Lengthening of QTc is the usual signal to indicate torsadogenic potential of a therapeutic agent. The ICH S7B guideline recommends that new chemical entities should be assessed for potential of delayed ventricular repolarization in animal models. The aim of this study was to determine a feasibility of using isolated failing heart rabbit to assess the QT-lengthening drugs in comparison with their effects on isolated normal heart rabbits. Heart failure was induced by ligation of the left anterior descending and descending branch of left circumflex coronary arteries. One month after ligation, all rabbits were anesthetized and the hearts were removed quickly, and they were perfused with the oxygenated Krebs-Henseleit solution to which escalating concentrations of QT-lengthening compounds were added. RR, QT, and QTc(F) were not significantly different, at rest, between failing and normal hearts. During baseline, dP/dtmax was lower and dP/dtmin was higher for failing hearts than for normals. In responses to all three QT-lengthening compounds, RR, QT and QTc(F) lengthened similarly in a dose-response manner in both the failing and normal hearts. Neither the failing nor the normal hearts developed fatal arrhythmias, torsades de pointes. Langendorff preparations of failing hearts are as good as normal isolated hearts and can be use to assess the potential of delayed ventricular repolarization of test articles.