The Electrophysiological Study Versus Electrocardiographic Monitoring (ESVEM) trial randomized 486 patients with spontaneous sustained ventricular tachycardia (VT), ventricular fibrillation (VF) or unmonitored syncope, who manifested reproducibly inducible sustained ventricular arrhythmias by provocative stimulation and 10 or more premature ventricular contractions per hour on Holter monitoring, to two groups treated with pharmacotherapy guided by suppression of stimulation-inducible VT/VF or suppression of spontaneous or exercise induced ventricular arrhythmias. There was no difference over four years of follow-up in the rates of recurrence of arrhythmias, arrhythmic mortality, cardiac mortality, or mortality from any cause between the two groups of patients but more patients (77%) received pharmacotherapy in the group treated on the basis of suppression of spontaneous arrhythmias than the group treated on the basis of electrophysiological study. In this trial, rates of recurrence of arrhythmias were higher (37% at one year and 66% at four years) than generally reported, but cardiac and arrhythmia mortality were comparable or lower than generally reported. Of the seven agents tested, six were sodium channel blockers (imipramine, mexiletine, procainamide, propafenone, pirmenol, and quinidine) and the other was sotalol. Sotalol had a significantly higher rate of efficacy predictions by EPS (35%) than the others (15%) and a comparable rate by Holter monitor. Sotalol was significantly more efficacious in preventing recurrences, arrhythmic mortality, cardiac mortality, and total mortality than the other agents and it was better tolerated. Probability of successful long term therapy with a sodium channel blocker tested by electrophysiological study was low (5% at one year).(ABSTRACT TRUNCATED AT 250 WORDS)