QA/ QC aspects of HbA1c analysis

Send This

Download

Comment

Favourite

more

Rating :

Rate It:

Embed :

Post a comment

Post Comment on Twitter Post Comment on SlideWorld

Comments:

Subscribe to follow-up comments

Notes

Hide Notes

Slide 1 :

Trefor Higgins FCACB, Director of Clinical Chemistry Dynacare Kasper Medical Laboratories Edmonton, Alberta QC and QA Aspects of HbA1c Testing

Slide 2 :

Outline Definition Formation QC Expectations Reality QA tools NACB Guidelines Calculated A1c Peak shape ? checks Hand approach

Slide 3 :

Definitions of HbA1c Original Definition: HbA1c: a peak on a chromatogram – reference range 4 – 6% IFCC Definition: HbA1c: Hb that is irreversibly glycated at one or both N-terminal valines of the ? chain - reference range 2 - 4%

Slide 4 :

Uses of HbA1c Analysis Monitor glycemic control patient management clinical guidance and audit clinical trial design Independent risk factor for MI in non-diabetic patients Estimate of recovery from infection in ICU

Slide 5 :

Formation of HbA1c

Slide 6 :

HbA1c Formation Non-enzymatic 2 Stage reaction First reaction is 100 times faster than second reaction HbA1c formation is irreversible

Slide 7 :

2002 National Academy of Clinical Biochemistry Guidelines Only methods certified by NGSP should be used Participation in CAP glycohemoglobin program Interassay CV < 5% (ideally < 3%) 2 control materials with different mean values Verify values below lower limit of reference interval by repeating testing Verify values above 15% by repeat testing Schiff base (labile HbA1c) should be removed before assay Reference range 4 - 6%

Slide 8 :

QC Aspects Precision Requirements for HbA1c assay 2002 NACB Guidelines Skeie’s approach patient expectations physician response Kolatkar’s approach Phillipou’s approach Physician expectation

Slide 9 :

Precision Requirements for HbA1c Methods 2002 NACB Guidelines Clin Chem 2002; 48: 436 – 472 Interassay CV 5% (ideally 3%)

Slide 10 :

Precision Requirements for HbA1c Methods Kolatkar et al Clin Chem 1994; 40: 1608-9 Intensive diabetes management requires very precise testing of glycohemoglobin . CV should be in the range of 2 - 3%.

Slide 11 :

Precision Requirements for HbA1c Methods Skeie et al. Clin Chem 2001; 47: 1212-7 Patients were asked to indicate change in HbA1c from 9.4% that they considered a true (real) change; from their responses we derived patient-derived quality specifications for HbA1c. Conclusion: HbA1c increasing CV 3.1% decreasing CV 3.2%

Slide 12 :

Skeie’s Approach (2) Skeie et al. Clin Chem 2005; 51: 1145 - 53 Asked physicians to assess impact of rising/falling HbA1c value on their method of treatment . Physicians expect CV 3-5%.

Slide 13 :

Physician Expectations: Diabetic endocrinologists were asked to state their expectations of reproducibility and significant differences between HbA1c results. The conclusion was that an analytical variation > 3% could give clinically significant HbA1c results. Precision Requirements for HbA1c Measurements

Slide 14 :

Precision Requirements of HbA1c Methods Phillipou’s approach. Clin Chem 1993; 39: 2305 - 8 Based on biological variation of 1.7% Practical working CV of 3%

Slide 15 :

Summary of Precision Requirements for HbA1c Measurements 1. NACB: ideally < 3% (<5% tolerated) 2. Kolatkar: CV should be < 2 - 3%. 3. Skeie: Patient-derived specifications 3.1% (HbA1c increasing) 3.2% (HbA1c decreasing) 4. Skeie: Physician response 3-5%. 5. Physician: preferably CV of < 3%. 6. Phillipou: practical CV of 3%.

Slide 16 :

Critical Differences for HbA1c at Different Analytical CVs at a HbA1c of 7%

Slide 17 :

Seasonal Variation in HbA1c Studies in the UK showed HbA1c is higher in January and February and lower in October. Possibly due to amount of food eaten

Slide 18 :

Requirements vs. Reality for Precision on HbA1c Testing CAP/Healthmetrix data Cembrowski approach Tran’s approach

Slide 19 :

Performance of HbA1c Methods on CAP Program

Slide 20 :

Precision Requirements for HbA1c Methods Cembrowski’s Approach: Diabetic Technology and Therapeutics 2003; 5: 975-8. Took replicate measurements of HbA1c from patients using different methods.

Slide 21 :

Average SD of HPLC (diamonds) and immunoassay (squares) vs. time between testing. The y-intercept represents the average analytic SD, while graphs represent both analytic and physiologic variation. Precision Requirements for HbA1c Measurements

Slide 22 :

Precision Requirements for HbA1c Measurements Cembrowski’s Conclusion: Precision for different methods vary substantially but some are acceptable (defined as 3% or less).

Slide 23 :

Precision Requirements for HbA1c MeasurementsDelta Checking HbA1c Values Tran et al - 33,000 differences from analyses of 150,048 HbA1c results 3 HPLC analyzers

Slide 24 :

Slide 25 :

Tran’s Conclusion At a HbA1c of 7.16%, the CV is 3.7%. This variation is remarkably low given that HbA1c were measured over 2 years on any one of 3 analyzers.

Slide 26 :

QA Aspects of HbA1c Analysis NACB guidelines for repeat testing – same v different analytical method monitoring of peak shapes calculated v measured HbA1c low hemoglobin Hb variants/HbF ? checks

Slide 27 :

NACB recommendations for repeat testing: below reference interval above 15% Use of same method will give same results: review CBC data use method based on different principle QA – NACB Guidelines

Slide 28 :

56 y f with fasting glucose 5.4 mmol/L, normal CBC HbA1c Results: Presence of ? chain variant interference on Bio-Rad VARIANT method

Slide 29 :

Chromatograms of patient with interfering ? chain variant on: HbA1c program ? thalassemia program

Slide 30 :

Five Steps in Identifying Presence of Variant - IRRCR Inspect chromatogram Repeat on same analyzer type Review cbc data Calculate HbA1c Repeat on analyzer based on alt. technology

Slide 31 :

On HPLC, changes in retention time and peak shape may indicate the presence of a hemoglobin variant. Monitoring Peak Shape

Slide 32 :

Slide 33 :

HbA1c in the presence of a hemoglobin variant has 2 problems: 1) the analytical problem 2) the physiological problem

Slide 34 :

Interferences we can see: hemolysis - K+, glucose, LD lipemia - many chemistries icterus - acetaminophen Interferences we cannot see: ascorbic acid - glucose, cholesterol, triglyceride, uric acid methods heterophile antibody - all immunoassays In general, we do not have a clue if these are present in a sample. Hemoglobin Variant Interference - Analytical

Slide 35 :

What ascorbic acid is to chemistry measurements and heterophile antibody is to immunoassay methods, hemoglobinopathies are to HbA1c analysis. Hemoglobin Variant Interference - Analytical

Slide 36 :

The Analytical Problem - HPLC a) variant co-elutes with HbA - glycated species co-elute b) variant co-elutes with HbA – glycated species (if any) does not c) variant and HbA inadequately resolved

Slide 37 :

The Analytical Problem - Immunoassay a) affinity of antibody for variant b) steric hindrance

Slide 38 :

Error in HbA1c by Immunoassay Patient 28 y f with known C trait At HbA1c of 6%, Variant II bias is +0.82% (+0.85 at 7%). Little (Clin Chem 2004; 50S6; D-12) Bias from true HbA1c value of 6.15%.

Slide 39 :

The Physiological Problem Normal red cell survival: 120 days ? thalassemia reported at 90 – 96 days Hemoglobin variants have decreased life cycle (HbS reported at 90 days) Preferential expression of normal HbA compared to variant

Slide 40 :

NGSP requires bias from HbC or HbS trait of < 10% to be not clinically significant. Bio-Rad VARIANT II meets this criterion. Higgins et al. Clin Chem 2006; 52(s): C-88 Mongia et al. Presented at Academy of Clinical Laboratory Physicians and Scientists 2006 (Abstract # 2006-58)

Slide 41 :

Effect of Presence of HbF % HbA1c vs. IFCC Reference Method Bias Plot - – samples with elevated HbF Effect of Elevated Fetal Hemoglobin on HbA1c Measurements: Five Common Assay Methods compared to the IFCC Reference Method. C. Rohlfing, S. Connolly, J. England, R. Little. Dept. of Pathology & Anatomical Sciences, U of Missouri School of Medicine, Columbia MO.

Slide 42 :

Effect of Presence of HbF Effect of Elevated Fetal Hemoglobin on HbA1c Measurements: Five Common Assay Methods compared to the IFCC Reference Method. C. Rohlfing, S. Connolly, J. England, R. Little. Dept. of Pathology & Anatomical Sciences, U of Missouri School of Medicine, Columbia MO. Bias vs. IFCC RM

Slide 43 :

Conclusions The G7 (when the HbF peak is properly recognized) and Variant II methods show minimal interference in HbA1c measurement up to HbF of ~30%. The DCA 2000, 2.2+ and CLC330/385 methods underestimate HbA1c in the presence of elevated HbF, especially at HbF levels > 20%. Physicians need to be aware of potential interferences when interpreting HbA1c results. Effect of Elevated Fetal Hemoglobin on HbA1c Measurements: Five Common Assay Methods compared to the IFCC Reference Method. C. Rohlfing, S. Connolly, J. England, R. Little. Dept. of Pathology & Anatomical Sciences, U of Missouri School of Medicine, Columbia MO.

Slide 44 :

Interference from HbF – DKML initial results

Slide 45 :

Limitations of HbA1c Analysis “Hemoglobinopathies may interfere with GHb analysis independent of their effects on erythrocyte survival.” “Moreover, hemoglobin variants cannot be identified by immunoassay.” Sacks DB. Clin Chem 2003; 49: 1245-7.

Slide 46 :

The Effect of Low Hemoglobin A positive bias occurs in samples with low Hb using the Tina-quant II HbA1c immunoassay. Clin Biochem 2004: 37; 735 (51)

Slide 47 :

Relationship of Glucose to HbA1c NGSP (glucose) = 1.98 (HbA1c) – 4.29 IFCC (glucose) = 1.84 (HbA1c) – 0.01 NGSP – DCCT method

Slide 48 :

ADA Guidelines Treatment goal normal or near normal with HbA1c < 7%. Correlation between HbA1c and mean plasma glucose levels.

Slide 49 :

Use of Calculated HbA1c ADA equation rearranged: HbA1c (calculated) = 0.5 (glucose) + 2.25 (SI units) HbA1c (calculated) = 0.0285 (glucose) + 2.15 (traditional units)

Slide 50 :

Calculated HbA1c Fasting HbA1c (calculated) = 0.90 HbA1c (measured) – 0.14 (r = 0.733) 2 hr pc HbA1c (calculated) = 1.65 HbA1c (measured) – 4.52 (r = 0.83) Random HbA1c (calculated) = 1.39 HbA1c (measured) – 3.13 (r = 0.68)

Slide 51 :

Use of Calculated HbA1c Best agreement found with HbA1c calculated from fasting glucose. Poorest agreement found with HbA1c calculated from random glucose.

Slide 52 :

ADA/CDA Recomendations for Measurement of HbA1c Frequency at least: 2 times/year in patients meeting treatment goals or with stable glycemic control 4 times/year in patients whose therapy has changed or who are not meeting glycemic controls

Slide 53 :

HbA1c in Edmonton Region Compared absolute differences in HbA1c values 3 months (+/- 2 weeks apart).

Slide 54 :

Slide 55 :

Slide 56 :

Conclusions Hemoglobin variants can challenge analytical methods. There are new QA tools that may be used. HbF interference +/-. Hb concentration may be a significant fact in immunoassay measurements. Imprecision of 3% desirable.

My diet and exercise...

Medical Aspects of S...

Mental Health Aspect...

An Analysis of Thera...

A generalized rankor...

Chromatographic anal...

Free Powerpoint Templates

	trefor.higgins@dynalifedx.com 5 Years ago.



1684 Views, 0 favourite
PowerPoint Presentation on QA/ QC aspects of HbA1c analysis or PowerPoint Presentation on A descrip more PowerPoint Presentation on QA/ QC aspects of HbA1c analysis or PowerPoint Presentation on A description of challenges in the analysis and interpretation of HbA1c analysis. less
More By User Flag as inappropriate Select your reason for flagging this presentation as inappropriate. If needed, use the feedback form to let us know more details. Cancel