Hyperinsulinemia has been shown to have strong and consistent associations with a cluster of cardiovascular risk factors. Yet the associations between hyperinsulinemia and coronary heart disease (CHD) have been weak, at best, and often inconsistent. Most previous studies have analyzed the insulin level using a radioimmunoassay method, which does not separate proinsulin from intact (true) insulin. New methods separating the two have demonstrated that proinsulin may be at least as strongly or even more strongly associated than intact insulin with a CHD-promoting risk factor profile. In this incident case-control study of a nondiabetic population, 67 cases of first acute myocardial infarction (AMI) were compared with 127 individually age- and sex-matched controls. Blood sampling was collected prior to disease outcome. Proinsulin and intact insulin levels were measured using highly sensitive two-site sandwich enzyme-linked immunosorbent assays (ELISAs). The highest quartile of proinsulin, in contrast to intact insulin, showed a greater than threefold increase in AMI compared with the lowest quartile, with an odds ratio (OR) and 95% confidence interval (CI) of 3.5 and 1.2 to 9.9, respectively. The increased risk of AMI persisted after controlling for total cholesterol, smoking status, diastolic blood pressure, and antihypertensive medication, and disappeared after additional control was used for the body mass index. High levels of proinsulin, even in a nondiabetic population, seem to be a strong and independent risk factor for AMI. The mechanism underlying the relationship may be direct via effects on fibrinolysis or, probably more plausibly, indirect, where proinsulin is a marker of an underlying metabolic disturbance.