Left ventricular (LV) ejection fraction (EF) is known to be related to prognosis after acute myocardial infarction (AMI), but its role alone and in combination with other factors in the definition of a high-risk group has not been adequately specified. Several recent multicenter studies emphasize that LVEF together with features of ventricular ectopic activity during ambulatory electrocardiography define a group at high risk for death for up to 3 years. However, these high-risk groups comprised only a small fraction of the population (less than 7.5%) and failed to include 75% or more (less than 25% specificity) of observed events. In our study, LVEF was determined close to the time of hospital discharge in 750 patients with AMI enrolled in a collaborative study. Used alone, an LVEF of less than 0.45 best defined a high-risk group (39% of the population) yielding 62% sensitivity and 64% specificity for total cardiac mortality by 1 year; it was 77% sensitive for sudden death alone. In a multivariate analysis together with other factors, LVEF was an independent predictor, but other markers of LV dysfunction entered before LVEF with similar sensitivity for total cardiac deaths, but with increased specificity (75%). When an LVEF of less than 0.45 was used together with the presence of complex arrhythmias to define a high-risk group (19% of the population), sensitivity decreased to 39% and specificity increased to 84%. Thus, LVEF is a simple and effective alternative to multivariate analysis for risk assessment after AMI.(ABSTRACT TRUNCATED AT 250 WORDS)