Insulin resistance, the metabolic syndrome, and risk of incident cardiovascular disease: a population-based study.

OBJECTIVES: The goal was to clarify if insulin resistance (IR) would predict cardiovascular disease (CVD) independent of the metabolic syndrome (MetSyn). BACKGROUND: Although the cause of MetSyn is not well defined, IR has been proposed to be an important cause. Only a small number of population-based studies have sought to clarify if IR predicts CVD independent of MetSyn. METHODS: This was a prospective Danish population-based study of 2,493 men and women, age 41 to 72 years, without major CVD at baseline. We defined MetSyn according to both the International Diabetes Foundation (IDF) and the National Cholesterol Education Program (NCEP) criteria, and we quantified IR by the homeostasis model assessment (HOMA-IR). Prevalence of MetSyn was 21% according to IDF criteria and 16% according to NCEP criteria. Accordingly, we defined IDF-HOMA-IR as belonging to the highest 21% of the HOMA-IR distribution, and NCEP-HOMA-IR as belonging to the highest 16% of the HOMA-IR distribution. RESULTS: Over a median follow-up of 9.4 years, the incidence of CV end points (CV death, nonfatal ischemic heart disease, and nonfatal stroke) amounted to 233 cases. In proportional hazard models, adjusting for age, gender, smoking, and low-density lipoprotein cholesterol, and with IDF-HOMA-IR and IDF-MetSyn included in the same model, the relative risk of an end point was 1.67 (95% confidence interval [CI] 1.22 to 2.29) for IDF-HOMA-IR and 1.16 (95% CI 0.84 to 1.60) for IDF-MetSyn. The corresponding figures for NCEP-HOMA-IR and NCEP-MetSyn included in the same model were 1.49 (95% CI 1.07 to 2.07) and 1.56 (95% CI 1.12 to 2.17). CONCLUSIONS: In this Danish study, both HOMA-IR and NCEP-MetSyn were independent predictors of incident CVD.