Laboratory Procedure Manual

Analyte: Glycohemoglobin

Matrix: Whole Blood

Method Tosoh G 7 HPLC
Glycohemoglobin Analyzer

as performed by: Fairview-University Medical Center

University Campus
Collaborative Studies Clinical Laboratory
Minneapolis, Minnesota

Contact: Dr. Michael Steffes

Important Information for Users

The Fairview -University Medical Center periodically refines these laboratory methods.
It is the responsibility of the user to contact the person listed on the title page of each
write-up before using the analytical method to find out whether any changes have been
made and what revisions, if any, have been incorporated.
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

Public Release Data Set Information

This document details the Lab Protocol for testing the items listed in the following table:

File Variable
Name SAS Label
Name

GHB_E LBXGH Glycohemoglobin (%)

There was an upgrade in instruments in June 2007. During the first six months of 2007
the laboratory used Tosoh 2.2 Plus and from the second six months forward the Tosoh G7
Automated HPLC System Glycohemoglobin Analyzer was used. The method used in the first half
of 2007 is described in a separate document

2
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

1. SUMMARY OF TEST PRINCIPLE AND CLINICAL RELEVANCE

A. Principle
The G7 Automated HPLC Analyzer – HbA1c Variant Analysis Mode uses non-porous
ion-exchange high performance liquid chromatography (HPLC) for rapid, accurate,
and precise separation of the stable form of HbA1c from other hemoglobin fractions.
Analysis is carried out without off-line specimen pretreatment or interference from
Schiff base.

The analyzer dilutes the whole blood specimen with Hemolysis & Wash Solution, and
then injects a small volume of this specimen onto the TSKgel G7 HSi Variant
Column. Separation is achieved by utilizing differences in ionic interactions between
the cation exchange group on the column resin surface and the hemoglobin
components. The hemoglobin fractions (designated as A1a, A1b, F, LA1c+, SA1c,
A0, and H-V0, H-V1, H-V2) are subsequently removed from the column by
performing a step-wise elution using the varied salt concentrations in the Elution
Buffers HSi Variant 1, 2, and 3.

The time from injection of the sample to the time the specific peak elutes off the
column is called Retention Time. The G7 Automated HPLC software has been written
so that each of the expected fractions has a window of acceptable retention times. If
the designated peak falls within the expected window, the chromatogram peaks will
be properly identified. When a peak elutes at a retention time not within a specified
window, an unknown peak (P00) results. If more than one peak elutes at times not
specified by the software windows, each is given a sequential P0x title. In order to
keep the peaks within their appropriate windows, it may be necessary to change how
fast the buffers are moving through the system by changing the pump flow rate.

The separated hemoglobin components pass through the LED photometer flow cell
where the analyzer measures changes in absorbance at 415 nm. The analyzer
integrates and reduces the raw data, and then calculates the relative percentages of
each hemoglobin fraction. The Total Area of the SA1c is divided by the sum of the
total areas of all peaks up to and including the A0 to obtain a raw SA1c percentage.
This uncorrected result is substituted as the “x” value in the linear regression formula
determined during calibration. The analyzer prints the final numerical results and
plots a chromatogram showing changes in absorbance versus retention time for each
peak fraction.

The Tosoh G7 Automated HPLC Analyzer – HbA1c Variant Analysis Mode is certified
by the National Glycohemoglobin Standardization Program (NGSP). The final
reportable result is traceable to the Diabetes Control and Complications Trial
(DCCT).

B. Clinical Relevence

Diabetes causes elevated levels of glucose to circulate in the blood. Maintaining

normal levels of blood glucose is part of the routine clinical management of diabetes.
Continuous and careful management of blood glucose levels prevents development
of serious long term complications resulting from vascular impairment such as
retinopathy, nephropathy, and neuropathy.

Although a fasting blood glucose measurement gives the clinician information about
the patient’s status over the last twelve hours, the stable HbA1c offers a more

3
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

accurate indication of the patient’s long-term diabetic control over the last two to
three months.

Glycohemoglobin is a general term for hemoglobin-glucose complexes in which

glucose is bound to the alpha and beta chains of hemoglobin. The most quantitatively
prevalent complex is called HbA1c, in which glucose binds to the N-terminus of the
beta chain of HbA.

HbA1c is nonenzymatically synthesized in two steps:

1. The glucose aldehyde group and the free amino group on the valine in the N-
terminus of the hemoglobin beta chain react to form the Schiff base, aldimine
(also known as labile HbA1c or LA1c).

2. A stable ketoamine form of the hemoglobin complex (SA1c) is then produced by a

reaction known as Amadori rearrangement.

The level of LA1c changes rapidly in response to changes in blood glucose

concentration. However, the level of the SA1c does not fluctuate significantly in
response to physiological factors. Consequently, the SA1c measurement provides a
better indication of the average glucose level over the previous two to three months
(the average red blood cell life span).

Hemoglobin → Hemoglobin → Hemoglobin

+ fast + slow +
Glucose ← Glucose Glucose
Aldimine (LA1c) Ketoamine (SA1c)

Formation of Labile and Stable Forms of A1c (LA1c and SA1c)

In the past, accurate measurement of SA1c was possible only after removing LA1c by
pretreatment. As with the Tosoh A1c 2.2 Plus Glycohemoglobin Analyzer, the Tosoh G7
Automated HPLC Analyzer – HbA1c Variant Analysis Mode can individually resolve SA1c
and LA1c on the chromatogram without manual pretreatment, allowing accurate
measurement of SA1c directly

2. SAFETY PRECAUTIONS

Follow all procedures and policies in the Fairview-University Medical Center Laboratory
Safety Manual. Consider all specimens as potentially infectious.

Sodium azide can react with copper and lead plumbing to form explosive metal azides.
On disposal, flush reagents with a large volume of water to prevent the buildup of azides.

3. COMPUTERIZATION; DATA SYSTEM MANAGEMENT

NHANES SA1c% results are entered unto a spreadsheet provided electronically by
WESTAT, Inc for NHANES.

To access the spreadsheet click on My Computer → Z drive → User → Dep Labs →

Collab Studies → NHANES → Glyhb 004.

Choose the file named with the corresponding box number.

4
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

Enter the analysis date, run number, technologist’s initials, SA1c%, and result comment
code.

The spreadsheet will be sent electronically by the contact person.

4. SPECIMEN COLLECTION, STORAGE, AND HANDLING PROCEDURES; CRITERIA

FOR SPECIMEN REJECTION

Samples are collected and processed in mobile examination centers according to

NHANES protocols.

Specimens are packaged and shipped on cold packs or dry ice according to the
established schedule.

Specimens are shipped via Federal Express for delivery directly to Collaborative Studies
Clinical Laboratory.

Shipments for NHANES will arrive on Tuesdays and/or Wednesdays. The shipments will
consist of two boxes, one with frozen gel packs containing HbA1c specimens and one
with dry ice containing frozen glucose and insulin specimens. These shipments will be
recorded on the shipping log located in a blue 3 ring binder labeled NHANES Shipping
Log in the receiving area.

Included in the shipping box for HbA1c (glycohem) specimens are a shipping manifest, a
Federal Express airbill for return shipment, frozen gel packs, and a box or boxes of
HbA1c(glycohem) specimens (vessel/vial number 004). Record the appropriate
information on the shipping log. Check the specimen numbers in the box against the
manifest. Write the received date on top of the box. Bring the specimens to the HbA1c
desk. File the manifest in the blue 3 ring binder labeled NHANES Shipping Manifests
located in the receiving area. Remove all labels from the shipping box and attach the
provided airbill for return shipment. Weigh the boxes on the scale in L237 to complete
the information on the airbill. Bring the boxes to the Fairview dock.

A venous whole blood specimen collected in EDTA is required. Tubes containing

heparin, potassium oxalate or sodium flouride are acceptable. Whole blood specimens
o
are stable up to fourteen days stored at 2-8 C or up to eight hours at room temperature
before analysis. Prior to analysis, mix each patient specimen by gentle inversion to
ensure homogeneity.

Fingerstick capillary specimens collected using the Bio-Rad Sample Preparation Kit are
an acceptable alternative to venous whole blood collection and provide enhanced stability
during sample storage and transportation. Samples prepared as directed are stable for 2
weeks stored at room temperature or four weeks stored at 2-8oC.
Optimum sample volume: 1 mL whole blood
Minimum sample volume: 50 uL whole blood (for specimens of volume less than 1
mL whole blood, a manual pre-dilution (1:250) must be
prepared)

5. Procedures for Microscopic Examinations

Not applicable for this procedure.

6. EQUIPMENT AND INSTRUMENTATION, MATERIALS, REAGENT PREPARATION,

CALIBRATORS (STANDARDS), AND CONTROLS

5
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

A. Instrumentation
(1) A1c G7 HPLC Glycohemoglobin Analyzer. Part # 019327, with 90
sample loader, Par # 019475. Tosoh Medics, Inc., 347 Oyster Pt. Blvd.,
Suite 201, So. San Francisco, Ca 94080.

(2) Labquake Rotator. Catalog no. 415-110, Labindustries, Inc., 620 Hearst
Avenue, Berkeley, CA 94710-1992.

(3) Auto Dilutor, model AD-7, Catalog No. 196-7393, Bio-Rad clinical
Division, 4000 Alfred Noble Drive, Hercules, CA 94547.

B. Materials

(1) TSKgel Glyco HSi Variant Column. Part # 019680, Tosoh Medics, Inc.
Guaranteed for 2500 counts; replace as necessary (as indicated by
appearance of chromatograms). Stable indefinitely when stored at 4-15oC
away from direct sunlight. Use only with column-matched buffers (first letter
of buffer lot must match last letter of column lot). When a new column is
installed, analyze 5 duplicates after calibrating and analyzing controls. Also
record the previous results on the protocol page. The results must agree
within established duplicate range.

(2) Filter element, 5/pkg. Part # 019506, Tosoh Medics, Inc. Replace at or
before 400 injections (do not exceed 400 injections) or when pressure rises
above 150 kg/cm2 (15 Mpa).

(3) Thermal paper for A1c G 7, 10 roll/box. Part # 019563. Tosoh Medics, Inc.

(4) Sustaining tube, Dilution Sample, 10 mL, 50/pkg, Tosoh Medics, Inc.

(5) Adapter Ring, Dilution Sample, 5/pkg, Tosoh Medics, Inc.

(6) Adapter Ring, Sample Rack: 12 mm, 13 mm, and 14 mm .

(7) Sample Vials, 100/pkg. Tosoh Medics, Inc. (For preparing dilutions of whole
blood samples.)

(8) Microcentrifuge tubes, 0.6 mL polypropylene (with caps), 500/bag, Stock #

CX15559. University Stores. (For preparing aliquots of controls for frozen
storage.)

(9) DIAMAT HbA1c Sample Preparation Kit, Cat. No. 196-1026, Bio-Rad
Laboratories, Clinical Division, 4000 Alfred Nobel Drive, Hercules, CA
94547. Samples prepared as directed in the Instruction Manual are stable
o
for 2 weeks at room temperature or 4 weeks at 2-8 C. Includes supplies
sufficient for 100 test samples:

Sample Preparation Vials, 100/kit, each contain 1 mL of an aqueous

solution of EDTA and potassium cyanide (0.25 mmol/L). Store at 5-
30oC.

6
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

Capillaries, one glass dispenser vial containing 100 sodium-heparinized

capillary tubes (5 uL). Reorder box of 20 vials (50 capillary tubes/vial),
Cat. No. 195-1053, Bio-Rad Laboratories, Clinical Division.

Capillary tube holder, one holder for manipulating 5 uL capillary tubes.

Reorder box of 20 holders, Cat. No. 196-1054. Bio-Rad Laboratories,
Clinical Division.

Labels, 4 sheets of 25 blank labels each.

Instruction Manual.

(10) Supply Line Filters for Buffer Lines. Part#018723(1/pkg).

(11) Sampling Needle – Std. Part#019500(1 each).

(12) Sampling Needle – Side hole. Part#019791(1 each)

(13) HLC-723RP- G7 Reporting Software. Part#021209(1 each)

C. Reagent Preparation

(1) Elution Buffer HSi Variant No. 1, (S) Part # 019552 (1 x 800 mL). Tosoh
Medics, Inc. Succinic acid buffer, contains less than 0.06% sodium azide
as a preservative. Unopened buffer is stable until expiration date printed
o
on label. Once open, (S) buffer is stable for three months. Store at 4-25 C.
Use only with other column-matched buffers (first letter of buffer lot
matches last letter of column lot). When a new lot number of buffer is
installed, analyze 5 duplicate samples at the beginning of the run. Record
the previous results on the protocol page also. The results must agree
within the established duplicate range.

(2) Elution Buffer HSi Variant No. 2, (S) Part # 019553 (1 x 800 mL). Tosoh
Medics, Inc. Succinic acid buffer, contains less than 0.06% sodium azide
as a preservative. Unopened buffer is stable until expiration date printed on
label. Once open, (S) buffer is stable for three months. Store at 4-25oC.
Use only with other column-matched buffers (first letter of buffer lot
matches last letter of column lot). When a new lot number of buffer is
installed, analyze 5 duplicate samples at the beginning of the run. Record
the previous results on the protocol page also. The results must agree
within the established duplicate range.

(3) Elution Buffer Hsi Variant No. 3, (S) Part # 019554 (1 x 800 mL). Tosoh
Medics, Inc. Succinic acid buffer, contains less than 0.06% sodium azide
as a preservative. Unopened buffer is stable until expiration date printed on
label. Once open, (S) buffer is stable for three months. Store at 4-25oC.
Use only with other column-matched buffers (first letter of buffer lot
matches last letter of column lot). When a new lot number of buffer is
installed, analyze 5 duplicate samples at the beginning of the run. Record
the previous results on the protocol page also. The results must agree
within the established duplicate range.

(4) Elution Buffer His Variant No. 1, Part # 021446 (1 x 1500 mL). Tosoh
Medics, Inc.

7
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

(4) Hemoolysis & Wash Solution, (S) Part # 018431 (1 x 800 mL. Tosoh
Medics, Inc. Contains deionized water, EDTA and Triton X and contains
less than 0.12% sodium azide as a preservative. Once open, (S) buffer is
stable for three months. Store at 4-25oC.

D. Standards Preparation

HbA1c Calibrator Set: Calibrator 1 (5 x 4 mL) and Calibrator 2 (5 x 4 mL). Part #

018767, Tosoh Medics, Inc. Buffered human red blood cells, 2 mg/mL human
hemoglobin, and 0.5 mM EDTA as preservative. Un-reconstituted calibrator set
is stable stored at 4-8oC until expiration date printed on label.

(1) Remove caps from the calibrator vials.

(2) Reconstitute Calibrators 1 and 2 using a volumetric pipette to add 4.0 mL

Type I Reagent Grade water to each vial.

(3) Replace respective caps and mix thoroughly by inversion. Record dates
of reconstitution and expiration on vial labels, then promptly store upright
at 4-8oC. Always return calibrators promptly to refrigerator--do not leave
vials at room temperature for an extended period. Store reconstituted
calibrators upright at 2-8°C for up to seven days.

Calibrator Lot Validation: Each new lot of calibrators must be evaluated against
the current lot prior to putting into use. (Evaluation against whole blood
calibrators may be performed as needed – see Note 9). Analyze each level in
duplicate within the same run over a period of two to three days (include both
instruments) to verify that manufacturer-assigned values are valid. First,
calibrate the run using the current lot of calibrators and analyze the controls.
Analyze the new lot calibrators as unknowns immediately after the controls,
running each level in duplicate. Record the values obtained including analyses
from both instruments. When the tally is complete, calculate a mean to confirm
the assigned bottle value or to determine new assigned values. Prior to
analyzing patient specimens, verify that analysis of current lot of controls against
the new lot calibrators produces results that fall within established control ranges.

New lots of calibrator may be evaluated as necessary against whole blood

calibrators obtained from the NGSP CPRL at the University of Missouri (UMO
calibrators). Perform this procedure when validation of new lot calibrators
against current lot calibration does not confirm manufacturer-assigned values.

(1) Obtain aliquots of UMO whole blood calibrators and UMO controls. First
calibrate both instruments with UMO calibrators using their assigned values.
Analyze the UMO controls, the current lot of in-house controls and both
levels of new lot calibrator in duplicate as unknowns. Verify that the controls
fall within their respective QC limits (preferably within 1 SD). Evaluate the
results of the new lot calibrators against their manufacturer-assigned values.

(2) Next calibrate both instruments with the new lot of Tosoh calibrators using
the manufacturer-assigned values. Analyze the UMO controls and in-house
controls and verify that they fall within their respective QC limits (preferably
within 1 SD). If control results are acceptable, the assigned values may be
used. If controls do not fall within established ranges, repeat analysis of new
lot calibrators against UMO calibrators to establish new assigned values.

8
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

(3) Additionally, the most recent set of NGSP Monthly Monitoring samples may
be thawed and analyzed on both instruments and the results compared with
results obtained using the current lot calibrators.

E. Preparation of Quality Control Material

Two levels of glycated hemoglobin control (Normal and Elevated) are analyzed in
duplicate (or more) with each batch. See QC charts for controls currently in use
and established ranges. Controls are prepared from whole blood drawn from a
normal (Normal) and a diabetic (Elevated) individual. Stable indefinitely stored at
–70oC.

Collect six 10-mL potassium-EDTA tubes from one normal or one diabetic
volunteer depending on the control level to be prepared. Mix well by gentle
inversion then pour blood into a 100-mL beaker containing a small magnet and
place the beaker into a bucket containing wet ice. Place bucket on a magnetic
stirrer set on low speed. Aliquot ~ 100 uL into 0.6 mL polypropylene
microcentrifuge tubes with caps. Continue to add ice to the bucket as needed to
keep beaker chilled. During preparation, aliquots may be held in an insulated
bucket filled with ice until placed into boxes to be stored at –70oC (chest freezer).

At the start of each week, take one week’s supply of controls from the stock
supply and place in the working -70oC freezer.

Evaluate the new lot of controls according to QC/QA guidelines to establish

temporary ranges prior to placing into clinical use.

College of American Pathologists (CAP) specimens and Monthly Monitoring

specimens from the University of Missouri are used for proficiency testing. The
TOSOH 2.2 Plus instruments are NGSP certified.

7. CALIBRATION AND CALIBRATION VERIFICATION PROCEDURES

The analyzer has a two-point automatic calibration function. Studies have shown the
calibration to be stable for at least seven days if the system is calibrated and maintained
according to the procedures provided in this guide and the G7 Automated HPLC Analyzer
Operator’s Manual.

Perform weekly calibration of the instrument on Mondays prior to analysis of controls and
patient samples. Calibration must also be performed after repeated control failure, major
maintenance or service has been performed or whenever a new column is installed.

A. Verify that there is sufficient volume of Elution Buffers and Hemolysis & Wash
Solution.
Replace if necessary.

B. Check analyzer status.

-If analyzer is off, press the POWER key. The analyzer begins its 10.8 minute
warm-up.
-If analyzer is in STAND-BY mode, proceed to step 3.

C. Select CALIB from the MAIN screen. Once selected, it will be reverse highlighted. If
it is necessary to change the current calibration values, select MENU, then
PARAMETER. Press CALIB-1 and enter the assigned value for Calibrator 1. Press
CALIB-2 and enter the assigned value for Calibrator 2.

9
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

D. Pipette at least 800 µL of each calibrator into sample cups. Place the sample cups in
the rack with Calibrator 1 in position 1 (on the left) and Calibrator 2 next to it in
position 2. Place dilutions made from the current lot of controls in positions 3 and 4.
Place an empty rack on the loader to signal the end of the run.

E. Press the START key to begin the calibration. The analyzer samples Calibrator 1
three times and Calibrator 2 two times. The analyzer discards the first measurement
of Calibrator 1, and uses the remaining four measurements to calculate factors A and
B. Upon completion of the calibration, the CALIB button will automatically go off.
Following a successful calibration, patient and control samples will be calculated
using the new factors. Record the new calibration parameters on the protocol page.
Allow the instrument to analyze the controls and evaluate them before placing more
specimen racks in the sample loader. If controls exceed acceptable limits, re-
calibrate.

8. PROCEDURE OPERATING INSTRUCTIONS; CALCULATIONS; INTERPRETATION

OF RESULTS

A. Procedure

(1) Turn on the autodilutor. On Mondays, prime the system with Hemolysis &
Wash Solution by placing the inlet tubing into the reagent reservoir and
secure the cap. (The autodilutor is stored in deionized water over the
weekend.) Remove the spacers under each syringe. Set the mode switch
to CONT and tap the remote switch on the probe to initiate continuous
reagent dispensation. Let syringes cycle 4-5 times to purge tubing and
syringes of air and to prime the system with reagent. When the syringes are
on the upstroke, set mode switch back to MAN mode (manual dispense).
Install the 5% spacer for the 100 µL sample syringe and the 50% spacer for
the 2.5 mL reagent syringe. For specimens with a low hematocrit, dilute the
specimen using the 10% sample syringe spacer.

(2) Daily Setup procedure:

Consult both the Daily Maintenance and the As Needed Maintenance Logs
prior to starting analysis each day (see Maintenance instructions at end of
procedure). Check off each required task as it is performed and initial the
log.

Both analyzers are currently programmed to warm-up automatically at 7:00

AM Monday through Friday. However, if necessary to manually initiate
warm-up, press the POWER key located on the operation panel to switch on
the analyzer. This will initiate a 10.8-minute warm-up sequence. The
analyzer Status screen displays the analyzer’s current operating mode
(WARMING-UP). Check for leaks and record pump pressure during the
10.8-minute WARMING-UP sequence.

To check flow rate at a time other that during the WARMING-UP sequence,
first verify that the analyzer is in STAND-BY mode. Press the ‘down arrow’ on
the MAIN screen. The valves operate as toggle switches. . The valves
operate as toggle switches. Press the SV-1, Sv-2 and or SV-3 valve keys
appropriately until valve status line appears as follows: O X X. Press PUMP
to start the pump. Wait a few seconds for flow rate to stabilize, then record.
Press PUMP again to stop the pump. Press the ‘down arrow’ to return to the
MAIN screen.

10
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

Note: The main power switch is located on the lower left side of the analyzer
and must remain ON at all times. If the main power has been interrupted or
switched off, the application software and default parameters must be re-
loaded by inserting the System Disk in the floppy drive with the main power
switch on and pressing the analyzer’s POWER. Remove the current data
disk from the analyzer and insert the analyzer’s System disk. The analyzer
automatically loads the program and begins the WARMING-UP sequence.
After the LED on the drive goes out, remove the System Disk and return to a
safe place. Replace the current data disk in the drive. Be sure to re-enter
the current calibration parameters (and re-calibrate if necessary).

(3) Generate a parameter printout to bracket the run by pressing MENU on the
MAIN screen, then press UTILITY, then PARAM PRINT. Verify that
parameters are set correctly by comparing them with the example posted
inside the right door of the analyzer. Pay close attention to the current
values posted for CALIB-1 and CALIB-2. Press ‘EXIT’ to return to the MAIN
screen.

(4) If instrument calibration is not required, run both calibrators as

unknowns in positions 1 and 2, followed by dilutions made from the current
lot of controls in positions 3 and 4. Place the rack in the left compartment of
the sample loader. Place a second (empty) rack behind this rack. Press the
START key. Allow the instrument to analyze the calibrators and controls
and evaluate them before more specimen racks in the sample loader. If
controls exceed acceptable limits, re-calibrate.

(6) Protocol pages:

A day’s analysis load may consist of one or more runs. It acceptable to
analyze samples in one continuous run or in several shorter runs.

(a) Controls: Begin with analysis of both controls in the first run. Alternate
analysis of control levels in each subsequent run and at least once on
each protocol page. Controls must be analyzed in duplicate (at least)
each day. Evaluate against established limits.

(b) Patient samples: Record accession numbers and/or CIDs on the protocol
page in the order in which the samples are to be analyzed.

(c) Calibrators (as unknowns): Analyze Calibrator 1 and Calibrator 2 as

unknowns at least once per day within the batch and again to bracket all
samples at the end of the day’s batch. Acceptable range for calibrators
analyzed as unknowns are + 0.2 of assigned values.

(d) Between batch duplicate: Analyze a specimen from the previous day as
a duplicate at some point in the current day’s run. Result must agree
within + 0.2 of the previous value.

(e) Within batch duplicate: Analyze a specimen from the beginning of the
run again at the end of the run. Result must agree within established
duplicate range.

NOTE: The analyzer is programmed to begin a 10 minute wash mode

immediately following the completion of the last sample. Once the WASH

11
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

cycle begins, you must allow it to proceed to completion (approximately 10

minutes). If washing is insufficient, column lifespan will be reduced and the
result for the next sample could be affected. If there is no further input from
the operation panel while the analyzer is in the STAND-BY mode, after 3
hours the analyzer will shut down automatically.

(7) Prepare samples for analysis:

Controls. Thaw aliquots and vortex briefly. Using the autodilutor, prepare
hemolysates by diluting 5 uL (5% sample spacer) of each control with 1.25 mL
(50% reagent spacer) Hemolysis & Wash Solution. Wipe the probe tip after
drawing up sample and again after dispensing into a labeled vial. Place the
sample vials in the rack.

Whole Blood. Ensure that the stopper on the tube is properly seated and that
the barcode label is vertically aligned; re-affix barcode label vertically on tube
if necessary. Mix each patient specimen several times by gentle inversion (or
place briefly on a rotator). Then place the specimen tube in the rack in order
from left to right according to its rack and position number as recorded on the
protocol page. Align its barcoded label so that it faces the barcode reader
(i.e., facing away from you in the rack as it’s loaded on the instrument).

Note: Blood cells will begin to settle out as the tubes sit on the instrument
waiting to be measured. This cell sedimentation over a period of
approximately 5 hours does not affect the HbA1c result.

Low-volume samples (< 1.0 mL whole blood in tube). Using the autodilutor,
prepare hemolysates by diluting 5 uL (5% sample spacer) of each low-volume
patient sample with 1.25 mL Hemolysis & Wash Solution (50% reagent
spacer). Wipe the probe tip after drawing up sample and again after
dispensing into labeled sample vial. Place the sample vial in the rack.
(Minimum dilution volume dispensed or pipetted into a sample vial is 300 uL.)

HbA1c Sample Preparation Vials. Remove caps prior to sampling! Place the
Prep vial in the rack using an adapter tube.

B. Operation

(1) Place racks in ascending order into the sample loader. Place an empty
rack after the last rack to be processed. Presses START key to begin
analysis. The racks will be moved automatically along the sample loader.
The analyzer will prime the fluid lines with buffer for 4.4 minutes, then
analyze samples at 2.2-minute intervals.

IMPORTANT: Keep tubes in the sample rack until the whole rack is
processed and printed reports are available and have been reviewed

(2) Processing automatically stops when the analyzer detects an empty rack
(or 10 sequential empty spaces). When measurement ends, the analyzer
will enter the WASH mode in which it washes the column by pumping
buffer for 10 minutes, then enters STAND-BY mode again. Once it has
started, always allow the WASH sequence to go to completion!

12
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

(3) If there is no further input from the operation panel while the analyzer is in
STAND-BY mode, after 3 hours (Off Time setting), the analyzer shuts itself
off automatically.

C. Special Method Notes

(1) To avoid an error condition during calibration, place Calibrators 1 and 2 in

the first sample rack in positions 1 and 2 respectively.

(2) Each master reagent lot number of G7 HSi Elution Buffers supplied by
Tosoh is performance matched to the corresponding TSKgel G7 HSi
Variant Columns. Following any announced change in supplied Tosoh
TSKgel G7 HSi Variant Columns, contact Tosoh to determine suitability of
existing elution buffers. Note: Hemolysis & Wash Solution may be used
with any master lot of column/buffers since it is not necessary that it be
matched to the other components.

(3) The reagents must be at room temperature (15-25°C) prior to use.

(4) Never pour or transfer reagent from one bag or container to another;
results may be compromised.

(5) When replacing the column, run at least three whole blood samples to
prime the column then recalibrate the system before running quality control
materials. Refer to the G7 Automated HPLC Operator’s Manual for specific
instructions.

(6) Replace the filter element if pressure is greater than 9 MPa or after 400
injections.

(7) If the column is not to be used for more than one week, remove it from the
analyzer and seal the ends with the protective plugs and store in cool place
at 4-15°C. Avoid direct sunlight.

(8) At least once a day check the waste container to ensure that there is
enough space remaining to accommodate a run. If necessary, empty
container and add about one liter of fresh 5% sodium hypochlorite solution
(household bleach).

(9) New lots of calibrator may be evaluated as necessary against whole blood
calibrators obtained from the NGSP CPRL at the University of Missouri
(UMO calibrators). Perform this procedure when validation of new lot
calibrators against current lot calibration does not confirm manufacturer-
assigned values.

a. Obtain aliquots of UMO whole blood calibrators. First calibrate both

instruments with UMO calibrators using their assigned values. Analyze
the current lot of in-house controls and both levels of new lot calibrator
in duplicate as unknowns. Verify that the controls fall within their
respective QC limits (preferably within 1 SD). Evaluate the results of
the new lot calibrators against their manufacturer-assigned values.

b. Next calibrate both instruments with the new lot of Tosoh calibrators
using the manufacturer-assigned values. Analyze the UMO controls

13
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

and in-house controls and verify that they fall within their respective QC
limits (preferably within 1 SD). If control results are acceptable, the
assigned values may be used. If controls do not fall within established
ranges, repeat analysis of new lot calibrators against UMO calibrators
to establish new assigned values.

c. Additionally, the most recent set of NGSP Monthly Monitoring samples

may be thawed and analyzed on both instruments and the results
compared with results obtained using the current lot calibrators.

(10) Following major maintenance, calibrate the system. Run controls plus
5 duplicate samples. Results should agree within 0.2%. Record these
results on the Rgt/Column Change Sheet found in the Rgt Change,
Calibrators, etc book.

D.. Calculations

How the Analyzer Calculates Calibration Factors

A(slope) = (Cal2A – Cal1A) B(intercept) = Cal2A – (Cal2M x A)

(Cal2M – Cal1M)

where Cal1A and Cal1M are assigned and measured values for
Calibrator 1, and Cal2A and Cal2M are assigned and measured
values for Calibrator 2.

Calibration Acceptability Criteria

When the calibration procedure is completed, the analyzer automatically accepts

or rejects the calibration results. If the calibration is unsuccessful, recalibration
will be required. A Calibration Error message appears and the run aborts if:

.-The two SA1c% results for Calibrator 1 differ by 0.3% or more.

-The two SA1c% results for Calibrator 2 differ by 0.3% or more.
- Any of the four calibrator results differs from its assigned value by ±30% or
more.

E. Interpretation of Results

For detailed information about interpreting the results, see the G7 Automated HPLC
Analyzer Operator’s Manual.

An unusual chromatogram can result either from an abnormal sample, a procedural error,
an instrument malfunction or sample-handling problem.

In general, review and question any chromatogram with the following characteristics:

1. The SA1c value is outside the normal range established by your laboratory.
2. The SA1c peak is not detected.
3. Total area reported is less than 500 or greater than 3000.
4. An unidentifiable peak (i.e., P00, P01, …) appears before the A0 peak.
5. The retention time for the SA1c peak is less than 0.62 or greater than 0.82.
If a result displays any of these characteristics, repeat the sample.

14
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

If the repeated sample also displays unusual characteristics, then it is appropriate to

evaluate whether the unusual result is due to an abnormal sample, a procedural error, an
instrument malfunction or a sample-handling problem. For further information, see
Troubleshooting in Chapter 6 of the Operator’s Manual.

Important! Do not report any result that displays the characteristics above without
appropriate review. The presence of any of these criteria may indicate that there is a
problem or the sample may be inappropriate for measurement by this method.

Barcoded samples are scanned automatically by the analyzer and the CID number
appears on the chromatographic printout in the ‘SAMPLE ID’ field. If a barcode is
unreadable or unavailable, the rack and position numbers of the sample appear in this
field instead. In such cases, always record the accession number or Lab ID on the
chromatogram. Record %SA1C value from the tape onto the protocol page. Be sure to
note any abnormal peak(s) (abnormal variants or POO peaks) on the protocol page.

1. Quality control evaluation:

-Calibrators (as unknowns): Acceptable range for calibrators analyzed as unknowns
are + 0.2 of assigned values.

-Controls: Values must fall within established ranges for each level.

-Batch duplicates: Results must agree within + 0.2 of the previous batch’s value.

2. Dilution studies demonstrate that the assay is linear from a Total Area of 500 to 3000.
In general, review and question any chromatogram with the following characteristics:

-The SA1C value is below 4.0%. Repeat the sample to confirm. Consult a
supervisor before reporting (< 4.0 %).

-Total area reported is less than 500 or greater than 3000. Repeat dilution using
appropriate reagent or sample syringe spacer to obtain area results within this range.

-The SA1C peak is not detected. Repeat the sample to confirm. Do not report
results. Consult a supervisor.

-An unidentifiable peak (P00, P01, ...) peak appears before the A1A or between the
A1A and the A0 peaks. Do not report results. Check for clots in the sample; re-
analyze. Consult a supervisor before reporting results.

NOTE: If a repeated sample also displays unusual characteristics, then it is

appropriate to evaluate whether the unusual result is due to an abnormal sample, a
procedural error, or sample-handling problem.

SA1C - Report % HbA1c (SA1C) to one decimal place.

F – Observe elevated HbF peak between A1B and LA1c+ peaks. Levels of fetal
hemoglobin (HbF) up to 15% do not affect test results because HbF is completely
resolved by the analyzer. For adult patients with HbF > 15%, send the specimen
to the VA Hospital (see instructions in ‘Other hemoglobin variants’ section).

H-VAR –Hemoglobin variants (for example, HbS and HbC and other)

15
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

-HbS (heterzygous) – HbS appears as an H-V1 peak following the A0 peak and
there is no carryover observed in the chromatograms that follow. HbS
(heterozygous) does not interfere with quantitation of HbA1c. Report the HbA1c
result with the following coded comment: C7672 (Abnormal hemoglobin variant
observed).

-HbC – HbC appears as an H-V2 peak that follows the A0 peak. There is no carryover
observed in the chromatograms that follow. HbC (heterozygous) does not interfere
with quantitation of HbA1c. Report the HbA1c result with the following coded
comment: C7672 (Abnormal hemoglobin variant observed).

-Other hemoglobin variants – may appear as a POO peak or H-VAR. Consult a

supervisor. Bring the sample to the Primary Care Lab for analysis on the DCA 2000.
Report the result with the comment: HGBVC (Abnormal hemoglobin variant observed.
HbA1c measured by the DCA 2000 immunoassay method.)

9. REPORTABLE RANGE OF RESULTS

Elevated levels of HbA1c suggest the need for more aggressive treatment of glycemia.
The American Diabetes Association recommends that a primary goal of therapy should
be

HbA1c < 7%, and that physicians should reevaluate the treatment regimen in patients
with HbA1c values consistently above 8%.

LINEAR RANGE (AMR): 3.0 – 19.0 %

Report results falling outside this range as <3.0 or >19.0 %.

CLINICALLY REPORTABLE RANGE (CRR): 3.0 – 19.0 %

.

10. QUALITY CONTROL (QC) PROCEDURES

Two levels of glycated hemoglobin control (Normal and Elevated) are analyzed in
duplicate (or more) with each batch. Westgard rules are followed as outlined in the
general laboratory Quality Control and Quality Assurance procedure. Controls are
analyzed at the beginning of a run, periodically throughout, and at the end of a run.
Controls are prepared from whole blood drawn from a normal (Normal) and a diabetic
o
(Elevated) individual. Stable indefinitely stored at –70 C. Controls should be diluted with
Hemolysis & Wash Solution to obtain a Total Area on the chromatogram in the range of
500-3000. If the value of one or more control specimens is out of the acceptable range,
recalibrate the system and rerun the controls before testing patient samples.

Collect six 10-mL potassium-EDTA tubes from one normal or one diabetic volunteer
depending on the control level to be prepared. Mix well by gentle inversion then pour
blood into a 100-mL beaker containing a small magnet and place the beaker into a
bucket containing wet ice. Place bucket on a magnetic stirrer set on low speed. Aliquot
~ 50 uL into 0.2 mL polypropylene microcentrifuge tubes with caps. Continue to add ice
to the bucket as needed to keep beaker chilled. During preparation, aliquots may be held
o
in an insulated bucket filled with ice until placed into boxes to be stored at –70 C (chest
freezer). At the start of each week, take one week’s supply of controls from the stock
supply and place in the working -70oC freezer.

Evaluate the new lot of controls according to QC/QA guidelines to establish

16
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

ranges prior to placing into clinical us

Quality control evaluation:

Calibrators (as unknowns): Acceptable range for calibrators analyzed as unknowns are +
0.2 of assigned values.
Controls: Values must fall within established ranges for each level.
Batch duplicates: Results must agree within + 0.2 of the previous batch’s value.

11. REMEDIAL ACTION IF CALIBRATION OR QC SYSTEMS FAIL TO MEET

ACCEPTABLE CRITERIA

If control values are out of the acceptable range, recalibration is required. Reanalyze any
patient samples after recalibration.

12. LIMITATIONS OF METHOD; INTERFERING SUBSTANCES AND CONDITIONS

Dilution studies demonstrate that the assay is linear from a Total Area of 500 to 3000.

Patients who possess no hemoglobin A (i.e., homozygous variants, double

heterozygotes, etc.), will not have HbA1c. Therefore, the level of glycemic control in
these patients cannot be measured by this method.

Levels up to 15% Hemoglobin F (fetal hemoglobin) do not interfere with the accurate
determination of the percent of HbA1c in the normal range population.

HbA1c results cannot be reported in patients with heterozygous hemoglobinopathies in

which the non-A hemoglobin elutes prior to the A0 peak. In such situations, the results
may be falsely increased or decreased depending upon the retention time of the
abnormal hemoglobin. Peak areas of hemoglobins eluting after the A0 are not included in
the calculations of Total Area on the G7 Automated HPLC System. Therefore, the
presence of such hemoglobin does not interfere with the calculation of the HbA1c result.
Because HbS, HbD, and HbC elute after the A0, glycemic control for heterozygous
patients with Hemoglobin AS, AD and AC may be accurately monitored using HbA1c on
the G7 Automated HPLC System when the Variant Analysis Mode is used.

Various substances other than sugars can form adducts with hemoglobin, thereby
altering its charge characteristics. Falsely elevated results can occur if these adducts co-
migrate with the stable HbA1c. Literature cites examples including individuals with opiate
addiction 2, lead poisoning, uremia and alcoholism 3, as well as those receiving large
amounts of aspirin (acetylated hemoglobin)4. Clinically, the most important of these
possible interfering adducts occurs in uremia 3. The interference correlates with the BUN
(blood urea nitrogen) and appears to result at least in part from the carbamylation of
hemoglobin by urea-derived cyanate (carbamylated hemoglobin).5 Studies on the G7
Automated HPLC in the Variant Analysis Mode indicate that the carbamylated and
acetylated hemoglobin derivatives are eluted in the labile fraction. No interference in
stable HbA1c results from these derivatives has been observed up to concentrations of
5.5% of carbamylated hemoglobin and 2.7% of acetylated hemoglobin.

The results obtained from this assay should be used in conjunction with other data (for
example, signs and symptoms, duration of diabetes, results of other tests, age of patient,
clinical impressions, degree of adherence to therapy, etc.). This test should not be used
to diagnose diabetes mellitus. The performance characteristics for diagnosis have not yet
been established.

17
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

Abnormal Red Cell Survival

The life span of red blood cells is shortened in patients with hemolytic anemias,
depending upon the severity of the anemia. As a consequence, specimens from such
patients will exhibit decreased glycohemoglobin levels compared to patients with
expected red cell life span.

The life span of red blood cells is lengthened in polycythemia or post-splenectomy

patients. Specimens from such patients may exhibit increased glycohemoglobin levels 6.
For these types of patients, reference range values will not apply.

13. REFERENCE RANGES (NORMAL VALUES)

REFERENCE RANGE: 4.3 – 6.0 % (DCCT/EDIC normal range)

14. CRITICAL CALL RESULTS (“PANIC VALUES”)

Early Reporting Results for NHANES:

Notify the NHANES Medical Officer of any SA1c% results greater than 6.9%. The
contact person will report these results as soon as possible.

15. SPECIMEN STORAGE AND HANDLING DURING TESTING

Any specimens not analyzed on the day of arrival in the laboratory are stored in the
refrigerator (4°C - 8°C). Upon completion of analysis, specimens are stored for 1 week.
NHANES specimens are frozen at -70°C and discarded after 1 year.

16. ALTERNATIVE METHODS FOR PERFORMING TEST OR STORING SPECIMENS IF

TEST SYSTEM FAILS

The laboratory has 2 instruments for performing glycohemoglobins. If neither instrument

is available for use, the specimens are stored at 4°C until testing can be performed.

Unopened and stored at 2-8°C, the HbA1c Calibrator and Control Sets are stable until the
expiration date printed on the label. After reconstitution, calibrators are stable for one
week when stored at 2–8°C. Reconstituted Controls are stable for five (5) days when
stored at 2–8°C.

Unopened Elution Buffers HSi Variant (S) 1, 2, and 3 are stable until the expiration date
printed on the label. After opening, Elution Buffers (S) is stable for three months. Store at
4-25°C.

Unopened Hemolysis & Wash Solution is stable until the expiration date printed on the
label. After opening, Hemolysis & Wash Solution is stable for three months. Store at 4–
25°C.

The unopened TSKgel G7 HSi Variant column is stable indefinitely when stored at 4–
15°C away from direct sunlight. Columns are warranted for 1500 injections. If columns
are performing well based on acceptable quality control results and chromatogram
quality, they may be used past 1500 injections until they no longer meet performance
criteria.

18
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

17. TEST RESULT REPORTING SYSTEM; PROTOCOL FOR REPORTING CRITICAL

CALLS (IF APPLICABLE)

NHANES SA1c% results are entered unto a spreadsheet provided electronically by

WESTAT, Inc for NHANES.
To access the spreadsheet click on My Computer → Z drive → User → Dep Labs →
Collab Studies → NHANES → Glyhb 004.
Choose the file named with the corresponding box number.
Enter the analysis date, run number, technologist’s initials, SA1c%, and result comment
code.

The spreadsheet will be sent electronically by the contact person.

Early Reporting Results for NHANES:
Notify the NHANES contact person of any SA1c% results greater than 6.9%. The contact
person will report these results as soon as possible.

18. TRANSFER OR REFERRAL OF SPECIMENS; PROCEDURES FOR SPECIMEN

ACCOUNTABILITY AND TRACKING

All shipments are recorded on the NHANES Shipping Log upon receipt. Actions taken
during the course of analysis, result reporting, and specimen retention are also recorded
on the log.

19
 Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

19. SUMMARY STATISTICS AND QC GRAPHS

Summary Statistics for Glycohemoglobin by Lot

Standard Coefficient of
Lot N Start Date End Date Mean Deviation Variation
XII 111 1/17/2007 5/30/2008 5.422 0.081 1.5
XIIa 111 1/17/2007 5/30/2008 10.905 0.122 1.1
XIII 56 6/3/2008 2/11/2009 5.357 0.074 1.4
XIIIa 56 6/3/2008 2/11/2009 10.598 0.080 0.8

2007-2008 Glycohemoglobin Quality Control

12
XIIa
XIIIa

10

6 XII

0
1/17/2007 4/27/2007 8/5/2007 11/13/2007 2/21/2008 5/31/2008 9/8/2008 12/17/2008

20
Glycohemoglobin in Whole Blood using Tosoh G 7 HPLC
NHANES 2007-2008

REFERENCES

1. G7 Automated HPLC Analyzer Operator’s Manual, TOSOH Bioscience, Inc., Inc. 2002.
2. Coriello A, Giugliano D, Dello Russo P, Sgambato S, D’Onotrio F. Increased glycosylated
hemoglobin A1 in opiate addicts. Evidence for hyperglycemic effect of morphine. Diabetologia
1962;22:379.
3. Goldstein DE, Little RR, Wiedmeyer HM, England JD, and McKenzie EM. Glycated
hemoglobin: methodologies and clinical applications. Clin Chem 1986;32;B64-B70.
4. Nathan DM, Francis TB, Palmer JL. Effect of aspirin on determinations of glycosylated
hemoglobin. Clin Chem 1983;29:466-9.
5. Fluckiger R, Harmon W, Meier W, Loo S, Gabbay KH. Hemoglobin carbamylation in
uremia. N Eng J Med 1981;304:823-7.
6. Tze et al. Hemoglobin A1c – An Indicator of Diabetic Control. J of Pediatrics 1978, 93:1316.
7. American Diabetes Association, Standards of medical care for patients with diabetes
mellitus (Position Statement). Diabetes Care. 1998;21 (Suppl. 1):S23-S31.
G7 Automated HPLC Analyzer
8.Trivelli LA, Ranney HM, Lai H-T. Hemoglobin components in patients with diabetes mellitus.
NEJM 1971; 284(7):353.
9.Bunn HF, Gabbay KH, Gallop PM. The glycosylation of hemoglobin: relevance to diabetes
mellitus. Science 1678; 200:21-7.
10.Cerami A, Koenig RJ. Hemoglobin a1c as a model for the development of sequelae of
diabetes mellitus. TIBS 1978; Apr:73.

21