A1C for diagnosis of diabetes   (July 12, 2009)

For years, physicians and patients have trusted the A1C lab test to judge if a patient's diabetes is well-controlled or not; diabetes physicians and diabetes organizations have urged people with diabetes to get their number under 7 (or in some cases, even lower). A1C, also called hemoglobin A1c, and sometimes abbreviated HbA1c, measures the amount of glucose that's hooked to hemoglobin in red blood cells, and gives an estimate of how the blood glucose has been doing the past 2 to 3 months.

But until recently, the possible use of the A1C test to diagnose diabetes has been considered a no-no. Diagnosis of diabetes has been based almost exclusively on fasting blood glucose values of 126 mg/dl (7 mmol/L) or greater. Symptoms (if present), fancy glucose testing with oral glucose tolerance tests, family history, positive tests for urine glucose, and elevated A1C values have all been considered supportive of the diagnosis, but if the glucose isn't 126 or more, the diagnosis cannot be established with certainty.

Using the A1C to diagnose diabetes may become a recommended way to make the diagnosis. An international expert committee is recommending the A1C assay as the new test for the diagnosis of diabetes; their report is published in the journal Diabetes Care this month: International Expert Committee Report on the Role of the A1C Assay in the Diagnosis of Diabetes

The experts point out that the A1C assay is an accurate, precise measure of chronic glucose levels and correlates well with the risk of diabetes complications, and has several advantages over laboratory measures of glucose, especially that it does not need to be collected fasting.

They recommend that diabetes should be diagnosed when A1C is at or above 6.5%. They suggest that the diagnosis should be confirmed with a repeat A1C test, unless the patient is symptomatic with plasma glucose levels above 200 mg/dl (above 11.1 mmol/l). They point out that if A1C testing is not possible (because of cost or convenience), previously recommended diagnostic methods (fasting or 2-hour post-prandial glucose levels, with confirmation) would remain as acceptable ways to make the diagnosis.

The experts make it clear that their recommendations have not (yet) been accepted by the major diabetes organizations (American Diabetes Association, the European Association for the Study of Diabetes, and the International Diabetes Federation) who assembled the committee. Nor is it likely that these organizations will suddenly switch gears and endorse this report's recommendations in the near future. The American Diabetes Association issued a press release and "responded to the Expert Committee's report by endorsing in principle the use of A1C testing to diagnose diabetes. The Association will also establish a task force to explore the implications of this report including how best to implement its recommendations." Task forces take time, and so I wouldn't expect an ADA approval in the next few months -- maybe next year?

In the meantime, should physicians order an A1C test to make the diagnosis of diabetes? Well, they already are. The only issue is whether two or more elevated A1Cs (6.5 or higher) would be appropriate to nail the diagnosis if the blood glucose level was almost but not quite in the diabetic range. If a patient were to have elevated A1C values and not-quite-high BG levels, the patient is clearly at high risk, and meal planning and lifestyle interventions would be appropriate, as well as teaching the patient home glucose monitoring, and following the situation closely over the ensuing months.

[Editor's Note: By 2010, other organizations were agreeing with this concept. For example, see American Association of Clinical Endocrinologists Approves New Diagnosis for Diabetes.]