Coronary Artery Calcification in Type 2 Diabetes and Insulin Resistance: the Framingham Offspring Study
Risk for clinical atherosclerosis is increased in people with type 2 diabetes (DM). We assessed subclinical atherosclerosis (SCA) of the coronary arteries using electron beam computed tomography (EBCT) in subjects classified with normal or impaired glucose tolerance (NGT or IGT) or DM, or with or without insulin resistance (IR). Glucose tolerance was categorized by DM therapy (diagnosed [Dx] DM) or with an oral glucose tolerance test (IGT and new DM; 1997 ADA criteria) and IR as >90th percentile fasting insulin in subjects with NGT, in subjects of the Framingham Offspring Study attending the 5th exam (1991-95). A subset free of clinical CVD were selected for EBCT in 1998-99 from age-stratified sex-specific quintiles of the Framingham coronary heart disease risk score. Coronary SCA was quantified by Agatston scores, with the top quartile of score distributions defined as presence of SCA. We used logistic regression to calculate odds ratios (OR) and 95% confidence intervals (CI) for IGT and DM (Dx, new, or both), or IR, predicting SCA. Models were adjusted for age and sex, these and smoking, total/HDL cholesterol ratio and systolic blood pressure, and finally for IGT and DM together with IR. Of 325 subjects aged 31-73 years, 51% were men; 11.2% had IGT and 9.9% had DM (2.8% Dx DM); 14.5% had IR. Compared with NGT, subjects with IGT tended to be more likely (adjusted OR 1.5, 95% CI 0.7-3.4) and those with DM were significantly more likely (2.7, 1.2-6.1) to have coronary SCA. In age-,sex-adjusted models those with IR were more likely to have coronary SCA compared with those without IR (2.1, 1.0-4.2), but further risk factor adjustment weakened this effect. In adjusted models including IR, DM remained associated with risk for coronary SCA, (2.8, 1.2-6.7); Dx DM (6.0, 1.4-25.2) having a larger effect than new DM (2.1, 0.8-5.5). We conclude that subjects with DM have an elevated burden of coronary SCA independent of IR. IR or CVD risk factors may partially explain effects of new DM. Although sample size limits analytic power, there was a trend for subjects with subclinical glucose intolerance to have increased coronary SCA, supporting the hypothesis of a common DM/CVD risk state.