A point-counterpoint highlighted the continuing controversy over which is the more accurate and practical way to identify patients with the widely under-diagnosed disease: HbA1c or the traditional tests — fasting plasma glucose and the oral glucose tolerance test.
The debate at the World Congress on Controversies to Consensus in Diabetes, Obesity and Hypertension showed that, at least in this area, controversy is much closer at hand than consensus.
At issue was the statement issued last June during the American Diabetes Association’s annual meeting by a joint committee of the ADA, the International Diabetes Federation, and the European Association for the Study of Diabetes. It called for clinicians to rely on HbA1c instead of the traditional tests in diagnosing Type 2 diabetes because it gives more reproducible results and because it’s easier to administer for both clinicians and patients.
Jaakko Tuomilehto, MD, of the University of Helsinki in Finland, presented the case against HbA1c as a primary diagnostic tool in the debate.
- Thirty years of clinical experience and research using the traditional tests will be difficult or impossible to interpret against a new standard.
- Although efforts are under way to standardize HbA1c assay methods, many labs are not in compliance, making the measurement unreliable; point-of-care tests are even more variable.
- HbA1c levels can vary with patients’ ethnicity, smoking status, and other illnesses such as anemia and hemoglobin defects.
- Correlation between HbA1c and the traditional standards is questionable.
- HbA1c may not predict risk of some complications, such as coronary artery disease, as well as the traditional tests can.
- Glucose is the toxic agent in diabetes, and therefore it makes sense to focus on it for making the diagnosis.
He also argued that, because HbA1c lags changes in blood glucose levels, relying on it will delay diagnoses in many patients. “I think this is the big issue,” Tuomilehto said. “We are going in the wrong direction.”
Other research presented here supported his position. For example, researchers showed that, using the recommended HbA1c cutoff of 6.5% as the standard for a Type 2 diabetes diagnosis, more than 30% of patients were misdiagnosed.
Joo-Pin Foo, MD, and colleagues, used results from two-hour oral glucose tolerance testing (OGTT) as the reference standard (199mg/dL.cutoff) in a sample of 90 individuals of unknown diabetes status. They found that 20% of those with HbA1c values of 6.5% or higher failed to meet the OGTT standard. Another 11% who qualified for a Type 2 diabetes diagnosis by the OGTT results had HbA1c values below 6.5%.
Women were particularly vulnerable to false-positive results with the HbA1c test, and there was a trend toward increased false positives among those with higher body mass index, Foo and colleagues reported. Older individuals and men tended to be more prone to false negatives.
“Demographic and metabolic characteristics may have a role in affecting the accuracy of HbA1c,” they concluded.
Tuomilehto made a similar point, suggesting that HbA1c can vary according to numerous factors that are not yet well understood. The expert panel’s recommendation, he declared, ” is completely premature.” He also noted that the World Health Organization is now reviewing its own position statement on Type 2 diagnosis management. Its current version, issued in 2006, states that HbA1c is “not suitable” for diagnostic purposes.
As a member of the committee addressing diagnostic issues, Tuomilehto said the new document was still being finalized. But he assured attendees that “HbA1c is not going to be recommended.”
On hand to defend HbA1c-based diagnosis was a member of the expert panel, Enzo Bonora, MD, PhD, of the University of Verona in Italy. He said when he first joined the committee, he was dead-set against the idea. But after seeing the evidence, he said, “I changed my mind.”
Bonora ticked off several rationales for the change:
- HbA1c is a better measure of chronic glycemic activity than snapshot glucose-level testing.
- Most diabetic complications are predicted as well by HbA1c compared with the traditional tests, better in some cases.
- HbA1c has lower inherent biological variability compared with fasting plasma glucose.
- HbA1c captures individual susceptibility to glycation, which determines the risk of complications from a given level of glycemia.
- No fasting is required and the results are not affected by stress, recent exercise, or short-term changes in diet.
- Assays for HbA1c are easier to perform, though more costly, and are not subject to some of the problems associated with glucose measurement.
Bonora argued that results of glucose testing in a research setting may not translate well to clinical practice. Glucose disappears from blood samples at a rate of 5% to 7% per hour because blood contains elements that break it down.
As a result, if a blood sample is not assayed promptly, falsely low readings will result. The problem occurs even when blood samples are collected in tubes containing anti-glycolytic agents, he said. That and other variability issues with glucose testing mean that the lack of correlation with HbA1c does not necessarily undermine the utility of HbA1c, he suggested.
“There is no gold standard” for diagnosing Type 2 diabetes, Bonora said. Another problem, he said, is that fasting plasma glucose or OGTT results must be confirmed with a second test. This requirement doubles the hassle involved with these tests for clinicians and patients.
He acknowledged that HbA1c is not reliable under some circumstances, but noted that, in those cases, the traditional tests are still available.
- Explain to interested patients that an international panel assembled by the world’s three leading diabetes organizations has recommended that an HbA1c level of 6.5% or higher is sufficient to diagnose a person with Type 2 diabetes.
World Congress on Controversies to Consensus in Diabetes, Obesity and Hypertension; Foo J-P, et al “Demographic and metabolic profile may influence specificity of HbA1c in diagnosing diabetes mellitus” CODHy 2010; Poster 1.
Reprinted from http://www.diabetesincontrol.com/index.php?option=com_content&view=article&id=9375&catid=1&Itemid=17