The aim of this study was to analyze the relationship between serum cholesterol level and all causes mortality in men who sustained a first myocardial infarction. The cholesterol distribution 3 months after the infarction was established. Ten annual cohorts (n = 1,204) were followed for a maximum period of 11 years. Secondary risk factors were comparable among the groups of the serum cholesterol distribution quintiles according to a multiple logistic prognostic function based on left heart failure, atrial fibrillation, breathlessness on infarction, maximum S-ASAT, relative heart size, and a history of hypertension. When all ages were analyzed together, the total mortality was higher in the upper cholesterol quintiles (P = 0.02). This association was confirmed when analyzed with Coxian adjustments for age, change in smoking habits after infarction, and the previously mentioned prognostic function. When broken down by age (less than or equal to 49, 50-59, greater than or equal to 60) and period of follow-up (3-24 and 25-84 months), the association between mortality and cholesterol quintiles was confined to patients under 50 years during the late follow-up period (P = 0.01), whereas there was no association for the other age groups.