HbA Doesn’t Aid Risk Prediction in Nondiabetic Patients

Glycated hemoglobin (HbA1c) does not provide additional information on cardiovascular risk in patients without diabetes or cardiovascular disease (CVD), suggests a meta-analysis in the Journal of the American Medical Association.

The meta-analysis included individual-level data on 294,998 participants, all initially without known diabetes or CVD, from 73 prospective cohort studies. Glycated hemoglobin level was evaluated as a predictor of initial cardiovascular events in patients in different 10-year cardiovascular risk categories: low, less than 5 percent; intermediate, 5 percent to less than 7.5 percent; or high, 7.5 percent or greater. The analysis included measures of risk discrimination and reclassification.

The data included 20,840 fatal and nonfatal CVD events—13,237 coronary heart disease and 7603 stroke outcomes—at a median follow-up time of 9.9 years. After adjustment for some conventional cardiovascular risk factors, the slope of the association between HbA1c and CVD risk was approximately J-shaped. There was little effect of further adjustment for total cholesterol and triglyceride levels or estimated GFR. The association was attenuated by adjustment for high-density lipoprotein cholesterol and C-reactive protein.

Risk discrimination was little improved by the addition of HbA1c data to a model incorporating conventional cardiovascular risk factors, and net reclassification improvement was not improved at all. The results were similar in all 10-year CVD risk categories. The additional risk information from HbA1c was similar to or greater than that provided by fasting, random, or postload plasma glucose levels.

Higher levels of glycemia have been linked to increased CVD risk, suggesting a role of HbA1c for cardiovascular risk assessment in asymptomatic, nondiabetic adults. However, the new analysis showed limited value of adding HbA1c to conventional models for predicting initial CVD events. The authors call for further studies to evaluate the significance of the “consistent J-shaped associations between various glycemia measures and CVD incidence” [The Emerging Risk Factors Collaboration. Glycated hemoglobin measurement and prediction of cardiovascular disease. JAMA 2014; 311:1225–1233].