The incidence of atherosclerotic coronary artery disease (CAD) was assessed in 4,559 male participants (aged 40 to 64 years) from the Prospective Cardiovascular Münster study, over a 6-year follow-up period. During this time, 186 study participants developed atherosclerotic CAD (134 definite nonfatal myocardial infarctions and 52 definite atherosclerotic CAD deaths including 21 sudden cardiac deaths and 31 fatal myocardial infarctions). Univariate analysis revealed a significant association between the incidence of atherosclerotic CAD and high-density lipoprotein (HDL) cholesterol (p less than 0.001) and triglycerides (p less than 0.001). The relation to HDL cholesterol remained after adjustment for other risk factors. By contrast, the relation between the incidence of atherosclerotic CAD and triglycerides disappeared if, in a multivariate analysis by means of a multiple logistic function, cholesterol or HDL cholesterol were taken into account. However, the data suggested that hypertriglyceridemia is a powerful additional coronary risk factor, when excessive triglycerides coincide with a high ratio of plasma low-density lipoprotein cholesterol to HDL cholesterol (greater than 5.0). Even though the prevalence of this subgroup was only 4.3%, it included a quarter of all atherosclerotic CAD events observed.