The relationship of serum lipoprotein and apolipoprotein concentrations to angiographically determined coronary artery disease was investigated in 105 consecutive male survivors of myocardial infarction under the age of 45. Concentrations and composition of lipoproteins, lipid indexes, and nonlipid risk factors (tobacco consumption, hypertension, reduced glucose tolerance, and obesity) were related to a recently developed scoring system for semiquantitative estimation of diffuse coronary atheromatosis, as well as to the number and severity of significant coronary artery stenoses. The concentrations of cholesterol in very low-density lipoprotein (VLDL), low-density lipoprotein (LDL), and high-density lipoprotein (HDL), in combination with serum triglyceride or VLDL triglyceride level, comprised the best set of independent discriminatory lipid variables between patients and control subjects. In the patients, LDL cholesterol and apolipoprotein B levels showed strong relationships to the extent and severity of coronary atheromatosis but not to the number and severity of distinct coronary stenoses. HDL2 cholesterol concentration correlated inversely with the coronary atheromatosis score, whereas other variables reflecting HDL concentration and composition or VLDL lipids were not independently related to any of the coronary scores. The LDL triglyceride level, an index of intermediate-density lipoprotein (IDL) accumulation, was significantly correlated to the coronary atheromatosis score in univariate analysis. Nonlipid risk factors were correlated neither to coronary atheromatosis nor to severity of stenoses. Stepwise multiple regression analyses of data adjusted for age, cumulative tobacco consumption, and weight indicated that 18% of the variation in the coronary atheromatosis score could be accounted for by levels of apolipoprotein B. Addition of other lipoprotein variables or the nonlipid variables hypertension and glucose tolerance did not significantly increase the value of R2. When ratios of lipoprotein lipids and apolipoproteins were included in the regression model, the highest multiple correlation coefficient was obtained with the LDL/HDL cholesterol ratio alone (R2 = .22). The present data demonstrate the importance of elevated LDL cholesterol and apolipoprotein B concentrations for the development of coronary atheromatosis in young male survivors of myocardial infarction. The lack of correlations between the levels of lipoprotein lipids and serum apolipoproteins and the severity of coronary stenoses suggests that mechanisms other than disturbances of lipoprotein metabolism may be involved in the progression of more advanced coronary lesions.