This prospective study sought to determine whether programmed ventricular stimulation before hospital dismissal in patients who had received a loading dose of amiodarone would identify those at risk for recurrent ventricular arrhythmias. Between January 1985 and January 1989, 64 patients (55 men and 9 women; mean age, 64 years) with a history of sustained ventricular tachycardia (VT) or ventricular fibrillation were referred to our institution for electrophysiologic testing. Of these patients, 52 had coronary artery disease, 11 had dilated cardiomyopathy, and 1 had hypertrophic cardiomyopathy. Of the 64 patients, 47 had baseline tests while no drugs were administered and repeated electrophysiologic testing after 10 days of amiodarone loading (1.2 g/day). The other 17 patients had no baseline study because of instability of their arrhythmias but underwent electrophysiologic testing after amiodarone loading. Follow-up ranged from 7 to 1,536 days (mean, 652 days). During the follow-up period, recurrent arrhythmias were detected in 22 patients. Of the 64 patients, 14 had suppression of VT. Of 43 patients in whom the cycle lengths of VT were determined both at baseline and after amiodarone therapy, 20 had an increase of 100 ms or more, and 23 had no substantial change. The mean ejection fraction was 31%. Of a total of 16 deaths in the series, 8 were sudden. Suppression of VT during amiodarone therapy suggested a lower rate of fatal and nonfatal recurrent arrhythmias, but the difference was not statistically significant. An increase in the cycle length of VT did not predict an improved outcome. The age of the patient and the presence of a left ventricular aneurysm were slightly predictive of mortality.(ABSTRACT TRUNCATED AT 250 WORDS)