Quality Requirements and Standards

Recent coverage of estimated average glucose (eAG) and the rise of HbA1c as the method for diagnosis of diabetes raises an obvious question: Are HbA1c methods and results of better quality than glucose methods and results? Dr. Westgard looks at the numbers.

Pre-Analytic vs. Analytic Issues:
HbA1c for screening and diagnosis of diabetes?

• Convenience first!
• Review of abbreviations
• Analytic traceability reveals complexity!
• Are test results equivalent?
• Why must analytical issues take a backseat?

December 2008

I have been following the story of HbA1c and the recent recommendations that HbA1c values be expressed as “estimated average glucose” (eAG, Clin Lab News, October 2008) and that HbA1c now be used as a test for screening and diagnosis, not just as a measure of long-term glycemic control (Clin Lab News, December 2008). Some may explain this evolution as part of evidence-based laboratory medicine, but I also suspect it is partly as result of “scientific marketing,” i.e., recommendations based on the preferential interests of an involved group of manufacturers and scientists, who have good intentions and strong commitments to making improvements in healthcare. However, there often is a trade-off between pre-analytic and analytic concerns, which I believe can be illustrated in this case.

Convenience first!

In this case, the age-old standards for Fasting Plasma Glucose (FPG) and the Oral Glucose Tolerance Test (OGTT) are no long considered convenient:

“…drawbacks of the tests include the requirement for patients to fast at least 8 hours beforehand and the need for a confirmatory test on another day. ‘If a doctor sees a patient and tells them they need to come back for an OGTT of FPG, it’s often inconvenient and many simply don’t come back…’”

And the argument for use of HbA1c is then that it is more convenient:

“Perhaps the greatest argument in favor of HbA1c as a screening and diagnostic test are that it can be performed at any time, involves a simple blood draw, and is highly reproducible. The test also reflects long-term glycemic levels and is not as subject to short-term lifestyle changes as OGTT or FPG…”

Thus, pre-analytic factors take priority, along with a defense of analytic performance of HbA1c:

“While standardization of HbA1c was a significant problem at the time the landmark Diabetes Control and Complications Trial (DCCT) was published in 1993, the National Glycohemoglobin Standardization Program (NGSP) has been instrumental in standardizing and establishing a true reference method for HbA1c. Today, more than 99% of labs that measure HbA1c in the U.S. used NGSP-certified methods. In addition, CAP proficiency testing criteria will be steadily tightened over the next 4 years with a reduction from the current ± 12% to ± 6%. Furthermore, based on results of the A1c-Derived Average Glucose (ADAG) study, which demonstrated a linear relationship between estimated average daily glucose (eAG) and HbA1c, ADA now recommends that laboratories provide physicians with both eAG and HbA1c.”

Review of abbreviations

Just to keep things clear, here are the current abbreviations being used by CLN in this discussion:

• ADA = American Diabetes Association
• ADAG = A1c-Derived Average Glucose study
• CPG = Casual Plasma Glucose test
• DCCT = Diabetes Control and Complications Trail
• eAG = estimated Average Glucose (calculated from HbA1c result)
• FPG = Fasting Plasma Glucose test
• GHb – Glycated Heoglobin
• HbA1c = Hemoglobin A1c test
• IFG = Impaired Fasting Glucose
• IFCC = International Federation for Clinical Chemistry
• IGT = Impaired Glucose Tolerance
• NGSP = National Glycohemoglobin Standardization Program
• OGTT = Oral Glucose Tolerance Test
• RPG = Random Plasma Glucose test

Analytic traceability reveals complexity!

Glucose is a simple analyte with well-established methodology and traceability. Analytical methods are primarily enzymatic with hexokinase predominating over glucose oxidase methodologies.

For HbA1c, different methods measure glycated hemoglobin (GHb) with different sensitivities and specificities. There are four measurement principles and some 30+ different methods. Traceability is a complex task requiring a global network of laboratories to maintain and monitor the relationships between national reference methods (such as the US NGSP method) and an international definitive method by the IFCC. Because of significant differences between different manufacturer’s methods, it has been necessary for manufacturers to establish calibration factors that lead to comparability results from the NGSP method. They do this by testing a group of 40 patient samples and comparing results with those obtained by the NGSP. If 95% of the results agree within ± 0.85% GHb, the method becomes NGSP certified.

Given the established equivalence of HbA1c and glucose, i.e., 1% HbA1c is equivalent to 14-15 mg/dL glucose, agreement within ± 0.85% would be the same as 12-13 mg/dL glucose. When compared to the glucose CLIA criterion for acceptable performance (6 mg/dL or 10%, whichever is greater), this is not a very demanding criterion for certification. Furthermore, CAP’s current 2008 PT criterion for GHb is 12%, which is more lax than the CLIA criterion for glucose and the NSGP criterion for certification. Note that the CAP criterion was 15% in 2007 and is scheduled to tighten to 10% in 2009, 8% in 2010, and 6% in 2011. Evidently there is some perception by CAP that the quality of HbA1c testing needs to be improved.

Thus, it seems clear that there are more significant analytical issues with HbA1c testing than with glucose testing. Comparability of glucose test results is straightforward, whereas manufacturer’s must go through a calibration/certification process to provide comparability of HbA1c results.

Are test results equivalent?

The proposed criteria for screening of diabetes are as follows:

• FPG ? 100 mg/dL
• HbA1c > 6.0%
• RPG ? 130 mg/dL

The proposed criteria for diagnosis are as follows:

• FPG ? 126 mg/dL
• HbA1c > 6.5%
• RPG ? 200 mg/dL

If you now recall the equation for equivalence between glucose and HbA1c (CLN, Oct 2008):

eAG(mg/dL) = 28.7*HbA1c – 46.7

That means that the screening cutoff for HbA1c (6.0) is equivalent to 125 mg/dL (not 100 mg/dL as for FPG) and the diagnostic cutoff is equivalent to 140 mg/dL (not 126 mg/dL as for FPG).

Now I’m getting worried! If HbA1c can be used to estimate an equivalent glucose concentration, then why aren’t the new HbA1c criteria for screening and diagnosis of diabetes equivalent to the earlier criteria for FPG? Obviously, things are not as simple and straightforward as the new recommendations suggest.

Why must analytical issues take a backseat?

There has been a big investment in HbA1c by manufacturers, professional organizations, and research scientists. There certainly has been progress as a result of global and national standardization initiatives. And there certainly are still are problems remaining to be solved – many of which are analytical issues that affect the reliability of test results.

We have discussed the performance of tests for glycated hemoglobin on this website in some detail (see here, here and here for recent analyses). On the basis of CAP proficiency testing results, it is clear there still are significant biases between methods and that differences between values such as 6.0% and 6.5% can be readily attributed to an analytical method selected by a laboratory. Glucose testing is much more reliable and differences between 100 mg/dL and 126 mg/dL for FPG will be much more clinically significant.

Nonetheless, evidence-based medicine seems content to ignore scientific evidence of analytical variability for the sake of convenience. But take note, as discussed in the CLN article, these new recommendations are not likely to take hold outside the US! Our high-technology healthcare values certain measurement technology above the analytical certainty of results. Again our healthcare system seems to be subject to market forces rather than guided by quality of care!