Using data from the Coronary Artery Surgery Study (CASS) registry, we evaluated the relationship between cholesterol levels measured at enrollment and the following events: all-cause mortality, cardiac death, fatal myocardial infarction (MI), and nonfatal MI. Only patients with a significant coronary artery disease (at least one lesion with stenosis > or = 50%) were considered for this study. Results presented for mortality are for a period of up to 11.5 years and those for MI are for a maximum of 8 years of follow-up. Analyses were performed for each type of event and for each subgroup: women (n = 1861) and men (n = 10,941) under age 65, and women (n = 426) and men (n = 1144) age 65 or older. After adjusting for important covariates, cholesterol level was not associated with cardiac or all-cause mortality. No relationship between cholesterol level and fatal or nonfatal MI could be demonstrated except for men under age 65. However, in this subgroup the risk of MI was highest for those with low or middle cholesterol levels. The data show that in patients with angiographically determined coronary artery disease, cholesterol level is not a statistically significant risk factor for death or MI over the follow-up period in CASS.