in the
joints which had had the sham operations. As the time of onset is known, this experimental
model in a larger species enables a study to be made of the biochemical as well as the mor-
phological changes in the early stages of OA.
Related references (1) Pond N, Nuki G. Experimentally induced osteoarthritis in the dog.
Annals of Rheumatic Diseases 1973; 32:387–8 (Abstract).
(2) McDevitt C, Muir H, Pond N. Canine articular cartilage in natural
and experimentally induced osteoarthritis. Biochemical Society
Transactions 1973; 1: 287–8 (Abstract).
(3) McDevitt C, Muir H. Biochemical changes in the cartilage of the
knee in experimental and natural osteoarthritis in the dog.
Journal of Bone and Joint Surgery 1976; 58(B): 94–101.
(4) Brandt KD, Braunstein EM, Visco DM. Anterior cruciate ligament
transection in the dog: a bona fide model of canine osteoarthri-
tis, not merely of cartilage injury and repair. Journal of
Key message
An experimental model of OA in a large animal results in changes that are indistinguishable
from those of the naturally occurring disease, and the model can be used to study the early
biochemical changes of OA.
Osteoarthritis 271

One of the many serious problems in studying OA in humans is the fact that one never
knows how, why or when it started. The onset of the clinical disease is insidious, and often
appears to have been triggered by an event (such as a serious joint injury) that occurred
many years previously.
In this paper, McDevitt and colleagues provide a full, definitive description of what
became known as the ‘Pond-Nuki model’ of OA, OA in the dog induced by a stab incision to
divide the cruciate ligament. This was the first good animal model of OA which allowed
investigators to document early changes at any stage after the initiating event, in a species of
sufficient size to allow good tissue samples to be available, and one in which the disease
occurs naturally. This model, and others like it, have been used extensively to study OA over
the last two decades.
This paper describes elegant studies of the phenomenon. The authors used some controls
in which a stab incision was made but the ligament not divided, to show that the OA resulted
from the ligament transection, and was not just the result of a surgical assault on the joint.
They also compared the artificial arthritis with the naturally occurring disease in the same
species. Finally, they showed how biochemical changes could be correlated with histological
ones, heralding a new era in the investigation of OA. One of the key findings in this paper
was the demonstration of increased hydration of the articular cartilage in the early phases of
the disease.
Strengths
1. Careful experimental design, with good controls.
2. Correlation of histology with biochemical changes in the cartilage.
3. Analysis of the outcome at various time points after the insult.
Weaknesses
1. Relatively small numbers of animals described.
2. Lack of clarity on the exact site of the sections taken for biochemical analysis.
3. Biochemical changes only studied after a relatively short time period (16 weeks
maximum).
Relevance
Introduction of a new, relevant model of human OA, in which early changes could be
described and in which morphological changes can be correlated with biochemical ones.
272
Paper 8
The prediction of the progression of joint space

narrowing in osteoarthritis of the knee by bone
scintigraphy
Authors
Dieppe P, Cushnaghan J, Young P, Kirwan J
Reference
Summary
Objectives: To test the hypothesis that bone scintigraphy will predict the outcome of OA of the
knee joint.
Methods: 94 patients (65 women, 29 men, mean age 64.2 years) with established OA of one
or both knee joints were examined in 1986, when radiographs and bone scan images (early
and late phase) were obtained. The patients were recalled, re-examined, and had further
radiographs taken in 1991. Paired entry and outcome radiographs were read by a single
observer, blind to date order and other data. Scan findings and other entry variables were
related to outcome. Progression of OA of the knee was defined as an operation on the knee
or a decrease in the tibiofemoral joint space of 2 mm (0.08 in) or more.
Results: Over the 5 year study period 10 patients died and nine were lost to follow-up.
Fifteen had an operation on one or both knees (22 knees). Of the remaining 120 knees
(60 patients) analysed radiographically, 14 (12%) had progressed in the manner defined. Of
32 knees with severe scan abnormalities, 28 (88%) showed progression, whereas none of the
55 knees with no scan abnormality at entry had progressed. The strong negative predictive
power of scintigraphy could not be accounted for by disease severity or any combination of
entry variables. Pain severity predicted a subsequent operation, but age, sex, symptom dura-
tion, and obesity had no predictive value.
Conclusions: Scintigraphy predicts subsequent loss of joint space in patients with established OA
of the knee joint. This is the first description of a powerful predictor of change in this disease.
The finding suggests that the activity of the subchondral bone may determine loss of cartilage.
Related references (1) McAlindon TE, Watt I, McCrae F, Goddard P, Dieppe P. Magnetic
resonance imaging in osteoarthritis of the knee: correlation
with radiographic and scintigraphic findings. Annals of
(2) McCrae F, Shouls J, Dieppe P, Watt I. Scintigraphic assessment of
osteoarthritis of the knee joint. Annals of Rheumatic Diseases
1992; 51:938–942.
(3) McCarthy C, Cushnaghan J, Dieppe P. The predictive role of
scintigraphy in radiographic osteoarthritis of the hand.
Osteoarthritis and Cartilage 1994; 2:25–28.
Osteoarthritis 273
Key message
If a bone scan shows inactive subchondral bone in the knee joint in a patient with OA, that
joint will not progress radiographically over the next 5 years.

Bone scintigraphy is a technique generally used to detect bony metastases; it indicates areas of
high bone turnover. In a series of papers by this group, scintigraphy was used to show that the
subchondral and marginal bone of osteoarthritic joints is active, that this activity turns on and off
(substantiating the idea of Kellgren and Moore that the activity of OA was phasic, Figure 17.2),
and then by correlating scintigraphic, clinical and radiographic findings with time, showed that
scintigraphic activity was predictive of radiographic progression in both the knee and the hand.
Prior to this, fuelled by the early descriptions of OA as a disease of articular cartilage by
Nichols and Richardson and others, there had been a general assumption that cartilage is
the main tissue affected in this disease, and that cartilage should be the main target for ther-
apeutic interventions. However, this was one of the first pieces of empirical research to impli-
cate bone pathology as one of the major factors in the pathogenesis of progressive OA. It
helped the development of a paradigm change in our thinking about OA, which is now
regarded as a disease process of the whole joint organ in which bony, synovial and capsular
changes are as important as cartilage damage. It has also helped switch attention from carti-
lage to bone in thinking about therapeutic targets. Subsequently, MRI has shown similar
bone changes, as outlined in McAlindon, et al. (1).
Figure 17.2 Late phase (3 hour) bone scans

of the knee of a patient with OA taken
1 year apart. The first scan (1987) shows
extensive retention of isotope in the tibial
subchondral bone, indicating high turnover.
One year later, most of this activity has
disappeared and isotope retention is limited
to the ‘rim pattern’ described by McCrae
and colleagues, indicating activity in
osteophytes only. This illustrates the
phasic nature of the activity of OA.
Strengths
1. A relatively large and long-term prospective study.
2. Careful attention to detail in scoring scans and radiographs and in the analysis of variables
that might affect progression.
Weaknesses
1. The definition of radiographic progression and non-progression is somewhat arbitrary.
2. The use of knee replacement as a surrogate for progression may be erroneous.
3. The patients were a highly selected group, coming from a hospital practice.
Relevance
Empirical research to show that bone pathology is important in the progression of OA. The
first, and to date the strongest predictor of OA outcome.
275
CHAPTER 18
Crystal deposition diseases

Daniel J McCarty
Introduction
Progress in any field proceeds by a series of modest steps forward. Each step is influenced by
steps preceding it and influences those that follow. Although I have been involved in
research into arthritic diseases that involve deposition of microcrystals in joint tissues for
nearly 40 years, I find that choosing six to eight papers out of thousands published in the
past century a daunting and somewhat subjective task. My first choice of papers was (immod-
estly) my paper with JL Hollander reporting the definitive identification of monosodium
urate (MSU) crystals in gouty arthritis. The discovery of calcium pyrophosphate dihydrate
crystals and ‘pseudogout’ was a direct result of this work. Both papers have been transferred
by our editors to the joint fluid section of this volume leaving me with fresh choices.
The synthesis, characterization and demonstration of a dose-related inflammatory
response to MSU crystals fulfilled Koch’s postulates in a non-infectious disease (Paper 1).
The discovery of the first cell-derived chemotactic factor and the principal mechanism of
action of colchicine, perhaps the oldest known effective drug, were derived from this work
(Paper 2). Gout is arguably the most treatable disease in all of medicine because MSU crystal
deposition is possible only from saturated urate solutions. Saturation is markedly influenced
by local temperature. Solubilization of MSU crystals from cooler body parts often requires
serum levels of 0.24 mmol/l (4 mg/dl) or less. Effective control of hyperuricemia and MSU
crystal deposition followed the introduction of effective, relatively non-toxic, drugs which
inhibited renal tubular resorption of uric acid (probenecid, Papers 3a and b) or inhibited
the synthesis of uric acid by blocking xanthine oxidase (allopurinol Paper 4).
As MSU crystal formation is driven by a systemic rise in uric acid levels, CPPD crystal depo-
sition requires locally increased inorganic pyrophosphate (PPi) levels in synovial fluid and
joint tissues. Plasma PPi levels are normal in patients with this condition. Just as effective con-
trol of hyperuricemia and gout was preceded by quantitative studies of uric acid metabolism,
effective control of CPPD crystal deposition requires such studies of PPi metabolism.
Generation of PPi from extracellular adenosine triphosphate (ATP) by ectonucleoside
triphosphate pyrophosphohydrolase in synovial fluid was quantitatively sufficient to account
for the PPi production rate in vivo as estimated by PPi pool size and turnover rate in human
knee joints, and for the PPi quantities elaborated by human cartilage in organ culture
extrapolated to the estimated mass of cartilage in human knee joints (Paper 5).
The originators of the term ‘chondrocalcinosis’ found that most joints with such calcified
cartilage became severely degenerative over time (Paper 6). Joints not affected as part of the
‘nodal osteoarthritis’ pattern such as the wrist, carpus, elbow, shoulder, hip and metacarpal
phalangeal often showed such changes. Lastly, it seemed appropriate to recognize the role of
the kidney tubules in both the aetiology of hyperuricemia and as targets for drugs used to
lower serum uric acid. Our current understanding is based on a four component model
(Paper 7).
276
Paper 1
Acute arthritis in man and dog after intrasynovial

injection of sodium urate crystals
Authors
Faires JS, McCarty DJ
Reference
The Lancet 1962, 2:682–685
Summary
Synthetic monosodium urate monohydrate crystals identical by x-ray diffraction powder pat-
tern to crystals purified from a gouty tophus were sterilized and injected into the knees of
two normal (non-gouty) subjects and into the stifle joints of 20 healthy dogs. Swelling, ten-
derness, warmth and redness of overlying skin and finally pain ensued (three-legged gait in
dogs). Crystals were identified in synovial fluid neutrophils. Inflammatory intensity and neu-
trophilic response were directly proportional to the dose of injected crystals.
Characterization of the suspected aetiologic agent, reproduction of the signs and symp-
toms in subjects not affected with the disease, and recovery of the aetiologic agent from the
lesions fulfilled Koch’s postulates for proof of aetiology, albeit in a metabolic, rather than an
infectious disease.
Related references (1) Seegmiller JE, Howell RR, Malawista SE. The inflammatory
response to injected sodium urate. Journal of the American
Medical Association 1962; 180:469–475. (Independent study
showing acute inflammation after injection of synthetic
monosodium urate crystals into joints of gouty patients.)
(2) Freudweiler M. Studies on the nature of gouty tophi. Deutches
Archiv fir Klin Med 1899; 63:266. (Showed that synthetic sodium
urate crystals caused inflammation in a normal human and in
several animal species.)
(3) McCarty DJ, Brill J. Annals of Internal Medicine 1964; 60: 486–505.
(An abridged translation of the work of Freudweiler and His
with comments.)
Key message
This paper established that sodium urate crystals cause acute inflammation. The classic work
of Freudweiler and his mentor, Wilhelm His Jr, appears to have been lost to posterity, per-
haps due to the untimely death of the former at age 31 years. We discovered it during an
intensive search of the older literature. The work of Seegmiller, et al. confirmed the inflam-
matory response to urate crystals, but release into the joint space of endogenous tophaceous
urates could not be ruled out (1).
Crystal deposition diseases 277

It explains why long-term lowering of serum urate results in far fewer acute attacks of gout.
Also led to the discovery of the mechanism of action of colchicine.
Figure 18.1 Clinical

All joints tapped dry response in dogs to
3-legged intrasynovial injection of
gait crystalline sodium urate.
Warmth
20 mg (2)
erythema
Tenderness 15 mg (2)
10 mg (1)
Tense effusion
6.5 mg (8)
Detectable Control (6)
effusion
0 2 4 6 8 10 12
Hours
Strength
A dose-dependent relationship was established between monosodium urate crystals and an
acute inflammatory reaction as seen in acute gouty arthritis. Koch’s postulates remain the
gold standard for proof of aetiology.
Weakness
The injected material was sterile as it had been autoclaved but no proof that the crystals had
not entrapped pyrogen was provided. The limulus assay for pyrogen had not yet been devel-
oped. The standard test at that time was development of an elevated rectal temperature in
rabbits after intravenous injection of the test material. This was not done either. But even if it
had been done, it would not have been sufficiently sensitive to detect amounts of pyrogen
sufficient to cause joint inflammation.
Relevance
Demonstration of the phlogistic properties of urate crystals.
278
Paper 2
Polymorphonuclear leukocyte motility in vitro.

IV. Colchicine inhibition of chemotactic activity
formation after phagocytosis of urate crystals
Author
Phelps P
Reference
Arthritis and Rheumatism 1970; 12:l–9
Summary
Colchicine, then regarded as a highly specific treatment, and since the 1930s as a prophylac-
tic agent for acute gout, blocked the release of a cell-derived chemotactic factor in doses as
low as 10-10M. Other actions of the drug such as inhibition of phagocytosis or cell adhesive-
ness require concentrations 10-6M.
Related references (1) Phelps P. Polymorphonuclear leukocyte motility in vitro. III.

Possible release of a chemotactic substance following phagocy-
tosis of urate crystals. Arthritis and Rheumatism 1969; 12:197–204.
(2) Wallace SH, Omokoku B, Ertel NH. Colchicine plasma levels –
implications as to pharmacology and mechanisms of action.
American Journal of Medicine 1970; 48:443–448.
(3) Spilberg I. Colchicine and pseudogout. Arthritis and Rheumatism
1980; 23:1062–1063.
(4) Alvarellos A, Spilberg I. Colchicine prophylaxis in pseudogout.
Key message
Colchicine, at concentrations easily achieved in plasma by usual doses and known to be effec-
tive in the suppression of gouty inflammation, almost completely abolished the release of a
chemotactic glycopeptide by polymorphonuclear leukocytes that had phagocytosed particu-
lates such as microcrystals. These data explained: (a) why the drug is effective in non-gouty
conditions such as pseudogout and calcific periarthritis; (b) why the drug was more effective
when given intravenously (10-fold greater plama concentration), and (c) why the drug is not
an effective general antiinflammatory agent.

This study showed that colchicine is specific therapy for inflammation induced by the endo-
cytosis of particulate matter by polymorphonuclear leucocytes and not, as long thought, spe-
cific for gouty inflammation per se. It represented the death knell of the last surviving ‘diag-
nostic’ therapeutic trial. It also explained the mechanism of action of the oldest known drug.
Herbal colchicine was known in ancient Asia Minor as ‘hermodacyl’ (the finger of Hermes).
Strengths
1. Very clever experimental design enabling simultaneous measurement of chemotaxis, ran-
dom cell motility and crystal phagocytosis.
2. The effect of varying drug dose was determined using a crystal dose-response curve for
each of these variables.
Weakness
No in vivo correlation of suppression of chemotactic factor formation in patients with acute
gouty arthritis was attempted. This was partially addressed in subsequent experiments using
the canine model.
Relevance
Demonstrated the mode of action of colchicine in crystal-induced arthritis.
280
Paper 3a
The clinical and metabolic effects of Benemid in

patients with gout
Authors
Talbot JH, Bishop C, Norcross BM, Lockie LM
Reference
Transactions of the Association of American Physicians 1951; 64:372–377
Paper 3b
Benemid (p-di-n-propylsulfamyl)-benzoic acid as

uricosuric agent in chronic gouty arthritis
Authors
Gutman AB, Yü TF
Reference
Transactions of the Association of American Physicians 1951; 64:279–288
Summary
Independent simultaneous reports demonstrating the uricosuric properties of probenecid,
the first effective practical treatment for hyperuricemia. Aspirin in large (*5 g/day) doses
had been used previously for this purpose but salicylism was frequently severe.
Related references (1) Garrod AB. Observations on certain pathological conditions of the
blood and urine in gout, rheumatism and Brights’ disease.
Transactions of the Medical-Chirogical Society London 1848; 31:83–98.
(2) Burns JJ, Yü TF, Ritterband A, Perel JM, Gutman AB, Brodie BB. A
potent new uricosuric agent, the sulfoxide metabolite of the
phenylbutazone analogue G 25671. Journal of Pharmacology and
Experimental Therapeutics 1957; 119:418–426.
(3) Jain AK. Effect of single doses of benzbromarone on serum and
urinary uric acid. Arthritis and Rheumatism 1974;21:456–458.
(4) Gutman AB, Yü TF. Protracted uricosuric therapy in tophaceous
gout. The Lancet 1957; 2:1258–1260. (Follow-up study showing
control and even resolution of gouty tophi.)
(5) Loeb JN. The influence of temperature on the solubility of
monosodium urate. Arthritis and Rheumatism 1972; 15:189–192.
Key message
Both studies measured serum and urinary uric acid in gouty subjects using colorimetric
methods. Toxicity to probenecid was dose-related between 500 and 2000 mg/day. The
Gutman-Yü study showed that treatment induced acute gouty attacks and uric acid crystal-
luria, and that the effect of the drug could be reversed by increased protein intake
(increased de novo purine synthesis) and by aspirin ingestion in the range of 1–3 g/daily
(block of probenecid effect on urate tubular reabsorption). The Talbot study noted mobi-
lization of urate from a non-miscible pool, suggesting that tophaceous urates would be reab-
sorbed by lowering serum uric acid levels sufficiently.
L.G.
Figure 18.2 Suppressive
Serum urate mg (%)

10.0 effect of addition of
acetylsalicylic acid (ASA)
1250 (2.6 g daily) to Benemid
5.0
(2.0 g daily) on urinary
Urinary urate mg/24 hours
urate excretion in a case

1000 0
of chronic gout
maintained on a low
750 protein, low purine diet.
Each column represents
500 24 hours.
250
0
ASA
Benemid

Tophaceous deposits can cause destruction of bone, cartilage and other tissues, producing
chronic gouty arthritis. Probenecid represented the first relatively safe drug to rid the body
of established tophaceous masses and prevent new masses from forming.
Strength
Almost everything now known about probenecid, it’s mechanism, dose, toxicity, drug interac-
tions, and so on was presented in the initial two reports.
Weakness
Until the advent of the specific uricase spectrophotometric method for quantifying uric acid,
non-specific colorimetric methods were used. These measured reducing substances generally
but were sufficiently accurate for the type of clinical studies described here.
Relevance
The first definitive treatment for gout.
282
Paper 4
Effects of a xanthine oxidase inhibitor on thiopurine

metabolism, hyperuricemia and gout
Authors
Rundles RW, Wyngaarden JB, Hitchings GH, et al.
Reference
Transactions of the Association of American Physicians 1963; 76:126–140.
Summary
First report of the use of the hypoxanthine analogue allopurinol, a competitive inhibitor of
xanthine oxidase, to lower serum uric acid levels in gouty patients. Hitchings and Elion
shared the Nobel Prize for this and other work on antagonists of purine metabolism. Now
widely over-prescribed, allopurinol revolutionized the treatment of hyperuricemia, providing
a powerful means of reversing urate crystal deposition and the associated tissue destruction.
Related references (1) Seegmiller JE, Rosenbloom FM, Kelley WN. Enzyme deficit associ-
ated with a sex-linked human neurological disorder and exces-
sive purine synthesis. Science 1967; 155:1682–1684.
(2) Kelley WN, Greene ML, Rosenbloom FM, Henderson JF,
Seegmiller JE. Hypoxanthine-guanine phosphoribosyltrans-
ferase deficiency in gout. Annals of Internal Medicine 1969;
70:155–206.
(3) Becker MA, Meyer U, Wood AW, Seegmiller JE. Purine overpro-
duction in man associated with increased phosphoribosylpy-
rophosphate synthase activity. Science 1973; 179:1123–1126.
(4) Palella TD, Silverman LS, Schroll CT, Homa FH, Levine M, Kelley
WN. Herpes simplex virus-mediated human hypoxanthine-gua-
nine phosphoribosyltransferase gene transfer into neuronal
cells. Molecular and Cellular Biology 1988; 8:457–460. (Early in
vitro attempt at gene therapy of Lesch-Nyhan neuronal cells.
This syndrome is high on the list of severely incapacitating sin-
gle gene abnormalities and as such is a strong candidate for
attempts at normal gene insertion therapy.)
Key message
Allopurinol blocked conversion of hypoxanthine to xanthine and xanthine to uric acid, caus-
ing a rapid fall in serum urate levels.

The use of allopurinol, sometimes in combination with probenecid, achieved plasma urate
levels low enough to dissolve tophaceous deposits in nearly all patients. It was also quantita-
tively converted to its ribonucleotide by hypoxanthine-guanine phosphoribosyl transferase
(HGPRTase) thereby depleting the pool of phosphoribosylpyrophosphate (PRPP) a key sub-
strate in the de novo synthesis of inosinic acid, the parent purine precursor of uric acid.
Lastly, as its ribonucleotide, it acted as a pseudofeedback inhibitor of the enzyme catalyzing
the first step of inosinic acid synthesis. Allopurinol can produce serious side-effects but con-
sidering its potency and specificity, it is relatively benign.
Strengths
1. The precise mechanism of inhibitors of xanthine oxidase by this hypoxanthine analog was
known.
2. The uric acid balance studies were carefully done using specific (uricase) assays.
Weakness
The drop in serum uric acid was accompanied, as expected, by a fall in urinary uric acid
excretion that was not matched by an equivalent use in urinary xanthine and hypoxanthine
in most patients. This mystery was later solved (1, 2).
Relevance
Demonstration of the efficacy of allopurinol in rapidly lowering serum and urinary uric acid
levels.
284
Paper 5
Inorganic pyrophosphate generation from adenosine

triphosphate by cell-free synovial fluid
Authors
Park W, Masuda I, Cardenal-Escarcena A, Palmer DL, McCarty DJ
Reference
Summary
Quantitative studies of ATP hydrolysis to PPi by synovial fluid nucleoside triphosphate
pyrophosphohydrolase showed that this reaction could account for the formation of all
extracellular PPi in joint tissues. The limiting factor was the availability of the substrate extra-
cellular ATP. Down regulation of local PPi concentration may prevent or reverse calcium
pyrophosphate crystal deposition in cartilage and other articular tissues. By analogy with
urate crystals in gout, this may reverse or prevent the associated tissue damage. These data
explain why CPPD crystals are commonly found in association with osteoarthritis (OA) but
not rheumatoid arthritis (RA). Extracellular ATP is elevated in OA joint fluid and is very low
in fluid from inflamed joints. The rate of ATP conversion to PPi in rheumatoid joint fluid
may approach zero.
Related references (1) Camerlaine M, McCarty DJ, Silcox DC, Jung A. Inorganic
pyrophosphate pool size and turnover rate in arthritic joints.
Journal of Clinical Investigation 1975; 55:1373–1381.
(2) Howell DS, Muniz 0, Pita JC, Enis JE. Extrusion of pyrophosphate
into extracellular media by osteoarthritic cartilage incubates.
Journal of Clinical Investigation 1975; 56:1473–1480.
(3) Ryan LM, Wortmann RL, Karas B, McCarty DJ. Cartilage nucleo-
side triphosphate (NTP) pyrophosphohydrolase. I.
Identification as an ectoenzyme. Arthritis and Rheumatism 1984;
27:404–409.
(4) Masuda I, Halligan B, Barbieri J, Haas A, Ryan LM, McCarty DJ.
Molecular cloning and expression of a porcine chondrocyte
nucleotide pyrophosphohydrolase. Gene 1997; 197:277–287.
Key message
Conversion of ATP to PPi by joint fluid ecto-nucleoside triphosphate pyrophosphohydrolase
measured ex vivo accounts for all PPi production by synovial tissues.

If it is assumed that the biologic activity of CPPD crystals can result in destructive joint
changes then attempts should be made to prevent and/or dissolve them. The study cited
showed that conversion of ATP, at levels measured ex vivo, to PPi by joint fluid ecto-nucleo-
side triphosphate pyrophosphohydrolase under physiologic conditions was sufficient to
account for all PPi production by synovial tissues. Substrate availability, not ecto-enzyme, is
the limiting factor in the reaction. Control of joint tissue PPi level and CPPD crystal deposi-
tion will probably require control of extracellular ATP levels.
Strengths
1. Studies were quantitative using precise methods to control for PPi hydrolysis and genera-
tion of ATP products other than PPi.
2. PPi generated was compared with the results of previous intra-articular PPi pool and
turnover studies and with PPi quantities released from articular cartilage incubated in
organ culture.
Weaknesses
1. Studies were done in vitro using ATP levels measured ex vivo, i.e. in fluid obtained from
joints afflicted with various forms of arthritis.
2. It is possible that the ex vivo measured ATP levels were too high.
Relevance
Demonstration of the enzyme system responsible for extracellular pyrophosphate production
using quantitative methods.
286
Paper 6
Natural course of articular chondrocalcinosis

Authors
Zitnan D, Sitaj S
Reference
Arthritis and Rheumatism 1976; 19(suppl.):363–390
Summary
Longitudinal radiographic study of familial calcium pyrophosphate crystal deposition in
Hungarians living in Slovakia. The trait leading to cartilaginous calcification followed a
Mendelian autosomal dominant pattern with complete penetrance by the fourth decade of
life. Initially the affected joints appeared normal radiographically, except for the calcifica-
tion. Degenerative joint disease, sometimes very severe, followed inexorably over time.
Related references (1) Martel W, Champion CK, Thompson GR, et al. A roentgenological-
ly distinctive arthropathy in some patients with pseudogout syn-
drome. American Journal of Roetgenology, Radiation Therapy and
Nuclear Medicine 1970; 109:587–605.
(2) Menkes CJ, Simon F, Chourki M. Les arthropathies destruictices
de la chondrocalcinose. Revue Rhumatologie et Maladie
Osteoarticulaire 1973; 40:115–123.
Key message
The finding that degeneration followed calcification suggests that prevention or reversal of
crystal deposition might prevent or reverse destructive tissue changes as has been shown in
gout.
Figure 18.3 Acute

Charcot joint in a black
man with chondro-
calcinosis: left – antero-
posterior view; right –
lateral view. Acute
collapse of the medial
tibial plateau with an
auto-arthrogram due
to powdered bony
fragments and joint
rupture into the calf.
Clinically the process
resembled a severe
septic arthritis.

Extensive longitudinal studies of the phenotypic effects of a single gene mutation on joint
structure and function provides a unique insight into the natural history of a metabolic dis-
ease. It provides hope that effective therapy might be developed.
Strength
It is likely that the effects of a single gene mutation were studied as all patients were ethnic
Hungarians living in a single Slovak village.
Weakness
It is difficult to extrapolate the findings to sporadic cases of chondrocalcinosis or even affect-
ed families of other ethnic groups, e.g. French, Dutch, Japanese, German, English, Swiss.
Relevance
Associated chondrocalcinosis with the onset of a destructive arthropathy in joints but usually
affected by classical osteoarthritis.
288
Paper 7
Evidence for a post-secretory reabsorptive site for

uric acid in man
Authors
Diamond HS, Paolino JJ
Reference
Journal of Clinical Investigation 1973; 52:1491–1499
Summary
Innovative studies on human normal and gouty subjects using pharmacological inhibitors of
urate secretion (pyrazinamide or low dose aspirin) and of urate reabsorption (probenecid,
sulphinpyrazone, or high dose aspirin) or of both secretion and reabsorption (high dose
aspirin) clearly established that uric acid handling by the kidney was a four component sys-
tem. Pinpointed site of action of probenecid and sulphinpyrazone as inhibition of post-secre-
tory urate absorption.
Related references (1) Sorenson LB, Levinson DJ. Isolated defect in post-secretory reab-
sorption of uric acid. Annals of the Rheumatic Diseases 1980;
39:180–183.
(2) Simkin PA. Urate excretion in normal and gouty men. Advances in
Experimental Medicine Biology 1977; 76:41–45.
(3) Berger L, Yü TF. Renal function in gout. An analysis of 524 gouty
subjects including long term follow up studies. American Journal
of Medicine 1975; 59:605–613.
Key message
The renal handling of uric acid is complex. Many
investigators contributed evidence for complete
Glomerulus glomerular filtration, then complete tubular reabsorp-
tion followed by active tubular secretion. This three
100% Filtration component system became a four component system
with the evidence for post-secretory reabsorption. The
98–100%
observation is important because both low and high
reabsorption serum uric acid values can be explained by the effect
of organic acids on secretion or on post-secretory
Proximal tubule
0–2%
absorption. Probenecid, sulphinpyrazone and large
Secretion (5–6 g/day) doses of salicylates block reabsorption
50% whereas acetoacetic, beta hydroxybutyric, pyrazinoic
and lactic acids inhibit secretion.
40–44%
reabsorption
Figure 18.4 Schematic depiction of handling of uric acid by

6–12% Excretion the kidney showing the four components.

The data led to a model that explains the observed clinical effects of organic acids, including
those used as uricosuric drugs.
Strength
Carefully controlled studies were carried out in a clinical research centre using human sub-
jects.
Weakness
Only the uninhibited believe data derived from the use of inhibitors!
Relevance
Demonstration of the complex way in which urate is handled by the kidney.
290
CHAPTER 19
Osteoporosis
Cyrus Cooper
Introduction
Osteoporosis is defined as a reduction in bone mass and disruption of bone architecture,
resulting in reduced bone strength and increased fracture risk. Fragility fractures are the
hallmark of osteoporosis, and they are particularly common in the spine, hip and forearm.
These fractures show a steep age-related increase and have a major impact on the health of
elderly populations in the western world, causing significant morbidity and mortality. They
impose huge financial burdens on health services throughout the world. Demographic
changes and increasing life expectancy will lead to a dramatic increase in the number of peo-
ple suffering from fractures over the next few decades (Figure 19.1) unless more effective
action is taken to prevent the disorder.
In recent years, there has been significant progress in our understanding of the causes,
diagnosis and prevention of osteoporosis (Figure 19.2). Techniques have now been devel-
oped which permit prediction of future fracture risk, and the explosion of therapeutic
modalities available should significantly reduce fracture risk in future generations. These two
themes (assessment of fracture risk, and interventions to retard bone loss with ageing) com-
prise the major topics in this anthology of classic papers. However, they are by no means the
only advances which have been made in our understanding of osteoporosis over the last
three decades. The cellular basis of age-related bone loss has now been elucidated, and is
known to result from an increase in the rate of bone remodelling and an imbalance between
the activity of osteoclasts and osteoblasts. Bone remodelling occurs at discrete sites within the
skeleton and proceeds in an orderly fashion, with bone resorption always being followed by
bone formation, a phenomenon referred to as coupling. If these two processes are not
matched, the resulting remodelling imbalance results in irreversible bone loss. Predisposing
factors include oestrogen deficiency in the postmenopausal woman, age-related osteoblast
Figure 19.1 Estimated number

3500 of hip fractures (in thousands)
Women
1990 for men and women in
3000
2025 different regions of the
2050
Hip fractures, 000s
2500 world in 1990–2050.

2000
Women Men
1500
1000
Men
500
0
North America, Region Middle East, Asia,
Europe, Oceania, Latin America, and
and Russia Africa
Osteoporosis 291
senescence and secondary hyperparathyroidism, as well as a number of diseases and drugs

(most notably corticosteroids) which alter the rate of bone turnover. Other determinants of
bone strength include bone geometry (for example hip axis length), the accumulation of
fatigue damage, and loss of trabecular connectivity.
Over the last 30 years, significant progress has also been made in the development of
methods for the non-invasive assessment of bone mass. These include dual energy x-ray
absorptiometry, quantitative computed tomography and broadband ultrasound attenuation.
Dual energy x-ray absorptiometry is the most widely used technique because of its high
reproducibility, low radiation dose and ability to measure bone mineral density at both
appendicular and axial sites in the skeleton. Bone density, so measured, is a major determi-
nant of bone strength and fracture risk. The strength of this relationship is comparable to
that between blood pressure and stroke, and equates to an 8–12-fold difference in fracture
risk across the distribution of bone density in the population. Addition of risk factors to bone
mineral density values may lead to enhanced prediction of fracture; the most important of
these are a previous history of fragility fracture and, in the elderly, risk factors for falling.
Although a useful clinical adjunct in the management of patients with osteoporosis, mass
bone density screening of the general population at any age cannot be justified. Clinical indi-
cations for bone densitometry have now been clarified, which include important historical
risk factors such as previous low trauma fracture, radiological evidence of vertebral deformi-
ty, glucocorticoid therapy and oestrogen deficiency.
Preventive strategies against osteoporotic fracture can be applied throughout the life
course. They include non-pharmacological interventions to increase peak bone mass (nutri-
tion and exercise), to reduce age-related bone loss, to decrease the risk of falling, and to
reduce the impact of falls. In addition, a number of pharmacological interventions have
recently been developed which aim to reduce future fracture incidence by altering bone
turnover. These include calcium and vitamin D, the bisphosphonates, and the selective
oestrogen receptor modulators. Guidelines for the cost-effective use of these agents have now
been developed, and their implementation will lead to marked improvement in the morbidi-
ty and mortality attributable to osteoporotic fractures.
1000 BC 0 1000 AD 2000 AD
400 BC 700 AD 1826 AD 1850 AD

Hippocrates Saxon Sir Astley Osteoporosis
of Kos tomb wells Cooper
1940 1950 1960 1970 1980 1990 2000
1948 1963 1976 1987 1990–99

Albright SPA HRT DXA Calcium
Fluoride
Bisphosphonates
SERM
WHO
Figure 19.2 Historical evolution in the study of osteoporosis over three millenia.
292
Paper 1
Postmenopausal osteoporosis: its clinical features

Authors
Albright F, Smith PH, Richardson AM
Reference
Summary
This manuscript provides the first clinical description of postmenopausal vertebral osteo-
porosis, and links the disorder directly to oestrogen deficiency. Clinical data are presented
for 42 patients with osteoporosis, 40 of whom were postmenopausal women. Ten of these
had undergone a premature menopause, and one had primary hyperparathyroidism. The
serum calcium biochemistry of these subjects was essentially normal and the profile of frac-
tures encountered was typical, with a preponderance of vertebral deformities.
Related reference (1) Riggs BL, Khosla S, Melton U. A unitary model for involutional
osteoporosis: estrogen deficiency causes both type I and type II
osteoporosis in postmenopausal women and contributes to
bone loss in ageing men. Journal of Bone and Mineral Research
1998; 13:763–773.
Key message
This is the first detailed clinical description of postmenopausal vertebral osteoporosis and
directly links the disorder to oestrogen deficiency. The manuscript provides a basis for subse-
quent trials of hormone replacement therapy in the prevention of bone loss.

A number of key messages pertaining to the pathogenesis of osteoporosis were first alluded
to in this manuscript. The authors distinguished clearly between osteomalacia (a mineraliza-
tion abnormality in which calcium hydroxyapatite cannot be incorporated into osteoid
matrix) and osteoporosis (in which there is a reduced overall amount of bony tissue, with
adequate mineralization of the available matrix). The authors also postulate the importance
of oestrogen deficiency in the pathophysiology of postmenopausal bone loss, and propose
that hormone replacement therapy (HRT) might have a beneficial effect on calcium reten-
tion in the disorder. They go on to distinguish between age-related or involutional bone loss
(a consequence of disuse) and oestrogen deficiency. Finally, they outline important sec-
ondary causes of osteoporosis including thyrotoxicosis, hypercortisolism, and primary hyper-
parathyroidism.
Strength
The manuscript constitutes a clear and readable account of the clinical manifestations of
postmenopausal osteoporosis.
Osteoporosis 293
Weakness
The study is essentially a case series, relying on astute clinical observation and a synthesis of
the previous literature.
Relevance
This initial clinical description of the disorder complements the classical descriptions of hip
fracture published over a century earlier by Sir Astley Cooper, which documented that: (a)
incidence rates of osteoporotic fracture increase steeply with age; (b) that rates are higher
among women than among men; and (c) that the disorder has a predilection for skeletal
sites containing substantial amounts of trabecular bone. It provides a forerunner of the con-
trolled trials of HRT which convincingly demonstrated that oestrogen replacement following
the menopause retards bone loss while it is continued.
294
Paper 2
The loss of bone with age, osteoporosis, and fractures

Authors
Newton-John HF, Morgan DB
Reference
Summary
This classical manuscript brings together available literature three decades ago on measure-
ments of bone density between fracture cases and non-fracture controls. The authors realize
the implication of a universal loss of bone with age in normal persons, and suggest that
osteoporosis is simply an increase in the rate of age-related bone loss. Analysis of the data
leads to the conclusion that: (a) all persons lose bone with age, and (b) that the risk of frac-
ture is largely determined by the amount of bone. The authors demonstrate that the
increase in frequency of fractures with advancing age is explained by the universal loss of
bone with age.
Related references (1) Aaron JE, Gallagher JC, Anderson J, et al. Frequency of osteomala-
cia and osteoporosis in fractures of the proximal femur. The
Lancet 1974; 1:229–233.
(2) Nordin BEC, Peacock M, Aaron J, et al. Osteoporosis and osteoma-
lacia. Clinics in Endocrinology and Metabolism 1980; 9:177–204.
(3) Riggs BL, Melton U. Involutional osteoporosis. New England Journal
of Medicine 1986; 314:1676–1686.
Key message
The authors propose that fracture risk is largely accounted for by variation in bone mass in
the general population, and that the age-related increase in fracture rate is determined by
the normal loss of bone with age.
Osteoporosis 295

The authors construct one of the earliest systematic reviews, including the literature on: (a)
the age of onset of bone loss; (b) the magnitude of bone loss, and (c) the variance in bone
loss with age among men and women. The information is utilized to compare bone loss rates
in the peripheral and axial skeletons of each gender, and to explore the association between
bone loss and fracture. The findings confirm that all bones lose mineral with advancing age
and that bone loss commences at around 40–45 years among women and 50–65 years among
men. Greater rates of bone loss were found in women and bone loss continued throughout
later life. Estimation of fracture incidence rates from proportions of the population falling
below different bone density thresholds generated incidence curves which were almost iden-
tical to those derived from epidemiological studies conducted at the time. The authors con-
cluded, for the first time, that osteoporosis is not a discreet disease process which can be easi-
ly separated from normality. Rather, it is simply a term used to characterize individuals at the
extreme of the distribution of bone density in later life, whose fracture risk is corresponding-
ly increased. The pattern of the age-related increase in hip and wrist fracture was able to be
explained entirely on the basis of the pattern of bone loss with age.
Strengths
1. A wonderful review of the literature of its day.
2. Clear elucidation of several problems in the definition of osteoporosis which have
remained controversial ever since. These include the continuous distribution of bone
density in the general population, the variation in bone loss rates with advancing age, and
the absence of a threshold in the relationship between bone density and fracture.
Weakness
The mathematical explanation of the relationship between changing bone density and frac-
ture risk is obscure, but the reasoning behind the relationships is sound.
Relevance
This manuscript is ahead of its time in proposing that osteoporosis might be defined by
points on the distribution of bone density at any given age, but that fracture risk is likely to
increase smoothly with declining bone density. The clear distinction between osteoporosis
and fracture is therefore established; this relationship was enshrined in the World Health
Organization (WHO) definition of the disorder.
296
Paper 3
The diagnosis of osteoporosis

Authors
Kanis JA, Melton LJ, Christiansen C, Johnston CC, Khaltaev N
Reference
Journal of Bone Mineral Research 1994; 9:1137–1141
Summary
This manuscript provides the basis for the WHO diagnosis of osteoporosis. The article stems
from the deliberations of the WHO Study Group on osteoporosis in 1994, which generated
the first widely used threshold for osteoporosis definition on the basis of bone mineral mea-
surement (2.5 SD below the young normal mean value at the skeletal site of interest).
Related references (1) Consensus Development Conference. Diagnosis, prophylaxis and

treatment of osteoporosis. American Journal of Medicine 1993;
94:646–650.
(2) World Health Organization. Assessment of fracture risk in its applica-
tion to screening for postmenopausal osteoporosis. Technical Report
Series. Geneva: WHO, 1994.
(3) Melton U, Chrischilles FA, Cooper C, Lane AW, Riggs BL. How
many women have osteoporosis? Journal of Bone and Mineral
Research 1992; 7:1005–1010.
Key message
Osteoporosis is defined as a bone mineral measurement more than 2.5 SD below the young
normal mean value.

With the advent of accurate and precise non-invasive measurement methods of bone miner-
al, the issue as to whether osteoporosis might be defined exclusively on the basis of such mea-
surements, or whether a fracture was mandated, became highly controversial. This manu-
script, summarizing a lengthier WHO report, clearly established the rationale for defining
osteoporosis on the basis of bone mineral measurements alone. Prospective studies revealed
that the risk of fragility fractures increased progressively and continuously as bone mineral
density declined. A fracture threshold, at which the disorder could be defined, was not there-
fore readily apparent. The WHO study group based their choice of a threshold on the preva-
lence of clinically relevant associations; the threshold of 2.5 SD below the young normal
mean was appropriate to identify approximately 20% of postmenopausal women as having
osteoporosis (Figure 19.3). This threshold also had high specificity for prediction of future
fracture. Finally, the article provides estimates of the lifetime fracture risk among men and
women at age 50 years for the proximal femur (17.5% and 6% respectively), vertebra (15.6%
and 5.0% respectively) and wrist (16% and 2.5% respectively).
Osteoporosis 297
Figure 19.3 WHO definition of

140 A B C osteoporosis: (A) normal bone
+2 density, (B) osteopenia, and
120
+1
(C) osteoporosis.
Forearm BMD (%)
100 0
T score
-1
80
-2
60 -3
40 -4
-5
0
30 40 50 60 70 80
Age (years)
Strengths
1. Provides a succinct description of the WHO definition of osteoporosis.
2. Describes the importance of adopting a bone density threshold relating to the young nor-
mal mean.
3. As a decision threshold among younger postmenopausal women (e.g. 50–79 years), the
WHO definition remains extremely useful and provides a particularly helpful cost-effec-
tive cut-point at which relatively expensive drugs may be prescribed for long periods in
patients with the disorder.
Weaknesses
1. More recently, it has become clear that this definition has deficiencies when applied to
very elderly samples of the population (e.g. 80 years and over) where prevalence rates rise
to high levels.
2. It is also of unproven value in men.
Relevance
The WHO definition of osteoporosis was initially proposed as a tool with which to express
the frequency of the disorder in different populations. However, it has subsequently been
demonstrated that it also meets the requirements for a therapeutic threshold in women aged
50–79 years. It therefore underpins the clinical management of patients with this disorder,
particularly in the prescription of newer antiresorptive agents such as the bisphosphonates
and selective oestrogen receptor modulators.
298
Paper 4
Meta-analysis of how well measures of bone mineral

density predict occurrence of osteoporotic fractures
Authors
Marshall D, Johnell O, Wedel H
Reference
Summary
A meta-analysis of prospective cohort studies published between 1985 and 1994 was per-
formed to determine the ability of measurements of bone density in women to predict later
fractures. Eleven separate study populations, including 90 000 person-years of observation
time and over 2000 fractures, were included. All measurement sites were found to have simi-
lar predictive capacity (relative risk 1.5, 95% CI 1.4–1.6) for decrease in bone mineral densi-
ty, except measurements at the spine for prediction of vertebral fracture (relative risk 2.3,
95% CI 1.9–2.8), and measurements at the proximal femur for hip fracture (relative risk 2.6,
95% CI 2.0–3.5). The predictive ability of decrease in bone mass was similar to that of a 1 SD
decrease in blood pressure for stroke, and better than a 1 SD increase in serum cholesterol
concentration for cardiovascular disease.
Related references (1) Cummings SR, Black DM, Nevitt MC, et al. Bone density at various
sites for prediction of hip fractures. The Lancet 1993; 341:72–75.
(2) Ross PD, Davis JW, Epstein RS, Wasnich RD. Pre-existing fractures
and bone mass predict vertebral fracture incidence in women.
(3) Seeley DG, Browner WS, Nevitt MC, et al. Which fractures are asso-
ciated with low appendicular bone mass in elderly women?
Key message
A decrease in bone mineral density by 1 SD is associated with a doubling in the risk of future
fracture. This predictive ability is similar to that of blood pressure measurement for stroke,
and better than that of serum cholesterol concentration for cardiovascular disease.

Measurement of bone mineral density is a key method for the early identification of individu-
als at high risk of fracture, although it is only one of a number of risk factors for fracture.
Bone density is a continuously distributed variable in the general population, and the issue as
to whether screening programmes incorporating this measurement are effective, remains con-
troversial. This study provided a systematic review of 11 prospective studies in which measure-
ments of bone density among women were related to future fracture risk. The results were sur-
Osteoporosis 299
prizingly uniform, in suggesting that each standard deviation decrease in bone density results
in a doubling of the risk of future fracture (Figure 19.4). There was some evidence for
enhanced predictive capacity among measurements made at the site of fracture, but even
measurements made at distant sites retained significant positive predictive value. At a lifetime
fracture incidence of around 30%, bone
RR mineral measurements had 89% speci-
3 ficity, 34% sensitivity, 58% positive pre-
dictive value, and identified approxi-
mately 21% of all fractures. The results
2
clearly demonstrated that bone density
measurements could predict risk of frac-
1 ture, but that they were not sufficiently
predictive to warrant population-wide
screening. The results led to clarifica-
0
Radius tion of the precise circumstances in
proximal distal Hip Spine OS calcis
which bone density measurements were
justifiable, in the management of indi-
Figure 19.4 Predictive value of bone density at
vidual patients at the highest risk (see
different skeletal sites for future hip fracture.
Paper 3).
Strengths
1. This important and timely review clarified the predictive capacity of bone density for
future fracture.
2. Clearly demonstrated the lack of effectiveness of a mass screening programme using bone
densitometry.
3. Also clearly demonstrated the predictive capacity of bone density measurements in indi-
viduals at the highest risk.
Weaknesses
1. The data could not be extrapolated to men or to younger women, in whom further
research is needed.
2. Analysis did not include quantitative computed tomography, although ultrasound mea-
surements of the calcaneus were found to be predictive.
Relevance
Clarification of the relationship between baseline bone density measurement and future risk
of fracture helped to position this new technology appropriately within clinical guidelines for
the evaluation and treatment of osteoporosis. Bone density measurements are now widely
used in the management of patients at the highest risk, in whom knowledge of the results
alters clinical or therapeutic decision-making. The key clinical indications for densitometry
include: (a) prevalent vertebral deformity; (b) previous low trauma fracture; (c) use of corti-
costeroid therapy, and (d) oestrogen deficiency. When used in this setting, measurements
permit the most cost-effective use of therapy over prolonged periods of time with the objec-
tive of preventing fractures.
300
Paper 5
Hormone replacement therapy and risk of hip

fracture: population based case-control study
Authors
Michaelsson K, Baron JA, Farahmand BY, Johnell O, Magnusson C,
Persson PG, Persson I, Ljunghall S, on behalf of the Swedish Hip
Fracture Study Group
Reference
Summary
This population-based case-control study in six Swedish counties compared 1327 women
aged 50–81 years with hip fracture, and 3262 randomly selected controls. The objective was
to determine the relative risk of hip fracture associated with postmenopausal HRT, including
the effect of duration and recency of treatment, the addition of progestogen, the route of
administration, and the dose of oestrogen. Compared with women who had never used HRT,
current users had a marked reduction in risk of hip fracture (OR = 0.35, 95% CI 0.24–0.53).
The protective effect was markedly attenuated, and not statistically significant among former
users of hormone replacement (OR = 0.76, 95% CI 0.57–1.01). For every year of therapy, the
overall risk decreased by around 6%. Last use 1–5 years previously, with a duration of use
more than 5 years, was associated with protection; cessation of therapy for longer than 5
years results in removal of the protective effect of HRT. Transdermal hormone replacement
gave similar risk estimates to that from oral regimens.
Related references (1) Grady D, Rubin SM, Petitti DB, Fox CS, Black D, Ettinger B, et al.
Hormone therapy to prevent disease and prolong life in post-
menopausal women. Annals of Internal Medicine 1992;
117:1016–1037.
(2) Cauley JA, Seeley DG, Ensrud K, Ettinger B, Black D, Cummings
SR. Estrogen replacement therapy and fractures in older
women. Study of Osteoporotic Fractures Research Group.
(3) Kiel DP, Felson DT, Anderson JJ, Wilson PWF, Moskowitz MA. Hip
fracture and the use of estrogens in postmenopausal women.
New England Journal of Medicine 1987; 317:1169–1174.
Osteoporosis 301

Many studies have shown that HRT can reduce bone loss over relatively short periods of time
(3–5 years). Case-control and cohort studies performed two decades ago also suggested that
HRT was associated with a reduced risk of hip fracture. However, closer scrutiny of these data
suggested that anti-fracture efficacy might be concentrated among current users of HRT.
Risk estimates for older women, in whom longer periods had elapsed since cessation of HRT,
were found to be closer to unity. This case-control study, coupled with data from the United
States, confirmed that HRT was associated with a reduced risk of hip fracture among current
users, but clearly demonstrated an attenuation of this protective effect after cessation of use.
In particular, women who had ceased use more than 5 years previously were found to have
no significant residual protection. The results have great importance for the use of HRT in
the prevention of osteoporotic fracture. They suggest that treatment might be commenced
several years after the menopause without loss of efficiency in reducing fracture risk. They
also suggest that continuation of treatment, to the ages at which fractures occur, is necessary
for the intervention to have a major public health impact.
These findings countered the prevailing view at the time, that a decade of HRT immedi-
ately following the menopause would suffice to retard bone loss and prevent hip fractures
over the subsequent 20–30 years. They have led to substantial efforts to improve compliance
with long-term HRT, and to develop other oestrogen-like interventions for use in osteoporo-
sis and cardiovascular prevention.
Strengths
1. The study was of large size, population-based, and had a high response rate.
2. The validity of oestrogen replacement was of the highest quality and cross-checked against
various hospital registers.
Weaknesses
1. The principal limitations of the study were possible confounding and non-response bias.
2. There was also relatively low statistical power to explore the association between hip frac-
ture and different types of HRT.
Relevance
This paper confirmed that HRT should be continued for long periods of time in order to
obtain optimal protection against hip fracture. No overall hip fracture protection remained
after 5 years without HRT, although therapy could be initiated several years after the
menopause without loss of fracture protection. Oral and transdermal therapy were equally
effective in reducing the risk of hip fracture, and the addition of progestogen permitted a
lower dose of oestrogen.
302
Paper 6
Vitamin D3 and calcium to prevent hip

fractures in elderly women
Authors
Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S,
Delmas PD, Meunier PJ
Reference
Summary
This manuscript reports a randomized controlled trial of the effects of supplementation with
vitamin D3 and calcium on the frequency of hip and other non-vertebral fractures. Three
thousand two hundred and seventy healthy ambulatory women, mean age 84 years, were
included in the 18 month study. They were randomized to receive a supplement of calcium
1.2 g, and vitamin D 3 800 IU, or double placebo. The results confirmed a 43% lower
(p = 0.04) incidence of hip fracture and 32% lower (p = 0.02) incidence of non-vertebral
fractures among the women treated with calcium and vitamin D3. In this group, the mean
serum parathyroid hormone concentration decreased by 44% from baseline and the serum
25–hydroxyvitamin D concentration increased by over 50%.
Related references (1) Chapuy MC, Arlot ME, Duboeuf F, et al. Effect of calcium and
cholecalciferol treatment for three years on hip fractures in
elderly women. British Medical Journal 1994; 308:1081–1082.
(2) Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM.
Vitamin D supplementation and fracture incidence in elderly
persons – a randomized placebo controlled trial. Annals of
(3) Dawson-Hughes B, Harris SS, Kraw EA, Dallal GE. Effect of calci-
um and vitamin D supplementation on bone density in men
and women 65 years of age or older. New England Journal of
Medicine 1997; 337:670–676.
Key message
This large randomized controlled trial established that calcium and vitamin D were effective
in the primary prevention of hip fracture among elderly women resident in French nursing
homes.
Osteoporosis 303

The key contributors to fractures among elderly women are a reduction in bone mass and an
increased frequency of falls. Part of the decrease in bone density with advancing age can be
explained by increased parathyroid hormone secretion, resulting from vitamin D deficiency
and low calcium intake that are not compensated for by an increase in 1,25-dihydroxyvitamin
D production. This trial was the first study to demonstrate convincingly that supplementation
with calcium and vitamin D was able to normalize serum parathyroid hormone, retard bone
loss and reduce the rate of fractures among elderly women.
The study was followed by attempts to demonstrate whether similar anti-fracture efficacy
might be obtained from the use of vitamin D alone, as the addition of calcium reduces com-
pliance, and adds to expenses. Unfortunately, a follow-up study in The Netherlands, which
evaluated a supplement of 400 IU cholecalciferol daily among elderly men and women failed
to demonstrate effectiveness against hip fracture. The results of further studies are awaited
before the best modality for treatment of secondary hyperparathyroidism in the elderly can
be established. However, this manuscript led to the introduction of calcium and vitamin D as
the basal therapy for patients with osteoporosis in the placebo arms of subsequent random-
ized controlled trials, and confirmed the effectiveness of this intervention in patients at high
risk of future fracture.
Strengths
1. This was a pragmatic randomized controlled trial performed in the general population
with a clear result.
2. It has led to an important change in health policy, and confirms that pharmacological
intervention, even in the elderly, can lead to a reduction in fracture incidence.
Weakness
The extent to which the intervention acted on hip fracture incidence through an influence
on falling was less well studied.
Relevance
Calcium and vitamin D supplementation are now established as first-line treatment for osteo-
porosis. Although the magnitude of their effect is less than that observed for more potent
inhibitors of resorption such as the bisphosphonates, this trial clearly established that sup-
pression of secondary hyperparathyroidism can translate into a reduction in hip fracture
incidence. Whether similar effects can be obtained by delivery of vitamin D alone, or
whether combined calcium and vitamin D supplementation is required, remain the subject
of current trials.
304
Paper 7
Randomized trial of effect of alendronate on risk of

fracture in women with existing vertebral fracture
Authors
Black DM, Cumming SR, Karpf DB, Cauley JA, Thompson DE, Nevitt MC,
Bauer DC, Genant HK, Haskell WL, Marcus R, Ott SM, Torner JC, Quandt
SA, Reiss TF, Ensrud KE, for the Fracture Intervention Trial Research Group
Reference
The Lancet 1996; 348:1535–1541
Summary
Two thousand and twenty-seven women aged 55–81 years with low femoral neck BMD and
with at least one vertebral deformity on thoracolumbar radiographs, were enrolled in this 3
year placebo- controlled trial of alendronate. Women were followed up to identify incident
vertebral and non-vertebral fractures. The incidence of new morphometric vertebral frac-
tures among women in the alendronate group was markedly lower than that in the placebo
group (relative risk 0.53, 95% CI 0.41–0.68). The relative risk of hip fracture (0.49, 95% CI
0.23–0.99) and of wrist fracture (0.42, 95% CI 0.31–0.87), was also markedly lower among
women given alendronate.
Related references (1) Watts NB, Harris ST, Genant HK, et al. Intermittent cyclical
etidronate treatment of postmenopausal osteoporosis. New
England Journal of Medicine 1990; 323:73–79.
(2) Liberman UA, Weiss SR, Broll J, et al. Effect of oral alendronate on
bone mineral density and the incidence of fractures in post-
menopausal osteoporosis. New England Journal of Medicine 1995;
333:1437–1443.
(3) Harris ST, Watts NB, Genant HK, et al. Effects of risedronate treat-
ment on vertebral and non-vertebral fractures in women with
postmenopausal osteoporosis: a randomized controlled trial.
Journal of the American Medical Association 1999; 282:1344–1352.
Key message
This large randomized controlled trial confirmed that alendronate reduces vertebral and hip
fracture among women with low bone mass and pre-existing vertebral deformities.
Osteoporosis 305

Although randomized trials have shown increases in bone mass associated with several thera-
pies including oestrogens, calcitonin, calcitriol, sodium fluoride, and the bisphosphonates,
few have reported reductions in the incidence of vertebral fracture, and only one a reduction
in the incidence of hip fracture (see Paper 6). The aminobisphosphonate alendronate, has
been shown to increase bone mineral density at the spine and hip, and to reduce the risk of
radiographically-defined vertebral fracture among women with low BMD. The Fracture
Intervention Trial established that three years of alendronate therapy among post-
menopausal women with low bone mass and pre-existing vertebral fractures not only reduced
incident vertebral fracture, but also the incidence of hip and wrist fracture. The bisphospho-
nates (etidronate, alendronate and risedronate) have now been widely evaluated in the pre-
vention and treatment of osteoporosis, and have been shown to reduce the rate of bone loss
among healthy postmenopausal women; to reduce the incidence of fractures among elderly
patients with osteoporosis, and to be effective in the management of corticosteroid-induced
osteoporosis. The more potent bisphosphonates have marked effects on bone resorption,
and may improve bone microarchitecture and mineralization, as well as reducing bone
turnover. They have an established role in the prevention and treatment of postmenopausal
osteoporosis, as well as in the management of corticosteroid-induced osteoporosis.
Strengths
1. Established the necessary design for evaluation of treatments in osteoporosis.
2. Sufficiently large to establish clearly effectiveness of alendronate on the incidence of ver-
tebral fracture, and also provide some evidence on the risk of limb fractures.
Weakness
The stringent recruitment criteria to the study did not permit adequate evaluation of adverse
effects. It was only after release of the agent that the association with oesophageal discomfort
and ulceration were recognized, and strict guidance was issued as to the means whereby the
agents should be administered.
Relevance
The bisphosphonates are now routinely used in the management of osteoporosis. This trial
was among the first studies to establish their role in management. It also remains one of the
only studies to demonstrate that an anti-resorptive drug reduces the incidence of hip
fracture.
307
Section 3
Regional and
Miscellaneous
Disorders
308
CHAPTER 20
Back pain
Malcolm IV Jayson
Introduction
Back problems are extremely common. Epidemiological studies in the general population
have show a lifetime prevalence of significant episodes of back pain of 70–80% and experi-
ence of back pain within the month previous to the survery of 30–40%.
The back is an extremely complex structure with numerous potential sources of back pain.
This includes not only the nerves but also the thecal sheaths, blood vessels, ligaments,
periostium, muscles and other tissues.
Commonly it is difficult and often impossible to define the specific source of a patient’s
symptoms. Much pathology can be demonstrated but this frequently bears only a very poor
relationship to the development of back pain. Modern medical technology, including MRI
and CT scanning, technetium scans and neurophysiology, have frequently served to heighten
the confusion rather than provide specific answers in any individual case. Moreover we now
know that the perception of pain commonly depends upon central amplification so that
patients with trivial organic damage have an enhanced perception of the severity of their
pain and disability. This phenomenon of central amplification probably accounts for the
enormous increase in back disability seen in recent years.
With this background it is hardly surprising that our knowledge and understanding of the
pathogenesis of many back pain syndromes has evolved very slowly and it is only in recent
years that we have learned to distinguish the concepts of nociception, pain experience, dis-
tress and illness behaviour.
Although much original work on the pathology of disease on the spine was undertaken by
pathologists such as Schmorl and Junghanns (1932, see Related reference, Paper 1), there was
no clear identification of the disc as being responsible for back problems until the classic stud-
ies of Mixter and Barr (1934, Paper 1). There has been one or two previous descriptions of disc
prolapse causing nerve root damage but in general it was thought that protrusions from the
back of the disc were chondromas. Mixter and Barr re-examined pathological specimens in the
museum and showed that they were disc herniations and that they could damage the nerve
root. They described a series of patients treated successfully with surgery.
This paper led to an explosion of interest in the intervertebral disc and for many years there-
after orthopaedic surgeons uncritically operated on virtually every back pain patient whether or
not there was a defined protrusion and sciatica. There were some successes but also many disas-
ters. In part the problem was lack of appropriate diagnostic tools. Myelography is an invasive
procedure and often correlated poorly with symptoms. Many surgeons operated without previ-
ous specific indentification of disc protrusion and nerve root damage. Surgery for sciatica with
an appropriately defined disc prolapse has become a very successful operation.
The principal function of the spine is to support the trunk and at the same time allow flexibil-
ity of movements. The relationship between the structure and function of the intervertebral disc
is all important. Nachemson and Morris (l964, Paper 2) initiated physiological research into the
biomechanics of the disc showing that it was possible to measure the pressures within the disc in
life and during a wide variety of physical activity. In a long series of papers they established the
effects of posture and exercise in both healthy and degenerate discs. This work proved to be the
Back pain 309
foundation of modern ergonomics. Modern advice used in industry on the correct ways to lift
directly stems from these studies. This approach led to an improved design of seating. The Volvo
Car Company invested heavily in this research and used it to produce their ergonomic car seats.
These principles are now adopted by virtually all car manufacturers.
A more general approach to biomechanics has included a long series of studies of muscle
function and strength in various postures. Extensive research has been conducted on the biome-
chanics of the human spine and machines developed in efforts to improve quantification of
spine function.The results of these studies however, have been relatively disappointing. Despite
considerable investment in technology, the results of attempts to quantify spine function have
only correlated very poorly with back pain.We now know that there are other important patho-
genic mechanisms apart from disc prolapse. Kellgren (l938, see Chapter 21, Paper 8) showed
that nociceptive stimulation of soft tissues could lead to referred pain experienced in the lower
limbs. Far too often it is assumed that pain in the leg must mean nerve root damage. This study
was the first to identify pain in the lower limb referred from the soft tissues and to appreciate
that this is different from radicular pain. The actual mechanism of nerve root damage was origi-
nally thought to be due to direct trauma by the disc herniation pressing on the nerve. Hoyland,
et al. (l989, Paper 3) identified the significance of vascular damage and in particular venous
obstruction leading to nerve root swelling, fibrosis and neural atrophy. More recent studies have
shown two level compression within the spine causing isolated segments of poor perfusion.
Increasingly it is apparent that vascular damage plays an important part of the pathogenic
process and occurs as a consequence of both disc herniation and degenerative disease.
The disc itself may be a source of pain. There is a correlation between the severity of disc
degeneration and back pain, albeit rather poor. It is known that in the healthy disc innerva-
tion is restricted to the outer layers of the annulus fibrosus. One difficulty has been to identi-
fy why disc degeneration may lead to pain in some patients but not in others. The study by
Freemont, et al. (l997, Paper 4) identified the ingrowth of nociceptive fibres into the inner
annulus and the nucleus of painful degenerate disc, and indicates an important pathogenic
mechanism in the production of back pain.
Nevertheless, our understanding of the pathogenesis of many back pain problems has been
poor. Waddell (l987, Paper 5) drew attention to the distinction between back disability and back
pain. He pointed out that back pain is an extremely common and world-wide phenomenon, yet
the severity of disability associated with back pain appears very much a disorder of Western civiliza-
tion and has developed as a recent epidemic. He identified the role of psychosocial factors in exac-
erbating patients’ perception of the severity of the back pain problem, and leading to the degrees
of distress and disability which we commonly see. In this context, Bigos, et al. (l991, Paper 6) under-
took a prospective study in which he examined baseline data and followed-up workers in order to
identify pre-back pain factors predictive of future back episodes. As would be expected, workers
with a previous history of backache developed future back episodes. In addition, they also showed
that job dissatisfaction was a major predictive factor for future backache. Again this emphasizes the
role of psychological distress and other psychosocial factors in the back pain scenario.
The traditional management of back pain has been long periods of rest often with bed rest,
lying flat, and cautious and careful remobilization. This approach was challenged by Deyo, et al.
(l986, Paper 7). They showed that this advice is wrong and indeed may promote disability. They
introduced the concept of only a short period of bed rest and early mobilization and activation.
Since then others have shown that the best results are obtained by minimizing the period of rest
and restriction of activity, and returning to normal function as soon as possible. The benefits may
be due to a combination of physical effects on the spine and psychological effects in avoiding the
reinforcement of the concept of disability.This philosophy has been extended into the manage-
ment of chronic back pain. For patients for whom there is no specific solution, intensive rehabilita-
tion programmes including physical rehabilitation and psychological counselling seem to offer the
best way forward. Programmes incorporating this approach are now available in many centres.
310
Paper 1
Rupture of the intervertebral disc with

involvement of the spinal canal
Authors
Mixter WJ, Barr JS
Reference
Summary
This study reported a series of 19 cases of herniation or rupture of the intervertebral discs in
the cervical, dorsal and lumbo-sacral spine with the development of both spine pain and neu-
rological damage. Previously it was thought that the swellings from discs were neoplasms but
the authors identified that rupture of the disc is much more common and recommended
that the treatment of this condition is essentially surgical.
Related reference (1) Schmorl G and Junghanns H. Archiv und Atlas der normalen und
pathologischen Anatomie in typischen Röntgenbildern. 1932. Georg
Thiem, Liepzig.
Key message
Prolapse of nucleus pulposus and fracture of the annulus is a common cause of back
problems.

In a pathological study Schmorl and Junghanns (1) had demonstrated disc prolapse but had
not considered the clinical significance of these findings. Other authors had identified disc
lesions leading to both spinal cord and nerve root pressure but generally it was thought they
were due to chondromas, and apart from one or two individual case reports the phenome-
non of disc prolapse was not recognized. This study indicated the importance of disc hernia-
tion as a specific cause of spine problems and that operative removal of the prolapse could
be a successful form of treatment. It was this paper which led to an explosion of interest in
the surgical management of back problems. Following this publication there was a dramatic
increase in the amount of lumbar spine surgery. This was frequently undertaken uncritically
and for back pain rather than for sciatic pain. Much of this was performed in North
American private practice. For many patients surgical intervention was inappropriate. We
now know that surgery can be very effective for patients with clearly defined nerve root com-
pression and sciatic pain. However, the place of surgery for patients with back pain without
significant nerve root compression is still to be defined.
Back pain 311
Strengths
1. This study first drew attention to disc herniation as a diagnostic entity.
2. It was the first to provide specific diagnoses for some of our back problems and for some
patients provided an important therapeutic advance.
Weaknesses
1. We now know that there is a mass of disc pathology which correlates very poorly with back
symptoms.
2. The enthusiasm of the authors for the surgical approach led to many unnecessary opera-
tions.
Relevance
This paper was an important milestone in the modern development of our understanding of
back pain problems. Although it provided an important advance it also led to much unneces-
sary and inappropriate surgery.
312
Paper 2
In vivo measurements of intra-discal pressure

Authors
Nachemson A, Morris JM
Reference
Journal of Bone and Joint Surgery 1964; 46(A):1077–1092
Summary
Following pilot studies on cadaveric discs, the authors inserted a specially constructed needle
with a pressure-sensitive polyethylene membrane at the tip into normal lumbar discs (Figure
20.1) in a series of 16 patients with long-standing back problems at other levels, many of
whom had undergone previous spine surgery. They obtained basic data on the intra-discal
pressures at rest, and demonstrated increases on sitting, standing and during various physical
activities. They also were able to analyse the effects of the Valsalva manoeuvre, the applica-
tion of an inflatable corset and the changes associated with spinal fusion at adjacent levels.
They were able to determine the stresses on the disc in different positions and under differ-
ent conditions of rest. They also undertook discography and determined the size of the
nucleus, using this to develop equations calculating the total load on the disc in various pos-
tures and activities.
Figure 20.1 Experimental

design for measuring intradiscal
pressure in vivo.
Amplifier Pressure
transducer
Nitrogen
Calibration Water
Key message
Measurement of intradiscal pressure in living subjects at rest and in a variety of postures and
physical activities is a practical procedure of value for understanding the physiology of the
human spine.
Back pain 313

This was the first study of the physiology and biomechanics of the human spine in life. The
authors measured the intra-discal pressure during a variety of physical activities and showed
that this technique was safe. They were able to record the variations in pressure in various
postures, and in particular showed the increase associated with sitting, with lower values on
reclining.This study and a long series of subsequent papers analysed much more clearly the
effects of posture, weightlifting and so on. They led to a proper understanding of the
ergonomics of the back. Much of the current advice on posture, the provision of lumbar sup-
ports, particularly in car seats, and advice on the right and wrong ways to lift and carry have
been derived from this work.
Strength
For the first time in vivo measurements were made of intra-discal pressures and the changes
associated with various physical activities and postures.
Weakness
This particular study was restricted to patients who already had back problems elsewhere.
Although the study was restricted to the normal discs in these subjects, previous autopsy stud-
ies had shown radical effects on disc mechanics when there was evidence of disc degenera-
tion. In subsequent papers they addressed this issue measuring disc pressures in normal
subjects.
Relevance
This study was the stimulus for much research in ergonomics and in particular the effects of
posture and weightlifting, and in providing appropriate advice and support for patients with
back problems. This research provided a fundamental stimulus to the scientific understand-
ing of the function of the human spine.
314
Paper 3
Intervertebral foramen venous obstruction.

A cause of periradicular fibrosis
Authors
Hoyland JA, Freemont AJ, Jayson MIV
Reference
Spine 1989; 14:558–568
Summary
The authors undertook a cadaveric study of 160 lumbar foramina. They demonstrated that disc
herniation and osteophytic outgrowths into the intervertebral foramen may compress the neur-
al structure. but much more commonly were associated with compression and distortion of the
large venous plexus within the intervertebral foramen. In the absence of direct nerve compres-
sion the most severe neural changes were associated with venous compression, congestion and
resultant dilatation. Pathological changes within and around the nerve root complex included
oedema of the nerve roots, peri and intraneural fibrosis and focal demyelination. Inflammatory
cells were notably absent. The vascular changes within the thickened fibrous sheath around
damaged nerves included basement membrane thickening suggestive of endothelial cell injury.
There was an association between venous compression tissue fibrosis, endothelial injury distant
from the compression, and neuronal atrophy. It was thought that this was probably due to
ischaemia as a result of reduced venous outflow. These observations led the authors to propose
that venous obstruction may be an important pathogenic mechanism in the development of
peri- and intraneural fibrosis and subsequent nerve damage.
Key message
Degenerative disease of the spine and disc herniation are associated with nerve root damage.
This paper draws attention to venous obstruction with impaired nutrition as an important
pathogenic mechanism in addition to the direct effects of pressure.

The study drew attention to the importance of vascular damage in the pathogenesis of nerve
root symptoms. Previously the focus had been on the direct effects of pressure by an osteo-
phyte or a herniated disc on the nerve root. However, it is well recognized that the direct evi-
dence of pressure may not be obvious. At surgery engorged veins around the nerve roots are
often described. This particular study brought together these different observations and indi-
cated an important pathogenic mechanism. It has led onto much more detailed studies on
the role of vascular damage in back pain problems. It also provided an explanation for the
association of spinal problems with cardiovascular disease and with smoking. Both may be
associated with vascular damage and impairment of blood supply within the vertebral canal
and thus play important roles in mediating back pain problems. Further work from this
group together with others analysed the cellular content in proliferating tissue associated
with mechanical herniation. In particular there is cellular infiltration which does not include
polymorphs but which appears to be a reaction to nucleus pulposus and other tissues, again
contributing towards the pathogenic process.
Back pain 315
Strength
This is a systematic study carefully quantifying the degrees of obstruction of the interverte-
bral foramen and statistically relating these to evidence of venous obstruction and in turn to
fibrosis and neural damage.
Weakness
In a cadaveric study it is not possible to relate the venous obstruction and nerve root damage
to pain syndromes in life. Subsequently other workers have critically examined the role
impaired blood flow in vivo and demonstrated its importance.
Relevance
This study indicated important mechanisms of nerve root damage associated with mechani-
cal problems of the spine which are not a direct result of mechanical pressure on the nerves.
It led to consideration of alternative approaches to treatment based upon the evidence of
vascular damage.
316
Paper 4
Nerve ingrowth into diseased intervertebral

disc in chronic back pain
Authors
Freemont AJ, Peacock TE, Goupille P, Hoyland JA, O’Brien J, Jayson MIV
Reference
The Lancet l997; 350:178–181
Summary
The authors examined samples of intervertebral discs obtained at spine fusion examining
specimens from both painful degenerate discs and adjacent normal discs. Cadaveric controls
were also obtained. Immunohistochemical techniques were also used to test for general
nerve markers, Substance P and a protein expressed during axonogenesis (Growth asso-
ciated protein 43(GAP 43)). In healthy control discs, the nerve fibres were restricted to the
outer and middle thirds of the annulus fibrosus. However, in the subjects with chronic low
back pain the nerves commonly extended into the inner third of the annulus and into the
nucleus pulposus. The nerves were usually accompanied by blood vessels although some-
times isolated nerve fibres were seen at disc matrix. Non-associated fibres expressed GAP 43
but both vessel and non-vessel associated fibres expressed Substance P. When comparing
pain and non-pain levels in the spine, again nerve ingrowth was more frequent at the painful
levels.
Key message
The finding of isolated nerve fibres that expressed Substance P deep within diseased inter-
vertebral discs and their association with pain, suggests an important role for nerve growth
into the intervertebral disc in the pathogenesis of chronic low back pain.

There is a correlation between the presence of imaging evidence of disc degeneration and
back pain. However, this relationship is relatively poor. There are patients who may show
advanced damage in the spine and no pain and vice versa. Previous neurophysiological stud-
ies have concentrated on healthy discs and it was demonstrated that innervation is restricted
to the outer part of the disc. This study has shown that there is variable ingrowth of nocicep-
tive fibres into the inner annulus and the substance of the nucleus pulposus and that this
development correlates with the presence of pain. Differences in the growth of nerve fibres
within the disc may well explain why the development of pain varies so considerably between
different people.
Strength
This is the first systematic study using modern immunohistochemical techniques to examine
the distribution of nerve fibres within the disc and to relate these findings to clinical
symptoms.
Back pain 317
Weaknesses
1. It is difficult to obtain control discs for a study such as this. An adjacent unaffected disc
exposed at the time of spinal fusion for a damaged disc was used, but clearly the adjacent
disc may have pathology not easily identifiable.
2. Another set of controls were cadaveric spines. In these circumstances it is not possible to
determine accurately whether back pain has been present in life.
3. There is also a possibility the immunohistochemical appearances have changed following
death, although experimental work suggests that this is unlikely.
Relevance
This study has provided important data on the neurophysiological structure of the healthy
and damaged intervertebral disc.
318
Paper 5
A new clinical model for the treatment of

low back pain
Author
Waddell G
Reference
Spine 1987; 12:632–644
Summary
The author has produced a theoretical framework to understand the natural history of low
back pain. Most episodes of pain should be benign self-limiting conditions. Low back disability
as opposed to pain is a relatively recent Western epidemic. He critically examined the role of
medicine in that epidemic, and constrasted the traditional medical model of the disease with
a bio-psycho-social model(Figure 20.2) and used this to explain many clinical observations.
Although rest is the commonest treatment prescribed after analgesics, there is a little evidence
of long lasting benefit. There is little doubt about the harmful effects of prolonged bed rest.
Conversely, there is no evidence that activity is harmful. It does not necessarily make the pain
worse. Experimental studies have shown that controlled exercises not only restore function,
reduce distress and illness behaviour and promote apparent return to work, but actually
reduce pain. Clinical studies confirm the value of active rehabilitation in practice.
To achieve the goal of treating patients rather than spines we must approach low back dis-
ability as an illness rather than low back pain as a physical disease. As a result we must distin-
guish pain from disability and the symptoms and signs of distress and illness behaviour from
those of physical disease. Management should change from the negative recipe of rest for
pain to more active restoration of function.
Figure 20.2 An illustration

A conceptual model of pain A clinical model of illness of the bio-psycho-social
concept of illness. On the
left is the model devised
Pain behaviour Sick role
by Loeser, and on the
Suffering Illness behaviour right the Glasgow Illness
Pain Distress Model, with an
Nociception Physical problem adaptation by Waddell
to the problems of
chronic back pain.
Related reference (1) Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1956;
150:971–999.
Back pain 319
Key message
Low back disability is not the same as low back pain. The development of disability is a rela-
tively recent Western epidemic, and it is necessary to distinguish the symptoms and signs of
psychological distress and pain behaviour from those of physical disease. When assessing
patients one should not only consider the specific syndromes but also the need to treat the
whole person. In particular one should alter the philosophy of management so that the
patient takes an active role sharing responsibility for his or her own progress, and that there
is a change from rest to rehabilitation and restoration of function.

This was a fundamental publication drawing together a vast range of studies and provided a
reasoning for a holistic approach to back pain problems. He reviewed the epidemiology of
low back pain pointing out that over the years there has been a dramatic increase in the
numbers of working days lost due to back incapacities. In contrast in less developed coun-
tries, although back pain is almost universal, there was very little disability before the intro-
duction of Western medicine.
He identified the need to distinguish the physical problem from the resulting distress and
this in turn from illness behaviour and from the adoption of a sick role. This required a bio-
psycho-social analysis and indicates the importance of psychological and social influences in
modifying individual perception of and response to disease. The gate control theory by
Melzack and Wall (1) may provide a neurophysiological mechanism for this concept. On this
basis he described a series of symptoms and signs which may be used to identify magnified or
inappropriate illness behaviour, and which help to distinguish these factors from those
specifically related to physical disease. With this understanding he looked at the evidence
regarding the use of bed rest. He identified that protracted rest is harmful due to a wide vari-
ety of mechanisms and proposed a much more active approach in which patients’ participa-
tion in the rehabilitative process is fundamental.
Strengths
1. This study drew together a mass of evidence on the epidemiology of back pain and of
back disability.
2. It provided a theoretical construct enabling improved understanding of back disability
and indicating appropriate ways forward.
Weaknesses
1. Although providing an important framework for understanding disability associated with
back problems it has been used to explain virtually all back disability, with a result that
specific diagnoses may be missed.
2. In many patients there is a specific nociceptive problem but the psycho-social factors have
led to a greatly amplified response.
3. The study is useful in avoiding inappropriate intervention procedures when the psycho-
social factors are paramount, but equally it may lead to failure to identify specific underly-
ing problems which are capable of being remedied.
Relevance
This study provided the conceptual framework for much of our current understanding of the
persistence of back distress and disability.
320
Paper 6
A prospective study of work perceptions and

psychosocial factors affecting the report of
back injury
Authors
Bigos SJ, Battie M, Spengler DM, Fisher LD, Fordyce WD, Hanson TH,
Nachemson AN, Wortley MD
Reference
Spine 1991; 16:1–6
Summary
A longitudinal prospective study of 3020 aircraft employees to identify risk factors for report-
ing acute back pain at work. Pre-morbid data including individual physical, psychosocial and
work place factors were recorded over slightly more than 4 years of follow-up. Two hundred
and seventy-nine subjects reported back problems. Other than a history of recurrent or
recent back problems, the factors found to be most predictive of subsequent reports of back
pain in a multivariate model were work perceptions and certain psychosocial responses. In
particular subjects who ‘hardly ever’ enjoyed their jobs were 2.5 times more likely to report
their injury (Figure 20.3). The subjects scoring highest on the Scale 3 (HY) of the Minnesota
Multiphasic Personality Inventory (MMPI) were 2.0 times more likely to report back pain
than subjects with the lowest scores. A history of lack of job task enjoyment, high MMPI
Scale-3 score, and a history of back treatment showed that subjects in the highest risk group
had 3.3 times the numbers of reports compared with the lowest risk group.
Figure 20.3 A diagram relating

100
to the work APGAR (job
% not reporting back injury
enjoyment) versus subsequent

95
job injury. The graph compares
the percentage of subjects
90
reporting subsequent back
problems to their responses to
85 Almost always
the statement ‘I enjoy the tasks
Some of the time
80
involved in my job’. Those who
Hardly ever
‘hardly ever’ enjoyed their job
75 tasks were at highest risk.
0 6 12 18 24 30 36 42 48
Exposure time (months)
Key message
Psychological factors and in particular job dissatisfaction are important predictors of the
future development of back problems.
Back pain 321

Back pain is one of the commonest causes of loss of work. The numbers of working days lost
has rapidly escalated in recent years. Until this study was produced it was thought that the
development of back pain was principally related to mechanical problems associated with
work. Much effort went into ergonomic analyses of the workplace situation and magnitudes
of weights lifted and so on. Despite this the incidence of work loss continues to increase. This
study challenged that view in emphasizing the role of the psychosocial factors. It was well
established that a history of a previous back problem is an important risk factor. It has been
known for some time that patients with back pain may become distressed. Here for the first
time it was shown that job dissatisfaction by itself will predict future back pain.
With this knowledge, epidemiological surveys of back pain and in particular prospective
studies are recognized as incomplete if they do not assess the psychosocial background.
Failure to get on with one’s supervizing officer, the compensation system and medico-legal
claims, operant conditioning, and the role of family and friends play important roles in
encouraging disability. This study emphasized the multifaceted nature of back problems in
industry and has helped explain why previous prevention efforts which focussed purely on
physical factors have been unsuccessful. Jobs that are perceived to be a burden, unenjoyable,
unfulfilling and providing few assests strongly influence the development of future back
problems. People who enjoy their jobs or who are in less emotional distress tend to under-
report back injury.
Strengths
1. This was a prospective study in which the baseline data was recorded prior to the develop-
ment of back pain.
2. It was then possible to relate the development of back problems back to the pre-morbid
data.
Weaknesses
1. This study analysed the report of back injury rather than the development of back pain.
There is an important difference between the two and it would have been valuable to
determine the development of back problems and analyse this data separately from those
who actually report and took time off as a result.
2. The subjects all worked in the Boeing Aircraft factory. It was stated that the study was per-
formed in a diverse, highly sophisticated manufacturing industry with the job tasks tend-
ing not to be extremely stressful for the back. The data with regard to this particular point
is not provided, and the back injury claim rate was found to be similar in fact to other
American industries where back stresses are considered to be higher.
3. The study analysed the reports of back injury. Intuitively one would think that psychoso-
matic factors would be more important in the development of chronicity of back prob-
lems. This has not been analysed.
Relevance
This study has indicated the importance of psychosocial factors in predicting the future
occurrence of reports of back injury.
322
Paper 7
How many days of bed rest for acute low back pain?
Authors
Deyo RA, Diehl AK, Rosenthal M
Reference
New England Journal of Medicine l986; 315:1064–1070
Summary
Two hundred and three patients with mechanical low back pain were randomly asigned to
treatment of either 2 day’s bed rest (Group I) or 7 day’s bed rest (Group II). Although com-
pliance with the recommendation of bed rest was variable, patients assigned to group I
missed 45% fewer days off work than those assigned to group II (3.1 versus 5.6 days,
p = 0.01). No differences were observed in other functional, physiological or perceived out-
comes. For many patients without neuromotor defects, shorter periods of 2 day’s bed rest
may be appropriate. If widely applied, this policy may substantially reduce absenteeism from
work and the resulting indirect costs of low back pain for both patients and employers.
Key message
For patients with acute low back pain, shorter periods of bed rest promoted more rapid and
better recovery than prolonged periods of bed rest.

The traditional management of acute episodes of back pain required prolonged periods of
recumbency with the patient lying flat, usually on a firm bed, and only allowed up for toilet
or washing. At the same time analgesics were provided to provide pain relief. The patient
remained on bed rest until the acute back episode resolved. Mobilization was then undertak-
en very cautiously with great care being taken to protect the back.This was the first study to
challenge this approach to management, and indicated that prolonged periods of immobility
promoted the perpetuation of symptoms and disability. It has led to a rethink about the man-
agement of back pain.
It has been followed by a succession of studies confirming these findings and indicating
the importance of maintaining physical activity and remaining at work. The more recent
studies show that patients with acute back pain should remain at work if at all possible. If
they need to be off work it should be for the shortest period, and if they need rest in bed it
should be for a limited period of time, generally 2 days or less. Prolonged periods of bed rest
are now recognized as promoting the perpetuation of disability. A number of different mech-
anisms are thought to play a part. These include the importance of motion in healing
injured soft tissues and joints; the development of stiffness in relation to immobility;
improvement of the arterial blood supply and venous drainage in association with spine
movement; the development of muscle weakness with prolonged inability, and the psycho-
logical effects of prolonged rest including a fear of injuring the back.
Back pain 323
Strengths
1. A controlled study comparing two regimes of bed rest for patients with acute back pain.
2. The assessors were blinded to the treatment assignment.
Weaknesses
1. This study included patients with acute and chronic pain, some had previous acute
episodes, many had evidence of nerve root irritation.
2. This type of research requires a much larger study examining separately the effects of
acute and chronic back pain, and patients without and with sciatica.
3. Although treatment with 2 and 7 day’s bed rest was recommended, compliance was
limited particularly in the latter group. In terms of practical application, a specific pre-
scription of 2 or 7 day’s bed rest is very difficult as many patients modulate their activities
in relation to their symptoms.
Relevance
This paper led to a fundamental shift in the approach to management of back pain. In partic-
ular it was responsible for the change from recommending prolonged periods of bed rest and
protection of the back to maintaining normal activity if at all possible. Failing that, limited
periods of inactivity and return to normal function as soon as possible are now recommended.
325
CHAPTER 21
Soft tissue disorders

Cathy Speed and Brian Hazleman
Introduction
Soft tissue rheumatic complaints encompass disorders of tendon, ligament, bursa, fascia,
joint capsule and isolated lesions of muscle. Some chronic pain syndromes such as fibromyal-
gia are also included. Although they represent a significant proportion of new patient con-
sultations in rheumatology, they remain poorly understood with respect to their pathologies,
epidemiology and management. The diverse spectrum of these often complex disorders is
frequently underestimated.
Some earlier observations on specific disorders have, until recently, been overlooked. It is
only now, with the use of sensitive imaging techniques such as magnetic resonance imaging
that the importance of these early observations has been highlighted. This is illustrated in
those papers by Smith (Paper 1) and Rathbun and McNab (Paper 2), relating to rotator cuff
pathologies, which are a spectrum of commonly encountered complaints. In contrast, there
have been clinical papers which have had a significant influence upon opinions relating to
soft tissue complaints. Such is the case with the papers presented here by Neer (Paper 3),
Steinbrocker (Paper 5) and Brain, et al. (Paper 6).
Other more scientifically-based papers have helped vastly to improve our understanding of
some soft tissue complaints and were an early recognition of the complexity of the regions
that are affected in soft tissue complaints. These include those papers by Kellgren (Paper 8),
Upton and McComas (Paper 7) and by Lucas (Paper 4). Soft tissue disorders represent a sig-
nificant socio-economic burden as a result of lost productivity and costs of health care.
However the emphasis upon the relevance of soft tissue complaints in the workplace has
shifted from true industrial injury as described in 1945 by Hunter, et al. (Paper 9), to less spe-
cific complaints with major legal and financial implications for the employer, as emphasized
by Hadler in 1986 (Paper 10).
These classic papers help to emphasize not only the complexities of a wide range of com-
plaints, but also of the many challenges which remain to be met in their understanding. Soft
tissue rheumatology, as Dixon described over 20 years ago, remains ‘the great outback of
rheumatology’ and truly is a vast and exciting frontier.
326
Paper 1
Pathological appearances of seven cases of injury of

the shoulder joint; with remarks
Author
Smith JG
Reference
London Medical Gazette 1834; 14: 280
Summary
Smith was credited with the first report on rotator cuff tears, which he published in 1834.
This paper consisted of a report of five post-mortem cases in which rotator cuff pathology
was found in seven shoulders. In each case a tear of subscapularis was demonstrated and in
two of these this was the sole rotator cuff pathology. One of these latter two cases involved a
full thickness tear, the other partial thickness. The age in some cases was not stated, but
included patients in the range of 30–56 years of age.
Smith’s findings were as follows:
• In one case global tear of the cuff was described.

• In one case a partial isolated tear of the subscapularis tendon was described and in anoth-
er an isolated full thickness tear of this tendon was noted.
• In one case full thickness tears of the tendons of supraspinatus, infraspinatus and teres
minor was demonstrated, with a partial thickness tear of subscapularis.
• In three shoulders, tears of subscapularis and supraspinatus were noted.
• In all seven shoulders, pathology of the tendon of the long head of the biceps was noted,
involving either subluxation or rupture, with or without scarring and reattachment.
• Laxity of the capsule and ligaments of the shoulder and stretching of the cuff tendons or
excessive movement of the humeral head was reported in two cases.
• Subdeltoid bursal thickening or open communication with the glenohumeral joint was
described in four shoulders.
• Exostoses, covered with an ‘enamel like’ substance on the undersurface of the acromion
was described in two cases and similar findings also noted on the humeral head in one
case.
• A ‘fracture’ of the acromion, forming an ‘artificial joint’ with fibrocartilage and fibroliga-
mentous capsule was described in three cases.
Related references (1) Hauser EDW. Avulsion of the tendon of the subscapularis muscle.
Journal of Bone and Joint Surgery 1954; X(A):139–141.
(2) Nevasier RJ, Nevasier TJ, Nevasier JS. Concurrent rupture of the
rotator cuff and anterior dislocation of the shoulder in the
older patient. Journal of Bone and Joint Surgery 1988;
70(A):1308–1311.
(3) Gerber C, Krushell RJ. Isolated rupture of the tendon of the sub-
scapularis muscle: clinical features in 16 cases. Journal of Bone
and Joint Surgery 1991; 76(B);371–380.
(4) Ticker JB, Warner JJP. Single tears of the rotator cuff. Orth Clin N
Amer 1997; 1(28):99–116.
Soft tissue disorders 327
Key message
First description of rotator cuff tears (alone or in combination) and associated lesions.
Rotator cuff pathology can involve one or more of the tendons of the rotator cuff. Isolated
tears of subscapularis can occur. Involvement of the tendon of the long head of the biceps is
common. Laxity of the glenohumeral joint should always be considered in patients with rota-
tor cuff pathology.

This paper is considered to be the first to describe tears of the rotator cuff and contains a
description of many of the important features of rotator cuff pathology, ranging from global
cuff rupture to partial thickness tears and involving multiple portions of the cuff or tendons
in isolation.
Whilst isolated tears of the other components of the rotator cuff (supraspinatus, infra-
spinatus and teres minor) have received much attention, it was not until 1954 that isolated
tears of subscapularis were reported (1) and subsequently only rarely did further case reports
appear (2). It was not until the last decade that series of such tears have been reported (3).
Such series have assisted in elucidating indicators of subscapularis tears, such as the age of
the patient (with a non-degenerate cuff), a history of trauma involving hyperextension or
external rotation of the adducted arm and pain with the arm in positions below shoulder
level as well as during overhead activities (4). Specific features on examination include an
increased range of passive external rotation, weakness of internal rotation and a positive lift
off test (3, 4).
Involvement of the tendon of the long head of the biceps in rotator cuff disorders is now
noted to be common and is related to the intimate association between the cuff and the
biceps tendon as the latter runs to insert onto the supraglenoid tubercle of the scapula.
Pathology of the rotator cuff should be expected in cases of rupture of this tendon. Capsular
and ligamentous laxity of the glenohumeral joint is now well recognized and may result in
instability of the shoulder joint and subsequent cuff pathology. It may be related to trauma
(unilateral instability) or congenital laxity (multidirectional instability). Subdeltoid bursal
changes are now looked for routinely during ultrasonography in the evaluation of the
patient with rotator cuff pathology.
Smith’s findings of the ‘enamel’ covered exostoses on the undersurface of the acromion
presumably represented osteoarthritic changes and/or impingement, the latter also being
the likely aetiology to such findings on the head of the humerus. Lastly, was the ‘fracture’ of
the acromion described by Smith simply the acromioclavicular joint?
Strengths
1. Alerted medical community to a wide range of rotator cuff pathologies.
2. Noted some of the associated pathological changes that can occur in rotator cuff lesions.
Weaknesses
1. No background to the cases was available.
2. Macroscopic pathology only.
3. Interpretation of associated pathological changes was limited.
Relevance
Credited as the first published report of rotator cuff tears. Noted associated factors.
Identified isolated tears of subscapularis.
328
Paper 2
The microvascular pattern of the rotator cuff

Authors
Rathbun JB, Macnab I
Reference
Journal of Bone and Joint Surgery 1970; 52(B):540–553
Summary
Rathbun and Macnab performed cadaveric studies to examine the vascular supply of the
supraspinatus tendon and to compare this with that of other tendons of the rotator cuff. A
histological examination of each tendon was also performed. The rotator cuff vascular bed
was filled with radio opaque dye and the dye was allowed to harden. The rotator cuff was dis-
sected and soft tissue radiography with a beryllium window allowed the microvasculature to
be demonstrated. This technique was performed with the arm of the cadaver at the side and
then repeated for the opposite arm with the shoulder abducted.
With the arm adducted, an area of relative avascularity was seen in the supraspinatus ten-
don near its point of insertion and in the superior portion of the infraspinatus tendon, but
not in the other tendons of the rotator cuff. Avascularity was also noted in the tendon of the
long head of the biceps. Histological studies showed degenerative changes, calcification and
rupture in these areas of avascularity, while healthy tendons were seen where blood supply
was good. With the arm in abduction, the vessels within the supraspinatus tendon filled
almost completely, but an avascular zone appeared in the subscapularis tendon. The authors
suggested that with the arm in the adducted and neutrally rotated position, the zones of avas-
cularity were secondary to a ‘wringing out’ phenomenon. The authors also postulated that
the areas of tendon degeneration were preceded by avascularity.
Related references (1) Riley GP, Harrall RL, Constant CR, et al. Tendon degeneration
and chronic shoulder pain: changes in the collagen composi-
tion of the human rotator cuff tendons in rotator cuff tendini-
tis. Annals of Rheumatic Diseases 1994; 53:359–366.
(2) Mosley HF, Goldie I. The arterial pattern of the rotator cuff of the
shoulder. Journal of Bone and Joint Surgery 1963; 45(B):780.
(3) Swiontkowski M, Iannotti JP, Boulas JH, et al. Intraoperative
Assessment of Rotator Cuff Vascularity Using Laser Doppler Flowmetry.
St Louis: Mosby, Year Book, 1990:208–212.
(4) Sigholm G, Styf J, Korner L, Herberts P. Pressure recording in the
subacromial bursa. Journal of Orthopaedic Research 1988;
6:123–128.

The ‘critical zone’ of the supraspinatus tendon is the anterior portion near its point of inser-
tion that appears to be particularly prone to ruptures and calcification. Many of these lesions
are degenerative in nature (tendinosis), with underlying pathological changes including
alteration of collagen fibre type distribution, with a relative increase in type III collagen,
fibrovascular proliferation and microtears (1). The pathogenesis of such lesions is unclear,
but vascularity was long considered to play a central role. Other factors may include age,
genetics and mechanical impingement.
Prior to the publication of this paper in 1970, the critical zone had been described as a
zone of relative avascularity. Although Moseley and Goldie noted in 1963 that the vascular
supply to the cuff was rich in anastamoses between tendinous and osseous vessels, many con-
tinued to consider avascularity to be the primary cause of rotator cuff tendinopathies (2).
Rathbun and Macnab were the first to postulate that these findings may have been an artifact
of positioning. Subsequent studies by others using laser doppler demonstrated substantial
flow and increased flow in the area of cuff tears (3). It is therefore apparent that other fac-
tors play a significant role in the development of rotator cuff tendinopathies. These may
include raised subacromial pressure has been demonstrated with flexion of the shoulder (4).
Strengths
1. Elegant demonstration of vascular supply of the rotator cuff.
2. Rational theory to explain findings.
3. Had an important impact upon theories of the pathogenesis of rotator cuff tendinopathy.
Weaknesses
1. Cadaveric studies.
2. Number of cadavers in the study is not stated.
3. No quantitation of blood supply possible.
Relevance
Avascularity is not the sole factor in the development of rotator cuff tendinopathies. Poor
vascular supply to portions of the cuff may occur with the arm in specific positions.
330
Paper 3
Anterior acromioplasty for the chronic impingement

syndrome in the shoulder. A preliminary report
Author
Neer CS
Reference
Journal of Bone and Joint Surgery 1972; 54(A)[1]:41–50
Summary
Neer described the importance of the anterior third and undersurface of the acromion in
the impingement syndrome. He based these findings on cadaveric studies and surgical obser-
vations. In cadaveric studies he noted that, with elevation of the arm in internal or external
rotation, the critical area of the tendons of the rotator cuff (supraspinatus, occasionally in
association with the infraspinatus and tendon of the long head of the biceps) pass under the
coraco-acromial ligament or the anterior process (but not the posterior two-thirds) of the
acromion. Neer extrapolated these findings to develop a surgical approach to the manage-
ment of 50 patients with partial or complete tears of the rotator cuff or persisting impinge-
ment after lateral acromioplasty, which at that time was the standard surgical approach to
treatment.
He followed up 47 of these patients for a mean of 2.5 years. A satisfactory outcome was
defined as full use of the shoulder, less than 20 degrees of limitation in full overhead exten-
sion and at least 75% of normal strength. This was noted in 15 of 16 partial thickness tears,
19 of 20 full thickness tears and four of 11 patients who had previously had lateral
acromionectomies.
Key message
Description of importance of anterior subacromial impingement in rotator cuff pathologies,
associated radiological findings, and surgical approach to management.

In this paper, Neer popularized the concept of impingement in rotator cuff lesions. He
described three different stages of the impingement syndrome:
1. Reversible oedema and haemorrhage, usually in the younger patient.

2. Fibrosis and tendinitis, particularly in 25–40 year old patient.
3. Bone spurs and tendon rupture, particularly in the patient >40 years of age.
From his cadaveric and operative studies, Neer recognized the importance of the anterior
third and undersurface of the acromion in the impingement syndrome. He noted that, with
elevation of the arm in internal or external rotation, the ‘critical area’ of the tendons of the
rotator cuff (see below) pass under the coraco-acromial ligament and/or the anterior
process, but not the posterior 2/3, of the acromion.
Neer also made several other points in this paper:
• He emphasized the importance of non-surgical management of rotator cuff tendinitis.

• It is extremely difficult to distinguish clinically between partial and full thickness rotator
cuff tears.
• He described what has come to be recognized as Neer’s test: subacromial injection of
xylocaine in partial thickness tears to evaluate the effects of impingement.
• Neer emphasized the importance of recognizing that, in pathologies of the tendon of the
long head of the biceps, operative intervention should also address subacromial impinge-
ment.
• Radiological findings of cysts and/or sclerosis of the greater tuberosity and anterior acro-
mial osteophytes were noted.
Since Neer’s initial description of anterior acromioplasty, several workers have reported
their favourable results using this technique and it remains popular. Arthroscopic acromio-
plasty is also now popular but whether it is superior to open surgery is still unclear.
Strengths
1. Clear rational interpretation of a major mechanism in the development of rotator cuff
lesions.
2. Extrapolation of hypothesis to the development of an operative approach to management
of these conditions.
3. Had a significant impact upon the understanding and management of rotator cuff tears
secondary to impingement.
Weaknesses
1. Based on the author’s observations from cadaveric studies and a cohort of patients with
impingement syndrome, most with rotator cuff tears.
2. The details of these patients are not given.
3. Post-operative follow-up was limited.
4. Outcome based on a subjective scale of ‘satisfactory’ versus ‘unsatisfactory’.
5. No randomized controlled trial was performed.
Relevance
Had a significant impact upon the understanding and management of rotator cuff tears
secondary to impingement. Impingement occurs primarily at the anterior third of the
acromion and its related undersurface. This area must be focussed upon if surgical interven-
tion is considered.
332
Paper 4
Biomechanics of the shoulder joint

Author
Lucas DB
Reference
Archives of Surgery 1973; 107:425–432
Summary
Reviewed the functional anatomy and biomechanics of the glenohumeral joint, basing their
review on the work of Inman, et al. (1944) (1) and Codman (1934) (2). This review empha-
sized the importance of the ‘force couple’ in abduction of the shoulder. This is the mecha-
nism whereby two muscle groups act to abduct the arm, the deltoid acting to elevate the arm
whilst the rotator cuff and long head of the biceps act to stabilize the humeral head. This
review also emphasized the concept of scapulohumeral rhythm, whereby rotation of the
scapula occurs with glenohumeral abduction in order to permit the deltoid to maintain an
optimum length-tension ratio during most of the abduction range.
Related references (1) Inman VT, Saunders J bde CM. Observations on the function of
the clavicle. Clin Med 1946; 65:158–166.
(2) Codman EA. The Shoulder: Rupture of the Supraspinatus Tendon and
Other Lesions In or About the Subacromial Bursa. Boston: Thomas
Todd, 1934.
(3) Poppen NK, Walker PS. Forces at the glenohumeral joint in abduc-
tion. Clinical Orthopaedics 1978; 58:165.
Key message
Combined action of the deltoid and rotator cuff and biceps allows abduction of the arm by
the formation of a force-couple. The deltoid elevates the arm whilst the rotator cuff stabilizes
the humeral head in the glenoid.

The glenohumeral joint is not an inherently stable joint. It is the most mobile joint in the
body, potentially sacrificing stability for this wide range of movement. This review highlight-
ed the fact that abduction by deltoid requires the head of the humerus to be stabilized in the
glenoid fossa by contraction of the muscles of the rotator cuff. It was not until 5 years after
the publication of this review that Poppen and Walker demonstrated the importance of the
combined action of both supraspinatus and deltoid in the initiation and the first 30 degrees
of abduction of the arm (3). If the cuff is insufficient (for example torn or weak), then the
humeral head migrates superiorly and impinges upon the cuff and undersurface of the
acromion. In addition the glenohumeral joint will be unstable, emphasizing the importance
of the rotator cuff as a dynamic stabilizer. This concept allows some insight into the cause of
many rotator cuff tendinopathies. Repetitive tasks involving use of the shoulder can result in
fatigue of the musculature of the cuff and loss of the stabilizing effect. This in turn can lead
to impingement, with repetitive microtrauma leading to tendon lesions.
The relationship between the movement at the glenohumeral joint and rotation of the
scapula during abduction is an important component of normal shoulder function and is
described as scapulohumeral rhythm. Normal scapulohumeral rhythm involves abduction of
the arm to 90 degrees prior to any scapular rotation taking place. Loss of scapulohumeral
rhythm is an important clinical sign in many shoulder disorders and re-establishing this
rhythm a major goal in shoulder rehabilitation.
Strengths
1. Clarification of previous workers’ research.
2. Emphasized the importance of the force couple in elevation of the arm and the role of
the rotator cuff as dynamic stabilizers.
3. Emphasized the importance of scapular rotation and scapulohumeral rhythm.
Weaknesses
1. This was not original work but summarized the findings of others.
2. The biomechanics of the shoulder complex is complicated and many aspects remain
unclear.
Relevance
Stabilization of the humeral head in the glenoid by the rotator cuff is of paramount impor-
tance in normal shoulder function. Rotator cuff tendinopathies can result from, or be the
cause of, loss of this stabilizing action.
334
Paper 5
The shoulder hand syndrome

Author
Steinbrocker O
Reference
Summary
Steinbrocker presented and discussed six cases (five females, one male) of shoulder pain
with ipsilateral swelling of the hand, followed by trophic changes in five of these six. Prior
diagnoses had been made in these cases, including rheumatoid arthritis (RA), connective tis-
sue disease, periarthritis and septic arthritis. He postulated that these findings were due to a
vascular and/or neurological disturbance or to sympathetic dysfunction.
He also listed three stages of evolution. Firstly, an acute or chronic onset of shoulder pain
and generalized swelling and stiffness of hand and fingers, appearing simultaneously or
sequentially. Vascular disturbance was reported in an unspecified number. A history of possi-
ble minor trauma was reported in only one of the six patients in this group. Secondly, over
3–6 months resolution of pain and dysfunction of shoulder and resolution of swelling of
hand, but more pronounced stiffness and flexion deformity of the fingers and notable
osteopenia of the hand and shoulder on plain x-ray. The third stage involves trophic changes
of the hand and flexion contractures of the fingers which may persist for years.
Recovery in this series ranged from 10 months to 7 years, the latter case being only par-
tially recovered when Steinbrocker published this report. Although he noted that Sudeks
atrophy, causalgia, reflex sympathetic dystrophy and post-traumatic osteoporosis are all sug-
gested by this clinical picture, he considered these cases to present more acutely than
Sudek’s and a history of trauma was generally lacking.
Key message
Shoulder hand syndrome is a syndrome of pain, tenderness, and swelling of the hand in asso-
ciation with a painful stiff shoulder. This is now considered to be a form of algodystrophy.
Early diagnosis and intervention are important to prevent progression.

Steinbrocker was the first to recognize the shoulder hand syndrome, the clinical picture of
tenderness, pain, and swelling of the hand in association with shoulder pain and/or adhesive
capsulitis. The onset may be insidious or acute and many cases are related to a medical event
such as a myocardial infarction or cerebrovascular accident or to (often minor) trauma. As
Steinbrocker noted, diagnostic confusion may occur. Shoulder hand syndrome is now con-
sidered to be a form of algodystrophy. The stages that Steinbrocker described are still consid-
ered relevant. Stage I may begin with shoulder pain and stiffness or this may be noted after
the peripheral symptoms. Swelling, pain, tenderness, hyperpathia, allodynia, vasomotor and
sudomotor changes are noted and may persist for up to 6 months. Trophic changes appear
in Stage II and soft tissue atrophy, contractures and osteopenia occur in Stage III. Early diag-
nosis and intervention are important to prevent progression from Stage I. The diagnosis is
often a clinical one, but thermography will show asymmetry of cutaneous temperature, while
radionuclide studies will show an increased uptake in the blood flow and pool phases and a
delay in soft tissue uptake in most early cases. Osteopenia on plain radiography and localized
reductions in bone mineral density on DEXA scanning are seen in chronic cases. Treatment
includes education, reassurance and encouragement to mobilize actively the affected limb.
Pain relief by the use of analgesics, non-steroidal anti-inflammatory drugs (NSAIDs), oral cor-
ticosteroids, tricyclic agents, propanalol, dimethyl sulphoxide, sympathetic nerve blocks and
calcium channel blockers have all been advocated.
Strength
Described a clinical syndrome with cases as illustrations.
Weaknesses
1. No confirmatory investigations.
2. No discussion of treatment.
Relevance
First description of shoulder hand syndrome.
336
Paper 6
Spontaneous compression of both median nerves in

the carpal tunnel. Six cases treated surgically
Authors
Brain WR, Wright AD, Wilkinson M
Reference
The Lancet 1947; 1:277–282
Summary
This paper described non-traumatic bilateral carpal tunnel syndrome and had a significant
impact on the understanding of hand pain. Prior to this paper, median nerve compression at
the wrist had been described almost uniquely in association with trauma. In addition, the
classical symptomatology had not been clarified, with some confusion existing as to the rea-
sons for the relative lack of sensory findings. Brain, et al. described six cases involving middle
aged or elderly women presenting with pain and tingling in the hand in the distribution of
the median nerve. In some cases there was a history of increased use of the hand. Wasting
and weakness of the thenar muscles was noted, with diminished light touch and two-point
discrimination in the median nerve distribution in some. At operation the median nerve was
swollen and pink, with surrounding oedema. Surgical release of the carpal tunnel improved
the symptoms in all cases although there remained some sensory and/or motor
symptoms/impairment in all.
Related references (1) Stopford JSB. British Medical Journal 1926; 1:1028.
(2) Repaci M, Torrieri F, Di-Blasio F, Uncini A. Exclusive electrophysi-
ological motor unit involvement in carpal tunnel syndrome.
Clinical Neurophysiology 1999; 110(8):1471–1474.
Key message
This paper described non-traumatic bilateral carpal tunnel syndrome that led to a significant
improvement in the understanding of hand pain. The clinical picture of pain and tingling in
the distribution of the median nerve with thenar wasting and weakness but only minor senso-
ry findings was emphasized. Operative findings of a swollen, thickened median nerve below
the carpal ligament indicated a ‘neuritis’. Carpal tunnel release was highlighted as an effec-
tive approach to management of this condition.

Prior to this paper median nerve compression at the wrist, though recognized, was poorly
understood. It was considered that such complaint was associated with significant local trau-
ma and was therefore usually unilateral. The symptomatology associated with median nerve
compression, particularly the relative lack of sensory symptoms, led to confusion and it had
been previously postulated that only the motor branch was involved. In this paper the classi-
cal symptoms and signs of bilateral carpal tunnel syndrome were described in a typical group
of patients. Impairment of light touch, pinprick and two-point discrimination was noted and
subsequently these sensory findings, particularly the latter, have been relied on clinically to
gauge the severity of the problem, although correlation with median nerve sensory conduc-
tion findings has not been consistently demonstrated. The authors pointed out that their
findings were in keeping with previously described stages of compression of a mixed nerve
(1). Subsequent studies have confirmed that pure motor fibre involvement is rare (2).
The role of repetitive use of the hand in carpal tunnel syndrome has been confirmed in
some epidemiological and experimental studies (1, 2). However, in reviewing epidemiologi-
cal studies mainly from Finland, Hadler suggested that there was no increased incidence of
carpal tunnel syndrome in shop assistants, assembly line workers nor in light mechanical
industry. Carpal tunnel syndrome is a prescribed disease. That is, the disease is considered to
be a risk of specific occupations (work involving the use of hand-held vibration tools in the
case of carpal tunnel syndrome), providing no other predisposing factors are present in that
individual. Other disorders can be associated with carpal tunnel syndrome. These include
inflammatory arthritides including those induced by crystals, connective tissue diseases, local
space occupying lesions, infections and metabolic and endocrine disorders. Pregnancy and
renal dialysis may also be associated with the complaint.
The authors discuss the differential diagnosis of carpal tunnel syndrome and its
distinguishing features. Differentiation from progressive muscular atrophy, syringomyelia,
cervical spine and costoclavicular syndromes and brachial plexus lesions are all discussed.
They point out that flexion of the wrist does not always increase compression on the nerve,
perhaps explaining the lack of consistency of Phalen’s test. Surgical management, involving
carpal tunnel release, in such patients is described and remains popular today. This tech-
nique had been described earlier by others, but to release the nerve after compression relat-
ing to trauma.
Strengths
1. Presented a review of literature and some of the sources of confusion on the subject of
hand pain.
2. Presented classical cases and provided a clear rationale to explain the clinical pictures.
3. Discussed the differential diagnosis and distinguishing features.
4. Provided information on each case including investigations, operative findings and
follow-up.
Weaknesses
1. Electrodiagnostics were performed only on one case.
2. Observational study of only six cases.
Relevance
Significantly contributed to the understanding of carpal tunnel syndrome. Described typical
cases of idiopathic cases and treatment by carpal tunnel release.
338
Paper 7
The double crush in nerve entrapment syndromes

Authors
Upton AR, McComas AJ
Reference
The Lancet 1973; II:359–361
Summary
This paper discussed the anomalies which exist in some cases of carpal tunnel syndrome,
specifically that there may be no apparent precipitant, that proximal symptoms or a coexist-
ing ulnar nerve syndrome may be present, obvious nerve pathology at surgery may be lack-
ing, poor correlation between surgical findings and the extent of the symptoms, failure of
some cases to respond to carpal tunnel release and the finding of slowing of impulse conduc-
tion proximal to the site of compression.
The authors studied this further in 220 patients presenting with numbness or tingling in
the hand, some with weakness of intrinsic hand muscles. All had motor unit population stud-
ies in the thenar and hypothenar muscles, motor and sensory conduction studies and con-
centric needle EMG of muscles innervated by C4-T1 nerve roots. These investigations indicat-
ed that 85/220 patients had carpal tunnel syndromes, 24 had ulnar neuropathies and six had
both. Seventy percent of these had evidence of a cervical root lesion, on the basis of radiolog-
ical evidence of a cervical vertebral abnormality, neck symptoms, a history of whiplash, clini-
cal evidence of a dermatomal sensory abnormality or EMG evidence of denervation. The
authors postulated that single axons, having been compressed (or excessively stretched) in
one region, become susceptible to damage at another site, due to serial impairment of axo-
plasmic flow.
Related references (1) Narakas AO. The role of thoracic outlet syndrome in the double
crush syndrome. Ann Chir Main Memb Super 1990;9(5):331–340.
(2) Cassavan A, Rosenberg A, Rivera LF. Ulnar nerve involvement in
carpal tunnel syndrome. Arch Phys Med Rehabil 1986;
67(5):290–292.
(3) Dellon AL. Musculotendinous variations about the medial humer-
al epicondyle. J Hand Surg Br 1986; 11(2):175–181.
(4) Golovchinsky V. Double crush syndrome in lower extremities.
Electryogr Clin Neurophysiol 1998; 38(2):115–120.
(5) Richardson JK, Forman GM, Riley B. An electrophysiological
exploration of the double crush hypothesis. Muscle Nerve 1999;
22(1):71–77.
(6) Dahlm LB, Archer DR, McLean WG. Axonal transport and mor-
phological changes following nerve compression. An experi-
mental study in the rabbit vagus nerve. J Hand Surg Br 1993;
18(1): 106–110.
(7) Lundborg G, Dahlm LB. Anatomy, function and pathophysiology
of peripheral nerves and nerve compression. Hand Clin 1996;
12(2):185–193.
Key message
Compression of a nerve root may occur at multiple levels.

The hypothesis helps to explain the anomalies that are seen in the groups of patients
described above and emphasizes the importance of looking for additional lesion(s) in
patients with peripheral nerve entrapments. Proximal lesions include not only nerve root
entrapment but also thoracic outlet syndrome (1). Anatomical variations in the arm such as
the presence of fibrous bands and accessory or overdeveloped musculature have been
demonstrated (2), which may explain the double crush syndrome. Electrophysiological stud-
ies have also supported the existence of such a condition (3).
Double crush syndrome has subsequently been described in the lower extremities, with
peripheral entrapment syndromes occurring in conjunction with proximal nerve lesions at
the lumbar spine (4).
While the concept of double crush syndrome has achieved wide clinical acceptance, others
remain sceptical. Richardson, et al. (5) hypothesized that, since median sensory response is of
C6/C7 origin and the median motor response is primarily of C8 origin, then C6 and/or C7
cases would demonstrate an increased frequency of median mononeuropathy by sensory cri-
teria and C8 cases by motor criteria. However, neurophysiological studies failed to support
this hypothesis.
Strengths
1. Clearly written and provides a concise summary of the anomalies seen in patients with
carpal tunnel syndrome.
2. Provides a rational explanation for the clinical picture(s) seen in this group of patients.
Weaknesses
1. The basis for diagnosis of a cervical root lesion is of uncertain validity, as it is unclear as to
how many of the stated criteria for a cervical root lesion were required. Many of these cri-
teria are non-specific and insufficient to make the diagnosis. However electrophysiological
evidence in 81 patients was provided.
2. As the paper was written before MRI was readily available it is not possible to substantiate
the diagnoses of cervical disc lesions with imaging studies.
3. The hypothesis that impaired axoplasmic flow after initial nerve injury results in an
increase in the susceptibility at a second site is elegant but unsubstantiated. However, it
has been subsequently demonstrated that morphological and biochemical changes can
take place in a compressed neurone (6). These may include alterations in intraneural
microcirculation and nerve fibre structure, and in vascular permeability with resulting for-
mation of oedema and deterioration in nerve function (7).
Relevance
Provides a rational explanation for the clinical picture(s) seen in this group of patients.
340
Paper 8
Observations on referred pain arising from muscle

Author
Kellgren JH
Reference
Clinical Science 1938; 3:176–190
Summary
Kellgren investigated the pain patterns produced by stimulation of deep muscle, fascia and
tendon by injection of small quantities of hypertonic saline. Stimulation of muscle bellies
gave rise to diffuse pain often at a distance from the point stimulated. He concluded from
this part of his study that diffuse pain from a given muscle may be confused with pain arising
from deeper structures. Stimulation of muscle fascia and of tendon was found to be more
sensitive and with a more localized distribution of pain. Kellgren then investigated whether
the distribution of the referred pain followed a spinal segmental pattern and reported this to
be the case. This pattern differs from the segmental innervation of the skin. Kellgren investi-
gated areas of the distribution of referred tenderness on muscle stimulation. He found that
referred pain from muscle is associated with referred tenderness from deep structures.
Kellgren postulated that the mechanism of reference could be due to a common pathway
shared by pain arising from muscular and deep somatic structures.
Related references (1) Simons DG. Muscle pain syndromes (I). American Journal of Physical
Medicine 1975a; 54(6):288–311.
(2) Simons DG. Muscle pain syndromes (I). American Journal of Physical
Medicine 1975b; 55(1): 15–42.
(3) Travell J. Myofascial trigger points: a clinical view. Advances in Pain
Research and Therapy 1976; 1:919–926.
(4) Fricton JR. Myofascial pain. In: Bailliere’s Clinical Rheumatology 1994;
8(4):857–880.
Key message
Pain arising from skeletal muscle can be diffuse and mild to moderate in nature. It follows a
spinal segmental pattern. It may be confused by pain arising from deeper somatic structures.

Kellgren’s paper opened the chapter on the concept of diffuse pain arising from muscle and
the field has since developed significantly. Specific regional non-inflammatory muscle pain
syndromes are now recognized under the term myofascial pain. Kellgren described areas of
muscle that he stimulated which gave rise to diffuse referred pain as muscle points. Other
workers later took the concept described by Kellgren and developed the theme by describing
trigger points (1–3). Trigger points are defined as a localized deep tenderness in a taut band
of skeletal muscle that is responsible for the pain in the zone of reference and if treated will
resolve the resultant pain. The zone of reference is defined as the area of perceived pain
referred from the irritable trigger point, and is usually localized over the trigger point and
may spread out to a distant site.
Myofascial pain has specific clinical characteristics including the presence of trigger points
in a taut band of muscle, pain in a zone of reference, associated factors such as gastrointesti-
nal, visual and otological disturbances, paraesthesiae and dermatographia. Contributing fac-
tors to myofascial pain include trauma, whiplash, psychological stress, the performance of
repetitive tasks, physical illness and sleep disturbance (4). Investigations are normal and the
pathophysiology is not understood, since no specific pathological lesion has been identified.
Treatment involves pain control, stretching, muscle conditioning, postural realignment,
acupuncture and trigger point injections, tricyclic agents, counselling and addressing under-
lying factors.
Strengths
1. Important observations in the pattern of pain arising from muscle.
2. Showed that pain distributions identified by the subjects were reproducible.
Weaknesses
1. Observational study on small numbers.
2. Unclear as to the variation in pain distribution between subjects.
Relevance
The paper identified the distribution and nature of pain arising from muscles and led to the
development of the concept of myofascial pain.
Figure 21.1 The distribution of pain experienced

after stimulation of the gluteus medius muscle.
Points of injection
Muscle pain
342
Paper 9
Clinical effects of the use of pneumatic tools

Authors
Hunter D, McLaughlin AIG, Perry KMA
Reference
British Journal of Industrial Medicine 1945; 2:10–16
Summary
This paper summarized the literature on the clinical effects of pneumatic tools and present-
ed findings of their own series of cases. Osteoarthritis (OA) in the upper limbs and decalcifi-
cation of the carpus were discussed but the emphasis and significance of the paper relates to
a condition described at that time as ‘white finger’, now considered part of the hand-arm
vibration syndrome (HAVS). Clinical features of numbness, stiffness and pallor of the fin-
gers, in particular the fingers upon which the tool is resting are described. Recurrence on
exposure to cold some years after cessation of use of pneumatic tools is noted. Characteristic
associations, such as involvement of the non-dominant hand, exacerbation by cold tempera-
tures and increased incidence and severity with tools that vibrate at frequencies between
2000–3500 strokes/minute were reported. Approaches to prevention are suggested, most
importantly restricting the use of tools with vibration frequencies in the range of 2000–3500
strokes/minute. Other approaches, including warming the compressed air and shock
absorbers are also mentioned, but are noted to have been tried with little success.
Related references (1) International Standards Organization. Guide for the evaluation of
human exposure to whole body vibration. ISO 2631. Geneva: ISO,
1978.
(2) British Standards Institution. British Standards guide to measurement
and evaluation of human exposure to vibration transmitted to the
hand. BSI 6842. London: BSI, 1987.
(3) International Standards Organization. Guidelines for the measurement
and the assessment of human exposure to hand transmitted vibration.
ISO 5349. Geneva: ISO, 1986.
(4) American National Standards Institute. Guide for the measure-
ment and evaluation of human exposure to vibration transmit-
ted to the hand. ANSI S3–S4. New York: ANSI, 1986.
(5) Gemn G, Pyykko I, Taylor W, Pelmear PL. The Stockholm work-
shop scale for the classification of cold-induced Raynaud’s phe-
nomenon in the hand-arm vibration syndrome (revision of
Taylor-Pelmear Scale). Scandinavian Journal of Work
Environmental Health 1987; 13:275–278.
(6) Dupuis H, Riedel S. Experience on the reversibility of the vibration
induced white finger disease. Cent Eur J Public Health 1995; 3
(suppl.):19–21.
(7) Petersen-R, Andersen M, Mikkelsen S, Nielsen SL. Prognosis of
vibration induced white finger: a follow up study. Occup Environ
Med 1995 Feb; 52(2):110–115.
Key message
The syndrome of vibration white finger, later known as part of the HAVS, is a significant,
incapacitating disorder related to the use of vibrating hand-held tools. Other musculoskele-
tal conditions such as hand and wrist OA can also arise due to the use of such machinery.

The HAVS, consisting of vascular, neurological, and articular disorders (OA) occurring in
the upper limbs of vibration-exposed workers as described in this paper, is now well recog-
nized. It became an industrial prescribed disease in 1985. Vibration can be described by its
frequency, amplitude, acceleration and direction in addition to the exposure time. The
authors reported higher risk with specific frequencies and the medical risks are considered
to be directly related to the severity of the vibration and the total vibration exposure time.
However, there are too few epidemiological data to enable reliable conclusions to be drawn
about precise exposure-response relationships for any of the complaints which form part of
the HAVS. As a result, there are four current recommended industry standards for tools: BSI
6842 (1987) (2), ISO 5349 (1986) (3), ANSI S3–34 (1986) (4), ACGIH (1992–93) (4).
In the early literature describing HAVS, dating from the early 1900s, it was given numer-
ous labels, including vibration white finger (VWF), spastic anaemia, dead fingers, traumatic
vasospastic disease and occupational Raynaud’s phenomenon. In this paper Hunter, et al.
emphasized the importance of differentiating VWF from primary Raynaud’s phenomenon.
In VWF, persistent paraesthesia and pain are common in the hands and arms and these
occur independently from the ‘white attacks’. There are no stages of hyperaemia or cyanosis.
Subjective reports can correlate poorly with objective findings. The severity of HAVS can be
assessed using numerous scales, including the Stockholm classification (5). However there is
currently no simple, reliable objective diagnostic test available.
The aetiology of the disorder remains unclear but appears to be related to blood vessel
and nerve (including mechanoreceptor) damage, which occur simultaneously but indepen-
dently of each other.
The reversibility of HAVS has been studied. A reduction in or cessation of exposure to
vibration may have a beneficial effect on finger-blanching symptoms and the cold response
of digital vessels but this may not occur for some years (6). Further research has implied that
continued work with high vibration hand-held tools, smoking, other circulatory diseases, and
low age at the time of diagnosis have an unfavourable influence on the prognosis (7).
Strengths
1. Clearly written, with practical messages.
2. Support their literature review with a large series of their own cases.
Weaknesses
1. Little discussion of pathogenic mechanisms.
2. Literature review and report of a case series rather than randomized controlled trial.
Relevance
Interpretation of the literature, supported by a large case cohort study, on the syndrome now
recognized as HAVS, which became an industrial prescribed disease 40 years after the publi-
cation of this paper.
344
Paper 10
Industrial Rheumatology. The Australian and

New Zealand experience with arm pain and
backache in the workplace
Author
Hadler NM
Reference
Medical Journal of Australia 1986; 144:191–195
Summary
In a critique of the concept of occupational soft tissue ‘disease’, Hadler provides an overview
of the clinical and socio-political environments that confound the management of soft tissue
complaints in the work place. He uses backache and arm pain to illustrate the salient points.
From the clinicians’ viewpoint, the diagnosis of work incapacity is straightforward and made
when the patient states that he/she cannot work. There then follows the period of determi-
nation as to whether a specific disorder is present and a management strategy is devised,
which may include alteration of factors associated with the workplace.
Socio-political and legal aspects have significantly added to the complexity surrounding
work incapacity. These issues address whether the relationship of cause (work) and effect
(incapacity) exists and if so what are the specific characteristics of each. Such issues are laden
with emotional, social, financial and bureaucratic factors that the clinical care of the patient
can become lost in the complexities of the label ‘work incapacity’. Hadler points out that
there is no evidence that back pain in the workplace is any more common than in the gener-
al population (1). He also comments upon the increasing frequency in claims for industrial
compensation in Australia, despite significant ergonomic improvements in the workplace
and the findings that arm pain in the workplace is less common than in the general popula-
tion in England and in the USA. Hadler emphasizes the need to consider not only physical
activities (static muscle contractions, repetitive activity and direct loading), but also the
broader working environment (social, financial, emotional and psychological factors, poor
working conditions and insufficient supervision) in the evaluation of musculoskeletal com-
plaints in the work place.
Related reference (1) Hadler NM. Industrial rheumatology: clinical investigations into
the influence of the pattern of usage on the pattern of regional
musculoskeletal disease. Arthritis and Rheumatism 1977;
20:1019–1025.
Key message
Hadler emphasizes the need for the medical profession to focus upon the clinical responsi-
bilities and the search for an explanation for work-related disorders such as arm pain.

In 1977 Hadler coined the term ‘industrial rheumatology’ in order to highlight the need for
the clinician to address discrete musculoskeletal complaints which led to work incapacity. In
this paper, written 9 years later, he noted that this field has been profoundly affected by sur-
rounding socio-political issues, in particular financial compensation. The clinician’s role in
care of the patient with a musculoskeletal complaint which may be related to work, includes
diagnosis, identification of the cause(s) if possible, and management strategies. These may
include alteration of ergonomic factors and consideration whether cessation of work is nec-
essary and if so, what effects will it have on the patient. However these ideals are often
obscured by the issue of ‘who is to blame’ and financial compensation. Hadler emphasizes
the need for the medical profession to focus upon the clinical responsibilities and the search
for an explanation for work-related disorders such as arm pain. While use-associated arm
pain has been given many labels, the term ‘repetitive strain injury’ is frequently coined, infer-
ring damage. However, no conclusive pathological findings have been noted. This emotion
(and legal) laden term should therefore be avoided.
Strengths
1. Review of the salient clinical and socio-political issues surrounding musculoskeletal com-
plaints and the workplace.
2. Strong reminder to the clinician to focus upon his/her role of care for the patient.
Weaknesses
1. Personal opinion.
2. Unclear in parts.
Relevance
Hadler is one of the few rheumatologists who have been able to consider regional back pain
syndromes such as back and arm pain within the biopsychosocial model of disease, as
opposed to the purely biomedical model. As a result, he recommended abolition of the term
‘repetitive strain injury’.
346
CHAPTER 22
Fibromyalgia
Hugh Smythe
Introduction
Fibromyalgia is complex but we have learned much about it in the last three decades, espe-
cially in the last 10 years. I began with a list of 53 ‘classic’ papers that I thought were essential
to cover all relevant aspects of the subject, but have been obliged to reduce these to six main
papers. I regret the omission of essential contributions by colleagues whose friendship I value
highly. I have limited my selections to a common theme, based around the observations on
referred pain of Kellgren and his colleagues (1, 2, 3). It is clear that the pain of fibromyalgia
arises neither from the tissues complained of, nor from the unsuspected but characteristic
sites of deep pain and tenderness found by the examiner. It is also clear that neural mecha-
nisms are involved in affecting the severity and chronicity of pain, and the presence of associ-
ated symptoms. So in dealing with fibromyalgia we must deal with referred pain and tender-
ness and with amplifying factors.
Beginning their work in 1936, Kellgren and Lewis produced pain by injecting hypertonic
saline into their own deep somatic tissues, and those of blindfolded volunteers. The pain
experienced was referred, interpreted as arising from structures remote from the stimulus.
They later injected interspinous ligaments and found that pain was referred distally in distri-
butions that were different from segmental dermatomes and the term ‘sclerotomes’ was sub-
sequently used (4). The referred pain was accompanied by referred tenderness. The work of
Kellgren was confirmed by others (5, 6), but was subsequently largely forgotten by rheuma-
tology, as well as by medicine in general. But Kellgren recognized the all-important fact that
much pain could arise from the spine, saying ‘There is little doubt that in many cases
obscure pains in the chest and abdomen are due to diseases of the spine. Thus in every case
it is well to preserve an open mind at the outset and to examine carefully all the structures
which, by virtue of their segmental innervation, could give rise to the pain under considera-
tion’. Kellgren’s chapter in Copeman’s Textbook of Rheumatic Diseases (3) summarizes the early
work and thoughtfully develops those aspects of relevance to the clinician and is my first
‘classic’ reference that I believe should be read by all physicians*.
More recently, the pioneering experimental work of Moldofsky and his colleagues led to
the modern definition of fibromyalgia (Paper 1), based on the findings of tender spots and
sleep disturbance. This work caused a lot of controversy and discussion about the reality of
fibrositis or fibromyalgia, largely laid to rest by the work of the American College of
Rheumatology (ACR) who produced robust classification criteria (Paper 2). Such criteria
help patient classification for the research studies needed into pathogenesis and treatment.
Papers 3 and 4 are experimental studies into the pathogenesis of fibromyalgia, each of which
suggest that spinal and other more central mechanisms are important. The final group of
papers (Papers 5a–c) bring us back, full circle, to the work of Kellgren, as they indicate that
the spine is the likely source of pain in fibromyalgia.
Fibromyalgia 347
*Editors’ note: Four different chapter authors selected the pain work of Kellgren as a ‘classic’
relevant to their area of expertise. One of the original papers is reviewed in the chapter on
soft tissue disorders. In the other chapters, to avoid duplication, we have either omitted the
reference and/or pointed out its importance through the introduction, as in this case.
References
1. Kellgren JH. Observations on referred pain arising from muscle. Clinical Science 1938;
3:175–190.
2. Kellgren JH. Deep pain sensibility. The Lancet 1949; 1:943–949.
3. Kellgren JH. Pain. In: Textbook of Rheumatic Diseases (Copeman W, ed.). 3rd edition.
Edinburgh and London: Churchill Livingstone, 1964.
4. Travell J, Bigelow NH. Referred somatic pain does not follow a single segmental pattern.
Federation Proceedings 1946: 5:106.
5. Inman VT, Saunders JbdeCM. Referred pain from skeletal structures. Journal of Nervous
and Mental Diseases 1944; 99:660–667.
6. Hockaday JM, Whittey CWM. Patterns of referred pain in the normal subject. Brain 1967;
90:485–496.
348
Paper 1
Two contributions to understanding of the

‘fibrositis’ syndrome
Authors
Smythe HA, Moldofsky H
Reference
Bulletin of Rheumatic Diseases 1977; 28:928–931.
Summary
There is a definable entity to which the traditional description of the ‘fibrositis syndrome’
may be applied. The purpose of this paper is to describe our observations and opinions in
order to stimulate others to challenge and possibly modify them.
The patients in our studies showed the invariable association of symptoms of chronic
aching, a non-restorative sleep pattern with marked morning stiffness and fatigue, the EEG
finding of alpha intrusion in non-REM sleep, and localized tenderness at 12 or more of 14
specific sites. These associations constitute a set of criteria for diagnosis of ‘fibrositis’ more
rigorously defined than those we have previously proposed and used. These criteria were not
fulfilled in many subjects with other diffuse soft tissue pain syndromes, such as post-exercise
myalgia, and myalgia accompanying viral infections. The ‘fibrositis syndrome’ is not merely
synonymous with chronic widespread soft tissue pain.
Related references (1) Moldofsky H, Chester WJ. Pain and mood patterns in patients with
rheumatoid arthritis. Psychosomatic Medicine 1970; 32:309–318.
(2) Moldofsky H, Scarisbrick P, England R, Smythe H. Musculoskeletal
symptoms and non-REM sleep disturbance in patients with
‘fibrositis syndrome’ and healthy subjects. Psychosomatic Medicine
1975; 37:341–351.
(3) Moldofsky H, Scarisbrick P. Induction of neurasthenic muscu-
loskeletal pain syndrome by selective sleep stage deprivation.
Psychosomatic Medicine 1976; 36: 35–44.
(4) Lentz MJ, Landis CA, Rothermill J, Shaver J. Effects of slow wave
sleep disruption on musculoskeletal pain and fatigue in middle
aged women. Journal of Rheumatology 1999; 26:1586–1592.
Key message
Fibrositis is a distinctive condition characterized by tender spots and disturbed sleep as well
as pain and fatigue.
Fibromyalgia 349

This paper, which was a brief summary of experimental studies published in the related ref-
erences, brought fibrositis to the attention of modern rheumatologists. There were two criti-
cal observations. First the sleep disturbance. The finding of alpha intrusion in non-REM
sleep was unexpected and was the first association of a measurable abnormality strongly asso-
ciated with the pain syndrome. The second was the tender spots. Where did the list come
from? Briefly preference was given to points that were generally unknown to patients but
commonly found (Figure 22.1). Many of these sites had been attributed to bursitis, ten-
donitis, costochondritis or other local conditions not confirmed by local pathology. The
sharply defined and remarkably predicatable location permitted definitions repeatable by
other observers.
Figure 22.1 A diagram of some of the

tender spots found in patients with
fibromyalgia and first described in this
paper (From Dieppe P, Bacon P, Bamji
A, Watt I (eds.), Slide Atlas of
Rheumatology, London: Gower, 1987.)
Strengths
1. A superb summary of a lot of excellent experimental work.
2. The comparison of ‘fibrositis’ patients with others with generalized pain.
3. Findings that have stood the test of time.
Weaknesses
1. The studies were small in number.
2. The methods were not all fully validated at this time.
3. The inferences were controversial.
Relevance
The paper that ushered in the modern concept of fibromyalgia.
350
Paper 2
The American College of Rheumatology 1990

criteria for the classification of fibromyalgia;
report of the multicentre trial committee
Authors
Wolfe F, Smythe HA, Yunus MB
Reference
Summary
To develop criteria for the classification of fibromyalgia we studied 558 consecutive patients:
293 patients with fibromyalgia and 265 control patients. Interviews and examinations were
performed by trained, blinded assessors. Control patients for the group with primary
fibromyalgia were matched for age and sex, and limited to patients with disorders that could
be confused with primary fibromyalgia. Control patients for the group with secondary-con-
comitant fibromyalgia were matched for age, sex and concomitant rheumatic disorders.
Widespread pain (axial plus upper and lower-segment plus left- and right-sided pain) was
found in 91.6% of all patients with fibromyalgia and in 69.1% of all control patients. The
combination of widespread pain and mild or greater tenderness in >11 of 18 tender sites
yielded a sensitivity of 88.4% and a specificity of 81.1%. Primary fibromyalgia patients and
secondary-concomitant fibromyalgia patients did not differ statistically in any major study
variable, and the criteria performed equally well in patients with and those without concur-
rent rheumatic conditions. The newly proposed criteria for the classification of fibromyalgia
are: (1) widespread pain in combination with (2) tenderness at 11 or more of 18 specific ten-
der point sites. No exclusions are made for the presence of concomitant radiographic or lab-
oratory abnormalities. At the diagnostic or classification level the distinction between prima-
ry fibromyalgia and secondary-concomitant fibromyalgia (as defined in the text) is
abandoned.
Related references (1) Campbell SM, Clark S, Tindall EA, Forehand ME, Bennett RM.
Clinical characteristics of fibrositis. A ‘blinded’ controlled study
of symptoms and tender points. Arthritis and Rheumatism 1983;
26:817–824.
(2) Wolfe F, Allen M, Bennett RM. The fibromyalgia syndrome: a con-
sensus report on fibromyalgia and disability. Journal of
Key message
Criteria for the classification of fibromyalgia have been produced from empirical research on
carefully selected patient groups.
Fibromyalgia 351

There was much of clinical value in this study. A history of headache, numbness, irritable
bowel or bladder, and dysmenorrhoea was more common in the fibromyalgia group. The
best sites were trapezius, lateral elbow, 2nd rib, low cervical, and inner knee. Some earlier cri-
teria had showed high specificity but low sensitivity (1, 2), and were of value in studies requir-
ing a high degree of diagnostic certainty. However, the value of these criteria were in the
identification of groups of patients for entry into research studies, uncontaminated with pre-
judgments about cause. For these purposes they have served very well, permitting thousands
of studies involving many disciplines in many countries.
Strengths
1. Large, well conducted clinical study.
2. Good description of many of the clinical features of fibromyalgia.
3. Excellent classification criteria for research studies.
Weaknesses
1. Relatively poor specificity mean that they are not useful for individual patient diagnosis.
2. Ceiling and floor effects mean that they are of limited value in determining relative
severity and associated disability.
Relevance
Prior to this work there had been great debate as to the nature and diagnostic features of
fibromyalgia. These criteria are now the accepted ones for classifying the condition.
352
Paper 3
Elevated levels of substance P and high incidence

of Raynaud phenomenon in patients with
fibromyalgia: new features for the diagnosis
Authors
Vaeroy H, Helle R, Forre O, Kass E, Terenius L
Reference
Pain 1988; 32:21–26
Summary
In 30 patients with diagnosed fibromyalgia, the cerebrospinal fluid (CSF) level of immunore-
active substance P (SP) was investigated. Compared to normal values (9.6 +/- 3.2 fmol/ml),
all patients had elevated levels of SP (36.1 +/- 2.7 fmol/ml, range 16.5–79.1). SP levels were
significantly higher in patients who were smokers. We propose elevated CSF levels of SP and
Raynaud phenomenon as characteristic features of fibromyalgia with potential as diagnostic
markers of the disease, and further that smoking might be an aggravating factor for its patho-
genesis and development.
Related references (1) Russell IJ, Orr MD, Littman B, Vipraio GA, Alboukrek D, Michalek
MJ, Lopez Y, MacKillip F. Elevated cerebrospinal fluid levels of
substance P in patients with the fibromyalgia syndrome. Arthritis
and Rheumatism 1994; 37:1593–1601.
(2) Smythe HA, Gladman A, Mader R, Peloso P, Abu-Shakra M.
Strategies for assessing pain and pain exaggeration: controlled
studies. Journal of Rheumatology 1997; 24:1622–1629.
(3) Bradley L, Alarcon GS, Sotolongo A, Weigent DA, Alberts KR,
Blalock JE, Kersh BC, Domino ML, De Waal D. Cerebrospinal
fluid levels of substance P are abnormal in patients with
fibromyalgia regardless of traumatic or insidious pain onset.
Arthritis and Rheumatism 1998; 41(suppl.9): S256 (abstract).
(4) Welin M, Bragee B, Nyber F, Kritansson M. Elevated substance P
levels are contrasted by a decrease in meta-encephalin-arg-phe
levels in CSF from fibroymalgia patients. Journal of
Musculoskeletal Pain 1995; 3(suppl.1): 4 (abstract).
Key message
Cerebrospinal levels of SP are elevated in patients with fibromyalgia.
Fibromyalgia 353

This was the first study describing increased levels of a pain neurotransmitter. Subsequent
studies (1–4) have confirmed the increase in CSF SP. It is an attractive idea that SP release is
due to pain of peripheral origin, but other possibilities exist – either the SP or the release
stimulus could come from higher levels. It still remains to be confirmed that the (pain) stim-
ulus for the release of SP originates peripheral to the central nervous system.
Strength
The first description of a subsequently confirmed and interesting observation.
Weaknesses
1. The use of control values from the literature, not their laboratory.
2. Unjustified speculation about cold hands.
3. Technical concerns.
Relevance
The first description of a chemical (neurotransmitter) abnormality in patients with
fibromyalgia, suggesting possible new approaches to treatment as well as diagnosis.
354
Paper 4
Diagnostic epidural opioid blockade in primary

fibromyalgia at rest and during exercise
Authors
Bengtsson M, Bengtsson A, Jorfeldt L
Reference
Pain 1989; 39:170–180
Summary
Nine patients with primary fibromyalgia participated. The patients were studied prior to, and
immediately after four identical periods of exercise (bicycle ergometer) each performed 30
minutes after injection with saline, repeated saline, an opioid, and naloxone. All substances
were given epidurally, except for naloxone, which was given intravenously. Finally, with the
patients resting in bed, lignocaine was injected epidurally. Phsiological variables, general
exertion, dyspnoea, lower extremity exhaustion, pain and tender points in the lower half of
the body were examined.
Resting pain and tender points diminished slightly after the opioid injection. Lignocaine
completely abolished resting pain and tender points. Lower extremity exhaustion was
reduced by the opioid. General exertion and dyspnoea were unaffected by the opioid. In
conclusion the results support the hypothesis that the pain in fibromyalgia is of peripheral
nocioceptive or spinal origin. We raise the hypothesis that the fatiguability is, at least partly,
due to inhibition because of pain.
Key message
Epidural injections of lignocaine abolish the pain and tender points of fibromyalgia.
Strength
Well designed experimental study.
Fibromyalgia 355
Weaknesses
No control patients with generalized pain but without fibromyalgia.
Relevance
The fact that the pain and tender spots can be abolished by epidural lignocaine suggests that
the pain may arise from spinal or more peripheral sites.
356
Paper 5a
Increased rates of fibromyalgia following

cervical spine injury
Authors
Buskila D, Neumann L, Vaisberg G, Alkalay D, Wolfe F
Reference
Paper 5b
Function of the hypothalamic-pituitary-adrenal axis

in patients with fibromyalgia and low back pain
Authors
Griep EN, Boersma JW, Eef GWM, Lentjes A, de Kloet ER
Reference
Paper 5c
The C6-7 syndrome – clinical features and

treatment response
Author
Smythe HA
Reference
Fibromyalgia 357
Summary
These three papers are grouped together to make a common point. After cervical injury, all
of the extra tender points required to meet criteria for fibromyalgia were in the upper body
(Buskila, et al. 1997). The patients studied by Griep, et al. were selected from patients with
chronic low back pain. In those meeting criteria for fibromyalgia the extra points were pre-
dominantly in the lower body (Griep, et al. 1998). The patients in the C6–7 syndrome paper
were all previously documented to have tenderness at the upper body sites included in the
ACR criteria for fibromyalgia (91 of 151 met all criteria). They had lost the upper body ten-
derness with support to the lower neck throughout sleep. They remained symptomatic, with
a different pattern of upper body tender points; and most responded to adjusted strategies
(Smythe 1994). Together these three papers point to specific regions of the spine as the site
of origin of the referred pain in fibromyalgia.
Key message
Fibromyalgia may arise from spinal problems, and effective management may depend on
treatment of spinal abnormalities.

These papers indicate that the spine is the origin of the pain in fibromyalgia.
Rheumatologists are not comfortable with neck and back problems. The literature indicates
that our colleagues in other disciplines equally fail to help those with chronic neck and back
problems so that referral is not yet a useful option. Failure to identify and address these
problems may account for treatment failures; fitness programmes cannot succeed if the
patient does not have a safe neck and low back.
Strengths, weaknesses and relevance

These papers are concerned with selected groups of patients, and the findings need to be
confirmed by others. The proof of the intriguing concept that spinal problems may underlie
fibromyalgia in many patients will come if other groups can show that fibromyalgia can be
treated through management strategies aimed at spinal problems.
359
CHAPTER 23
Geographical disorders
Adewale Adebajo
Introduction
Rheumatic disorders are not the same all over the world. Rheumatology research in the 20th
century has been dominated by work from the developed world, resulting in an inevitable
emphasis on the diseases as seen in the countries of origin of the researchers. But much can
be learnt by examining rheumatic disorders in other parts of the world, where there are
some different diseases, as well as a different expression of some of the most common forms
of arthritis. These differences can have either genetic or environmental causes.
The first paper in this selection describes a classical epidemiological study carried out in
the developing world (the Philippines) by the WHO-ILAR-COPCORD group, a study group
formed to aid understanding of the importance of musculoskeletal diseases throughout the
world. This model epidemiological study emphasizes the high frequency of rheumatic com-
plaints in the developing world.
The next 3 papers are concerned with the altered expression of common rheumatic dis-
eases in different countries. Paper 2 is concerned with gout, which is particularly common in
Polynesians; the importance of this particular contribution is that it suggests that there is a
strong genetic component to this susceptibility. Paper 3 is about rheumatoid arthritis (RA) in
Africa. This is a fascinating subject, which has been investigated by several groups. The cho-
sen paper is one of the earliest contributions, indicating that RA, along with other ‘autoim-
mune disorders’ is relatively uncommon in Africa. Subsequent work has suggested that this is
in part related to rural (as opposed to urban) living, and that things might now be changing,
with the prevalence of RA currently increasing in some parts of the developing world, where-
as it may be decreasing in the developed world. These investigations are of potential impor-
tance to our understanding of possible environmental triggers to RA. Paper 4 discusses geo-
graphical variations in another common form of arthritis, osteoarthritis (OA). This classic
contribution was one of the first to point out that hip OA is relatively uncommon in Hong
Kong Chinese, in spite of the fact that Heberden’s nodes (sometimes used as a surrogate
marker of a genetic predisposition to the disease) are common in this racial group. The
authors speculate that squatting may be an important factor in protection of the hip.
The final three papers in this selection concern diseases which are specific to certain
regions of the world. There are several such disorders, but we have limited our choice to
three; Kashin-Beck disease (Paper 5) and Mseleni disease (Paper 6) are two ‘endemic’ forms
of OA, first thought to be due to dietary toxins, but now thought more likely to be forms of
dysplasia or avascular necrosis, and to be more likely to be genetic rather than environmen-
tal in origin. Ross River arthritis (Paper 7) is an example of a form of infectious arthritis lim-
ited in distribution by the fact that it depends on the presence of a mosquito born virus.
However, this classic paper draws attention to the fact that travel and migration mean that we
have to be aware of the possibility of such disorders anywhere in the world now.
There are many other forms of arthritis limited to certain regions of the world that might
have been included, such as Blount’s disease. But this selection includes what we believe to
be classic papers, as well as giving a reasonable overview of the spectrum of rheumatic disor-
der seen in different countries, and the reasons for these differences.
360
Paper 1
Rheumatic disease in a Philippine village II: WHO-

ILAR-APLAR COPCORD Study. Phases II and III
Authors
Wigley R, Manahan L, Muirden KD, Caragay R, Pinfold B,
Couchman KG, Valkenburg HA
Reference
Rheumatology International 1991; 11:157–161
Summary
A survey of rheumatic complaints in a remote village area of the Philippines was carried out.
In phase one a screening questionnaire identified 269 adults (comprising 131 males and 138
females) as having rheumatic symptoms out of 950 adults (comprising 482 males and 468
females) studied. In phase two, 234 (87%) of positive respondents were re-questioned using
a more detailed proforma. This detailed proforma identified 196 adults with peripheral joint
pains, 67 adults with neck pain and 137 adults with back pain. One third of adults attributed
their symptoms to work and 127 adults had to stop work because of their complaints.
Disability including an inability to carry loads affected 1.8% of the study population. It was
observed that 82% of adults with rheumatic complaints still required help for their symp-
toms. In Phase three of this study, 166 adults with rheumatic symptoms were medically exam-
ined. Osteoarthritis of the knee was found in 25 adults and 17 adults had Heberden’s nodes.
There were 16 adults with epicondylitis, 16 adults with rotator cuff pain and 35 adults were
diagnosed as having levator scapulae insertion pain.
Related references (1) Darmawan J, Muirden KD, Wigley R, Valkenburg HA. Arthritis
community education by leather puppet shadow play in rural
Indonesia. Rheumatology International 1992: 12:97–101.
(2) Wigley R. Primary prevention of rheumatic disease. Journal of
Key message
This paper is the landmark study of the WHO/ILAR Community Orientated Programme for
the Control of Rheumatic Diseases (COPCORD). This paper has provided the template for
further community studies around the world.
Geographical disorders 361

The WHO/ILAR (COPCORD) was born out of a recognition by the World Health
Organization (WHO) that rheumatic diseases in general are insufficiently appreciated and
understood. In particular the social and economic impact of rheumatic diseases, particularly
in the developing world, is largely unknown due to an information deficit. In partnership
with the International League against Rheumatism (ILAR), the WHO believes that crucial to
tackling this issue is a need to overcome the information deficit.
COPCORD surveys have now been conducted in the Asia/Pacific region, South America
and similar projects in Africa are scheduled. The main principles behind the COPCORD
projects are that such surveys should be applicable to the majority of the world’s population,
which means rural communities in developing countries. Furthermore, they should also be
community-orientated rather than hospital-based. These surveys should emphasize existing
resources and have a low cost structure. They should consider disease prevention and avoid
ignoring traditional non-western health care methods.
This paper reflects these core COPCORD principles and provides a reference for the
design of such COPCORD surveys. The paper describes the three phases involved in the sur-
vey, consisting of an initial simple screening questionnaire followed by a more detailed pro-
forma and finally, clinical examination of positive responders together with the use of avail-
able investigations. This study also draws attention to the concept of rheumatic disease pre-
vention and control in the community.
Strengths
1. Provides a reference guide for community surveys of rheumatic complaints in developing
countries.
2. It describes very clearly the difficulties encountered in carrying out these surveys in devel-
oping countries.
Weaknesses
1. A small section makes reference to children which is confusing as the methodology was
aimed at adults.
2. No economic data is included.
3. The authors were unable to provide a reliable estimate of population prevalence.
Relevance
This landmark COPCORD study has enabled clinicians in various parts of the world to con-
duct community surveys to determine the frequency and pattern of rheumatic complaints in
their region, and to make comparisons across regions. This study also raised the concept of
rheumatic disease prevention and control in the community. Information derived from these
studies are crucial for health service planning and the allocation of health resources, particu-
larly in the developing world (1, 2).
362
Paper 2
Hyperuricaemia, gout and diabetic abnormality

in Polynesian people
Authors
Prior IA, Rose BS, Harvey HP, Davidson F
Reference
The Lancet 1966; 1:333–338
Summary
The prevalence of hyperuricaemia and gout in three groups of Polynesians living in different
environmental conditions in the Pacific were studied. The three groups consisted of New
Zealand Maoris and two groups of Maoris living in the Cook Islands. One group of Cook
Island Maoris lived in Raratonga, an urban population, and the other group lived in
Pukapuka, a rural population. Overall, 755 New Zealand Maoris, 471 Raratongans and 379
inhabitants of Pukapuka were studied. More than 40% of both males and females in all three
groups were hyperuricaemic. The attack rate for gout amongst the males was 10.2% among
the New Zealand Maoris, 2.5% among Raratongans and 5.3% among the Pukapukans. Ten
of the 38 gouty males also had diabetes which developed after the onset of gout in all but
one patient. Obesity and hypertension were common among the New Zealand Maoris with
gout, but not among the Cook Islanders. These findings suggest the influence of important
genetic factors among Polynesians contributing to hyperuricaemia, gout, and diabetes.
Related reference (1) Zimmet PZ, Whitehouse S, Jackson L, Thomar K. High prevalence
of hyperuricaemia and gout in an urbanised Micronesian popu-
lation. British Medical Journal 1978; 1:1237–1239.
Key message
The high prevalence of gout in Polynesians is probably genetic in origin.

This is the first study of Maoris with the same ethnicity living in three different environmen-
tal conditions. Despite their different environmental exposure all three populations had a
high prevalence of hyperuricaemia suggesting that genetic factors were largely responsible.
This study provided the first firm epidemiological evidence to indicate that genetic factors
may be involved in the hyperuricaemia and occurrence of gout observed among the
Polynesian people. Prior to this time it was presumed that the remarkably high prevalence of
hyperuricaemia and clinical gout among the Polynesians was due solely to environmental fac-
tors associated with westernization, including changes in diet such as increased calorie
intake, increased fat intake and alcohol. In particular these environmental factors were
believed to be responsible for the high level of hyperuricaemia and clinical cases of gout
found among the New Zealand Maoris.
The second population group which the authors studied consisted of Maoris
(Raratongans) who had lived under town conditions for 10 years or more and had been
exposed to a western diet and other environmental features of the western life. The third
population group consisted of the entire population of Pukapuka whose inhabitants were
very isolated and their way of life had not changed notably over many years. Thus they were
on subsistence-type living with a traditional rural diet.
Whilst the highest prevalence of clinical gout was found among the New Zealand Maoris,
the observation of a significantly increased level of hyperuricaemia among the Raratongans
as well as the Pukapukans, indicates a strong genetic predisposition to hyperuricaemia
among Polynesians despite the different living conditions. These observations have led to
further studies of hyperuricaemia and gout among this ethnic group (1) with a focus on a
reduced renal clearance of uric acid.
Strengths
1. The hypothesis is clear and the message is simple.
2. Comprehensive demographic details are given about the communities studied.
Weaknesses
1. Potential pathophysiological mechanisms involving genetic factors are not adequately
discussed.
2. Potential genetic factors are not discussed.
Relevance
This study of the prevalence of hyperuricaemia and gout further emphasizes the value of
studying the same ethnic population exposed to different environments as a way of deter-
mining the relative importance of genetic and environmental factors in the causation and
pathogenesis of diseases. This information is important in the development of public health
measures for the prevention of gout.
364
Paper 3
Autoimmune disease in parasitic infections

in Nigerians
Author
Greenwood BM
Reference
The Lancet 1968; 2:380–382
Summary
In order to determine the prevalence of autoimmune diseases in West Africa, all patients
with autoimmune conditions seen at the University College Hospital, Ibadan, Nigeria, over a
10 year period (1957–1966) were reviewed. Forty-two patients were diagnosed as having RA,
two patients as having systemic lupus erythematosus (SLE) and one patient with systemic
sclerosis. Other non-rheumatological conditions identified during this period were eight
patients with ulcerative colitis, six patients with myasthenia gravis, four patients with perni-
cious anaemia and one patient with Hashimoto’s thyroiditis. Overall there were only 104
patients with autoimmune diseases out of the total of 98 454 patients studied. No patient with
primary Sjögren’s syndrome or polyarteritis nodosa was found. The number of patients with
RA observed during this period was compared with expected values based on figures
obtained for England and Wales. Based on this comparison, the expected frequency of
patients with RA was six times the number actually observed in the West African population.
Related references (1) Muller AS, Valkenburg HA, Greenwood BM. Rheumatoid arthritis
in three West African populations. East African Medical Journal
1972; 49:73–83.
(2) Adebajo AO, Reid DM. The pattern of rheumatoid arthritis in
West Africa and comparison with a cohort of British patients.
Quarterly Journal of Medicine 1991; 80:633–640.
Key message
This paper was the first to provide evidence to support anecdotal reports suggesting that
autoimmune diseases in general, and RA in particular, was uncommon in West Africa. This
paper was also the first to raise the possibility of a protective effect by tropical infections such
as malaria, on the occurrence of autoimmune diseases.

Prior to this study, there had been anecdotal reports suggesting that RA was rare in Africa in
general, and in West Africa in particular. This study was the first to show in a systematic man-
ner that autoimmune diseases including RA was less common in West Africa than in Europe.
This study has provided the foundation for further studies looking at the prevalence of RA in
sub-Saharan Africa. These studies have included recent studies in West Africa which confirm
that although the prevalence of RA may be on the increase, the disease still remains relatively
uncommon. This study is the first to raise the possibility of a protective effect from tropical
infections such as malaria. This hypothesis has been put forward as a possible explanation for
the lower prevalence of RA in West Africa.
This fascinating hypothesis has recently been revisited in the light of genetic studies show-
ing that certain human leucocyte antigen (HLA) haplotypes such as HLA DRB1* 1302-
DQB1* 0501 are associated with protection against malaria. In contrast, HLA-DR4 and HLA-
DR1 (both associated with susceptibility to RA in most populations) are relatively uncommon
in the region. These observations taken together with Greenwood’s findings, suggest that
malaria (which is endemic in West Africa), has led to the selection of those HLA haplotypes
which protect against the development of malaria with a concomitant decrease in HLA
haplotypes associated with the development of autoimmune diseases such as RA. The poten-
tial link between RA and tropical infections in West Africa remains the subject of ongoing
studies (1, 2).
Strengths
1. Provides strong evidence to support previous anecdotal reports indicating that RA was
uncommon in certain populations in sub-Saharan Africa.
2. Provides the first review of the prevalence of autoimmune diseases in Africa.
Weaknesses
1. The methodology was poor and the nature of the catchment population was not ade-
quately described.
2. The study suffered from the methodological problems commonly associated with retro-
spective hospital-based studies, such as that of patient selection and ascertainment bias.
Relevance
This study provided evidence that RA was uncommon in sub-Saharan Africa when compared
with Europe and North America. This study has contributed to other studies world-wide
including ongoing studies looking at the complex relationship between genetic and environ-
mental factors in the aetiology of RA. Whilst the possibility of RA being due to an infectious
agent has long been a widely considered hypothesis, this study was the first to indicate that
directly or indirectly, certain infections may actually protect against the development of RA.
This study also provides some evidence for the belief that RA is on the increase in Africa and
Asia but on the decrease in Europe and North America. It provides a reference for ongoing
longitudinal studies to explore this possibility.
366
Paper 4
Osteoarthritis of the hip and other joints in

Southern Chinese in Hong Kong
Authors
Hoaglund F T, Yau A C, Wong WL
Reference
Journal of Bone Joint Surgery 1973; 55(A):545–557
Summary
A random study of the finger joints, hips and knees of 500 hospitalized Hong Kong Southern
Chinese patients above the age of 54 years was carried out in September 1967. Radiographs
of the hands, pelvis and knees were obtained in all patients. The first 211 patients (112 men
and 99 women) underwent a history and physical examination. 98% of these patients indicat-
ed that they regularly used the squatting position, particularly for toileting purposes which
involved maximum hip flexion and 10–30 degrees of hip abduction. Among the 211 patients
examined, 30% of the men and 31% of the women had evidence of Heberden’s nodes.
There was an increase in the presence of Heberden’s nodes with increasing age in both men
and women. The prevalence of radiological evidence of moderate and severe OA involving
the distal interphalangeal joints was comparable to that of previous British studies. Those
Chinese patients who had OA involving their finger joints had a much higher incidence of
OA involving other joints when compared to those patients whose finger joints were not
involved.
Related reference (1) Kellgren JH, Lawrence JS. Osteo-arthrosis and disc degeneration
in an urban population. Annals of Rheumatic Diseases 1958;
17:388–397.
Key message
This was the first description of a study in a non-Caucasian population which looked at clini-
cal and radiological features of OA in a systematic fashion using a large cohort of patients. In
particular, the study draws attention to potential relationships between ethnic differences
and the pattern of joint involvement in OA.

This paper was the first to document in a detailed manner clinical and radiological aspects of
OA in a non-Caucasian population. This paper has increased awareness of the potential ben-
efits which could be derived from studying the pattern of OA in different racial groups. The
study arose from the observation that hip surgery in the Hong Kong Chinese is rarely per-
formed for conditions other than joint destruction due to sepsis or tuberculosis.
The authors found that the prevalence of Heberden’s nodes among their study popula-
tion was similar to that which had previously been reported for British populations. This con-
trasted with their finding that OA of the hip was significantly less common in their study pop-
ulation, than had been reported among Caucasian populations. Thus, this paper confirmed
using a significant population size and detailed radiological examinations, that OA of the hip
was indeed uncommon in the Chinese population.
In this paper, the authors raise interesting hypotheses as to why OA of the hip might be
less common in the Chinese. They suggest that squatting, particularly for toilet needs and
the extreme range of motion required, might protect the hip joint against the development
of OA among the Chinese. In addition they also point out that predisposing factors such as
congenital dysplasia of the hip and slipped capital femoral epiphysis appear to be less com-
mon among the Chinese and that this might be an additional explanation of the low preva-
lence of OA of the hip in this population.
These hypotheses have served to stimulate other studies looking at racial differences in
OA (1). The majority of these studies have used a combination of clinical assessment as well
as the detailed radiological assessments described in this paper.
Strengths
1. An example of astute clinical observations leading to a prospective study.
2. It provides detailed information for the first time on OA in a non-Caucasian population.
Weaknesses
1. This population was hospital-based and consequently subject to possible ascertainment
bias. However, the authors felt there was minimal ascertainment bias due to poor health
provision or the high cost of hip surgery, as these were not significant problems in their
population.
2. It would have been of greater interest if this study had been extended to all adults rather
than only adults above the age of 54 years.
Relevance
This paper draws attention to the usefulness of studying racial differences in the pattern of
joint involvement in OA. It points out that genetic factors, environmental factors and in par-
ticular socio-cultural factors, may all influence the prevalence and pattern of joint involve-
ment in OA. These observations should help to shed further light on our understanding of
the aetiopathogenesis of OA
368
Paper 5
The clinical course of Kashin–Beck disease

Author
Nesterov AI
Reference
Summary
This paper reviews the clinical features and clinical course of Kashin–Beck disease. It
describes three grades of Kashin–Beck disease and includes case histories to illustrate each of
these subtypes. Overall, the disease is described as a chronic disabling condition character-
ized by generalized OA involving peripheral joints and the spine, but without systemic mani-
festations. Kashin–Beck disease occurs principally in childhood and results in growth distur-
bances. It is endemic in Eastern Siberia, Northern China and Northern Korea. The paper
discusses the aetiopathogenesis of this condition and in particular the possibility that it is due
to ingestion of cereal grain, grown in the affected regions of the world where the grain has
become infected with the fungus Fusaria sporotrichiella.
Related reference (1) Sokolov L. Endemic forms of osteoarthritis. Clinics in Rheumatic

Diseases 1985; 11:187–202.
Key message
This is the first review of Kashin–Beck disease. The review documents the history, clinical fea-
tures and aetiopathogenesis of the disease.

This first review of Kashin–Beck disease comprehensively covers all aspects of the condition
including its clinical and radiological manifestations. The paper documents the history of
Kashin–Beck disease from its earliest description in 1861, when features of the disease were
first noticed to be prevalent amongst the Cossacks who lived in the eastern area of Russia.
The review refers to multiple sources of data collection including the Soviet Health Service
and both the Institutes of the Academy of Medical Sciences and the Irkutsk Institute of
Orthopaedics. The review also draws heavily from personal observations.
Kashin–Beck disease is described from its insidious and often asymptomatic onset in child-
hood to its subsequent slow progression. The paper subdivides Kashin–Beck’s disease into
three subtypes; the mild form of the disease (referred to as the first degree), the moderate
form of the disease (referred to as the second degree), and the severe form of the disease
(referred to as the third degree). Illustrative case histories for each of these subtypes is given.
The mild form of the disease manifests as thickening and slight deformities of the interpha-
langeal or wrist joints, demonstrable both clinically and radiologically. Crepitus is frequently
present in the knees. The moderate form of Kashin–Beck disease is characterized by the
complete clinical and radiological picture of a deforming OA. In this phase of the disease,
the extremities become shorter as the joints become deformed. The patient experiences sig-
nificant joint pains and there is accompanying progressive disability. In its most advanced
form, Kashin–Beck disease is characterized by a combination of chronic deforming OA

together with more general clinical features such as muscular pain and weakness. Spinal
involvement may also occur, thereby further limiting mobility.
The paper discusses radiological and histological features of the disease, with radiographs
showing irregularity in often premature but focal ossification of the epiphyseal growth plate.
The authors draw the inference that the basic problem could reside in the growth plate of
articular cartilage due to nonspecific damage. The authors, however, also point out that his-
tological data suggests that articular cartilage shares with the growth plate a focal apparently
episodic necrosis of chondrocytes.
The paper discusses Kashin–Beck disease as having a possible environmental causation.
Evidence is put forward to support the hypothesis that Kashin–Beck disease is due to toxins
resulting from cereal grain grown in the affected areas of the world where the grain has
become contaminated with the parasitic fungus, Fusaria sporotrichiella. The contamination of
cereal grain with this fungus is believed to be due to the climatic conditions in these regions
of the world. The author points out that public health measures had already begun to lead to
a decrease in incidence of Kashin–Beck disease, and that elimination of new cases of the dis-
ease should eventually be possible.
Strengths
1. A comprehensive review describing all aspects of the condition.
2. Appropriate illustrative case reports are given.
3. The postulated cause of the disease is well described.
Weaknesses
1. No specific epidemiological data is given.
2. No data is given to substantiate the paper’s claim that the incidence of Kashin–Beck dis-
ease is falling.
Relevance
This excellent overview of Kashin–Beck disease brought this disease to the attention of the
rheumatology community outside the Soviet Union. Kashin–Beck disease was the first form
of endemic OA to be described, and its description has formed the basis for the search for
further forms of endemic OA world-wide (1).
370
Paper 6
Unusual hip disease in remote part of Zululand

Authors
Whittman W, Fellingham SA
Reference
The Lancet 1970; 1:842–843
Summary
An unusual form of hip disease was identified near the Mseleni Mission Station of Bombo
district, Zululand, South Africa. A random sample of households was carried out as part of a
community survey. Out of a total population of 236 men studied, 25% of men were affected.
Of 636 women studied, 66% of women were affected. Of 551 boys studied, 4% of boys were
affected and of 569 girls studied, 7% of girls were affected. The associated disability was mea-
sured by the need for a walking stick. Seven percent of men, 35% of women, 0.2% of boys
and 0.7% of girls required the use of a walking stick. Thirty-seven x-rays of the pelvis were
reviewed. The majority of these (24 radiographs) showed evidence of advanced OA.
Related reference (1) Yach D, Botha JL. Mseleni joint disease in 1981; decreased preva-
lence rates, wider geographical location than before, and
socioeconomic impact of an endemic osteoarthrosis in an
underdeveloped community in South Africa. International
Journal of Epidemiology 1985; 14:276–284.
Key message
This is the first description of Mseleni’s disease. This study adds Mseleni’s disease to the
small group of joint disorders referred to as endemic OA.

This is the first report of Mseleni’s disease and is the first form of endemic OA described
from the African continent. Prior to this, a similar condition known as Kashin–Beck disease
had been described. Kashin–Beck disease is commonly observed in South Eastern Siberia,
Manchuria, Northern China, the Shaanxi province of China and North Korea. The disease is
characterized by OA affecting both the peripheral joints and the spine. Mseleni’s disease
resembles dysplasia epiphysealis multiplix, and is found among the Mseleni area of Northern
Zululand in South Eastern Africa, near the border with Mozambique (1). It is a painful disor-
der most commonly affecting the hip joint although other joints, in particular the knees and
ankles, may be involved. This report emphasizes the progressive nature of the condition with
associated increasing disability leading to the need for walking aids and even an inability to
walk at all.
This paper draws attention to the radiological features of this condition. The authors
emphasize the radiological features of large osteophytes, protusio acetabuli, the deformity of
the femoral head with medial subluxation and the rarity of subarticular cysts. The authors
speculate on a nutritional deficiency or toxins similar to that postulated for Kashin–Beck dis-
ease, with potential candidates being veld foods, wild vegetables and palm beer.
Strengths
1. Clear and easy to read.
2. The excitement of identifying a new form of endemic OA is conveyed.
Weaknesses
1. The methodology was poor with no description of the overall catchment population.
2. No indication as to whether or not some individuals may have had OA due to other causes.
3. The demographic, clinical and serological features of those individuals with Mseleni’s dis-
ease was not adequately described.
Relevance
This first report of Mseleni’s disease added yet another form of endemic OA to the existing
literature. There is the possibility that the study of these conditions will serve to provide fur-
ther insight into the aetiopathegenesis of OA.
372
Paper 7
Ross River virus arthritis in Papua New Guinea

Authors
Scrimgeour EM, Aaskov JG, Matz LR
Reference
Transcripts of the Royal Society of Tropical Medicine and Hygiene 1987; 81:833–834
Summary
Over a 1 year period between 1980 and 1981 a survey was carried out of all patients present-
ing with acute polyarthritis at Port Moresby General Hospital located in the capital of Papua
New Guinea. Twenty-four Melanesian patients were identified and assessed for possible Ross
River virus infection. All of these patients had an acute onset of symmetrical polyarthritis
involving both large and small joints. In three patients, a diagnosis of Ross River virus infec-
tion was confirmed by appropriate serological tests. In three other cases serological tests were
suggestive of Ross River virus infection, but were not diagnostic. Synovial biopsies from the
knees of two patients with Ross River virus arthritis and one patient with suspected Ross River
virus infection were obtained. The biopsies showed mild chronic synovitis with synovial cell
hyperplasia.
Related reference (1) Nimmo JR. An unusual epidemic. Medical Journal of Australia 1928;
1:549–550.
Key message
This study confirms an association between acute polyarthritis and Ross River virus infection.
This study also confirmed the presence of Ross River virus among the population group of
Papua New Guinea.

This paper draws attention to the prevalence of arthritis caused by mosquito-born viruses in
the tropics and sub tropics. The paper also confirmed the association between polyarthritis
and Ross River virus infection (1). Synovial biopsy specimens further indicated that Ross
River virus infection can cause an inflammatory polyarthritis.The paper highlights the prob-
lem of the requirement of a serological service for the precise diagnosis of Ross River virus
arthritis. Unfortunately such a service is not available in many parts of the tropics, leading to
misdiagnosis and under diagnosis of the condition.
This paper was the first to identify patients with polyarthritis and Ross River virus infection
in Papua New Guinea. The paper describes the characteristic clinical features of Ross River
virus comprising sudden onset of fever and a symmetrical polyarthritis. The paper also points
out that treatment is essentially symptomatic. This paper draws attention to the potential for
the spread of the alpha virus across the region of the Pacific islands and the consequent pub-
lic health implications of Ross River virus infection.
Strengths
1. The serological aspects of Ross River virus infection are well described.
2. Good illustrative cases of Ross River virus infection are given.
Weaknesses
1. True prevalence estimates are not given as the study was hospital-based.
2. The demographic features of the hospital catchment population is not given.
3. No relationship to rainfall or other climatic variations is given.
Relevance
This study emphasizes Ross River virus infection as a cause of acute polyarthritis. The study
has provided evidence that Ross River virus arthritis occurs across the region of the Pacific
islands. With increasing migration and travel, this paper draws attention to the need to be
aware of Ross River virus infection as a cause of polyarthritis in anyone living in, or who has
visited the tropics or sub tropics.
375
CHAPTER 24
Associations of other systemic

diseases with arthropathies
Introduction
One of the important developments of the 20th century is the recognition and emphasis on
the fact that musculoskeletal manifestations can be caused by a wide variety of systemic dis-
eases, many of which are treatable, or as some are familial diseases, preventable in family
members. The editors have selected a small number of papers to highlight identification of
these associations. There are of course also many other examples. The classifications on the
rheumatic diseases as published in textbooks or the Primer on the Rheumatic Diseases can be ref-
erenced to show the many associations (1, 2). Careful clinical observers can almost certainly
help us identify still unrecognized associations and their implications.
Papers featured in this section focus on clinical observation most often with histopatholog-
ical studies to confirm or support new hypotheses. Do not forget there is still more to be
learned by biopsies of unexplained lesions. Application of new molecular and immunohisto-
chemical tools may even increase the value of biopsies. The recognition that clinical syn-
dromes seemed to differ from known diseases was important in identification of amyloid
arthropathy in dialysis patients and of the arthropathy in haemochromatosis. In these as well
as in sarcoidosis and acromegaly musculoskeletal symptoms had often previously been attrib-
uted to rheumatoid arthritis (RA) or osteoarthritis (OA) or left unexplained.
Evaluation of large series confirmed associations. In ochronosis musculoskeletal involve-
ment had long been noted but we elected to highlight the identification of the biochemical
defect. In most of these diseases studies now are using genetic and molecular techniques to
characterize further mechanisms and improve therapies but challenges remain. Sarcoidosis
is treated only symptomatically with corticosteroids. Is it really an infection? Despite identifi-
cation of the arthropathy of haemochromatosis, the joint disease remains the most resistant
aspect often not responding to even early phlebotomy. Results are somewhat more encourag-
ing with at least some aspects of early treatment of acromegaly. Gene or enzyme therapies
will eventually be coming for haemochromatosis and alcaptonuria-ochronosis.
References
1. Klippel J, Dieppe P (eds.). Rheumatology. 2nd edition. London: Mosby, 1998.
2. Schumacher HR (ed.). Primer on the Rheumatic Diseases. 11th edition. Atlanta: Arthritis
Foundation, 1999.
376
Paper 1
Clinical and pathological studies of joint

involvement in sarcoidosis
Authors
Sokoloff L, Bunim JJ
Reference
Summary
Five cases of sarcoidosis were presented in which polyarthritis was a conspicuous feature.
Each also had other systemic features of sarcoidosis. Synovial biopsies in three patients
showed typical non-caseating granulomas. Searches for other causes of granulomatous syn-
ovitis were unrevealing. Sarcoidosis was proposed to be a more frequent cause of arthritis
than previously recognized and it was proposed that previous cases might have been misdiag-
nosed as rheumatic fever or RA. An association with erythema nodosum was noted in one
case, and the frequent association with erythema nodosum was noted in some previous cases
with more transient arthritis.
Related references (1) Burman MS, Mayer L. Arthroscopic examination of knee joint:
report of cases observed in course of arthroscopic examination
including instances of sarcoid and multiple polypoid fibromato-
sis. Archives of Surgery 1936; 32:846–874. (The report by Sokoloff
and Bunim was not the first to document an association of
arthritis with sarcoidosis but it most clearly established the asso-
ciation.)
(2) Kaplan H. Sarcoid arthritis. Archives of Internal Medicine 1963;
112:925–935. (This report of arthritis in nine of 23 patients with
sarcoidosis in a US military hospital in Germany clearly separat-
ed the acute migratory arthritis seen with erythema nodosum as
having a good prognosis in contrast to the chronic disease.)
(3) Lofgren S. Primary pulmonary sarcoidosis. II. Clinical course and
prognosis. Acta Medica Scandinavica 1953; 145:424–431. (One of
the early reports of the transient acute sarcoidosis leading to
the commonly used eponymous designation.)
Key message
Sarcoidosis can cause an arthritis that is separate from the bone disease. Only one of these
cases had joint erosions despite a chronic course in some. This therefore needed to be distin-
guished from the other currently recognized diseases.
Associations of other systemic diseases with arthropathies 377

This paper was important as it reinforced previous isolated observations and established that
sarcoid arthritis can occur. This was one of a series of papers occurring around this time
emphasizing that many systemic diseases can have specific joint manifestations. Previously
there had frequently been a tendency to make two diagnoses such as still calling the arthritis
rheumatoid without detailed examination. This has been critical in making what we now call
RA a considerably cleaner concept.
Strength
Meticulous histological studies that suggested that mechanisms of arthritis might not be
identical in all cases. Some had only non-specific inflammation; similar cases continue to be
reported.
Weaknesses
1. This and other early observations were not yet able to separate out the Lofgren’s syn-
drome.
2. The African American patients in this series had very limited follow-ups at the time of the
report.
3. Details of their later courses were not reported.
4. Little was discussed about treatments but few alternatives were available at that time.
Relevance
We now recognize some 100 causes of arthritis and this is one that frequently enters diagnos-
tic considerations.
378
Paper 2
Acromegalic arthropathy
Authors
Bluestone R, Bywaters EGL, Hartog M, Holt PJL, Hyde S
Reference
Summary
A series of 42 consecutive hospitalized patients with acromegaly were studied for the pres-
ence of musculoskeletal problems. Peripheral joint abnormalities were found in 26. Carpal
tunnel syndrome was seen in 22 and was bilateral in 20. Twenty patients had back pain. Soft
tissue swelling was more common than joint effusion. There was no evidence of inflammato-
ry arthritis. Joint spaces were often initially wide on radiographs due to hypertrophied carti-
lage. Histologically there was active cartilage cell proliferation in the basal and middle layers
of cartilage presumably due to excess growth hormone. Four patients had severe arthrosis of
hips and/or knees. All had severe long-lasting acromegaly. Osteophytosis of the anterior
parts of the thoracic and lumbar vertebrae was often prominent.
Related references (1) Waine H, Bennet GA, Bauer W. Joint disease associated with
acromegaly. American Journal of the Medical Sciences 1945;
209:671.
(2) Kellgren JH, Ball J, Tutton GK. The articular and other limb
changes in acromegaly. Quarterly Journal of Medicine 1952;
21:405.
(3) Trainer PJ, Drake WM, Katznelson L, Freda PU, et al. Treatment of
acromegaly with the growth hormone-receptor antagonist
pegvisomant. New England Journal of Medicine 2000;
342:1171–1177.
Key message
Musculoskeletal problems occur at all stages of acromegaly. Effective treatment reversed the
carpal tunnel syndrome only. Despite the frequent back pain, mobility was often well main-
tained or even greater than normal probably due to thickened discs and lax hypertrophied
ligaments.

This and earlier smaller series noted above identified a potentially treatable cause of arthri-
tis. As better treatments now become available early diagnosis may not only resolve the soft
tissue problems but also prevent the later secondary OA. These observations also gave impor-
tant insights into effects of growth hormone on cartilage and bone. Acromegalic ulcers on
cartilage described earlier by Kellgren, et al. (2) were confirmed in more extensive pathologi-
cal studies. The dramatic bridging osteophytes illustrated are very similar to lesions later
described as DISH (diffuse idiopathic skeletal hyperostosis) that are probably also related to
growth hormone-like molecules. Ten patients had symptoms suggestive of mild Raynaud’s
phenomenon and two had severe morning stiffness causing potential confusion with other
rheumatic diseases. Fatigue is now also noted as a prominent symptom of acromegaly further
risking confusion.
Strength
X-rays were obtained on most and allowed correlation with extensive and detailed pathologi-
cal studies on three cases (two necropsies and one surgery).
Weaknesses
1. Disc and capsular calcifications were noted but not studied further. We still are not sure
about the relationship of chondrocalcinosis to acromegaly.
2. Soft tissues were described in less detail than bone and cartilage.
Relevance
This valuable resource on early as well as chronic musculoskeletal problems in acromegaly
has led to increased attention to joint symptoms and signs as clues to this diagnosis.
380
Paper 3
The nature of the defect in tyrosine metabolism

in alkaptonuria
Authors
LeDu BN, Zannoni VG, Laster L, Seegmiller JE
Reference
Journal of Biological Chemistry 1958; 230:251–256
Summary
This disease had been identified as a mendelian recessive trait by Garrod in 1902 who coined
the term ‘inborn error of metabolism’ suggesting that an enzyme defect would explain this
disease. This report described the definitive biochemical evidence and showed absence of
homogentisic acid oxidase in a liver from a patient with alkaptonuria and ochronosis. Liver
biopsy was performed during abdominal surgery for an oesophageal hiatus hernia. There was
no detectable homogentisic acid oxidase in the liver homogenate but all other enzymes in
the tyrosine oxidation pathway were the same as in normal human liver and liver from other
mammals.
Related references (1) Fernandez-Canon JM, Granadino B, Beltran-Valero de Bernabe D,

Renedo M, Fernandez-Ruiz F, Penalva MA, Rodriguez de
Cordoba S. The molecular basis of alkaptonuria. Nature Genetics
1996; 14:19–24. (This is one of a series of papers from this
group and others describing the cloning of the human
homogentisic acid oxidase gene and showed that it is the gene
responsible for alkaptonuria. The gene is located on chromo-
some 3q2. Missense mutations co-segregate with the disease.)
(2) O’Brien WM, Banfield WG, Sokoloff L. Studies on the pathogene-
sis of ochronotic arthropathy. Arthritis and Rheumatism 1961;
4:137–152. (Affinity of the cartilage for homogentisic acid or its
precursors is described.)
(3) Hunter T, Gordon DA, Ogryzlo MA. The ground pepper sign of
synovial fluid: A new diagnostic feature of ochronosis. Journal of
Rheumatology 1974; 1:45–53. (A clever practical observation that
may allow easy detection of this rare syndrome.)
Key points
Normal levels were demonstrated for all enzymes of the tyrosine oxidation system except for
homogentisic acid oxidase. There was no evidence that absence of enzyme was due to an
inhibitor.

This study set the stage for a series of studies on this autosomal recessive disorder showing
that homogentisic acid polymers deposited in cartilage and other tissues causing the disease
of ochronosis. This also set the stage for the continued genetic studies and established the
still unrealized potential for therapy of this devastating and still poorly managed disease.
Strength
Meticulous biochemical studies identified the single enzyme defect that had been suspected.
Weaknesses
1. Tissues other than the liver were not studied.
2. The exact nature of the defect (i.e. failure to make protein, or altered protein) was not
determined.
Relevance
This was an important step in developing the mechanisms of an uncommon genetic disease
that will eventually be treatable and preventable with enzyme or gene therapy.
Figure 24.1 Shards of pigmented

ochronotic cartilage embedded in
synovium.
382
Paper 4
Corticosteroid therapy associated with ischemic

necrosis of bone in systemic lupus erythematosus
Authors
Zizic TM, Marcoux C, Hungerford DS, Dansereau J-V, Stevens MB
Reference
Summary
In this prospective study 52% of 54 patients with systemic lupus erythematosus (SLE) had
ischaemic necrosis of bone. Ninety-three sites were involved with hips, knees and shoulders
affected most frequently. The dramatic differences between patients with and without
ischaemic necrosis were the mean daily doses of prednisone for the month with the highest
steroid dose and the highest consecutive 3, 6 and 12 months of therapy with the higher doses
favouring the development of the lesions. Duration of steroid therapy, total cumulative
steroid dose and mean daily doses in general were not significantly different between the two
groups. Patients with ischaemic necrosis were more likely to be cushingoid. A lower mean
prednisone dose was required to produce ischaemic necrosis in patients with Raynaud’s phe-
nomenon.
Related references (1) Harrington KD, Murray WR, Koontz SL, Belzer FO. Avascular
necrosis after renal transplantation. Journal of Bone Joint Surgery
1971; 53(A):203–X. (One of the early clues to the role of
steroids.)
(2) Adarraga DA, Sanchez-Martinez F, Caracuel MA, Escudero A,
Collantes E. A case of multiple osteonecrosis in a patient with
HIV. Journal of Clinical Rheumatology 2000; 6:41–44. (New causes
of AVN continue to be identified.)
(3) Lotke P. Ecker MI. Osteonecrosis of the knee. Orthopedic Clinics of
North America 1985; 16:797–808. (AVN at the knee occurs most
often in elderly women and can be a diagnostic problem.)
(4) Zizic TM, Hungerford DS, Stevens MB. Ischemic necrosis in sys-
temic lupus erythematosus. I. The early diagnosis of ischemic
necrosis of bone. Medicine 1980; 59:134–148. (Increased bone
marrow pressure is proposed as a common mechanism for AVN
and has led to one approach to early treatment with core
decompression.)
Key message
Higher peak steroid doses seem to increase the risk of AVN. This and the other related refer-
ences have helped make rheumatologists much more aware of AVN as a cause of joint symp-
toms.

Beginning to understand the relationship to corticosteroid dose in SLE had a variety of
impacts. Rheumatologists became better aware of ischaemic necrosis. This is also called AVN,
osteonecrosis (ON) or aseptic necrosis (AN). Rheumatologists and orthopaedists began
investigating its likely mechanisms in various states in work that continues to this time.
Evidence continues to suggest that minimizing peak steroid doses may decrease AVN. The
practice of high dosing continues as studies have not evaluated systematically the relative
effects of various doses on many manifestations of SLE.
Strength
A prospective study carefully analyzing the role of steroid regimens.
Weakness
1. Limitation of study to SLE making it difficult to isolate the pure role of steroids from an
additive effect of steroids and SLE.
2. Very narrow focus on steroids.
Relevance
Ischaemic necrosis remains a frequent complication of therapy of SLE. This and other com-
plications of corticosteroids now account for increasing morbidity in patients with SLE. A
diagnosis of ischaemic necrosis should always lead to a search for known underlying diseases
and consideration of mechanisms in each situation.
384
Paper 5
Hemochromatosis and arthritis

Author
Schumacher, HR
Reference
Abstract
Two patients with haemochromatosis and what appears to be a distinctive arthritis are presented.
Both showed prominent involvement of the hands, minimal evidence of inflammation, significant x-
ray changes and synovial haemosiderin deposition. Five others cases of a similar arthritis were
found in a review of 23 cases of idiopathic haemochromatosis. The arthritis is postulated to be the
result of iron excess damaging articular tissue in a yet undefined manner.
Summary
Two patients were described in detail and five others reviewed with osteoarthritis-like findings
but an unusual distribution that included the PIP and metacarpal phalangeal joints. All were
negative for rheumatoid factor. Synovial biopsies did not show inflammation but showed dra-
matic deposition of haemosiderin primarily localized to the synovial lining cells. Such iron
deposition had been previously described but had not been related to any joint symptoms.
Two patients without arthritis also had synovial tissue examined and also had iron deposition
so that such deposition did not invariably cause arthritis. Possible mechanisms were discussed.
X-rays showed generalized demineralization, irregular articular surfaces, subcortical cysts,
periarticular bony proliferation and joint space narrowing. Calcification was described at the
lateral meniscus of one knee, which was not commented upon, but turned out to be a find-
ing subsequently seen in many patients.
Related references (1) Schumacher HR. Articular cartilage in the degenerative arthropa-
thy of hemochromatosis. Arthritis and Rheumatism 1982;
25:1460–1468.
(2) Hamilton E, Williams R, Barlow KA, Smith PM. The arthropathy of
idiopathic hemochromatosis. Quarterly Journal of Medicine 1968;
37:171–182.
(3) Schumacher HR, Straka PC, Krikker MA, Dudly AT. The arthropa-
thy of hemochromatosis: recent studies. Annals of the New York
Academy of Science 1988; 526:224–233.
(4) Hamilton EDD, Domford AB, Laus JW, Williams R. The natural
history of arthritis in idiopathic hemachromatosis: progression
of the clinical and radiological features over 10 years. Quarterly
(5) M’Seffar A, Fornaiser VL, Foh IH. Arthropathy as the major clini-
cal indication of occult iron storage disease. Journal of the
American Medical Association 1977; 238:1825–1828.
Key message
Hemochromatosis is associated with an arthropathy that previously may have been confused
with RA or OA.

This arthropathy has now been identified to be an initial manifestation of haemochromatosis
in some cases, and thus it may lead to diagnosis and treatment of this disease before the life-
threatening complications of cardiomyopathy, cirrhosis and diabetes. The exact mechanisms
of the arthropathy have not been defined. The recent identification of the gene for most
familial haemochromatosis sets the stage for renewed investigations into mechanisms.
Strengths
1. An important observation in two cases and expanded with review of additional cases of
haemochromatosis for confirmation.
2. Documentation of the synovial deposition of iron predominantly in the lining cells in this
disease set the stage for series of observations confirming this important association.
Weaknesses
1. Did not examine implications for treatment of the joints that still remains unsatisfactory.
2. Did not examine bone or cartilage leaving mechanisms unresolved.
Relevance
It identified a new cause of arthropathy. One case and others since have been previously mis-
diagnosed as RA leading to inappropriate treatments. This identified an important familial
and metabolic disease associated with arthritis.
Figure 24.2 (left)

X-ray showing
classic findings of
haemochromatosis.
Figure 24.3 (right)

Iron in synovial
lining cells in a
patient with
haemochromatosis.
386
Paper 6
Synovial amyloidosis in patients undergoing

long-term hemodialysis
Authors
Bardin T, Kuntz D, ZingraffJ, Voisin MC, Zeimar A and Lansaman J
Reference
Summary
Synovial amyloid deposits were found in 18 patients with end-stage renal failure of various
aetiologies who had been treated with long term haemodialysis (mean 116 months). All had
carpal tunnel syndrome which was bilateral in 14; four patients also had flexor tenosynovitis
and two had destructive arthropathies that required surgical replacement of the hip.
Amyloid deposits were confirmed by light microscopy and in 6 by electron microscopy. Nine
patients had cystic radiolucencies of bone interpreted as likely due to amyloid.
Related references (1) Allieu Y, Asencio G, Mailhe D, Baldet P, Mion C. Syndrome du

canal carpien chez l’hemodialyse chronique: approche
etiopathogenique: apropros de 31 cas operas. Rev Chir Orthop
1983; 69:233–238. (An earlier identification of amyloid in
carpal tunnels of dialysis patients.)
(2) Munoz-Gomez J, Bergada-Barado E, Gomez-Perez R, Llopart-
Buisan E, Subias-Sobrevia E, Rotes-Querol J, Sole-Arques M.
Amyloid arthropathy in patients undergoing periodical
haemodialysis for chronic renal failure: a new complication.
Annals of Rheumatic Diseases 1985; 44:729–733, 1985. (Seven dial-
ysis patients had amyloid arthropathy but no amyloid in subcu-
taneous fat deposits.)
(3) Netter P, Kessler M, Bumel D, Hutin MF, Delones S, Benoit J,
Gaucher A. Aluminum in the joint tissues of chronic renal fail-
ure patients treated with regular hemodialysis and aluminum
compounds. Journal of Rheumatology 1984; 11:66–70. (Aluminum
might be another of the many factors involved in dialysis
arthropathy as well as in bone disease.)
Key message
Amyloid was convincingly established as the cause of most carpal tunnel syndromes in dialysis
patients and was also present in joint synovium and capsules. Juxta articular bone lucencies
seemed prominent and were proposed to be due to amyloid. Destructive arthropathies were
seen in large joints of four cases, erosive arthritis of the hands in two and destructive spine
lesions in seven. These were not explained by any histologic differences. One patient also
had the previously recognized apatite deposits. All patients had long durations of haemodial-
ysis with a mean of 116 months and the shortest duration of 74 months.

This and other related studies led to attempts to decrease this complication by examining
different dialysis membranes and different regimens. Shorter periods of dialysis before trans-
plantation can decrease the risk of amyloidosis. Recognition of amyloid as the cause of the
carpal tunnel syndrome can influence treatment. Local corticosteroids had, at best, transient
benefit in these patients. Histopathologic examination of unexplained lesions continues to
be important. Later studies confirmed that this amyloid is due to beta 2 microglobulin.
Strengths
1. Large well studied series with confirmation of the role of amyloid in a variety of problems
of chronic haemodialysis patients.
2. Extensive x-rays and pathologic studies.
Weaknesses
1. Bone biopsies were not obtained in this report to confirm the presence of amyloid
although this has subsequently been confirmed as a factor.
2. The fascinating destructive arthropathy in some patients remained unexplained.
Relevance
Shorter use of dialysis can decrease this complication if transplantation is available. The vari-
ety of problems that must be considered in dialysis patients with musculoskeletal problems
includes infections, apatite deposition in patients with elevated serum phosphate levels,
hyperparathyroidism, or manifestations of the primary disease leading to renal failure.
389
Section 4
Therapy
391
CHAPTER 25
Education
Ylva Lindroth
Introduction
Patient education has probably been practised as long as there have been patients and physi-
cians. The reason for the late appearance of patient education in the columns of scientific
publications is unclear. Today every practitioner knows that patients will not accept, let alone
carry out, a prescribed treatment regime unless it is explained to them. However, as recently
as 1976, the importance of patient education was neither acknowledged in medical textbooks
nor mentioned in texts on arthritis expressly written for patients. The proper field of medi-
cine was seen to be human biology. We viewed our patients as organisms composed of differ-
ent parts having different functions. The physician’s role was to locate the part that was not
fulfilling its function, and then repair that organ by employing external or internal medi-
cine–in short, the reductive model.
In the early 1980s, the reliability of objective, reproducible, numeric ‘hard’ data were chal-
lenged by interest in less confident, subjective ‘soft’ data. The result was a whole new world of
instruments attempting to measure something previously disregarded–the patient’s point of
view. Using these instruments, it became possible to evaluate the effect of patient education
not only on a patient’s knowledge and behaviour, but also on the outcome of the disease.
The first classic paper in the field of patient education did not appear until 1982 (Paper 1,
Mazzuca). It is a meta-analysis of publications on patient education in cases of chronic dis-
ease; arthritis, however, was not mentioned. Although there had been education pro-
grammes for patients with rheumatoid arthritis (RA) before, they had never been evaluated
and presented in scientific literature. Patients were instructed in the right and wrong way to
perform activities of daily living in order to circumvent forces that promoted deformity; but
the effect of these programmes was never analysed.
As the afflicted person’s role in treatment began to be emphasized, it was vital to assess a
patient’s perception of RA, and understand how it differed from those of the physicians. This
was first done first by Silvers, et al. in 1985 (Paper 2). The groundwork for patient education
was constructed around the question: what did the patients need to know? Lorig and her col-
laborators introduced their Arthritis Self-Management Program (ASMP) in 1985 (Paper 3).
This study was the first to present an evaluation of arthritis education in an experimental
model, namely, a prospective randomized controlled study. A further development of these
ideas, modified for another health care system, was presented by Taal, et al. in 1993 (Paper
4). Based on experience from a long tradition of patient education programmes in Lund,
Sweden and the ASMP from the USA an Australian programme was developed and evaluated
(Paper 5). This contained elements of problem-based education further developed. Other
means of presenting education cite an approach in which information and support were
delivered over the telephone (Paper 6). Looking toward the future, it is hoped that evidence-
based medicine will not only include randomized controlled studies, but will also validate the
phenomenology of disease. A classic paper of a qualitative method is presented (Paper 7).
Donovan, et al. tell us in their study what having arthritis means to the patients. A new way of
analysing and presenting data has opened our eyes to a broader view of the patient as an
important member of a team working to better the outcome of rheumatic disease.
392
Paper 1
Does patient education for chronic diseases

have therapeutic value?
Author
Mazzuca SA
Reference
Journal of Chronic Diseases 1982; 35:521–529
Summary
This is a review and meta-analysis of publications on patient education from 1970 to 1981.
Only such studies were included that were experimental, having a defined allocation of the
independent variable (i.e. instruction) to a treatment or control group. Dependent variables
were: (a) compliance with therapeutic regimen, (b) physiological progress toward therapeu-
tic goals, and (c) long-term health outcomes. The effect size of each relevant dependent vari-
able was calculated, with a positive result indicating the experimental group being superior
to control group. The differential effectiveness of the didactic (increase of knowledge) or
behavioural emphasis (promotion of self-management) of the intervention was determined.
Of 320 articles, 63 were viewed as experimental, 32 concerned chronic disease (30 of
which allowed the calculation of effect size), and 24 used some kind of randomization (in
two cases, altering the experimental conditions). Surprisingly, none of these publications
focussed on arthritis.
Related references (1) Haynes RB. Strategies for improving compliance. In: Compliance
with Therapeutic Regimens (Sackett DL, Haynes RB, eds.).
Balitmore: Johns Hopkins, 1976: 69–82.
(2) Glass GV. Integrating findings: The meta-analysis of research. In:
Review of Research in Education 5 (Schulman LS, ed.). Itasca,
Illinois, 1977: 157–173.
Key message
The literature clearly indicates that patient education has therapeutic value and can improve
the course of a chronic disease. Programmes that included behaviour-oriented strategies
were more successful in achieving positive change than were didactic interventions which
affected knowledge alone.

This article exemplifies the late appearance of publications specifically addressing patient
education. During his time of observation (1970–81) Mazzuca found only 320 articles
analysing effect of patient education. Twenty-two of these were experimental, randomized,
controlled studies dealing with chronic disease. The chronic diseases surveyed were hyper-
tension (10), other heart diseases (5), asthma (3), obesity (3), hyperlipidemia, glaucoma,
and renal disease (1 each), and uncategorized diseases (5). There were no evaluative studies
of patient education programmes where rheumatic disease was the subject under considera-
tion. At present, there are 4324 references to be found in Med Line during a 2 year period
Education 393
under the key word ‘patient education’. These articles focus on many diseases, including
arthritis. In comparison with past years, the trend is impressive.
Mazzuca poses the key question, ‘Does patient education have therapeutic value?’. He
declared that the pattern of diseases appears to have changed over the last 50 years from
acute illnesses to such chronic diseases as hypertension, diabetes, lung, and heart ailments.
Patient education had advanced the goal of more active participation on the part of the
patient in an attempt to maximize therapeutic benefits. However, although some reviews had
appeared documenting multidisciplinary patient education efforts and their results, it
remained impossible to estimate the degree of efficacy on health status in cases of chronic
disease. An earlier review by Haynes (1) had used an ad hoc rating system to find statistical
and clinical evidence that only providing a patient with information about a particular dis-
ease was less effective than interventions that included behavioural strategies. However,
Haynes dealt mostly with diseases of short duration and remarked that chronic disease would
present a far more challenging problem.
Mazzuca uses meta-analysis (2) to answer his own question. With this method, each test of a
dependent variable yields an estimated effect size, which can be readily interpreted as the change (in stan-
dard deviation units) attributable to the experiment’s intervention. He recommends that researchers
who want to interpret the clinical importance of their interventions compute the effect size
for their dependent variables.
The answer to Mazzuca’s leading question can now be seen to be a decisive ‘yes’. Patient
education does have therapeutic value. After searching in vain for literature on patient edu-
cation before the appearance of Mazzuca’s article, one takes encouragement from his paper
and realizes that patient education needs to be evaluated by the same scientific method we
employ in other areas of medical research.
Strengths
1. Mazzuca’s article provides a good review of publications on patient education.
2. The study’s objective methodology, and the author’s reasons for excluding many publica-
tions are carefully delineated.
Weakness
The interpretation of clinical relevance described in the discussion section appears some-
what tentative.
Relevance
Most authors presenting studies on patient education consider Mazzuca’s article a landmark
in the field. It poses a fundamental question as to the value of patient education, and goes
on to establish the relevance of this approach in the context of other methods of treating
chronic rheumatic disease.
394
Paper 2
Assessing physician/patient perceptions in

rheumatoid arthritis. A vital component in
patient education
Authors
Silvers IJ, Howell MF, Weisman MH, Mueller MR
Reference
Summary
This study has analysed the needs of patient education regarding the content, provider and
delivery methods perceived by 101 patients with RA and 28 rheumatologists. Ninety-two per-
cent of all patients surveyed wanted to learn more about arthritis. All of the physicians con-
ceded that patient education in arthritis management was beneficial, though less than 25%
of them had an up-to-date programme to offer their patients. Patients and physicians agreed
that certain topics were of importance to them both: medication, patient/physician commu-
nication, fraudulent practitioners, and the need to remain ambulatory. Rheumatologists
placed great value on psychosocial areas, activities of daily living, sexual concerns, and com-
munity resources, while patients rated a knowledge of the disease process, diagnostic proce-
dures, and questions of nutrition most important to them. Both patients and physicians
favoured individual over group sessions. There was a difference in opinion over the optimum
provider of education, with patients expressing more confidence in pharmacists and nutri-
tionists, whereas doctors preferred programmes that were led by trained volunteers (rather
than health professionals).
Key message
Patient education plays an important part of arthritis management by both arthritis patients
and physicians. Programmes must be carefully planned, implemented, and evaluated.
Related references (1) Green LW, Lewis FM, Levine DM. Balancing statistical data and
clinical judgements in the diagnosis of patient educational
needs. Journal of Community Health 1980; 6:79–91.
(2) Roter DL. Patient participation in the patient-provider interaction:
the effects of patient question asking on the quality of interac-
tion, satisfaction, and compliance. Health Education Monograph
1977; 5:281–315.

When this article appeared in 1985, patient education was practised on the basis of casual clinical
observations and individual case studies, often representing a stereotypical attitude toward
patients with chronic disease (1). The information given to such patients was what health profes-
sionals considered important for them to know. With the exception of studies on heterogeneous
groups of patients, there was no empirical data on what the actual needs of patients with RA were.
Education 395
This study considered the views of patients with defined RA, and those of rheumatologists as well.
The active role of patients in the management of their disease had begun to be recognized (2).
The authors found that patient education should be viewed as an important issue for both
patients and physicians. Earlier work had indicated that patients did not always understand
what their doctors told them; however, no substantial evidence could be found indicating
that patient education was considered necessary by the medical profession. Later investiga-
tions demonstrated that patients have great misconceptions about the nature of their disease
and its treatment, and confirmed the fact that they wanted to know more.
The authors also showed that patients and physicians had different views on the content of a
patient education programme (Figures 25.1 and 25.2). While the patients were most interested
in achieving more knowledge about the disease process and diagnostic procedures, the physi-
cians found personal, community, and psychosocial areas the most important topics for discus-
sion. This was in agreement with Mazzuka (Paper 1), who stated that programmes including
behavioural and psychosocial aspects of the disease were more successful than those dealing
with increased knowledge of the disease alone. However, as indicated in the study by Taal and
others (Paper 4), answers may change if questions about sensitive topics are phrased differently.
P) .001 Patients
100 100 Physicians
Patients P) .001 P) .001
90
Percentage of very to extremely
NS Physicians 90
Percentage of very to extremely
NS P) .001 P) .001
80 P) .015
P) .001 80
P) .001 P) .01
important ratings
70 NS
important ratings
NS NS 70
60
60
NS
50
50
40
40
30
30
20
10 20
10
ap al
N pist l
de d
ta itis
rk ial
up ist
a a
nis
lea e
er ic
er n
wo Soc
ur
O mac
p ain
en hr
th atio
Ph ist
pr A r
er
th hys
io
tiv
N
es r t
rit
ou Tr
P
ar
s
ur ity
oc se
n
ut
oc tic
ne cial
Se ing f
Co cern
io
liv o
cc
pr isea
so n
pr nos
t s
D s
D ss
rit
re mu
ily es
s
so
Ac ed
ce
es
e
ut
n
da iviti
ho
iag
co
m
gr
N
re
yc
x
Ps
Figure 25.1 Percentages of patients’ and Figure 25.2 Percentages comparing patients’
physicians’ ratings reported as extremely and physicians’ ratings reported as extremely
important for non-physician health important for various topics in arthritis patient
professionals involved in education. education.
Strengths
1. The study focuses on only one disease, RA, and one type of specialist, the rheumatologist.
2. Patients represent a broad range of ages (24–79 years), and the duration of their disease is
equally broad (1–49 years).
3. The doctors included came from different health care systems, such as teaching hospitals,
private practices, or rheumatology fellows.
Weaknesses
1. The response rate was low for patients (55%), although better for physicians (74%).
2. Only two out of 38 doctors surveyed were women.
Relevance
The importance of this paper lies in questions: what are the self-expressed needs of patients?
What do patients want to know, how do they prefer to have this information conveyed, and
by whom? What are the attitudes of physicians in this regard?
396
Paper 3
Outcomes of self-help education for patients

with arthritis
Authors
Lorig K, Lubeck D, Kraines RG, Seleznick M, Holman HR
Reference
Summary
A self-help education programme, the Arthritis Self-Management Program (ASMP), was eval-
uated in a randomized prospective study. The programme is a community-based patient edu-
cation course given by lay persons over six session. It is conducted for people with arthritis,
irrespective of cause. Discussions and exercises in groups were used to improve knowledge,
behaviour, and health status. Comparisons were made between the experimental and the
control group, and a longitudinal comparison with baseline was made after 20 months.
Outcome variables were patient knowledge, exercising, pain, disability, locus of control,
number of visits to physician, and participant’s satisfaction with physician.
After 4 months, there was an improvement in patient knowledge and a decrease in pain,
but no change in disability or number of visits to physician. After 20 months, knowledge and
frequency of exercise declined but still remained above base levels. Locus of control did not
change.
Related references (1) Lorig K, Fries J. The Arthritis Helpbook. Reading MA: Addison-
Wesley, 1990.
(2) Bandura A. Self-efficacy in human agency. Annals of Psychology
1982; 37:122–147.
Key message
The ASMP, taught by lay-leaders, can lead to increased patient knowledge, changed behav-
iour, and lessened pain. Improvements seen were still in evidence after 20 months.

When Mazzuca reviewed published articles on patient education, he could find none with
rheumatic disease as the target subject. This is the first study of patient education for individ-
uals with arthritis that fulfils his criteria for an experimental study. Since its appearance in
1985, it has become the classic paper for those working in the field of patient education. It
presents an inexpensive, replicable programme, an evaluation in a randomized, controlled
study, and a longitudinal follow-up, measuring changes in patient knowledge, behaviour, and
health status.
The Arthritis Helpbook (1) developed for the sessions has been the basis for similar pro-
grammes throughout the world, having stimulated health professionals in many countries to
institute such a course for patients with arthritis. When the authors later found that changes
in health status were not strongly associated with modifications in patient behaviour, they
sought reasons for this. They found that those with a documented improvement in their
Education 397
health status also had a feeling of increased personal control over their symptoms. The pro-
gramme was then redesigned to increase the self-efficacy of the participants, namely, their
capability and skill to achieve a desired outcome – something decried by Bandura (2).
The ASMP was originally based on a survey of 100 arthritis patients and 50 practising
rheumatologists. The topics of the six sessions were (a) anatomy and physiology, (b) effects
and uses of medication, (c) individualized exercise programmes, (d) cognitive pain manage-
ment, (e) nutrition, (f) patient/physician communication, and (g) problem solving disease-
related difficulties.
The programme was designed to be inexpensive and readily available. It is offered to all
individuals with arthritis, irrespective of diagnoses; it is community-based and thus given in
shopping malls, senior centres, churches, or libraries; and it is conducted by lay-leaders who
have been especially trained and are guided by a detailed protocol.
Before the ASMP programme appeared, patient education often took a didactic teaching
format: patients were lectured about their arthritis, with the occasional inclusion of slides if
medical students were being addressed. This publication introduced strategies for group dis-
cussions, the use of contracts, keeping diaries, and weekly feedback sessions. As a result, after
this publication, programmes became more interactive, encouraging the participation of
group members.
Strengths
1. The outline of the prospective, controlled study to evaluate the impact of the programme
on arthritis patients is well-planned and executed.
2. The establishment of patient education programmes for chronic disease sufferers, as a
useful part of any therapeutic regime, is a landmark.
Weaknesses
1. The assertion contained in the abstract, claiming that the programme is inexpensive and
well-received by patients, physicians, and other health professionals, receives no documen-
tation in the study.
2. Seventy-seven percent of the participants had osteoarthritis, only 10% had RA, and 12%
had arthritis of an unspecified kind. There is no report on how the different diagnoses
may have influenced the results of the study.
Relevance
A new model for approaching certain aspects of chronic illness, this paper offers thorough
documentation of the positive effects of arthritis patient education on patient knowledge,
health behaviour, and pain. It has encouraged the further development of other patient edu-
cation programmes, and pointed out the importance of designing one’s evaluation in accor-
dance with scientific method.
398
Paper 4
Group education for patients with

Authors
Taal, E, Riemsa RP, Brus HLM, Seydel ER, Rasker JJ, Wiegman O
Reference
Patient Education and Counseling 1993; 20:177–187
Summary
This Dutch education programme was partly based on the ‘Arthritis Self-Management
Course’ (ASMC) by Lorig and collaborators (1). The task of the programme was to
strengthen self-efficacy (2), facilitate the achievement of an individual’s expectations, and
change patient health behaviours, with the intention of improving health outcomes such as
pain, functional ability, and the psychosocial consequences of the disease.
There were some differences, as compared to the Lorig programme: it was taught by
health professionals instead of lay-leaders, and the participants all had RA. The programme
was evaluated by comparing the effect in an experimental group with a randomly assigned
control group. Results after 6 weeks, 4 months, and 14 months showed that patients in the
experimental group had increased knowledge of their disease, and had improved physical
exercise behaviour and self-efficacy function, as compared with the control group. No effects
were found on pain, disease activity, or the psychological and social aspects of the disease.
Related references (1) Lorig K, Lubeck D, Kraines RG, Seleznick M, Holman HR.
Outcomes of self-help education for patients with arthritis.
(2) Bandura A. Self-efficacy in human agency. Annals of Psychology
1982; 37:122–147.
Key message
A Dutch educational group programme, which was based on the ASMP by Lorig (1), resulted
in improvement with regard to functional disability, joint tenderness, health behaviour, self-
efficacy function, and knowledge. After 14 months, positive effects were still visible on self-
efficacy function, patient knowledge, and exercise behaviour.

Although a follow-up on the work done by Lorig and collaborators (ASMC), there are some
major differences here. The participants in the Dutch study formed a homogenous group
and shared a common diagnosis, RA. They were selected from the files of rheumatologists,
whereas the participants in Lorig’s study were recruited from public service and through the
mass media. Moreover, Lorig’s group was heterogeneous, most of them suffering from
osteoarthritis. Again, the leaders of the Dutch programme were health professionals, as
opposed to Lorig’s lay-leaders. The topics dealt with were generally the same in both pro-
grammes, although Taal’s group put greater emphasis on physical exercise and allotted extra
time to relaxation techniques. Furthermore, each person in the Dutch study was referred to
Education 399
a physiotherapist for individual instruction, and fewer classroom hours were spent teaching
facts. Written material on RA was handed out and information about other resources given.
In the discussions, the leaders gave guidelines on how to read and evaluate information
about different types of treatment, including alternative therapy. The main emphasis of the
programme, the enhancement of self-efficacy, was done by means of patient contracts, goal
setting, and feedback (similar to the ASMC). The results of the Dutch programme are com-
pared with those of Lorig and collaborators. The two studies report increased patient knowl-
edge and a change in exercise behaviour. A significant difference is that the Dutch pro-
gramme did not affect levels of pain or depression, but the principles of pain management
are similar. The authors give no explanation for this difference.
Taal and his collaborators found that changes in self-efficacy were only related to func-
tion, not to pain or other symptoms. Conversely, Lorig concluded that changes in self-effica-
cy were associated more with changes in health status than with changes in behaviour.
Strengths
1. The authors aimed at assembling a homogeneous group of including only RA patients
between the ages of 21 and 65 years, with a maximum use of second-line anti-rheumatic
medication for 8 years.
2. All patients went through a clinical examination which included the assessment of joint
tenderness and the taking of blood samples to measure disease activity.
Weaknesses
1. The response rate was low (only 75 of 140 who had been invited agreed to participate).
No information is given on the non-respondents.
2. A large number of tests were performed, making it possible that significant changes may
have occurred by chance.
Relevance
This paper can be seen as an indication of the further development of patient education.
Some ideas from the American programme by Kate Lorig and collaborators have been
applied. It does show, however, that each programme has to be adapted to the social struc-
ture and health care system of the country where it is introduced.
400
Paper 5
A controlled evaluation of arthritis education

Authors
Lindroth Y, Bauman A, Barnes C, McCredie M, Brooks PM
Reference
Summary
This prospective study evaluated an education programme for people with rheumatoid
arthritis (RA) and OA. The intervention group participated in a comprehensively planned
six session behaviour-based programme. A questionnaire was given to 100 patients and 95
matched but non-random controls before the programme, 1 month later, and at 3 and 12
months. The intervention group demonstrated improvements in knowledge, self-reported
health behaviour and disability scores at 12 months compared to the controls. No difference
was reported in symptoms, compliance with therapy, pain perception and locus of control.
Related references (1) Lorig K, Fries JF. The Arthritis Helpbook. Reading: Addison-Wesley,
1980.
(2) Althoff B, Nordenskiold U. Joint Protection. An Alleviating Way of
Living. Uppsala: Pharmacia, 1985.
Key message
Development of a successful patient education programme is facilitated by the involvement
of patients at the planning stage. A behavioural-based approach is likely to be most success-
ful, and can lead to improvements in knowledge and behaviour that are long lasting.

Much of the material available to patients for their ‘education’, as well as many of the pack-
ages developed by the professionals, take a didactic, authoritarian approach. One of the most
important features of the work described here related to the way in which the intervention
was developed. The authors undertook a comprehensive survey of the literature and took
note of related work from other centres and disciplines when planning their intervention.
They then took the important step of involving both the patients and therapists in a work-
shop at which the format of the programme was planned.
Another key feature of the programme described in this paper is the the emphasis on
changing behaviour of patients as opposed to an emphasis on knowledge, which predomi-
nates in many other packages. The outcomes as well as the intervention put major emphasis
on activities and behaviour and on the influence of arthritis on these aspects of the partici-
pants’ lives.
Education 401
Strengths
1. Careful planning, based on existing literature.
2. Involvement of patients and therapists at the planning stage.
3. Emphasis on health behaviour rather than knowledge.
Weaknesses
1. Neither the planning stage, nor the package of education itself are well described, so that
there is very little detail to help others who might want to do a similar study.
2. All outcome measures are self-report based, so that there were no other more ’objective’
ways of finding out whether the intervention group’s disease status or function improved.
3. The major weakness (acknowledged in the text) is the absence of a randomized con-
trolled trial, the comparative group were different in age and reported problems than the
study group.
Relevance
A landmark paper in the development of educational packages for patients with arthritis
because of the involvement of patients in planning and the demonstration of long-lasting
benefits on health behaviours.
402
Paper 6
Can the provision of information to patients with

osteoarthritis improve functional status?
Authors
Weinberger M, Theriney WM, Booher P, Katz BP
Reference
Summary
The aim of this study was to find out whether advising patients over the telephone could
affect their functional status in cases of chronic arthritis. Four hundred and thirty-nine
patients with osteoarthritis (OA) were randomly assigned to one of four groups receiving
information in different ways: (a) by telephone once a month, (b) in person at the clinic, (c)
at the clinic and by telephone, and (d) none (i.e. control group). The effect of these inter-
ventions was measured by changes in physical health, pain levels, and psychological well-
being.
Patients who received telephone intervention only, or who personally visited the clinic and
had telephone contact, had less pain and disability than the group of patients who only
attended the clinic. This may be explained by the fact that the number of contacts were con-
siderably more for those two groups that had ready access to their health care providers by
telephone. The authors also propose other explanations: a visit to a clinic may not be the
best forum for giving and receiving information, due to the pressure of time and the distur-
bances one encounters there. Telephone contact, on the other hand, could be made when it
suited the patient, and when the patient was feeling relaxed. In addition, it preserves the
patient’s anonymity. Finally, the telephone is an inexpensive means of giving information,
advice, and encouragement.
Key message
Repetitive information, personal advice, and individual encouragement may be conveniently
and inexpensively given to patients with OA over the telephone. In practice, this resulted in
improved physical health and reduced pain.

This paper presents a new way of conveying information and support to patients, i.e. by use
of the telephone. Osteoarthritis, according to the authors, is the most common chronic dis-
ease affecting adults. The cost of health care for these individuals is high and, with a view
toward the increasing age of the population in the future, the impact of this disease on
society is likely to increase. The common symptoms of OA are pain and loss of mobility. This
may lead to social isolation and feelings of depression. Traditional treatments have been
non-steroidal anti-inflammatory drugs, and exercise programmes. In spite of the fact that ear-
lier studies of patient education have documented the beneficial impact of social support,
there continue to be few publications evaluating the effect of this intervention on patients
with OA.
Education 403
According to this study, pain was the main problem for the patients with OA, physical dis-
ability caused the least discomfort. Those patients receiving information and support by tele-
phone had less pain and physical disability than other groups did. A surprising result was that
clinic intervention only had a detrimental effect on physical functioning! The authors offer
four possible explanations: first, these individuals had considerably fewer overall contacts
than the others did. Secondly, the competence of the telephone advisors was consistently
superior to that of the health care providers in the clinic. Thirdly, patients may have been
too ill to pay attention to information given at the clinic. There were also disturbances of
other patients seeking care in the clinic, and there was time pressure. This was avoided in the
telephone call which could be offered when it suited the patient. Finally a relaxed phone call
with some distance between participant and the care giver may have an advantage over a
face-to-face contact, which may even be threatening to some individuals. As telephone con-
tacts are an inexpensive means of communication, they can be made with greater frequency.
In fact, the mean number of such calls per patient was more than three times the number of
clinic visits by the same patient.
Strengths
1. The response rate was high, with 75% of those offered the opportunity agreeing to take
part in the study; the follow-up rate was equally so (89.7% of the initial number of partici-
pants).
2. The non-respondents and those who failed to follow-up had the same demographic char-
acteristics.
3. An ambitious interviewing programme, involving a great number of subjects, was success-
fully carried out.
Weaknesses
1. The patients taking part in this study were 70% black, and their level of education was
rather low.
2. The authors report in the abstract a ‘marginal improvement of psychological health’ with
a p-value of 0.10. Such a change, so small as to appear almost insignificant, may be a ran-
dom result.
3. The conclusion that telephone contact is an inexpensive yet effective way of communicat-
ing with arthritis patients should be documented by substantial evidence.
Relevance
The authors of this paper raise the issue of providing arthritis patients with social support,
information on coping with pain, and instruction on how to perform physical exercise by
phone. If this strategy proves to be effective, it would be an inexpensive addition to the total
care programme for the management of a very common, disabling disease.
404
Paper 7
The patient is not a blank sheet: lay beliefs and

their relevance to patient education
Authors
Donovan JL, Blake DR, Fleming WG
Reference
Summary
This is a qualitative study based on interviews of 32 patients with RA, both before and after
their first visit to a rheumatology clinic. It included observations made during these consulta-
tions. Interviews were tape-recorded, transcribed, and then analysed using a qualitative
method (1). The study describes each patient’s perception of arthritis, coping strategies, and
expectations in seeking clinical aid. Most of the individuals seen had learned to live with
their arthritis, although five of them stated that arthritis had totally disrupted their lives in
that they were cut off from the life they had lived before being afflicted. Patients did not talk
of symptoms per se (i.e. pain and stiffness) but of how pain had affected their lives. Although
most had found their own coping strategies, they continued to suffer in their effort to keep
going. They expected the doctor to give them information and suggest treatment, which they
then would decide whether to heed or not.
Related references (1) Glaser B, Strauss A. Grounded theory. Chicago: Aldine, 1967.
(2) Moll J. Medical communication: recent aspects and relevance to
rheumatology. In: Recent Advances in Rheumatology (Moll J,
Sturrock R, eds.). Bath: Churchill Livingstone, 1986.
Key message
When planning and evaluating patient education, one must bear in mind that any medical
advice given to a patient still has to pass through the filter of that person’s lay beliefs.
Patients are individuals, not blank sheets. The authors suggest a shift from providing didactic
information to facilitating more active participation of patients in their own therapy.

This article shows that patient education is more complex than generally assumed. Earlier
models of patient education, in which active instruction from rheumatologists and other
health professionals was accompanied by passive listening and the absorption of information
by patients, may not be enough. In order to be successful, patient education must be viewed
as a dialogue between health care provider and patient, i.e. two-way communication.
The method employed in this qualitative study may be new to some medical researchers,
although it is familiar in such other fields as sociology and anthropology (1). A small number
of individuals are interviewed, and their interviews analysed in detail. A tape recording docu-
ments the informal, semi-structured exchange, while the interviewer uses a checklist so that
all questions are asked of all subjects.
Education 405
The article gives the reader a different picture of how patients with arthritis go through
life. Great misconceptions can be avoided when one realizes that views on a disease may dif-
fer considerably between health professionals and patients. For physicians, illness is com-
monplace, symptoms are based on biomedical theories, and treatment is self-evident and log-
ical. For patients, on the other hand, illness is abnormal, symptoms are experienced in terms
of what one can and cannot do, and medical treatment is only one of several options, includ-
ing advice from friends, relatives, or alternative cures promoted in the media. Further, the
prevalent method of evaluating pain with a visual analogue scale (VAS) tends to reduce a
complex feeling to a simple sensation, whereas pain may be influenced by many factors that
cannot be accounted for by putting a mark on a 100 mm line. Conflict arising from these dif-
ferent perceptions of illness and treatment may lead to a guilty conscience in some patients
for non-compliant behaviour, even though the patient is acting on what subjectively feels like
logical reasoning. After the publication of this article, the importance of patients’ views on
what they wanted or needed to know became clearer. Communication is, by definition, a two-
way process (2). Understanding lay-beliefs that individuals bring with them into patient edu-
cation programme sessions began to be considered seriously.
Strength
The methodology employed results in a lucid presentation in which the tone is conversation-
al and the text, unencumbered by tables or diagrams, is easy to follow.
Weakness
It might have been preferable to outline the new methodology, rather than ask the reader to
contact the authors for details.
Relevance
The authors employ a qualitative approach to present information vital to the understanding
of disease from a patient’s point of view. The knowledge imparted here has been used in
planning patient education programmes. After this article appeared, such programmes
changed from lecture format, using slides to tell patients about their disease, to interactive
formats that often incorporated problem-solving strategies in which patients take an active
role in discussing their own individual problems, rather than problems which health profession-
als have selected as common to all patients with arthritis.
407
CHAPTER 26
Key drug developments

Peter M Brooks
Introduction
This century has seen the development of pharmacological therapy for rheumatic diseases.
The century began with the synthesis of salicylic acid, which in some ways spawned the mod-
ern pharmaceutical industry and led to many of the treatments we now take for granted in
rheumatic diseases. In the early part of the 1900s, salicylates were used extensively for pain
relief and as anti-pyretics, although it was not until the 1960s that they were really subjected
to clinical trials. Recognition that salicylates were associated with significant side-effects led to
the development of indomethacin, phenylbutazone, and other non-steroidal drugs (NSAIDs)
which rapidly became the mainstay of pain relief and anti-inflammatory activity. The seminal
observations by John Vane in the early 1970s that aspirin and other NSAIDs blocked the pro-
duction of prostaglandins opened a significant window on inflammation research.
Interestingly, in the last decade of this century the discovery that the cyclooxygenase pathway
is composed of two enzymes has led to the generation of a new class of specific cyclooxyge-
nase-2 inhibitors which seem to provide the benefits of anti-inflammatory effect with a
marked reduction in gastrointestinal adverse events.
The papers I have selected are those that demonstrated, often for the first time in a con-
trolled study, efficacy of a particular drug. Some of the papers, such as the one by Boardman
and Hart on the anti-inflammatory effects of salicylates, compare various doses whilst others
compare two active agents with a placebo as in the study on sulphasalazine in rheumatoid
arthritis (RA), in a double-blind comparison of sulphasalazine with placebo and sodium
aurothiomalate. The paper from David Henry’s group on gastrointestinal complications of
NSAIDs, brought together data from twelve studies of gastro-toxicity and performed a very
valuable meta-analysis showing that some NSAIDs were more toxic than others. The group of
papers on anti-rheumatic drugs begins with the classic study from Tom Fraser in Glasgow
who carried out one of the first controlled clinical trials in medicine comparing gold sodium
aurothiomalate with a placebo as part of the Empire Rheumatism Council Multi-centre Trial.
This is followed by the classic paper by Phillip Hench and his group at the Mayo Clinic
reporting the use of Compound E (hydrocortisone) in severe RA. Controlled studies on
penicillamine, sulphasalazine, and methotrexate are also reported. Inclusion of Fred Wolfe’s
study was important in that it illustrated the strength of large long-term databases in provid-
ing us with practical evidence of drug efficacy. This group of studies ends with the paper
from Maini’s group reporting the first randomized double-blind comparison of monoclonal
antibodies versus placebo in RA. This paper heralded a new age for rheumatology, that of
biologics, a group of drugs that will lead us into the next millenium, raising significant issues
of cost and the potential for long-term side-effects.
After perhaps two decades of activity in the 1960s and 70s when a large range of new
NSAIDs were introduced, rheumatology therapeutics went through a relatively quiescent peri-
od until the last 5 years. As we enter the next century, Rheumatologists do so surrounded by a
range of new anti-rheumatic drugs, specific Cox inhibitors, and biologics. It is an exciting time
for Rheumatologists and, more importantly, for patients with rheumatic diseases. Issues of
long-term safety and of the cost-effectiveness have, however, to be addressed as never before.
408
Paper 1
Clinical measurement of the anti-inflammatory

effects of salicylates in rheumatoid arthritis
Authors
Boardman PL, Hart FD
Reference
Summary
Paper demonstrates the anti-inflammatory properties of salicylates as measured by joint size.
Comparisons between placebo and prednisolone, paracetamol, and high- and low-dose salicy-
lates were made. Prednisolone and high-dose salicylates were effective in reducing joint size
whereas paracetamol and low-dose salicylates were not.
Related references (1) Fremont-Smith K, Bayles TB. Salicylate therapy in rheumatoid

arthritis. Journal of the American Medical Association 1965;
192:1133–1136.
(2) Calabro JJ, Paulus HE. Anti-inflammatory effects of acetyl salicylic
acid in rheumatoid arthritis. Clinical Orthopaedics and Related
Research 1970; 71:124–131.
(3) Capell HA, Rennie JAN, Rooney PJ, et al. Patients compliance: A
novel method of testing non-steroidal anti-inflammatory anal-
gesics in rheumatoid arthritis. Journal of Rheumatology 1979;
6:584–593.
Key message
This was the first paper to demonstrate clearly that high-dose salicylates were effective in the
treatment of the inflammation associated with RA. The paper also showed that low doses of
salicylate were not anti-inflammatory as was paracetamol.

At the time that the study was carried out there were very few NSAIDs available. Salicylates
were widely available and used for the treatment of inflammation and this study was designed
to compare placebo with 7.5 mg of prednisolone daily, 6 g of paracetamol, 5.3 g of salicylate
and 2.6 g of salicylate. Drug treatment was carried out for a 7 day period and the trial was a
cross-over design. In this study, jeweller’s rings were used to assess joint size – this methodolo-
gy was validated against a gauge for measuring joint sizes. Observations were also made
regarding the variation of joint size and grip strength, with grip strength increasing and joint
size decreasing between morning and evening measurements (Figure 26.1).
Key drug developments 409
Four 2 week cross-over studies were carried out comparing prednisolone with placebo,
paracetamol with placebo, high-dose of salicylate with low-dose and low-dose of salicylate with
placebo. Treatment periods were for 1 week with a direct cross-over to the alternative treat-
ment. Sequential analysis was used to show that prednisolone and high-dose salicylate were
both effective in reducing joint size. Grip strength also improved with prednisolone and
high-dose salicylate. Some side-effects were seen on the high-dose salicylate with the major
complaint being tinnitus.
This study showed a number of factors: (a) that high-dose salicylate was anti-inflammatory;
(b) that low-dose salicylate was not anti-inflammatory; (c) the study validated the use of a sim-
ple gauge for measuring joint circumference with the standard jeweller’s rings; and (d) the
study demonstrated a novel (at the time) clinical trial methodology for assessing anti-inflam-
matory properties of rapidly acting drugs.
Figure 26.1 Diurnal variation of joint

4500 0 size and grip strength in eight patients
with rheumatoid arthritis. Joint size is
measured in millimetres (+ indicates
Joint size
improvement). Grip strength is
measured in millimetres mercury,
Joint size
Grip
4000 +10 the sum for each time being plotted.
Grip
3500 +20
AM PM
Strengths
1. The endpoints were validated in comparison to standard methodology.
2. The paper clearly demonstrated efficacy with prednisolone and with high-dose salicylate.
Weakness
The cross-over design did not allow a wash-out between treatments.
Relevance
This was a pivotal trial in demonstrating the anti-inflammatory effects of aspirin.
410
Paper 2
Variability in risk of gastrointestinal complications

with individual non-steroidal anti-inflammatory
drugs: results of a collaborative meta-analysis
Authors
Henry D, L-Y Lim L, Garcia Rodriguez LA, Perez Gutthann S, Carson JL,
Griffin M, Savage R, Logan R, Moride Y, Hawkey C, Hill S, Fries JT
Reference
Summary
This paper presents a systematic review of 12 controlled epidemiological studies relating gas-
trointestinal complications to use of 14 NSAIDs. The relative risks of gastrointestinal compli-
cations with individual NSAIDs were ranked in relation to exposure to ibuprofen. Ibuprofen
ranked lowest or equal lowest for risk in 10 of the 11 studies reviewed. The low risk of serious
gastrointestinal complications of ibuprofen seems to be attributed mainly to lower doses of
the drug used in clinical practice.
Table 26.1 Comparison of comparative toxicity of range of drugs with use of ibuprofen as
reference for calculating relative risks.
Comparator No of studies Pooled relative 95% confidence P value

risk interval for pooled (heterogeneity)
relative risk
Ibuprofen – 1.0* –* –*
Fenoprofen 2 1.6 1.0 to 2.5 0.310
Aspirin 6 1.6 1.3 to 2.0 0.685
Diclofenac 8 1.8 1.4 to 2.3 0.778
Sulindac 5 2.1 1.6 to 2.7 0.685
Diflunisal 2 2.2 1.2 to 4.1 0.351
Naproxen 10 2.2 1.7 to 2.9 0.131
Indomethacin 11 2.4 1.9 to 3.1 0.488
Tolmetin 2 3.0 1.8 to 4.9 0.298
Piroxicam 10 3.8 2.7 to5.2 0.087
Ketoprofen 7 4.2 2.7 to 6.4 0.258
Azapropazone 2 9.2 4.0 to 21.0 0.832
*Reference category for calculating relative risk.

Related references (1) Singh, G, Romey DL, Morfeld D, Fries JF. Comparative toxicity of
non-steroidal anti-inflammatory agents. Pharmacology
Therapeutics 1994; 62:175–191.
(2) Langman MJS, Weil J, Wainwright P, et al. Risk of bleeding peptic
ulcer associated with individual non-steroidal anti-inflammatory
drugs. The Lancet 1994; 343:1075–1078.
Key message
This meta-analysis demonstrates that there are differences in the propensity for NSAIDs to
produce serious gastric complications. While some of these differences are related to dose
(particularly in relation to ibuprofen) there do seem to be some NSAIDs which cause less
gastric toxicity than others.

For many years it has been accepted that all NSAIDs are associated with gastric toxicity and
while this remains true, this meta-analysis does demonstrate that there are differences in the
potential of gastro-toxicity of NSAIDs. The authors had all produced major individual studies
in the gastric toxicity of NSAIDs and were quite rigid in the selection of studies for inclusion
in the meta-analysis. The study reported the relative risk with individual drugs and then pro-
vided a ranking of the individual NSAIDs to produce the data shown in the table. The impor-
tance of dose is emphasized in the paper, particularly in relation to ibuprofen, which tends
to be used in low doses.
The paper reaffirms the gastro-toxicity of NSAIDs as a major cause of morbidity and mor-
tality. The meta-analysis of the available epidemiological data does show wide differences
between individual drugs and the risk of producing gastrointestinal bleeding or ulcer perfo-
ration. Of the drugs in common use, ibuprofen and diclofenac ranked low in toxicity where-
as azapropazone, ketoprofen and piroxicam ranked high.
Strength
This study provides a meta-analysis of all the major studies looking at the relationship
between NSAID use and gastro-toxicity.
Weakness
Doses used for some of the drugs are not necessarily equipotent from an anti-inflammatory
or analgesic point of view.
Relevance
This study emphasizes the differential toxicity of NSAIDs and is important in clinical
practice.
412
Paper 3
N of 1 trials comparing a non-steroidal

anti-inflammatory drug with paracetamol
in osteoarthritis
Authors
March L, Irwig L, Schwarz J, Simpson J, Chock C, Brooks P
Reference
Summary
This study evaluated the efficacy of paracetamol and a NSAID for symptom relief in
osteoarthritis (OA) using n of 1 trial methodology. Patients completed three treatment cycles
of 2 weeks each of paracetamol (1 g twice daily) or diclofenac (50 mg twice daily) in identical
gelatin capsules. Fifteen patients completed the study and five withdrew early having made a
therapeutic decision. Eight of the 20 patients found no clear difference with symptoms being
adequately controlled by paracetamol while five preferred the NSAID. After 3 months nine
of the 20 patients had adequate symptom control with paracetamol alone. Figure 26.2 shows
the pain and stiffness scores in a patient who responded to NSAID.
Figure 26.2 Daily visual

100 Paracetamol analogue scores for
NSAID
pain and stiffness in
80
one patient.
Pain (mm)
60
40
20
0
100
80
Stiffness (mm)
60
40
20
0
0 14 28 42 56 70 84
Day
Related references (1) Williams WJ, Ward JR, Egger MJ, Neuner R, Brooks RH, Clegg DO,
et al. Comparison of naproxen and acetaminophen in a two
year study of the treatment of osteoarthritis of the knee. Arthritis
and Rheumatism 1993; 36:1196–1206.
(2) Dieppe E, Cushnagan J, Jasani MK, McRae F, Watt I. A two year
placebo-controlled trial of non-steroidal anti-inflammatory ther-
apy in osteoarthritis of the knee joint. British Journal of
(3) Dieppe PA Frankel SJ, Toth B. Is research into the treatment of
osteoarthritis with non-steroidal anti-inflammatory drugs misdi-
rected? The Lancet 1993; 341:353–354.
Key message
Significant numbers of patients with OA of the knee previously treated with NSAIDs can have
their pain adequately controlled with paracetamol for at least up to 3 months. N Of 1 studies
are useful for making treatment decisions in OA.

This study demonstrated once again that a significant number of patients with OA of the
knee can be controlled, at least in the short-term, with pure analgesics. It also demonstrated
that there is a subset of patients with OA of the knee who prefer NSAIDs. This study fits with
a number of others suggesting that between a quarter and a third of patients with OA can
obtain sufficient pain relief with pure analgesics. This study also demonstrated the usefulness
of n of 1 trail methodology in a disease such as OA where symptom control is important. N of
1 trials are particularly applicable to general practice and can be used for therapeutic deci-
sion making in an individual patient. The study also demonstrated that patients enjoyed this
type of trial in that they felt they gained a better understanding of their arthritis and became
more involved in the treatment.
Strength
Simple demonstration of n of 1 trial methodology to assist in treatment decisions.
Weaknesses
1. There were significant drop-outs (10) although five patients only did so after they had
made a therapeutic decision.
2. Some of the issues such as use of gelatin capsules to package the active components may
detract from the generalizability.
Relevance
The study demonstrates the relevance of simple trial methodology in determining treatment
decisions and the ease with which such a trial can be conducted. This trial methodology is
particularly relevant to general practice. The study also demonstrates the significant number
of patients with OA of the knee who could be treated with paracetamol rather than a NSAID,
reducing the potential for adverse drug reactions.
414
Paper 4
The effect of a hormone of the adrenal cortex

(17-hydroxy-11-dehydro-corticosterone-compound E)
and of pituitary adreno-corticotropic hormone on
rheumatoid arthritis – preliminary report
Authors
Hench PS, Kendall EC, Slocumb CH, Polley HF
Reference
Proceedings of the Staff Meetings of the Mayo Clinic 1949; 24:181–197
Summary
This study outlines the background to the development of compound E and its use in 14
patients with moderately severe or severe chronic polyarticular arthritis. The patients all had
severe disease.
Compound E was given by intramuscular injection. Placebo injections were also given to
the patients at irregular intervals. Administration of compound E led to a marked reduction
in erythrocyte sedimentation rate (ESR) and resolution of pain and significant disability.
Significant side-effects of acne and hirsutism were noted, but these were felt to be mild.
Related references (1) Empire Rheumatism Council. Multi-centre controlled trial com-
paring cortisone acetate and acetyl salicylic acid in the long-
term treatment of rheumatoid arthritis. Annals of Rheumatic
Diseases 1967; 16:277–289.
(2) Harris ED, Emkey RD, Nicholas JE, et al. Low-dose prednisolone
therapy in rheumatoid arthritis: a double-blind study. Journal of
(3) Kirwan JR and The Arthritis and Rheumatism Council Low-Dose
Corticosteroid Study Group. The effect of glucocorticoids on
joint destruction in rheumatoid arthritis. New England Journal of
Medicine 1995; 333:142–146.
Key message
This was the first study to demonstrate the anti-inflammatory effects of corticosteroids in
patients with RA. The responses to treatments were extremely dramatic and few side-effects
were reported.

This study heralded a new treatment for RA–corticosteroids, although at the time very small
amounts of corticosteroids were available (note the expression of regret from MERCK and
Co. Inc. ‘because of the exigencies of manufacture, no supplies of compound E are expected
for treatment or additional research until some time in 1950 at the earliest, at which time
supplies may still be exceedingly small’). The patients treated with compound E all had
severe disease with a number being temporarily bedridden or wheelchair bound. These
patients all responded dramatically to the treatment with a reduction in pain, swelling and in
many cases the return of the ability to walk unaided. Placebo injections were given on occa-
sions to the patients and the effect observed. In some cases (it is reported) even the doctor
was unaware that a placebo was being given. Few side-effects were reported during this study,
although some patients were noted to have developed hirsutism, a rounding of the face and
some transient abdominal pain. Oedema was also reported to occur on occasions. The gener-
al feeling however was that the medication was relatively safe. Two female patients were also
given small amounts of adrenocorticotropic hormone (ACTH) with dramatic reductions in
stiffness, pain on motion and articular tenderness. ESR was also notably reduced in these
patients (Figure 26.3). This clinical trial, albeit not well controlled, clearly demonstrated the
efficacy of corticosteroids in severe inflammatory RA and this led to the use of corticosteroids
which, of course, are still used today.
110 Figure 26.3 Effect of pituitary

100 ACTH on the sedimentation rate
90
of a 45-year-old woman with
80
moderately severe RA of 5 year’s
Sedimentation role
70
60 duration. The rate decreased
50
markedly during the use of ACTH,
40
30 then increased rapidly. Note the
20 ‘rebound’ between 26 February
10 and 10 March; thereafter the rate
0
100
ACTH increased again.
Mg
0
2 5 8 11 14 17 19 23 26 1 4 7 10
February March
1949
Strengths
1. This is the first time that this medication was used for the treatment of RA.
2. Patients were carefully observed.
Weaknesses
1. There was no real control.
2. Most of the patients were hospitalized and the effect of bed-rest had been significant.
3. In many cases the treatment was short-lived (less than 1 month) and it is perhaps surpris-
ing that so many adverse reactions were observed in this short-term trial.
Relevance
This was the first trial of corticosteroids in the management of RA and this heralded a new
management strategy for this condition. The report and subsequent research on corticos-
teroids in RA led to the award of the Nobel Prize to Dr Hench.
416
Paper 5
Gold treatment in rheumatoid arthritis

Author
Fraser TN
Reference
British Medical Journal 1945; 0:471–475.
Summary
This study evaluates the effect of intra-muscular gold (gold sodium thiomalate) in compari-
son to a placebo in patients with RA over a 12 month period. Of 103 patients entered, 57
received gold injections and 46 the inactive control. The intra-muscular (I.M.) injections
were given into the buttocks at weekly intervals at doses of 10, 20, and 50 mg followed by
nine injections of 100 mg to a total dose of 1 g. A second course in the same dosing schedule
was given to 13 patients receiving active drugs and 28 patients on placebo. Clinical improve-
ment occurred in 82% of patients receiving gold but only 45% of patients on placebo.
Adverse reactions (mainly skin rashes) were reported in 75% of those on gold but also by
37% of patients receiving placebo.
Related references (1) Empire Rheumatism Council: Research Subcommittee. Gold ther-
apy in rheumatoid arthritis. Report of a Multi-Centre
Controlled Trial. Annals of Rheumatic Diseases 1960; 19:95–119.
(2) Champion GD, Graham GG, Ziegler JB. The Gold Complexes.
Balliere’s Clinical Rheumatology 1990; 4:491–534.
(3) Furst DE, Levine S, Srinivasan R, et al. A double-blind trial of high
versus conventional dosages of gold salts for rheumatoid arthri-
tis. Arthritis and Rheumatism 1977; 20:1473–1480.
Key message
Intra-muscular gold (myocrisin) is effective in a group of patients with RA with acceptable
side-effects (primarily skin rashes). A significant proportion of patients with RA also show
both objective and subjective improvement over an observation period of 12 months. A rela-
tively high percentage of patients (37%) reported side-effects in the placebo arm.

Although gold therapy had been used in the treatment of polyarthritis since the 1920s, this
was probably the first reported controlled trial of gold salts and one of the first controlled tri-
als carried out in rheumatology. This study clearly demonstrates the beneficial effects of gold
treatment and also documented the major adverse reactions–skin rashes, mouth ulcers and
proteinuria. Interestingly, a significant number of patients receiving placebo also improved
and this was put down to the fact that both groups were receiving physiotherapy. Patients
were chosen if they had had the condition for 2–5 years, thus eliminating those who might
have a spontaneous remission or those long-standing cases where disease activity might be
difficult to change.
The author made the point that gold therapy was probably the best form of treatment for
RA available at the time but that this should be combined with general physiotherapeutic
and orthopaedic measures.
Strength
This was a controlled and blinded study carried out over a significant period of time.
Weakness
The randomization procedure was not described.
Relevance
This study set a benchmark at the time for the assessment of treatments in RA and the results
remain relevant to the present day.
418
Paper 6
Synthetic D-penicillamine in rheumatoid arthritis

– double-blind controlled study of a high and low
dosage regimen
Authors
Dixon AStJ, Davies J, Dormandy TL, Hamilton EBD, Holt PJL, Mason RM,
Thompson M, Webber JCP, Zutschi DW
Reference
Annals of the Rheumatic Diseases 1975; 34:416–421
Summary
This double-blind controlled multi-centre trail compared the therapeutic and adverse
effects of penicillamine in a dosage of either 600 mg or 1200 mg daily in comparison to a
group of patients receiving standard treatment and a minimal sub-effective dose of penicil-
lamine (12 mg daily). In this 24 week study doses of both 600 mg and 1200 mg of D-penicil-
lamine daily were superior to standard regimen therapy in RA. The higher dose was not
associated with a significantly greater therapeutic benefit but the frequency of adverse reac-
tions (skin rashes, blood dyscrasias and withdrawal from the trial) were higher in the
1200 mg dose group.
Related references (1) Multi-Centre Trial Group. Controlled trial of D-penicillamine in

rheumatoid arthritis. The Lancet 1973; 1:275–285.
(2) Situnayake RD, Grindulis KA, McConkey B. Long-term treatment
of rheumatoid arthritis with sulphasalazine, gold or D-penicil-
lamine: A comparison using life table methods. Annals of
(3) Ahern MJ, Hall ND, Case K, Maddison PJ. D-penicillamine with-
drawal in rheumatoid arthritis. Annals of Rheumatic Diseases
1984; 43:213–217.
Key message
The important finding of this study was that higher doses of D penicillamine (1200 mg daily)
were not associated with a superior therapeutic effect than 600 mg doses daily but were asso-
ciated with an increased frequency of adverse events.

This carefully conducted double-blind multi-centre study addressed the issue of doses of D-
penicillamine in patients with RA and its relationship to efficacy and side-effects. Previous
studies had demonstrated the efficacy of D-penicillamine in RA and suggested that there was
a relationship between dose and adverse events. This study clearly showed differences
between the control group and patients treated with both doses of D-penicillamine.
However, 67% of patients on the 1200 mg dose noted adverse reactions compared to 50% on
the 600 mg dose and 34% of patients on the control (Table 26.2). Serum-copper levels fell
significantly and to the same extent on both dosage regimens. This paper stresses the impor-
tance of using the lowest dose of penicillamine and that an adequate time (4–8 weeks)
should be given for clinical improvement to take place. X-rays of the hands and feet were
taken at day 0 and at 24 weeks and were assessed for the number and severity of erosions and
an overall impression of change. No trends were seen in either of the treatment groups.
Table 26.2 Responses of patients on D-penicillamine and placebo.
Control Daily D-penicillamine dose

600 mg 1200 mg
% improved 26 56 70
% patients with adverse effects 34 50 67
% ceasing treatment due to adverse effects 9 20 40
Strength
This was a double-blind study of two dose regimens of D-penicillamine and a placebo which
demonstrated a higher incidence of side-effects with a high dose of penicillamine.
Weaknesses
1. This study was of relatively short duration (24 weeks).
2. Mean differences on treatment are reported rather than raw data making clinical signifi-
cance hard to interpret.
Relevance
This was the first study to address the issue of doses of D penicillamine and its relationship to
side-effects and efficacy.
420
Paper 7
Sulphasalazine in rheumatoid arthritis – a

double-blind comparison of sulphasalazine
with placebo and sodium aurothiomalate
Authors
Pullar P, Hunter JA, Capell HA
Reference
Summary
Ninety patients with definite or classical RA were randomly allocated to receive sul-
phasalazine (enteric coated tablets of Salazopyrin), sodium aurothiomalate or sulphasalazine
placebo. Sulphasalazine was used in doses of up to 3 g daily. Gold sodium aurothiomalate
was administered I.M. at a dose of 50 mg per week after a 10 mg test dose up to a maximum
of 1 g or until a clinical response had been obtained, whereafter the frequency of injections
was reduced to the minimum necessary to maintain the clinical response. Disease activity was
assessed at 6 weekly intervals over a 24 week period. At 24 weeks, 18 patients were still taking
sulphasalazine, 18 patients sodium aurothiomalate and 14 placebo. The main reason for
stopping treatment were lack of effect (placebo 13, sulphasalazine two, and sodium auroth-
iomalate one patient) and adverse events (placebo two, sulphasalazine nine, and sodium
aurothiomalate 11 patients). The ESR, articular index and index of disease activity demon-
strated significant improvement in those patients on active treatment whereas no changes
were seen with placebo. The major toxicity for sulphasalazine was gastrointestinal intolerance
while rashes were the major cause for stopping the drug in the gold group. Acetylator status
was assessed in 28 of the patients treated with sulphasalazine, with 15 being slow acetylators
and 13 being fast acetylators. Five of the six patients who stopped taking sulphasalazine
because of gastrointestinal intolerance were slow acetylators whereas both patients who
stopped because of lack of effect were fast acetylators. The suggestion is made that toxicity
and lack of efficacy are related to acetylator phenotype.
Related references (1) Svartz N. Salazopyrin, a new sulfanilamid preparation. Acta Medica
Scandinavica 1942; 60:577–598.
(2) Sinclair RJG, Duffy JJR. Salazopyrin in the treatment of rheuma-
toid arthritis. Annals of Rheumatic Diseases 1948; 8:226–231.
(3) McConkey B, Amos R, Durham S, et al. Sulphasalazine in rheuma-
toid arthritis. British Medical Journal 1980; 280:442–444.
(4) Williams H, Ward J, Dahl S, et al. A controlled trial comparing sul-
phasalazine, gold sodium aurothiomalate, and placebo in
rheumatoid arthritis. Arthritis and Rheumatism 1988; 31:702–713.
Key message
Sulphasalazine appears to be an effective second line agent in the management of RA.

This is the first comparative study of sulphasalazine with another disease-modifying anti-
rheumatic drug (sodium aurothiomalate). This study clearly showed that the sulphasalazine
was effective in patients with active RA although the mean duration of disease was greater
than 6 years in all groups, with a range of up 35 years. This may have explained the variable
responses noted at the 24 week period (Table 26.3).
Table 26.3 Significance of improvement (p values) in variables at 24 weeks (Wilcoxon

(two-tailed) test)
Placebo Sulphasalazine Sodium

aurothiomalate
ESR NS <0.005 <0.005

Haemoglobin NS NS NS
Platelet count NS 0.001 NS
Rheumatoid factor NS NS 0.05
Articular index NS <0.001 0.001
Limbering up time NS NS NS
Pain score NS NS <0.005
Grip strength NS <0.005 NS
Index of disease activity NS 0.005 <0.005
Strengths
1. The study design was well conceived with the outcome measures being assessed by a nurse
who was blind to the nature of the therapy.
2. Adverse events were carefully assessed in the study.
3. Acetylator status in those patients receiving sulphasalazine was determined.
4. This study clearly demonstrates the efficacy of Salazopyrin in the treatment of RA with rel-
atively mild side-effects.
Weakness
The raw data for efficacy is not reported in the study and it is therefore hard to assess the
importance of the changes.
Relevance
Since the publication of this and other studies of sulphasalazine, the drug has become a stan-
dard second-line treatment in RA.
422
Paper 8
Comparison of low-dose oral pulse methotrexate

and placebo in the treatment of rheumatoid
arthritis – a controlled clinical trial
Authors
Williams HJ, Willkens RF, Samuelson CO Jr, Alarcon GS, Guttadauria M,
Yarboro C, Polisson RP, Weiner SR, Luggen ME, Billingsley LM, Dahl SL,
Egger MJ, Reading JC, Ward JR
Reference
Arthritis and Rheumatism 1985; 28(7):721–730
Summary
This paper describes a prospective, controlled, double-blind, multicentre trial comparing
placebo with methotrexate 7.5 mg weekly, taken as three 2.5 mg tablets 12 hours apart. One
hundred and ten of 189 patients entered completed 18 weeks of therapy. Methotrexate treat-
ment demonstrated statistically significant improvement over the placebo in the clinical vari-
ables of joint pain/tenderness and swelling, rheumatoid nodules, and patient and physician
assessments of disease activity (Figure 26.4).
During the trial, one-third of patients receiving methotrexate were withdrawn for adverse
drug reactions of which elevated liver enzymes were the most common. All adverse drug
effects resolved without sequelae.
Figure 26.4 Mean

0
change in selected
Painful/tender
-5
joint score
clinical variables by
Number of painful/
-10
tender joints
0 -15 treatment (placebo

-5
-10 -20 versus methotrexate)
-15 -25 at 6, 12, and 18
6 12 18 6 12 18
weeks of therapy.
0 0
swollen joints
Number of
joint score
-5 -5
Swollen
-10 -10
-15 -15
-20 -20
6 12 18 6 12 18
0 0
assessment
assessment
-.2 -.2
Physician
Patient
-.4 -.4
-.6 -.6
-.8 -.8
-1.0 -1.0
6 12 18 6 12 18
Weeks Weeks
Placebo
Methotrexate
Related references (1) Andersen EA, West SG, O’Dell JR, et al. Weekly pulse methotrexate
in rheumatoid arthritis: clinical and immunological effects of a
randomized, double-blind study. Annals of Internal Medicine
1985; 1–3:479–486.
(2) Weinblatt ME, Jonathan SC, Fox DA, et al. Efficacy of low-dose
methotrexate in rheumatoid arthritis. New England Journal of
Medicine 1985; 312:818–822.
(3) Kremer JM, Joong KL. The safety and efficacy of the use of
methotrexate in long-term therapy for rheumatoid arthritis.
Key message
This was the first double-blind trial of low-dose pulse methotrexate in RA to demonstrate effi-
cacy. The size of the study provided sufficient power to detect a difference between the two
treatments at the 5% significance level. One hundred and eighty-nine patients fulfilled the
entry criteria, but only 110 completed the 18 weeks of therapy, 41 withdrawing from the
placebo and 39 from methotrexate. Eighteen patients on methotrexate developed elevated
liver enzymes and five developed mucosal ulcers. The study showed that 60% of patients
were able to tolerate methotrexate but did identify an incidence of elevated liver function
tests. Pancytopenia was also identified in two patients. The conclusion of the study was that
methotrexate appeared to be effective in the treatment of RA, but did require close monitor-
ing for toxicity.
Strengths
1. This is a carefully constructed prospective, controlled, double-blind multicentre study.
2. Sufficient power to demonstrates differences between placebo and methotrexate.
Weakness
In terms of a lifetime of RA, it is a relatively short-term study.
Relevance
Methotrexate is still one of the most commonly used anti-rheumatic drugs and its long-term
efficacy seems to bear out the initial findings demonstrated by Williams, et al.
424
Paper 9
Termination of slow acting antirheumatic therapy

in rheumatoid arthritis: a 14 year prospective
evaluation of 1017 consecutive starts
Authors
Wolfe F, Hawley DJ, Cathey MA
Reference
Journal of Rheumatology 1990; 17(8):994–1002
Summary
This paper provides a very careful review of the treatment outcome with over 1000 new
administrations of slow acting anti-rheumatic drugs (SAARDs) in 671 patients followed over a
14 year period. The important observations from this study was the median time for discon-
tinuation of drugs and the reason for that discontinuation. The median time for discontinua-
tion for I.M. gold or auranofin, hydroxychloroquine or penicillamine was 2 years or less, but
was over twice that (4.25 years) for methotrexate (Figure 26.5). More patients had to cease
their therapy because of an adverse event than for lack of efficacy.
Figure 26.5 Product limit

1.0
(Kaplan–Meier) cumulative survival
1.9 Methotrexate
All other SAARDs
analysis of terminations because of
1.8
Cumulative probability of
lack of efficacy. Cases terminated

remaining on SAARDs
1.7
because of adverse reactions are
1.6
excluded from this analysis. All
1.5
other SAARD represent the
1.4
combined group of I.M. gold,
1.3 P = 0.007 auranofin, hydroxychloroquine,
1.2
and penicillamine. Difference in
1.1
cumulative survival is significant
0
0 1 2 3 4 5 6 7 8 9 10 11 12
at the 0.007 level (Mantel-Cox).
Years
Related references (1) Jessop JD, O’Sullivan MM, Lewis PA, Williams LA, Camilleri JP,
Plant MJ, Coles EC. A long-term five year randomized trial of
hydroxychloroquine, sodium aurothiomiolate, auranofin and
penicillamine in the treatment of patients with rheumatoid
arthritis. British Journal of Rheumatology 1998; 37:992–1002.
(2) Van de Putte LBA, Van Gestl AM, Van Riel PLCM. Early treatment
of rheumatoid arthritis–rationale, evidence and implications.
(3) Egmose C, Lund V, Borg G, Petersson H, Berg B, Brodin U, Trang
L. Patients with rheumatoid arthritis benefit from second line
therapy–five year follow-up of a prospective, double-blind,
placebo-controlled study. Journal of Rheumatology 1995;
22:2208–2213.
Key message
Specific antirheumatic drugs are still used widely for the management of RA. Despite their
usefulness, there has been growing concern about the side-effects of these agents and the
relative efficacy of them. This study was made possible by the establishment and careful fol-
low-up of a cohort of patients with RA who were started on a variety of slow-acting anti-
rheumatic drugs over a period of some 14 years. In this study, 671 patients were com-
menced on 1017 SAARD starts. These patients were all followed at regular intervals and
patients were only started once they had already received a second-line agent (this paper
reports the history of a second slow-acting anti-rheumatic drug). Over the period of the
study (1975–88) the type of drug varied with I.M. gold being commonly used initially, then
penicillamine and finally, methotrexate. Those patients commenced on hydroxychloro-
quine and auranofin seemed to have milder disease while those on methotrexate and peni-
cillamine were more likely to have increased levels of disability and a higher prednisolone
dose. The major outcome of this study was the length of time that patients remained on a
slow-acting antirheumatic therapy (Figure 26.6). Those patients on methotrexate remained
on therapy for significantly longer than those patients on I.M. or oral gold, hydroxychloro-
quine, or penicillamine. The reasons for discontinuation included both adverse reactions
and loss of efficacy. Discontinuation for adverse reactions was more common for all drugs
than discontinuation for lack of efficacy, but those patients taking methotrexate remained
on the drug for significantly longer periods of time than those on the other medications.
Very few (6 of 580) terminations were for remission. The major reasons for drop-out were
gastrointestinal and oral ulceration (on methotrexate), proteineuria (on I.M. gold), gastro-
intestinal (on auranofin) and haematological or renal reasons for penicillamine termina-
tion. Nearly a third of patients terminated therapy because of adverse reactions, while 10%
of terminations were due to a lack of efficacy.
Strengths
1. The paper presented a significant number of patients followed for a very long time
(14 years).
2. Provided practical data to assess and compare individual anti-rheumatic drugs, to assess
efficacy and contrast agents.
3. Provided comparative data on side-effects.
Weaknesses
1. The study failed to identify any clear predictors of discontinuation.
2. There were significant differences in the patients treated with different drugs (such as dis-
ease duration which was significantly longer in those on methotrexate than those treated
with gold), in the level of disability, and the dose of prednisolone. (Although these may
have been compensated for somewhat by the large groups involved, they need to be taken
into account in interpretation.)
Relevance
This paper demonstrated the importance of the development of long-term databases to
investigate therapeutic advances in the management of RA.
426
Paper 10
Randomized double-blind comparison of chimeric

monoclonal antibody to tumour necrosis factor
alpha (cA2) versus placebo in rheumatoid arthritis
Authors
Elliott MJ, Maini RN, Feldmann M, Kalden JR, Antoni C, Smolen JS, Leeb B,
Breedveld FC, Macfarlane JD, Bijl H, Woody JN
Reference
The Lancet 1994; 344:1105–1110
Summary
This was the first study of a monoclonal antibody to tumour necrosis factor (TNF) used in a
clinical trial. This randomized double-blind trial compared a single infusion of two doses
(1 and 10 mg/kg cA2 with placebo) in 73 patients with active RA. At the end of week four
using a Paulus 20% response, only four of 24 placebo recipients responded, while 11 of 25
treated with low-dose cA2 and 19 of 24 treated with high-dose cA2 fulfilled these criteria.
These results provided the first good human evidence that specific cytokine blockade can be
effective in RA.
Related references (1) Wallis WJ, Furst D, Strand V, Keystone E. Biologic agents in
immuno-therapy in rheumatoid arthritis–progress and perspec-
tive. Rheumatic Disease Clinics of North America 1998;3:537–565.
(2) Choy EHS, Kingsley GH, Panayi GS. Monoclonal antibody therapy
in rheumatoid arthritis. British Journal of Rheumatology 1998;
37:484–490.
(3) Moreland LW, Baumgartner SW, Schiff MH, et al. Treatment of
rheumatoid arthritis with a recombinant human tumor necrosis
factor receptor (p75)–Fc fusion protein. New England Journal of
Medicine 1997; 337:141–147.
(4) Elliott MJ, Maini RN, Feldmann M, et al. Repeated therapy with
monoclonal antibody to tumour necrosis factor (cA2) in
patients with rheumatoid arthritis. The Lancet 1994;
344:1125–1127.
Key message
This paper presents compelling clinical evidence that chimeric tumour necrosis alpha plays a
significant role in inflammation in RA. The study was extremely carefully conducted, using
two doses of TNF antibody and placebo in patients with RA. Significant reductions in clinical
features of swollen and tender joint count, pain and overall disease activity were matched
with reductions in C-reactive protein and ESR. Both clinical and laboratory parameters fell
very rapidly within the first week and remained low (in the high-dose patients) over a period
of 4 weeks. The primary endpoint of the study was the achievement in week four of a Paulus
20% response and this was achieved in 19 of 24 patients treated with high-dose. Using the
more stringent 50% Paulus criteria, over half of the high-dose patients responded. One of
the disappointing features of the study was the relatively short efficacy achieved, particularly
with low-dose, with clinical parameters returning towards normal at the end of the 4 week
treatment period. In this small study, minor side-effects were noted, although one patient
developed pneumonia, which was signified as possibly treatment related.

This study, the first of a series, demonstrated that monoclonal antibodies are useful in sup-
pressing pain and inflammation and defined a new direction for the treatment of RA.
Strength
This is the first study to demonstrate the efficacy of monoclonal antibodies in the treatment
of RA.
Weaknesses
1. There were a relatively small number of patients in the study.
2. It was only continued for a 4 week observation period.
Relevance
This study provides one of the benchmarks for the management of RA with monoclonal
antibodies.
428
CHAPTER 27
Exercise and rehabilitation

in arthritis
Michael V Hurley
Introduction
This section summarizes important studies that investigated the role of muscle dysfunction
and exercise in the management of patients with osteoarthritis (OA) and rheumatoid arthri-
tis (RA). The studies have been placed in the context of the research at the time they were
devised and published, and includes later work to show how the fields of research developed
from these early studies. It may be surprising how recent most of these ‘classic’ papers are.
The reason for this is that research interest and expertise have only relatively recently been
developed in the field of rehabilitation. However, the findings from these studies demon-
strate why and how patients with arthritis should exercise to limit the affects of arthritis.
The first two papers highlight why. They show the intimate, but usually disregarded, rela-
tionship between peri-articular muscles and intra-articular structures, and emphasize that
synovial joints are better viewed as functional units of movement composed of bone, cartilage,
nerves and muscles. Unfortunately we know very little about the arthritic changes that occur
to muscle. Since reversing muscle dysfunction is the main aim of exercise, it is essential to
know what changes occur to muscles and understand the implications of these changes, if we
are to decide whether patients with arthritis should bother exercising at all. Several papers by
Freeman and Wyke in the 1960s showed alteration of joint mechanoreceptors caused muscle
sensoriomotor dysfunction (Paper 1), which has subsequently been shown to be associated
with disability. The very recent paper by Slemenda, et al. (Paper 2) showed quadriceps weak-
ness may precede knee joint damage, and challenges the normal assumption that arthritis
causes muscle weakness, suggesting muscle dysfunction may cause arthritic damage.
Whether muscle dysfunction is a cause or a consequence of arthritis and results in disabili-
ty, muscle is the most ‘plastic’ component of the functional unit of movement and can be
manipulated relatively easily, therefore improving muscle function may ameliorate the
effects of arthritis. Three papers demonstrate exercise can improve muscle function,
decrease pain and increase function without exacerbating the condition. Minor, et al.’s paper
(Paper 3) suggests specific, sophisticated exercise regimes may not be essential, but any form
of increased physical activity is beneficial. In their paper Nordemar, et al. (Paper 5) showed
that long-term exercise did not cause long-term damage as has been feared. The final paper
by Chamberlain, et al. (Paper 4) demonstrates exercising at home is as effective as exercising
under supervision in hospital outpatient departments, which improves the cost-effectiveness
of exercise. It will also improve patients’ adherence, which is important since the size of the
patient population with arthritis is large and increasing. This paper also highlights the
importance of continuing to exercise to maintain improvements attained during exercise
and prevent patients regressing. Strategies need to be developed that will enable patients
with arthritis to obtain and maintain benefits from exercise.
Exercise and rehabilitation in arthritis 429
These ‘classic’ papers were not selected because they are definitive studies that are beyond
reproach (they have design flaws that may have inflated their results): these papers are ‘clas-
sics’ because they were innovative and initiated new lines of enquiry. Scientifically rigorous
studies must confirm their findings, and the regimes refined to enhance their effectiveness
and assess their costs, so that effective, affordable, clinically applicable regimes can be imple-
mented that will lead to better management of arthritis. These, however, will be variants of
the foundations laid down by the pioneering studies described here.
430
Paper 1
Articular reflexes at the ankle joint: An

electromyographic study of normal and abnormal
influences of ankle-joint mechanoreceptors upon
reflex activity in the leg muscles
Authors
Freeman MAR, Wyke B
Reference
Journal of Bone and Joint Surgery 1967b; 54:990–1001
Summary
In up to 17 neurologically intact, lightly anaesthetized cats, passive limb movement and
direct stimulation of articular mechanoreceptors and nociceptors altered the EMG activity of
the muscles acting across the joint. Variations in discharge from slowly and rapidly adapting
mechanoreceptors produced coordinated reflex changes in motor unit activity in limb mus-
cles, increasing activity in some muscle groups and decreasing activity in others. Intra-articu-
lar injections of local anaesthetic abolished the articular reflexes by destruction of the articu-
lar mechanoreceptors or interruption of the afferent fibres.
Related references (1) De Andrade JR, Grant C, Dixon ASJ. Joint distension and reflex
muscle inhibition in the knee. Journal of Bone and Joint Surgery
1965; 47(A):313–322.
(2) Ferrell WR, Baxendale RH, Carnachan C, Hart IK. The influence
of joint afferent discharge on locomotion, proprioception and
activity in conscious cats. Brain Research 1985; 347:41–48.
(3) Hurley M, Scott DL, Rees J, Newham DJ. Sensorimotor changes
and functional performance in patients with knee osteoarthri-
Key message
Discharge from articular mechanoreceptors contributes to posture and movement by reflex
coordination of muscle tone. Alteration to the articular environment may alter discharge
from the articular receptors resulting in abnormal posture and movement.

For some time articular receptors had been known to project to spinal and higher centres,
and they were considered to be involved in the perception of posture and movement, but the
role of articular receptors in the reflex control of muscle tone was unclear. This paper
demonstrated articular mechanoreceptors were involved in the coordinated activity and con-
trol of muscles across the joint. It established a link between what happens in the articular
environment and how this affects motor control via alpha-motor neurones and gamma-
motor neurones. The intimate relationship between the joint and muscle activation and con-
trol confirmed the findings of other studies from around this time (1). In addition it estab-
lished the important link between changes in the articular environment and the sensory
function of muscle. This relationship suggests a synovial joint should be considered a func-
tional unit comprised of the articular structures, neurones that innervate these structures
and the muscles that acts across the joint, effecting movement and protecting the joint via
neuromuscular protective mechanisms.
Alteration in motor control affecting posture and movement of conscious cats has been
reported in later studies (2). More recently the link between arthritic joint damage and mus-
cle sensorimotor dysfunction has gained increasing interest in terms of its consequences of
knee OA on muscle strength, motor control and proprioceptive acuity (3).
Strength
Thorough methodology from experienced research group.
Weaknesses
1. Animal work.
2. Small numbers of animals.
3. Possibility that anaesthesia may be interfering with findings.
Relevance
This study reinforced the intimate physiological relationship between joints and muscles:
arthritic damage of intra-articular structures will affect function of the peri-articular muscles,
and muscle dysfunction (weakness or proprioceptive deficits) will impair neuromuscular pro-
tective mechanisms, exposing the joint to damage.
432
Paper 2
Quadriceps weakness and osteoarthritis of the knee

Authors
Slemenda C, Brandt KD, Heilman DK, Mazzuca S, Braunstein FM,
Katz BP, Wolinsky FD
Reference
Summary
This cross-sectional study investigated the prevalence of muscle weakness, pain, function and
the extent of joint damage in 462 community-based volunteers aged 65 years and over. Joint
damage was assessed from radiographs taken of the knees, the Western Ontario and
MacMaster’s University Index for OA knees (a disease-specific questionnaire) was used to
assess pain and function, knee extensor and flexor strength and the lower limb lean tissue
mass were measured. Quadriceps, but not hamstring, weakness was common, and was pre-
sent in people who had normal lower limb muscle mass. In women with tibiofemoral OA
quadriceps weakness was evident even in the absence of pain. Quadriceps weakness was pre-
dictive of radiographic and symptomatic knee OA. Quadriceps weakness may be a risk factor
for knee pain, disability and the progression of joint damage.
Related references (1) Hurley M, Scott DL, Kees J, Newham DJ. Sensorimotor changes
and functional performance in patients with knee osteoarthri-
(2) McAlindon TE, Cooper C, Kirwan JR, Dieppe PA. Determinants of
disability in osteoarthritis of the knee. Annals of Rheumatic
Diseases 1993; 52:258–262.
(3) Pai Y-C, Rymer, WZ, Chang RW, Sharma L. Effect of age and
osteoarthritis on knee proprioception. Arthritis and Rheumatism
1997; 40:2260–2265.
Key message
Quadriceps weakness is common in patients with and without radiological evidence of knee
OA, and may be a factor in the aetiology of the condition.

Slemenda, et al.’s paper took an innovative look at the involvement of quadriceps muscle in
knee OA by considering the mechanism through which quadriceps exercise could reduce
the pain of knee OA. In their introduction and an accompanying editorial they highlighted
how little research had been performed into the changes in muscle that occur in knee OA,
and that the link between muscle weakness and knee pain was poorly understood. Using a
large community population this paper confirmed other studies demonstrating that quadri-
ceps weakness is a common finding, which is associated with knee OA and disability (1, 2). It
also showed that people with weak muscles had normal muscle mass suggesting the cause of
the weakness was due to muscle dysfunction rather than atrophy. But the most important
aspect of the paper was its proposal that muscle weakness preceded and predicted radiologi-
cal damage, thus muscle dysfunction might be a factor in the aetiology of knee joint disease.
They suggest this may occur by impairment of neuromuscular protective mechanisms that
play a vital role in shock absorption during gait. The role of muscle in arthritis is currently
gaining much more research attention (1, 3). Improving muscle function may be a strategy
for retarding or preventing the progression of joint damage.
Strengths
1. Large representative community population.
2. Well designed, well written report.
Weaknesses
1. Small numbers in some sub-groups limit the inferences that can be drawn from the joint
compartments affected.
2. Cross-sectional study, therefore what happens over time cannot be assessed.
Relevance
It was an innovative view of the role of quadriceps weakness in patients with knee OA. There
was a lamentable lack of research and understanding of the role muscles might play in the
cause and treatment of arthritis. The findings implicated muscle weakness as a factor in the
aetiology of knee OA. Since muscle is a very ‘plastic’ tissue, rehabilitation that improves mus-
cle function may be a strategy in the primary or secondary prevention of joint damage.
434
Paper 3
Efficacy of physical conditioning exercise in patients

with rheumatoid arthritis and osteoarthritis
Authors
Minor M, Hewett J, Webel R, Anderson S, Kay D
Reference
Summary
One hundred and twenty patients with symptomatic RA (n = 40) or OA (n = 80) in weight-
bearing joints were recruited and baseline assessment of their aerobic capacity, exercise
endurance, clinical disease activity, objective tests of functional performance, and the physi-
cal and psychological impact of arthritis were performed. The patients were randomized into
one of three exercises groups: (a) a land-based walking programme, (b) a pool-based exercis-
es programme, or (c) range of motion exercise. Groups (a) and (b) had an aerobic compo-
nent to see if the programmes increased aerobic fitness, and group (c) had no aerobic com-
ponent and acted as a control group. All the participants took part in supervised exercises
for 1 hour per week for 12 weeks. After 12 weeks the assessments were repeated, and the
patients were recalled for follow-up assessment 6 and 12 months from baseline. Immediately
following exercise, aerobic capacity and objective and subjective variables improved in both
aerobically exercising groups compared to control group, without exacerbation of disease
activity. At 12 month follow-up, 50–60% of the participants of all the groups reported per-
forming more than 60 minutes of physical activity (e.g. formal exercise, household or gar-
dening activities). The improvements of the aerobic exercise groups had been maintained,
but in addition the aerobic fitness of the control group had also improved and there were
now few differences between the aerobically exercised groups and the control group.
Related references (1) Fisher NM, Gresham GE, Abrams M, Hicks J, Horrigan D,
Pendergast DR. Quantitative effects of physical therapy on mus-
cular and functional performance in subjects with osteoarthritis
of the knees. Archives of Physical Medicine and Rehabilitation 1993;
74:840–847.
(2) Hurley MV, Scott DL. Improvements in quadriceps sensorimotor
function and disability of patients with knee osteoarthritis fol-
lowing a clinically practicable exercise regime. British Journal of
Rheumatology 1998; 37:1181–1187.
Key message
Exercise in RA and OA is effective.

This study was instigated because the advantages of regular activity and the dangers of inac-
tivity were well known, but there was concern about advising arthritic patients to exercise
because the benefits and risks of exercise for arthritis were contentious. Some studies had
shown exercise for arthritis was effective and safe, but these usually involved too few patients
and were poorly designed. This relatively large randomized controlled trial showed exercise
for arthritis was efficacious and did not exacerbate disease activity in the short/medium-
term.
An interesting but unexpected finding was the delayed improvement in aerobic capacity in
the control group. The authors hypothesize this may indicate that habitual performance of
any form of physical activity is more important than a specific exercise regime. Later reports
confirmed that long-term exercise in RA was safe (Paper 4), and studies of OA also showed
exercise was beneficial (1). However, although these research regimes were efficacious they
were prolonged and labour intensive, making them difficult, very expensive to implement
clinically, and have poor patient adherence. Recently exercise regimes have been devised
that are shorter but just as effective and safe, and clinically are more practicable (2).
Strengths
1. Randomized controlled trial.
2. Relatively large sample size for the initial intervention part.
Weaknesses
1. Incomplete follow-up due to the drop-out rate, this will decrease the statistical power of
the study at 12 month follow-up, and volunteer bias of those that remained in the study
will exaggerate these findings.
2. It is unclear how well the self-report exercise habits were verified.
3. From the report the actual method of randomization is unclear, whether an intention-to-
treat analysis was used, and whether the evaluators were blind to the interventions, and if
the patients were aware of the aim of the study.
4. The alpha-level was set at <0.05, but as no primary outcome variable was stated a priori and
many variables are reported, the chances of a type II statistical error increases.
Relevance
In spite of the limitations of some aspects of design and omissions in the report this study
demonstrated exercise for arthritis is efficacious and safe, and that if patients remained phys-
ically active these benefits are maintained. Furthermore these benefits could be attained irre-
spective of the type of activities that are performed, so patients could benefit from formal
exercises or with routine common activities of daily living, which will improve the likelihood
of patient adherence.
436
Paper 4
Physiotherapy in osteoarthritis of the knee.

A controlled trial of hospital versus home exercises
Authors
Chamberlain MA, Care G, Harfield B
Reference
International Journal of Rehabilitation Medicine 1982; 4:101–106
Summary
Forty-two patients with radiographic and clinical evidence of knee OA had their pain, func-
tion, range of movement, ability to lift weights and endurance assessed before and after a
rehabilitation regime. They were randomly allocated to receive either: (a) 12 sessions of
short wave diathermy and supervised exercise (three times a week for 4 weeks) in an outpa-
tient department and daily home exercises to be performed twice a day; or (b) three super-
vised sessions of the same exercises in the outpatient department and twice daily home exer-
cises. Both the groups were then discharged with the same advice to perform the exercises
daily at home and asked to record their exercise in an ‘exercise diary’. Some patients
(regardless of whether they were in group (a) or (b)) were told they would be recalled for re-
assessment after 12 weeks, while others were not told they would be recalled. Initially pain
decreased and function increased by similar amounts in groups (a) and (b). At 12 week re-
assessment, 10/14 patients who had been informed in advance they would be re-assessed
continued to exercise, and the improvements attained during the initial training phase of
the study had been maintained. Only 2/15 patients who were not told they would be re-
assessed had continued to exercise, and their improvements in pain and function attained
during the initial phase of the study had been lost.
Related references (1) Green J, McKenna F, Redfern FJ, Chamberlain MA. Home exercis-
es are as effective as outpatient hydrotherapy for osteoarthritis
of the hip. British Journal of Rheumatology 1993; 32:812–815.
(2) O’Reilly S, Muir K, Doherty M. Effectiveness of home exercises on
pain and disability from osteoarthritis of the knee: a random-
ized controlled trial. Annals of Rheumatic Diseases 1999; 58:15–19.
Key message
Simple daily exercises performed at home or in hospital improved pain and function in
these patients. These benefits were maintained if the patients continued exercising.

The study was prompted by the length of time physiotherapists spent teaching exercise to
large numbers of patients with knee OA, without any good indication that this use of
resources was beneficial. A pilot study had suggested with adequate advise home exercises
were as effective as supervised hospital treatment. This was borne out in the larger study
described above. Several studies were beginning to show the efficacy of exercise in arthritis,
but these regimes were prolonged and usually necessitated outpatient treatment. Home exer-
cises are arguably more useful because they avert the need for outpatient treatment which
will improve adherence to exercise regimes and cost-effectiveness.
It is also important because the size of the patient population with knee OA is so large,
resources are not available to enable them to be treated in outpatient departments. The
effectiveness of home exercise has been confirmed in a later study of hydrotherapy for hip
OA by the same research group (1), and in a recent larger study of knee OA (2). The study
also demonstrated that adherence with a regular home exercise programme can be
improved by ensuring patients can perform a simple exercise regime effectively and safely,
giving the patients written instructions, and reinforcing the importance of exercise with a
diary and follow-up appointment.
Strength
Randomized and blind evaluation.
Weaknesses
1. In some aspects the design and reporting of the study was poor: only one reference cited,
the randomization was poorly described, the presentation of the results makes it difficult
to assess the size of the changes to enable determining of whether the results were clini-
cally meaningful as well as statistically significant.
2. Some of the objective outcome measures may be inappropriate, invalid and insensitive.
3. Patient numbers were small.
Relevance
Exercise can effectively alleviate the cardinal symptoms of knee OA (pain and disability), but
because of limited resources and the size of the patient population long-term hospital outpa-
tient treatment is not feasible. This study demonstrated that home exercise can be just as
effective as supervised hospital treatment, cheaper and that relatively simple strategies can be
employed to improve adherence so that the exercise and hence the benefits can be maintain
in the longer term.
438
Paper 5
Physical training in rheumatoid arthritis:

a controlled long-term study
Authors
Nordemar R, Ekblom B, Zachrisson L, Lundqvist K
Reference
Scandinavian Journal of Rheumatology 1981; 10:17–30
Summary
In 23 patients with RA the effects of long-term exercising were observed and compared with
a control group (n = 23) who did not undergo regular exercise. Initially training consisted of
supervised outpatient group exercise, followed by discharge with a static exercise bike for use
at home, but over time other recreational exercise activities were encouraged, e.g. swimming,
golf, cycling or walking. The patients were followed regularly over 4–8 years, and during this
time the training group showed less progress in joint degeneration, better physiological tests,
improved clinical variables, reported less sick leave or hospitalized time, and better disease
outcomes than the control patients who did not exercise.
Related references (1) Stenstrom C, Lindell B, Swanberg P, Harms-Ringdahl K,

Nordemar R. Intensive dynamic training in water for rheuma-
toid arthritis functional class II: A long term study of effects.
Scandinavian Journal of Rheumatology 991; 20:358–365.
(2) Van den Ende CHM, Vliet Vlieland TPM, Munneke M, Hazes
JMW. Dynamic exercise therapy in rheumatic arthritis: a system-
atic review. British Journal of Rheumatologv 1998, 37:677–687.
(3) Bearne LM, Scott DL, Hurley MV. Exercise does not increase clini-
cal disease activity measures or plasma pro-inflammatory
cytokines in rheumatoid arthritis. Arthritis and Rheumatism 1999;
42(suppl.):530.
Key message
Exercise does not exacerbate disease and may improve some of the effects of the disease.
Long-term physical activity is not detrimental to the joints of people affected by RA, inactivity
can be.

There had been much controversy as to whether exercise was good or detrimental for
patients with inflammatory arthritis. Exercise was considered beneficial for the health of the
muscles, articular structures and the patient’s general health, but it was argued that exercis-
ing inflamed joints might increase pain and cause persistent inflammation and joint damage.
In clinical practice exercise of patients with well-controlled RA had shown the benefits of
exercise. This paper, and a later study by the group (1) confirmed exercise improved objec-
tive and subjective variables affected by the disease, increased function, with possible eco-
nomic benefits. Although the patients were at risk from injury, this risk was no greater than
other populations, and the benefits are likely to outweigh these risks.
A recent systematic review supports the benefits of dynamic exercise in patients with RA
(2), and in addition more evidence is accruing showing that neither clinical nor biochemical
(plasma concentration of pro-inflammatory cytokines) measures of disease activity are
increased by exercise (3). These reports refute the theoretical arguments that exercise in RA
is dangerous.
Strength
Longitudinal cohort study.
Weaknesses
1. Methodological deficits (lack of randomization, blinding, probability of strong influence
of volunteer bias) will have inflated some of the positive results of the study.
2. Training regime was not well described and varied from 4–8 years, with a mean of 5.4
years.
3. Relatively small patient numbers.
Relevance
If exercise is effective it must be performed regularly or the benefits will be lost. Fears that
long-term exercise may accelerate joint damage in inflammatory joint conditions appear to
be unfounded. Effective exercise for RA can be performed over a period of time without fear
of joint destruction.
441
INDEX
Classic Papers by chapter

1 Clinical outcome measurement
A table for the degree of involvement in chronic arthritis, 4
The pain chart, 6
Therapeutic criteria in rheumatoid arthritis, 8
Quantitation of the activity (manifestations) of rheumatoid arthritis, 10
2 Imaging
The skiagraphy of rheumatoid arthritis, 16
Typical alterations of the bones in gout, 18
A case of arthropathia psoriatica, 19
Abnormal calcium deposition inside the knee, a contribution to the question of the primary
‘meniscopathy’, 20
Computerized tomography of sacroiliitis, 22
Early detection of carpal erosions in patients with rheumatoid arthritis: A pilot study of magnetic
resonance imaging, 24
Rheumatoid arthritis with death from medullary compression, 26
Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films, 28
Radiological assessment of osteo-arthrosis, 30
3 Serological tests
On the occurrence of a factor in human serum activating the specific agglutination of sheep blood
corpuscles, 34
Presentation of two bone marrow elements: The ‘tart cell’ and the ‘LE’ cell, 36
Characteristics of a soluble nuclear antigen precipitating with sera of patients with systemic lupus
erythematosus, 38
Mixed connective tissue disease – an apparent distinct rheumatic disease syndrome associated with a
specific antibody to an extractable nuclear antigen (ENA), 40
Characterization of a soluble cytoplasmic antigen reactive with sera from patients with systemic lupus
erythematosus, 42
Identification of a nuclear protein (Scl-70) as a unique target of human antinuclear antibodies in
scleroderma, 44
Myositis autoantibody inhibits histidyl-tRNA synthetase: a model for autoimmunity, 46
4 Other investigations: synovial fluid analysis, arthroscopy and blood biochemistry

Examination of synovial fluid as a diagnostic aid in arthritis, 50
Identification of urate crystals in gout synovial fluid, 52
The significance of calcium phosphate crystals in the synovial fluid of arthritic patients: the ‘pseudogout
syndrome’. II. Identification of crystals, 54
Usefulness of synovial fluid analysis in the evaluation of joint effusions. Use of threshold analysis and likeli-
hood ratios to assess a diagnostic test, 56
Arthroscopy of the knee in rheumatic diseases, 58
The assessment of rheumatoid arthritis: a study based on measurements of the serum acute-phase
reactants, 60
Cartilage oligomeric matrix protein: a novel marker of cartilage turnover detectable in synovial fluid and
blood, 62
442
5 Clinical genetics
HLA-Dw4 in adult and juvenile rheumatoid arthritis, 66
Ankylosing spondylitis and HL-A27, 68
HLA-DR antigens and toxic reactions to sodium aurothiomalate and D-penicillamine in patients with
rheumatoid arthritis, 70
Transmission disequilibrium as a test of linkage and association between HLA alleles and pauciarticular-
onset juvenile rheumatoid arthritis, 72
The shared epitope hypothesis. An approach to understanding the molecular genetics of susceptibility to
HLA heterozygosity contributes to susceptibility to rheumatoid arthritis, 76
6 Rheumatoid arthritis
1958 revision of diagnostic criteria for rheumatoid arthritis, 82
Prevalence of rheumatoid arthritis, 84
The ameliorating effect of pregnancy on chronic atrophic (infectious rheumatoid) arthritis, fibrositis, and
intermittent hydrarthrosis, 86
Participation of monocyte-macrophages and lymphocytes in the production of a factor that stimulates
collagenase and prostaglandin release by rheumatoid synovial cells, 88
Appearance of anti-HLA-DR reactive cells in normal and rheumatoid synovium, 90
Measurement of patient outcome in arthritis, 92
When does rheumatoid arthritis start?, 94
Agglutination of erythrocytes coated with ‘incomplete’ anti-RH by certain rheumatoid arthritic sera and
some other sera, 96
Identification of immunoglobulins and complement in rheumatoid articular collagenous tissues, 98
Prognosis of rheumatoid arthritis. A prospective survey over 11 years, 100
Studies on the depressed hemolytic complement activity of synovial fluid in adult rheumatoid arthritis, 102
Orthopedic surgery in rheumatoid arthritis, 104
Peripheral vascular obstruction in rheumatoid arthritis and its relation to other vascular leisons, 106
7 Chronic arthritis in children

Methotrexate in resistant juvenile rheumatoid arthritis: Results of the USA–USSR double-blind, placebo-
controlled trial, 110
Revision of the proposed classification criteria for juvenile idiopathic arthritis: Durban 1997, 112
Iridocyclitis in Still’s disease: its complictions and treatment, 114
Prognosis in juvenile chronic polyarthritis, 116
8 Ankylosing spondylitis
Enthesopathy of rheumatoid and ankylosing spondylitis, 120
Ankylosing spondylitis, 122
Associations between ankylosing spondylitis, psoriatic arthritis, Reiter’s disease, the intestinal
arthropathies, and Behçet’s syndrome, 122
HLA-B27 related arthritis and bowel inflammation. Part 2 Ileocolonoscopy and bowel histology in patients
with HLA-B27 related arthritis, 124
A wider spectrum of spondyloarthropathies, 126
The European Spondyloarthropathy Study Group preliminary criteria for the classification of
spondyloarthropathies, 126
A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath
ankylosing spondylitis functional index, 130
Selection of instruments in the core set for DC-ART, SMARD, physical therapy, and clinical record
keeping in ankylosing spondylitis. Progress report of the ASAS Working Group, 130
9 Psoriatic arthritis
Psoriasis and arthritis, 136
Familial occurrence of psoriatic arthritis, 138
Index 443
Reduced synovial membrane macrophage numbers, ELAM-1 expression, and lining layer hyperplasia in
psoriatic arthritis as compared with rheumatoid arthritis, 140
Fibroblast function in psoriatic arthritis: I. Alternation of cell kinetics and growth factor responses, 142
The role of HLA antigens as indicators of progression in psoriatic arthritis (PsA): multivariate relative risk
model, 144
Mortality studies in psoriatic arthritis. Results from a single centre. II. Prognostic factors for death, 146
10 Reactive arthritis and Reiter’s syndrome

Contribution a l’étude d’une épidemie de dysenterie dans la Somme, 150
Isolation of bedsoniae from the joints of patients with Reiter’s syndrome, 152
Arthritis associated with Yersinia enterocolitica infection, 154
Reiter’s disease and HL-A27, 156
Followup study of patients with Reiter’s disease and reactive arthritis with special reference to HLA-B27,
158
Chlamydia trachomatis and reactive arthritis: the missing link, 160
Chlamydial rRNA in the joints of patients with chlamydia-induced arthritis and undifferentiated arthritis,
162
Molecular evidence for the presence of chlamydia in the synovium of patients with Reiter’s syndrome, 162
11 Lupus
Clinical manifestations of systemic lupus erythematosus. Computer analysis of 520 cases, 166
Therapy of lupus nephritis: controlled trial of prednisone and cytotoxic drugs, 168
Complement fixation with cell nuclei and DNA in lupus erythematosus, 170
Thrombosis, abortion, cerebral disease and the lupus anticoagulant, 172
The bimodal mortality pattern of systemic lupus erythematosus, 174
12 Scleroderma (systemic sclerosis)

Survival with systemic sclerosis (scleroderma). A life-table analysis of clinical and demographic factors in
309 patients, 178
The relationship of hypertension and renal failure in scleroderma (progressive systemic sclerosis) to
structural and functional abnormalities of the renal cortical circulation, 180
Scleroderma (systemic sclerosis): classification, subsets and pathogenesis, 182
scleroderma, 184
Connective tissue synthesis by scleroderma skin fibroblasts in cell culture, 186
Variable response to oral angiotensin-converting enzyme blockade in hypertensive scleroderma patients,
188
Diagnostic potential of in vivo capillary microscopy in scleroderma and related disorders, 190
13 Systemic vasculitis
Periarteritis nodosa: a critical review, 194
Nomenclature of systemic vasculitides: proposal of an international consensus conference, 196
Autoantibodies against neutrophils and monocytes: tool for diagnosis and marker of disease activity and
marker of disease activity in Wegener’s granulomatosis, 198
Cyclophosphamide therapy of severe systemic necrotizing vasculitis, 200
14 Sjögren’s syndrome
Zur kenntnis der keratoconjunctivis sicca: Keratitis filiformis bei hypofunktion der tränendrüsen, 204
Development of malignant lymphoma in the course of Sjögren’s syndrome, 206
Sjögren’s syndrome: a clinical, pathological and serological study of sixty-two cases, 208
Labial salivary gland biopsy in Sjögren’s disease, 210
erythematosus, 212
Primary Sjögren’s syndrome and other autoimmune diseases in families. Prevalence and immunogenetic
studies in six kindreds, 214
444
Gene interaction at HLA-DQ enhances autoantibody production in primary Sjögren’s syndrome, 216
Preliminary criteria for the classification of Sjögren’s syndrome: results of a prospective concerted action
supported by the European Community, 218
15 Myositis
Dermatomyositis with report of a case which presented a rare muscle anomaly but once described in man,
224
Treatment of dermatomyositis with methotrexate, 226
A computer-assisted analysis of 153 patients with polymyositis and dermatomyositis, 228
Mononuclear cells in myopathies, 230
Inclusion body myositis. Observations in 40 patients, 232
A new approach to the classification of idiopathic inflammatory mypathy: myositis-specific autoantibodies
define useful homogeneous patient groups, 234
16 Arthritis and infection

Histological changes in the knee joint in various infections, 238
Gonococcic arthritis: pathogenesis, mechanism of recovery and treatment, 240
Suppurative arthritis complicating rheumatoid arthritis, 242
Post-operative infection after total hip replacement with special reference to air contamination in the
operating room, 244
Allergic complications of meningococcal disease. II. Immunological investigations, 246
Human parvovirus-associated arthritis: a clinical and laboratory description, 248
Lyme arthritis: An epidemic of oligoarticular arthritis in children and adults in three Connecticut
communities, 250
Immunologic relation of streptococcal and tissue antigens. I. Properties of an antigen in certain strains of
Group A streptococci exhibiting an immunologic cross-reaction with human heart tissue, 252
A immunological relationship between the Group A streptococcus and mammalian muscle, 252
17 Osteoarthritis
‘Arthritis deformans’, 258
Heberden’s nodes: heredity in hypertrophic arthritis of the finger joints, 260
Generalized osteoarthritis and Heberden’s nodes, 262
Osteo-arthrosis: prevalence in the population and relationships between symptoms and x-ray changes, 264
Arthroplasty of the hip: a new operation, 266
The geometry of diarthrodial joints, its physiological maintenance and the possible signficicance of age-
related changes in geometry to load distribution and the development of osteoarthritis, 268
An experimental model of osteoarthritis: early morphological and biochemical changes, 270
The prediction of the progression of joint space narrowing in osteoarthritis of the knee by bone
scintigraphy, 272
18 Crystal deposition diseases

Acute arthritis in man and dog after intrasynovial injection of sodium urate crystals, 276
Polymorphonuclear leukocyte motility in vitro. IV. Colchicine inhibition of chemotactic activity formation
after phagocytosis of urate crystals, 278
The clinical and metabolic effects of Benemid in patients with gout, 280
Benemid (p-di-n-propylsulfamyl)-benzoic acid as uricosuric agent in chronic gouty arthritis, 280
Effects of a xanthine oxidase inhibitor on thiopurine metabolism, hyperuricemia and gout, 282
Inorganic pyrophosphate generation from adenosine triphosphate by cell-free synovial fluid, 284
Natural course of articular chondrocalcinosis, 286
Evidence for a post-secretory reabsorptive site for uric acid in man, 288
19 Osteoporosis
Postmenopausal osteoporosis: its clinical features, 292
The loss of bone with age, osteoporosis, and fractures, 294
The diagnosis of osteoporosis, 296
Index 445
Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures, 298
Hormone replacement therapy and risk of hip fracture: population based case-control study, 300
Vitamin D3 and calcium to prevent hip fractures in elderly women, 302
Randomized trial of effect of alendronate on risk of fracture in women with existing vertebral fracture, 304
20 Back pain
Rupture of the intervertebral disc with involvement of the spinal canal, 310
In vivo measurements of intra-discal pressure, 312
Intervertebral foramen venous obstruction. A cause of periradicular fibrosis, 314
Nerve ingrowth into diseased intervertebral disc in chronic back pain, 316
A new clinical model for the treatment of low back pain, 318
A prospective study of work perceptions and psychosocial factors affecting the report of back injury, 320
How many days of bed rest for acute low back pain?, 322
21 Soft tissue disorders

Pathological appearances of seven cases of injury of the shoulder joint; with remarks, 326
The microvascular pattern of the rotator cuff, 328
Anterior acromioplasty for the chronic impingement syndrome in the shoulder. A preliminary report, 330
Biomechanics of the shoulder joint, 332
The shoulder hand syndrome, 334
Spontaneous compression of both median nerves in the carpal tunnel. Six cases treated surgically, 336
The double crush in nerve entrapment syndromes, 338
Observations on referred pain arising from muscle, 340
Clinical effects of the use of pneumatic tools, 342
Industrial Rheumatology. The Australian and New Zealand experience with arm pain and backache in the
workplace, 344
22 Fibromyalgia
Two contributions to understanding of the ‘fibrositis’ syndrome, 348
The American College of Rheumatology 1990 criteria for the classification of fibromyalgia; report of the
multicentre trial committee, 350
Elevated levels of substance P and high incidence of Raynaud phenomenon in patients with fibromyalgia:
new features for the diagnosis, 352
Diagnostic epidural opioid blockade in primary fibromyalgia at rest and during exercise, 354
Increased rates of fibromyalgia following cervical spine injury, 356
Function of the hypothalamic-pituitary-adrenal axis in patients with fibromyalgia and low back pain, 356
The C6-7 syndrome – clinical features and treatment response, 356
23 Geographical disorders
Rheumatic disease in a Philippine village II: WHO–ILAR–APLAR COPCORD Study. Phases II and III, 360
Hyperuricaemia, gout and diabetic abnormality in Polynesian people, 362
Autoimmune disease in parasitic infections in Nigerians, 364
Osteoarthritis of the hip and other joints in Southern Chinese in Hong Kong, 366
The clinical course of Kashin–Beck disease, 368
Unusual hip disease in remote part of Zululand, 370
Ross River virus arthritis in Papua New Guinea, 372
24 Associations of other systemic diseases with arthropathies

Clinical and pathological studies of joint involvement in sarcoidosis, 376
Acromegalic arthropathy, 378
The nature of the defect in tyrosine metabolism in alkaptonuria, 380
Corticosteroid therapy associated with ischemic necrosis of bone in systemic lupus erythematosus, 382
Hemochromatosis and arthritis, 384
Synovial amyloidosis in patients undergoing long-term hemodialysis, 386
446
25 Education
Does patient education for chronic diseases have therapeutic value?, 392
Assessing physician/patient perceptions in rheumatoid arthritis. A vital component in patient education,
394
Outcomes of self-help education for patients with arthritis, 396
Group education for patients with rheumatoid arthritis, 398
A controlled evaluation of arthritis education, 400
Can the provision of information to patients with osteoarthritis improve functional status?, 402
The patient is not a blank sheet: lay beliefs and their relevance to patient education, 404
26 Key drug developments

Clinical measurement of the anti-inflammatory effects of salicylates in rheumatoid arthritis, 408
Variability in risk of gastrointestinal complications with individual non-steroidal anti-inflammatory drugs:
results of a collaborative meta-analysis, 410
N of 1 trials comparing a non-steroidal anti-inflammatory drug with paracetamol in osteoarthritis, 412
The effect of a hormone of the adrenal cortex (17-hydroxy-11-dehydro-corticosterone-compound E) and
of pituitary adreno-corticotropic hormone on rheumatoid arthritis – preliminary report, 414
Gold treatment in rheumatoid arthritis, 416
Synthetic D-penicillamine in rheumatoid arthritis – double-blind controlled study of a high and low
dosage regimen, 418
Sulphasalazine in rheumatoid arthritis – a double-blind comparison of sulphasalazine with placebo and
sodium aurothiomalate, 420
Comparison of low-dose oral pulse methotrexate and placebo in the treatment of rheumatoid arthritis – a
controlled clinical trial, 422
Termination of slow acting antirheumatic therapy in rheumatoid arthritis: a 14 year prospective evaluation
of 1017 consecutive starts, 424
Randomized double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha
(cA2) versus placebo in rheumatoid arthritis, 426
27 Exercise and rehabilitation in arthritis

Articular reflexes at the ankle joint: An electromyographic study of normal and abnormal influences of
ankle-joint mechanoreceptors upon reflex activity in the leg muscles, 430
Quadriceps weakness and osteoarthritis of the knee, 432
Efficacy of physical conditioning exercise in patients with rheumatoid arthritis and osteoarthritis, 434
Physiotherapy in osteoarthritis of the knee. A controlled trial of hospital versus home exercises, 436
Physical training in rheumatoid arthritis: a controlled long-term study, 438
447
Classic Papers by title

1958 revision of diagnostic criteria for rheumatoid arthritis, 82
A
A case of arthropathia psoriatica, 19
A computer-assisted analysis of 153 patients with polymyositis and dermatomyositis, 228
A controlled evaluation of arthritis education, 400
A immunological relationship between the Group A streptococcus and mammalian muscle, 252
A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath
ankylosing spondylitis functional index, 130
A new approach to the classification of idiopathic inflammatory mypathy: myositis-specific autoantibodies
define useful homogeneous patient groups, 234
A new clinical model for the treatment of low back pain, 318
A prospective study of work perceptions and psychosocial factors affecting the report of back injury, 320
A table for the degree of involvement in chronic arthritis, 4
A wider spectrum of spondyloarthropathies, 126
Abnormal calcium deposition inside the knee, a contribution to the question of the primary
‘meniscopathy’, 20
Acromegalic arthropathy, 378
Acute arthritis in man and dog after intrasynovial injection of sodium urate crystals, 276
Agglutination of erythrocytes coated with ‘incomplete’ anti-RH by certain rheumatoid arthritic sera and
some other sera, 96
Allergic complications of meningococcal disease. II. Immunological investigations, 246
An experimental model of osteoarthritis: early morphological and biochemical changes, 270
Ankylosing spondylitis, 122
Ankylosing spondylitis and HL-A27, 68
Anterior acromioplasty for the chronic impingement syndrome in the shoulder. A preliminary report, 330
Appearance of anti-HLA-DR reactive cells in normal and rheumatoid synovium, 90
Arthritis associated with Yersinia enterocolitica infection, 154
‘Arthritis deformans’, 258
Arthroplasty of the hip: a new operation, 266
Arthroscopy of the knee in rheumatic diseases, 58
Articular reflexes at the ankle joint: An electromyographic study of normal and abnormal influences of
ankle-joint mechanoreceptors upon reflex activity in the leg muscles, 430
Assessing physician/patient perceptions in rheumatoid arthritis. A vital component in patient education,
394
Associations between ankylosing spondylitis, psoriatic arthritis, Reiter’s disease, the intestinal
arthropathies, and Behçet’s syndrome, 122
Autoantibodies against neutrophils and monocytes: tool for diagnosis and marker of disease activity and
marker of disease activity in Wegener’s granulomatosis, 198
Autoimmune disease in parasitic infections in Nigerians, 364
B
Benemid (p-di-n-propylsulfamyl)-benzoic acid as uricosuric agent in chronic gouty arthritis, 280
Biomechanics of the shoulder joint, 332
C
Can the provision of information to patients with osteoarthritis improve functional status?, 402
Cartilage oligomeric matrix protein: a novel marker of cartilage turnover detectable in synovial fluid and
blood, 62
448
Characteristics of a soluble nuclear antigen precipitating with sera of patients with systemic lupus
erythematosus, 38
erythematosus, 42
Chlamydia trachomatis and reactive arthritis: the missing link, 160
Chlamydial rRNA in the joints of patients with chlamydia-induced arthritis and undifferentiated arthritis, 162
Clinical and pathological studies of joint involvement in sarcoidosis, 376
Clinical effects of the use of pneumatic tools, 342
Clinical manifestations of systemic lupus erythematosus. Computer analysis of 520 cases, 166
Clinical measurement of the anti-inflammatory effects of salicylates in rheumatoid arthritis, 408
Comparison of low-dose oral pulse methotrexate and placebo in the treatment of rheumatoid arthritis – a
controlled clinical trial, 422
Complement fixation with cell nuclei and DNA in lupus erythematosus, 170
Computerized tomography of sacroiliitis, 22
Connective tissue synthesis by scleroderma skin fibroblasts in cell culture, 186
Contribution a l’étude d’une épidemie de dysenterie dans la Somme, 150
Corticosteroid therapy associated with ischemic necrosis of bone in systemic lupus erythematosus, 382
Cyclophosphamide therapy of severe systemic necrotizing vasculitis, 200
D
Dermatomyositis with report of a case which presented a rare muscle anomaly but once described in man, 224
Development of malignant lymphoma in the course of Sjögren’s syndrome, 206
Diagnostic epidural opioid blockade in primary fibromyalgia at rest and during exercise, 354
Diagnostic potential of in vivo capillary microscopy in scleroderma and related disorders, 190
Does patient education for chronic diseases have therapeutic value?, 392
E
Early detection of carpal erosions in patients with rheumatoid arthritis: A pilot study of magnetic
resonance imaging, 24
Effects of a xanthine oxidase inhibitor on thiopurine metabolism, hyperuricemia and gout, 282
Efficacy of physical conditioning exercise in patients with rheumatoid arthritis and osteoarthritis, 434
Elevated levels of substance P and high incidence of Raynaud phenomenon in patients with fibromyalgia:
new features for the diagnosis, 352
Enthesopathy of rheumatoid and ankylosing spondylitis, 120
Evidence for a post-secretory reabsorptive site for uric acid in man, 288
Examination of synovial fluid as a diagnostic aid in arthritis, 50
F
Familial occurrence of psoriatic arthritis, 138
Fibroblast function in psoriatic arthritis: I. Alternation of cell kinetics and growth factor responses, 142
Followup study of patients with Reiter’s disease and reactive arthritis with special reference to HLA-B27, 158
Function of the hypothalamic-pituitary-adrenal axis in patients with fibromyalgia and low back pain, 356
G
Gene interaction at HLA-DQ enhances autoantibody production in primary Sjögren’s syndrome, 216
Generalized osteoarthritis and Heberden’s nodes, 262
Gold treatment in rheumatoid arthritis, 416
Gonococcic arthritis: pathogenesis, mechanism of recovery and treatment, 240
Group education for patients with rheumatoid arthritis, 398
H
Heberden’s nodes: heredity in hypertrophic arthritis of the finger joints, 260
Hemochromatosis and arthritis, 384
Histological changes in the knee joint in various infections, 238
Index 449
HLA-B27 related arthritis and bowel inflammation. Part 2 Ileocolonoscopy and bowel histology in patients
with HLA-B27 related arthritis, 124
HLA-DR antigens and toxic reactions to sodium aurothiomalate and D-penicillamine in patients with
HLA-Dw4 in adult and juvenile rheumatoid arthritis, 66
HLA heterozygosity contributes to susceptibility to rheumatoid arthritis, 76
Hormone replacement therapy and risk of hip fracture: population based case-control study, 300
How many days of bed rest for acute low back pain?, 322
Human parvovirus-associated arthritis: a clinical and laboratory description, 248
Hyperuricaemia, gout and diabetic abnormality in Polynesian people, 362
I
scleroderma, 44
Identification of immunoglobulins and complement in rheumatoid articular collagenous tissues, 98
Identification of urate crystals in gout synovial fluid, 52
Immunologic relation of streptococcal and tissue antigens. I. Properties of an antigen in certain strains of
Group A streptococci exhibiting an immunologic cross-reaction with human heart tissue, 252
In vivo measurements of intra-discal pressure, 312
Inclusion body myositis. Observations in 40 patients, 232
Increased rates of fibromyalgia following cervical spine injury, 356
Industrial Rheumatology. The Australian and New Zealand experience with arm pain and backache in the
workplace, 344
Inorganic pyrophosphate generation from adenosine triphosphate by cell-free synovial fluid, 284
Intervertebral foramen venous obstruction. A cause of periradicular fibrosis, 314
Iridocyclitis in Still’s disease: its complictions and treatment, 114
Isolation of bedsoniae from the joints of patients with Reiter’s syndrome, 152
L
Labial salivary gland biopsy in Sjögren’s disease, 210
Lyme arthritis: An epidemic of oligoarticular arthritis in children and adults in three Connecticut
communities, 250
M
Measurement of patient outcome in arthritis, 92
Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures, 298
Methotrexate in resistant juvenile rheumatoid arthritis: Results of the USA–USSR double-blind, placebo-
controlled trial, 110
Mixed connective tissue disease – an apparent distinct rheumatic disease syndrome associated with a
specific antibody to an extractable nuclear antigen (ENA), 40
Molecular evidence for the presence of chlamydia in the synovium of patients with Reiter’s syndrome, 162
Mononuclear cells in myopathies, 230
Mortality studies in psoriatic arthritis. Results from a single centre. II. Prognostic factors for death, 146
Myositis autoantibody inhibits histidyl-tRNA synthetase: a model for autoimmunity, 46
N
N of 1 trials comparing a non-steroidal anti-inflammatory drug with paracetamol in osteoarthritis, 412
Natural course of articular chondrocalcinosis, 286
Nerve ingrowth into diseased intervertebral disc in chronic back pain, 316
Nomenclature of systemic vasculitides: proposal of an international consensus conference, 196
O
Observations on referred pain arising from muscle, 340
On the occurrence of a factor in human serum activating the specific agglutination of sheep blood
corpuscles, 34
450
Orthopedic surgery in rheumatoid arthritis, 104

Osteoarthritis of the hip and other joints in Southern Chinese in Hong Kong, 366
Osteo-arthrosis: prevalence in the population and relationships between symptoms and x-ray changes, 264
Outcomes of self-help education for patients with arthritis, 396
P
Participation of monocyte-macrophages and lymphocytes in the production of a factor that stimulates
collagenase and prostaglandin release by rheumatoid synovial cells, 88
Pathological appearances of seven cases of injury of the shoulder joint; with remarks, 326
Periarteritis nodosa: a critical review, 194
Peripheral vascular obstruction in rheumatoid arthritis and its relation to other vascular leisons, 106
Physical training in rheumatoid arthritis: a controlled long-term study, 438
Physiotherapy in osteoarthritis of the knee. A controlled trial of hospital versus home exercises, 436
Polymorphonuclear leukocyte motility in vitro. IV. Colchicine inhibition of chemotactic activity formation
after phagocytosis of urate crystals, 278
Postmenopausal osteoporosis: its clinical features, 292
Post-operative infection after total hip replacement with special reference to air contamination in the
operating room, 244
Preliminary criteria for the classification of Sjögren’s syndrome: results of a prospective concerted action
supported by the European Community, 218
Presentation of two bone marrow elements: The ‘tart cell’ and the ‘LE’ cell, 36
Prevalence of rheumatoid arthritis, 84
Primary Sjögren’s syndrome and other autoimmune diseases in families. Prevalence and immunogenetic
studies in six kindreds, 214
Prognosis in juvenile chronic polyarthritis, 116
Prognosis of rheumatoid arthritis. A prospective survey over 11 years, 100
Psoriasis and arthritis, 136
Q
Quadriceps weakness and osteoarthritis of the knee, 432
Quantitation of the activity (manifestations) of rheumatoid arthritis, 10
R
Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films, 28
Radiological assessment of osteo-arthrosis, 30
Randomized double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha
(cA2) versus placebo in rheumatoid arthritis, 426
Randomized trial of effect of alendronate on risk of fracture in women with existing vertebral fracture, 304
Reduced synovial membrane macrophage numbers, ELAM-1 expression, and lining layer hyperplasia in
psoriatic arthritis as compared with rheumatoid arthritis, 140
Reiter’s disease and HL-A27, 156
Revision of the proposed classification criteria for juvenile idiopathic arthritis: Durban 1997, 112
Rheumatic disease in a Philippine village II: WHO-ILAR-APLAR COPCORD Study. Phases II and III, 360
Rheumatoid arthritis with death from medullary compression, 26
Ross River virus arthritis in Papua New Guinea, 372
Rupture of the intervertebral disc with involvement of the spinal canal, 310
S
Scleroderma (systemic sclerosis): classification, subsets and pathogenesis, 182
Selection of instruments in the core set for DC-ART, SMARD, physical therapy, and clinical record
keeping in ankylosing spondylitis. Progress report of the ASAS Working Group, 130
Sjögren’s syndrome: a clinical, pathological and serological study of sixty-two cases, 208
Spontaneous compression of both median nerves in the carpal tunnel. Six cases treated surgically, 336
Studies on the depressed hemolytic complement activity of synovial fluid in adult rheumatoid arthritis, 102
Index 451
Sulphasalazine in rheumatoid arthritis – a double-blind comparison of sulphasalazine with placebo and

sodium aurothiomalate, 420
Suppurative arthritis complicating rheumatoid arthritis, 242
Survival with systemic sclerosis (scleroderma). A life-table analysis of clinical and demographic factors in
309 patients, 178
Synovial amyloidosis in patients undergoing long-term hemodialysis, 386
Synthetic D-penicillamine in rheumatoid arthritis – double-blind controlled study of a high and low
dosage regimen, 418
T
Termination of slow acting antirheumatic therapy in rheumatoid arthritis: a 14 year prospective evaluation
of 1017 consecutive starts, 424
The ameliorating effect of pregnancy on chronic atrophic (infectious rheumatoid) arthritis, fibrositis, and
intermittent hydrarthrosis, 86
The American College of Rheumatology 1990 criteria for the classification of fibromyalgia; report of the
multicentre trial committee, 350
The assessment of rheumatoid arthritis: a study based on measurements of the serum acute-phase
reactants, 60
The bimodal mortality pattern of systemic lupus erythematosus, 174
The C6-7 syndrome – clinical features and treatment response, 356
The clinical and metabolic effects of Benemid in patients with gout, 280
The clinical course of Kashin–Beck disease, 368
The diagnosis of osteoporosis, 296
The double crush in nerve entrapment syndromes, 338
The effect of a hormone of the adrenal cortex (17-hydroxy-11-dehydro-corticosterone-compound E) and
of pituitary adreno-corticotropic hormone on rheumatoid arthritis – preliminary report, 414
The European Spondyloarthropathy Study Group preliminary criteria for the classification of
spondyloarthropathies, 126
The geometry of diarthrodial joints, its physiological maintenance and the possible signficicance of age-
related changes in geometry to load distribution and the development of osteoarthritis, 268
The loss of bone with age, osteoporosis, and fractures, 294
The microvascular pattern of the rotator cuff, 328
The nature of the defect in tyrosine metabolism in alkaptonuria, 380
The pain chart, 6
The patient is not a blank sheet: lay beliefs and their relevance to patient education, 404
The prediction of the progression of joint space narrowing in osteoarthritis of the knee by bone
scintigraphy, 272
The relationship of hypertension and renal failure in scleroderma (progressive systemic sclerosis) to
structural and functional abnormalities of the renal cortical circulation, 180
The role of HLA antigens as indicators of progression in psoriatic arthritis (PsA): multivariate relative risk
model, 144
The shared epitope hypothesis. An approach to understanding the molecular genetics of susceptibility to
The shoulder hand syndrome, 334
The significance of calcium phosphate crystals in the synovial fluid of arthritic patients: the ‘pseudogout
syndrome’. II. Identification of crystals, 54
The skiagraphy of rheumatoid arthritis, 16
Therapeutic criteria in rheumatoid arthritis, 8
Therapy of lupus nephritis: controlled trial of prednisone and cytotoxic drugs, 168
Thrombosis, abortion, cerebral disease and the lupus anticoagulant, 172
Transmission disequilibrium as a test of linkage and association between HLA alleles and pauciarticular-
onset juvenile rheumatoid arthritis, 72
Treatment of dermatomyositis with methotrexate, 226
Two contributions to understanding of the ‘fibrositis’ syndrome, 348
Typical alterations of the bones in gout, 18
452
U
Unusual hip disease in remote part of Zululand, 370
Usefulness of synovial fluid analysis in the evaluation of joint effusions. Use of threshold analysis and
likelihood ratios to assess a diagnostic test, 56
V
Variable response to oral angiotensin-converting enzyme blockade in hypertensive scleroderma patients,
188
Variability in risk of gastrointestinal complications with individual non-steroidal anti-inflammatory drugs:
results of a collaborative meta-analysis, 410
Vitamin D3 and calcium to prevent hip fractures in elderly women, 302
W
When does rheumatoid arthritis start?, 94
Z
Zur kenntnis der keratoconjunctivis sicca: Keratitis filiformis bei hypofunktion der tränendrüsen, 204

Classic Papers in Rheumatology (PDFDrive)

Uploaded by

Copyright:

Available Formats

Classic Papers in Rheumatology (PDFDrive)

Uploaded by

Document Information

Original Title

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

Classic Papers in Rheumatology (PDFDrive)

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What information can be found in the references section?

What information can be found in the references section?

Classic Papers in Rheumatology

Classic Papers in Rheumatology

First published in the United Kingdom in 2002

Tel: +44 (0) 20 74822202

ISBN 0-203-21423-4 Master e-book ISBN

ISBN 0-203-27085-1 (Adobe eReader Format)

Distributed in the rest of the world by

Project management, design and layout: Top Draw Design

Section 3: Regional and Miscellaneous Disorders

Muhammad Asim Khan

A table for the degree of involvement in

Stage I: Soft tissue changes.

Stage I: Slight symptoms.

Why it’s important

The pain chart

Why it’s important

Figure 1.1 Keele described the

Therapeutic criteria in rheumatoid arthritis

Why it’s important

Functional Classification. The Steinbrocker Functional Classification allowed patients to be

Class I: Complete. Ability to carry on all usual duties without handicaps.

Quantitation of the activity (manifestations) of

Why it’s important

The skiagraphy of rheumatoid arthritis

Why it’s important

Figure 2.1 A radiograph of the hand in RA

Three early descriptions of the radiology of arthritis

Typical alterations of the bones in gout

A case of arthropathia psoriatica

Related reference (1) Belot X, Chaperon MMJ. Radiographies de lesions et deformations

Abnormal calcium deposition inside the knee,

Key message of the three papers

Why they are important

Figure 2.2 Radiographs showing the typical

Computerized tomography of sacroiliitis

Why it’s important

Early detection of carpal erosions in patients with

Related reference (1) Ostergaard M, Hansen M, Stoltenberg M, Lorenzen I. Magnetic

Why it’s important

Rheumatoid arthritis with death from

Why it’s important

Radiographic evaluation of rheumatoid arthritis

Related references (1) Larson A. Radiological grading of rheumatoid arthritis, an interob-

Why it’s important

Figure 2.3 Representation of the

Grade 0 Grade 1 Grade 2

Grade 3 Grade 4 Grade 5

Radiological assessment of osteo-arthrosis

Why it’s important

On the occurrence of a factor in human serum

Why it’s important

Presentation of two bone marrow elements:

Why it’s important

Characteristics of a soluble nuclear antigen

Why it’s important