The objective of this study was to examine whether fasting serum insulin is a predictor of coronary heart disease in high-risk US men, and whether any such predictive role explains the enhanced cardiovascular risk seen in subjects with the apolipoprotein (Apo) E 3/2 phenotype. This was a nested case-control study of participants in the Multiple Risk Factor Intervention Trial. Ninety-four subjects who died from coronary heart disease (post-trial follow-up) and 114 case patients with myocardial infarction (during trial) were compared to control subjects (n = 414) matched (1:2) by age, center, randomization date, and intervention group. Overall, fasting serum insulin at baseline was not associated with case-control status. (Means for cases versus controls: 16.8 and 16.6 microU/mL), although serum insulin showed significant correlations with low-density-lipoprotein cholesterol, triglycerides, and uric acid. When stratified by the three Apo E phenotypes, 3/2, 3/3, 3/4, a significant association of fasting insulin with case-control status was seen for Apo E 3/2 individuals (19.9 versus 14.5 microU/mL; P = 0.02) but not for those with the other two phenotypes. Though fasting insulin is not a risk factor overall in this high-risk male population, it appears to contribute to cardiovascular risk in those with the Apo E 3/2 phenotype but does not explain the increased risk seen in these subjects. This new finding, if confirmed, may throw further light on the role of insulin in atherosclerosis.