Contribution of fasting hyperinsulinemia to prediction of atherosclerotic cardiovascular disease status in 293 hyperlipidemic patients.

In a cross-sectional study of 293 nondiabetic patients (169 men and 124 women) referred for the diagnosis and treatment of hyperlipidemia, our specific aim was to determine whether fasting serum insulin independently contributes to the prediction of atherosclerotic cardiovascular disease (ASCVD) status. Of the 169 men and 124 women, 65 (38%) and 44 (35%), respectively, had ASCVD with at least one of the following: unstable angina, myocardial infarction (MI), angioplasty, coronary artery bypass graft (CABG), claudication, transient ischemic attack, or ischemic stroke. In addition, 42% and 38% had fasting hyperinsulinemia (> or =20 microU/mL). Fasting serum insulin of 20 microU/mL or higher was very common in women (59% to 100%) and men (67% to 88%) when hypertension, obesity, top-decile triglyceride (TG), and bottom-decile high-density lipoprotein cholesterol (HDLC) were concurrent in various combinations. ASCVD events (present or absent) were dependent variables in a stepwise logistic regression model with explanatory variables including age, gender, race, hypertension, cigarette smoking, ASCVD in first-degree relatives at age 55 years or less, Quetelet Index, fasting serum insulin, a gender x insulin interaction term, anticardiolipin antibodies (ACLAs) IgG and IgM, total cholesterol to HDLC ratio, TG, lipoprotein(a) [Lp(a)], and homocysteine. The risk odds ratio for ASCVD (109 events and 184 nonevents) for subjects with top-decile insulin (vthe bottom nine deciles) was 3.71, with a 95% confidence interval (CI) of 1.62 to 8.9 (P = .002). For patients with MI and/or CABG and/or angioplasty ([MCA] 63 events and 184 nonevents), the risk odds ratio for top-decile insulin versus the rest was 5.07 (95% CI, 1.83 to 14.8, P = .002). For patients with MCA at age 55 or less, the gender x insulin interaction term was significant (P = .0004); the risk odds ratio for men with top-decile insulin was 13.28 (95% CI, 3.82 to 51.65, P = .0001). Hyperinsulinemia is very common in nondiabetic hyperlipidemic women and men. Fasting serum insulin, a crude, simple, practical, and inexpensive measure, independently and uniformly improved the prediction of ASCVD status beyond traditional risk factors and lipid variables in patients referred for treatment of hyperlipidemia.