Despite the fact that a role for thyrotropin-releasing hormone (TRH) in seizure modulation has been consistently hypothesized, the exact nature of this role remains unclear. In this study, we investigated the effects of TRH administration on seizure threshold and seizure duration in 13 depressed inpatients undergoing electroconvulsive therapy (ECT). In a balanced order crossover design, an intravenous bolus of 0.4 mg TRH or placebo was administered immediately before anesthesia, during the first two sessions, in a series of bilateral ECT. In both of these sessions, a threshold titration procedure was applied by using gradual increments of the electrical charge delivered until seizure elicitation, a procedure that has been safely used in the past. Seizure threshold was defined as the lowest energy level required for induction of a grand mal seizure, by use of this titration procedure. Seizure duration was estimated both by simultaneous EEG recording and by the cuff method. Results showed that neither seizure threshold, nor seizure duration (either by cuff or by EEG) differed between the TRH and the placebo conditions, regardless of the order in which TRH or placebo were administered in the two ECT sessions. This was the case regardless of whether the patients had at baseline a blunted TSH response to TRH or not. Our findings do not support a role for TRH on seizure modulation, at least when TRH is administered exogenously. Such an effect, if it exists, could be obscured, however, by several factors, including pharmacokinetics.