Electrophysiologic studies were performed in 134 patients (87 males, mean age 59 years) with unexplained syncope. Seventy-one patients had organic heart disease (ischaemic in 50). Electrophysiologic studies revealed conduction abnormalities and tachyarrhythmias that could account for syncope in 40 patients (30%). Thirty-seven (93%) of these patients received pacing or antiarrhythmic therapy compared with 23 (24%) of the remaining 94 patients who had a negative study and received empiric therapy (P less than 0.0001). Risk of having an abnormal electrophysiologic study was greater in patients with underlying heart disease (P less than 0.05). During a mean follow-up of 22 +/- 17 months, 26 patients (19%) either had recurrent syncope (22 patients) or died (four patients) suddenly. Men had a higher incidence of recurrent syncope than women (26% vs 6%, P less than 0.005). Other clinical characteristics, electrophysiologic findings, final diagnosis and therapy at discharge were not predictive of outcome. We conclude that (1) 19% of patients investigated for syncope will have a recurrent event, (2) female gender may be an independent predictor of favourable outcome.