Droperidol and haloperidol have demonstrated efficacy and safety in the treatment of acute delirium in critically ill patients. We conducted MEDLINE and manual searches of literature published from 1966–1996 to identify articles describing conduction disturbances associated with the drugs. The objectives were to describe the proposed mechanisms of acquired long QTc interval syndrome and torsades de pointes, and to recommend how critically ill patients receiving these agents should be monitored. We found 11 published reports of conduction disturbances associated with intravenous administration of droperidol or haloperidol. The majority of cases occurred in critically ill patients prescribed more than 50 mg/24 hours of either agent. Of the 18 patients described, 13 (72%) had a history of cardiovascular disease. Based on the small number of available case reports, it seems reasonable to suggest that the incidence of adverse cardiovascular effects due to droperidol and haloperidol is small. The mechanism of butyrophenone-induced QTc interval prolongation is not known, but is presumed to involve abnormal ventricular repolarization and the development of early after-depolarizations. Before initiating therapy with droperidol or haloperidol in critically ill patients, a baseline QTc interval and serum magnesium and potassium concentrations should be measured. If the baseline QTc interval is 440 msec or longer, and they are receiving other drugs that may prolong the QTc interval or they have electrolyte disturbances, a butyrophenone antipsychotic should be prescribed with caution. All critically ill patients receiving droperidol or haloperidol should undergo electrocardiogram monitoring and QTc interval measurement; special attention should be given to those receiving doses greater than 50 mg/24 hours, as these patients appear to be at greatest risk for development of conduction disturbances. Based on the currently available literature, in any critically ill patient receiving droperidol or haloperidol therapy whose QTc interval lengthens by 25% or more over baseline, therapy should be discontinued or the dosage reduced.