Aims/hypothesis: Our aim was to investigate the predictive power of a panel of variables in glucose and insulin metabolism for the incidence of stroke or transient ischaemic attacks (TIA). We hypothesised that proinsulin and insulin resistance contributes to an increase of risk for fatal and non-fatal stroke/TIA, independently of diabetes and established risk factors.
Methods: The study is based on the Uppsala Longitudinal Study of Adult Men cohort. The examinations were performed at age 70 years.
Results: In 1,151 men free from stroke at baseline, 150 developed stroke or TIA during a median follow-up of 8.8 years. In unadjusted Cox proportional hazards analyses, a 1 SD increase of a predictor variable was associated with an increased risk for stroke/TIA, e.g. plasma insulin (HR 1.19, 95% CI 1.01-1.40), fasting intact proinsulin (HR 1.28, 95% CI 1.09-1.49); whereas a 1 SD increase in insulin sensitivity measured by the euglycaemic insulin clamp method decreased the risk for stroke/TIA (HR 0.81, 95% CI 0.68-0.96). The predictive values of fasting intact proinsulin and insulin sensitivity endured but not that of plasma insulin when adjusting for diabetes. In models adjusting for diabetes, hypertension, atrial fibrillation, electrocardiographic left ventricular hypertrophy, serum cholesterol and smoking, proinsulin remained as a significant predictor of later stroke/TIA (HR 1.22, 95% CI 1.00-1.48) whereas clamp insulin sensitivity did not (HR 0.87, 95% CI 0.71-1.07).
Conclusions/interpretation: Fasting intact proinsulin level and insulin sensitivity at clamp predicted subsequent fatal and non-fatal stroke/TIA, independently of diabetes in elderly men whereas fasting insulin did not.