Previous studies have reached conflicting conclusions about whether cardiac arrest due to ventricular tachycardia (VT) or ventricular fibrillation (VF) in acute myocardial infarction (AMI) is of long-term prognostic significance. The mortality rate in 849 patients with confirmed AMI was analyzed. The mortality rate during the initial hospitalization was higher for patients in whom VT/VF occurred (27% vs 7%, p less than 0.001). The in-hospital mortality rate for patients with primary VT/VF, that is, VT/VF occurring in the absence of hypotension or heart failure, was similar to that of patients who did not have VT/VF (8% vs 7%, difference not significant), whereas the rate for patients with secondary VT/VF was significantly greater than that for patients with no VT/VF (51% vs 7%, p less than 0.001). The timing of occurrence of VT/VF also influenced mortality: Patients in whom VT/VF occurred more than 72 hours after admission had a higher in-hospital mortality rate than did patients in whom it occurred within 72 hours (57% vs 20%, p less than 0.05). All cases of primary VT/VF occurred within the first 72 hours of admission. The long-term mortality rate for hospital survivors was not significantly different for patients who had had VT/VF during acute infarction compared with those who had not (19% vs 21%) (mean follow-up 32 months). Thus, cardiac arrest due to ventricular tachyarrhythmia was associated with a higher in-hospital mortality rate but was not a prognostic factor among hospital survivors. Patients resuscitated from primary VT/VF, which characteristically occurs early after AMI, do not have an adverse prognosis.