Joint distribution of non-HDL and LDL cholesterol and coronary heart disease risk prediction among individuals with and without diabetes.

OBJECTIVE To assess coronary heart disease (CHD) risk within levels of the joint distribution of non-HDL and LDL cholesterol among individuals with and without diabetes. RESEARCH DESIGN AND METHODS We used four publicly available data sets for this pooled post hoc analysis and confined the eligible subjects to white individuals aged > or = 30 years and free of CHD at baseline (12,660 men and 6,721 women). Diabetes status was defined as either "reported by physician-diagnosed and on medication" or having a fasting glucose level > or = 126 mg/dl at the baseline examination. The primary end point was CHD death. Within diabetes categories, risk was assessed based on lipid levels (in mg/dl): non-HDL <130 and LDL <100 (group 1); non-HDL <130 and LDL > or = 100 (group 2); non-HDL > or = 130 and LDL <100 (group 3); and non-HDL > or = 130 and LDL > or = 100 (group 4). Group 1 within those without diabetes was the overall reference group. RESULTS Of the subjects studied, approximately 6% of men and 4% of women were defined as having diabetes. A total of 773 CHD deaths occurred during the average 13 years of follow-up time. A Cox proportional hazard model was used to estimate the relative risk (RR) of CHD death. Those with diabetes had a 200% higher RR than those without diabetes. In a multivariate model, CHD risk in those with diabetes did not increase with increasing LDL, whereas it did increase with increasing non-HDL: RR (95% confidence interval) for group 1: 5.7 (2.0-16.8); group 2: 5.7 (1.6-20.7); group 3: 7.2 (2.6-19.8); and group 4: 7.1 (3.7-13.6). CONCLUSIONS Non-HDL is a stronger predictor of CHD death among those with diabetes than LDL and should be given more consideration in the clinical approach to risk reduction among diabetic patients.