The study group consisted of 26 patients with a history of documented Torsade de Pointes (TdP) who were divides into 3 groups according to the causes of TdP. Group I consisted of 5 patients with congenital long QT syndrome. Group II consisted of 15 patients with TdP caused by antiarrhythmic drugs. Group III consisted of 6 patients with TdP caused by bradycardia resulting from third degree atrioventricular block. The QT interval was determined from a 12-lead electrocardiogram. Monophasic Action Potential (MAP) was recorded by a 6 F USCI electrode catheter. Isoproterenol infusion resulted in TU abnormality in all patients in Group I and induced a hump at phase 3 slope of MAP in all 3 patients tested. The QT interval change before and after IA administration was significantly larger in Group II patients compared to those without TdP (0.132 +/- 0.062 vs 0.029 +/- 0.31 sec, less than 0.005). Injection of 100 mg. of disopyramide in 2 patients in Group II resulted a hump at phase 3 slope of the MAP in both of them. The QT prolongation associated with decreasing the pacing rate from 70 to 50/min was significantly larger in patients with Group III compared to patients with bradycardia but without TdP (0.02 +/- 0.04 vs 0.07 +/- 0.05 sec, p less than 0.005). The results suggests: 1) different approaches are necessary for evaluation of TU abnormalities in patients with TdP according to the causes of TdP, 2) MAP might be a useful method for evaluating TU abnormality in patients with TdP.