Forty-four patients with primary ventricular fibrillation or recurrent ventricular tachycardia were stabilized on an antiarrhythmic drug regimen before electrophysiologic study and 24 to 72 hours of ambulatory electrocardiographic monitoring were performed. The long-term predictive value of these two tests was then compared retrospectively for a 12 to 32 month (mean 18) follow-up period, during which all patients continued receiving the same antiarrhythmic drug regimen. Electrophysiologic testing induced ventricular tachycardia (greater than or equal to 3 beats) in 26 patients; 23 had a poor clinical outcome (positive predictive value 88%), defined as sudden death or sustained ventricular tachycardia. In 18 patients with a negative electrophysiologic test, only 1 had a poor clinical outcome (negative predictive value 94%). Ambulatory electrocardiographic monitoring accurately predicted outcome in 7 of 10 patients with a positive recording (positive predictive value 70%), defined as three or more consecutive ventricular extrasystoles, and in 17 of 34 patients with negative ambulatory monitor recordings (negative predictive value 50%). The long-term predictive accuracy of the electrophysiologic study was significantly higher than that of the ambulatory electrocardiographic monitor (p less than 0.001). Electrophysiologic studies offer advantages over ambulatory electrocardiographic monitoring in this high risk patient group, providing a high degree of accuracy in predicting the long-term clinical response to antiarrhythmic drugs for at least 18 months.