July 25th, 2011
Adding HbA1c Measurements Improves CV Risk Prediction in Diabetics
Current risk prediction models classify diabetes as equivalent to established cardiovascular disease. Now, a new report from the Women’s Health Study and the Physician’s Health Study II suggests that adding HbA1c measurements to the model can improve risk prediction and lead to downward classification of some diabetics.
In a paper published in Archives of Internal Medicine, Nina Paynter and colleagues analyzed data from 24,674 women and 11,280 men (685 women and 563 men were diabetic at baseline). Cardiovascular events occurred in 125 diabetic women and 170 diabetic men. In women, including HbA1c levels improved the C statistic by 0.177 (P<0.001) and improved the accuracy of the risk-category assignment. The results were less dramatic in men, but the C statistic improved by 0.039 (P<0.02), and the accuracy of the risk-category assignment also improved.
The authors note that the presence of diabetes by itself did not indicate a 10-year CV risk greater than 20% for all patients, implying that their findings “are consistent with previously published studies suggesting that not all diabetic patients are at high risk of future vascular events.” They point out that their “results may be particularly helpful in light of current discussion around treatment choices for diabetic patients for prevention of CVD, including use of statins and aspirin.”
In an accompanying comment, Mark Pletcher writes that “there is little doubt that the primary findings” of the study are true, and that using HbA1c measurements would allow downward reclassification of some diabetic patients. However, he questions whether this will result in actual health benefits. Noting that diabetics have a substantially elevated lifetime risk for CV events, he asks: “Will it really benefit diabetic persons with low or medium short-term risk to withhold statins?”