Ten-hour electrocardiographic (ECG) monitoring was performed on 289 survivors of myocardial infarction (MI) at 2 weeks, at monthly intervals for 6 months and then at 9 and 12 months after MI. Four hundred thirty episodes of ventricular runs were recorded on 88 patients (30%). The clinical features during the acute phase of MI of these 88 patients were evaluated prospectively and compared with those of patients who did not have runs. Both groups were similar with respect to age, sex, and preexisting coronary risk factors. However, patients with runs had higher peak serum enzyme levels and a higher prevalence of congestive heart failure, cardiomegaly and left ventricular hypertrophy. They also had more frequent atrial premature complexes, premature ventricular complexes at a rate of 6/minute or more, and ventricular conduction defects during the acute infarction. The presence of runs within 3 months after the MI was predictive of the presence of runs during the period 4-12 months after MI. Using a multivariate logistic analysis patients could be divided into quartiles of risk for posthospital runs on the basis of features noted during the acute phase. The prevalence of runs ranged from 4% in the lowest quartile to 49% in the highest one. Although the rate of sudden death was not different in each quartile of risk of having ventricular runs, patients in the highest quartile had a significantly higher mortality rate (16.7%) than those in the lowest one (5.6%). We conclude that severe cardiac disease manifested by poor left ventricular function, high serum enzyme levels, and certain types of cardiac arrhythmias during acute MI are associated with an increased prevalence of ventricular runs during the posthospital phase. Our study suggests that the higher mortality rate in the highest quartile of risk of having ventricular runs is related to severe cardiac disease rather than to the presence of the runs.