Glycated hemoglobin threshold for diabetes diagnosis questioned
MedWire News: The proposed use of a specific glycated hemoglobin (HbA1c) threshold to identify Type 2 diabetes could miss the majority of patients who are diagnosed using oral glucose tolerance tests, a study suggests.
Among individuals identified with Type 2 diabetes based on two oral glucose tolerance tests during follow up of the Finnish Diabetes Prevention Study, up to 60% would have remained undiagnosed using HbA1c levels of at least 6.5% (48 mmol/mol).
This threshold was recently proposed by an International Expert Committee as a diagnostic tool for diabetes, researchers in the above study note.
A recently published position statement by the American Diabetes Association has also, accordingly, recommended HbA1c as alternative diagnostic criteria for the diagnosis of diabetes.
However, Pia Pajunen (National Institute for Health and Welfare, Helsinki) and colleagues say: "Our results show that changing the diagnostic method for diabetes from glucose measurement to HbA1c is not without major problems."
The team compared the diagnostic efficacy of annual HbA1c measurements with the oral glucose tolerance test in 172 men and 350 women taking part in the Diabetes Prevention Study, all of whom were overweight and had impaired glucose tolerance at baseline.
HbA1c of at least 6.5% as a diagnostic criterion for Type 2 diabetes had a sensitivity of 35% for women and 47% for men compared with diagnosis based on two consecutive oral glucose tolerance tests.
The corresponding sensitivity of HbA1c of at least 6.0% (42 mmol/mol) was 67% and 68%.
Individuals with HbA1c of at least 6.5% and diabetes based on the oral glucose tolerance test were more likely to be obese and had higher fasting glucose and 2-hour glucose levels than those who had a diabetic oral glucose tolerance test but HbA1c levels lower than this.
There were no differences in the predictive performance of baseline fasting glucose, oral glucose tolerance test and HbA1c.
Reporting in the journal Diabetic Medicine, the authors say: "This study provides evidence that agreement between HbA1c and oral glucose tolerance test-based diagnosis of Type 2 diabetes in persons with impaired glucose tolerance is limited."
MedWire (www.medwire-news.md) is an independent clinical news service provided by Current Medicine Group, a trading division of Springer Healthcare Limited. © Springer Healthcare Ltd; 2010
By Anita Wilkinson