BACKGROUND: Electroconvulsive therapy (ECT) is performed under general anesthesia (GA). GA could reduce the vigorous hemodynamic response including hypertension and tachycardia during ECT. This may be beneficial in patients with cardiovascular diseases. On the other hand, however, many intravenous and inhalational anesthetic agents potentially suppress the seizure and shorten the duration. The aim of this study is to elucidate whether propofol is superior to sevoflurane in maintaining the duration of electroencephalographic (EEG) seizure evoked by Thymatron System while stabilizing hemodynamic responses. METHODS: Thirteen patients (male:female = 3:10) were included in this cross-over study. The patients underwent 54 ECT sessions in total. The sessions were randomized into two groups. In the propofol group (P group), the patients received intravenous propofol 1 mg x kg(-1) on induction of anesthesia, whereas those in the sevoflurane group (S group) were induced with sevoflurane at 5% inspired concentration. In either group, after loss of consciousness, the patients received 1 mg x kg(-1) of suxamethonium. When muscle fasciculation faded away, the electrostimulus was delivered. The duration of EEG seizure, heart rate (HR) and noninvasive mean arterial blood pressure (MAP) were recorded. RESULTS: The duration of EEG seizure was longer in P group (34.6 +/- 15.5 sec versus 23.6 +/- 12.0 sec in S group, P = 0.006). There was no significant difference in hemodynamic parameters (MAP, HR) between the two groups. CONCLUSIONS: The duration of EEG seizure in ECT was significantly longer with propofol anesthesia than sevoflurane anesthesia. Propofol may provide more benefits than sevoflurane as the anesthetic for ECT.