This article is more than 5 years old. Information may no longer be current.

Debate: Should labs report an estimated average glucose with HbA1c?

Issue: July 2009

ByZachary T. Bloomgarden, MD, MACE, FASPC

Add topic to email alerts

Receive an email when new articles are posted on

Please provide your email address to receive an email when new articles are posted on .

Subscribe

Added to email alerts

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

Back to Healio

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Back to Healio

A debate was held at the American Diabetes Association’s 69th Scientific Sessions in June on the question of whether estimated average glucose should be reported with HbA1c. From my perspective, the answer is no.

The methodology of HbA1c standardization adopted in 2001 by the International Federation of Clinical Chemists is based on a synthetic N-terminal valine glycated hemoglobin A1 beta chain. This ‘true’ HbA1c value is only about three-quarters of what was previously considered HbA1c with biologically derived standards. In 2004, an ADA/European Association for the Study of Diabetes/International Diabetes Federation working group suggested a compromise approach between the new HbA1c and the now considered inaccurate, older HbA1c standard, which included other hemoglobin-derived compounds. The group recommended that HbA1c be reported as an estimated average glucose level equivalent to the HbA1c.

During the debate, David Nathan, MD, took the position that such estimated average glucose reporting with HbA1c will serve a useful purpose. To set diagnostic criteria for diabetes, the 1997 ADA Expert Committee on the Diagnosis and Classification of Diabetes used evidence that HbA1c, fasting glucose and two-hour post-load glucose are all associated with retinopathy. Nathan cited similar evidence from the Diabetes Control and Complications Trial and United Kingdom Prospective Diabetes Study, “cementing HbA1c as the means by which we measure our targets.” The goal of treatment is an HbA1c <7%, observing that physicians are “guiding people based on HbA1c,” he added.

Nathan also provided evidence from the A1C-Derived Average Glucose (ADAG) study of more than 500 healthy individuals and those with type 1 and type 2 diabetes (83% white). Participants performed continuous glucose monitoring two to three days per month as well as preprandial and postprandial self-monitoring of blood glucose for four months; researchers obtained approximately 2,500 continuous glucose monitoring values and 20 self-monitored blood glucose values per patient. The researchers calculated an estimated average glucose formula: 2.87 x HbA1c – 46.7, with r² 0.84.

I argued the contrary position, noting that the level of HbA1c may vary to a greater or lesser extent with different degrees of erythrocyte turnover, erythrocyte glucose entry and hemoglobin glycation and deglycation — all of which may be under the control of factors other than circulating glucose levels. For example, when two twins are concordant for diabetes, their HbA1c levels are, unsurprisingly, rather similar. But, HbA1c levels also show a correlation between twins discordant for diabetes, which suggests a strong non-glycemic heritable component.

Further, hyperglycemia is directly associated with reduced erythrocyte survival, suggesting a mechanism by which higher glucose levels might not have proportionate increase in HbA1c. HbA1c similarly underestimates glycemia in patients with HIV and those on hemodialysis. HbA1c (but not fructosamine) increases with age, controlling for factors including glycemia. HbA1c levels are also higher in blacks than in non-Hispanic whites.

I cited an analysis of 623 insulin-treated patients with type 2 diabetes who measured capillary blood glucose before and after breakfast, lunch and dinner and at bedtime on three days during the two weeks prior to HbA1c measurement. Frequency analysis of the glucose group vs. HbA1c revealed important differences. For example, of 224 patients with mean plasma glucose of 110 mg/dL to 140 mg/dL, 10% had an HbA1c level <6% and 10% had an HbA1c level >8.1%. Conversely, comparison of HbA1c grouping with mean plasma glucose revealed that of 260 patients with HbA1c between 6.5% and 7.5%, 10% had a mean level <115 mg/dL while 10% had mean glucose >171 mg/dL. Approximately 20% of estimated average glucose values, then, differed rather substantially from measured mean glucose values in the patients, according to the unpublished data.

This phenomenon of intrapatient variability of hemoglobin glycation is well-demonstrated in the DCCT one-day, seven-sample capillary glucose vs. HbA1c database of more than 250,000 blood glucose values and 72,000 HbA1c values. Continuous glucose monitoring of average glucose — although strongly correlated with HbA1c — also leads to error when applied to individual patients. The discrepancy between HbA1c and mean glucose may be clinically detrimental. 'High glycators' in the ACCORD study would, if assigned to a treatment protocol dictating additional medication for HbA1c regardless of glycemia, develop hypoglycemia. The study showed that both hypoglycemia and the intensive protocol, which utilized sulfonylureas and insulin, increased mortality.

I concluded that HbA1c reflects the phenomenon of hemoglobin glycation being proportional to mean glycemia, and that the measurement is certainly useful in understanding the glycemic exposure of patients with diabetes; however, one should not expect all patients to glycate hemoglobin to the same degree. Just as the estimated glomerular filtration rate has allowed greater appreciation of renal function based on creatinine by accounting for age, body weight and ethnicity, it will be important to develop and validate new approaches to understand the glycemic exposure of patients with diabetes; to understand HbA1c-derived estimated average glucose in the context of age, ethnicity, anemia and renal function; and, perhaps, to move beyond HbA1c.

Zachary T. Bloomgarden, MD, is an Endocrine Today Editorial Board member and Clinical Professor in the Department of Medicine at Mount Sinai School of Medicine.

For more information:

• Bloomgarden ZT. Diabetologia. 2008;51:1111-1114.
• Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 1997;20:1183–1197.
• Gallagher E. Journal of Diabetes. 2009;1:9-17.
• Gerstein HC. N Engl J Med. 2008;358:2545-2559.
• Herman WH. Diabetes Care. 2007;30:2453-2457.
• Kilpatrick ES. QJM. 1996;89:307-312.
• Kim PS. Diabetes Care. 2009 Jun 5.
• McCarter RJ. Diabetes Care. 2004;27:1259-1264.
• Nathan DM. Diabetes Care. 2008;31:1473-1478.
• Pani LN. Diabetes Care. 2008;31:1991-1996.
• Peacock TP. Kidney Int. 2008;73:1062-1068.
• Selvin E. Diabetes Care. 2009;32:828-833.
• Snieder H. Diabetes. 2001;50:2858-2863.
• Virtue MA. Diabetes Care. 2004;27:931-935.
• Wilson DM. Diabetes Care. 2008;31:381-385.