Prolongation of the QTc interval of the electrocardiogram (ECG) is used as a surrogate marker for a rare, but life threatening, ventricular arrhythmia known as torsades de pointes (TdP). However, the clear link between QTc prolongation and the arrhythmogenic risk has not been demonstrated unequivocally. In the present review article, we examine (a) the current understanding of electrophysiological and pharmacological mechanisms linking changes in action potential (AP) properties with proarrhythmia and (b) the value of the isolated, paced Langendorff-perfused female rabbit heart model (Screenit system) in predicting the torsadogenic potential of drugs in man. The Screenit system records monophasic action potentials (MAPs) from which the following parameters are evaluated: action potential duration (APD), conduction, instability (indicative of beat to beat APD variability), triangulation (indicative of changes of Phase 3 repolarization), and reverse-use dependency (indicating that the APD is more prolonged at slow heart rates). So far, over 16,000 experiments have been conducted, including approximately 300 dedicated tests to evaluate, in a blinded manner, approximately 70 clinically used drugs. The drugs tested covered a wide range of compounds from various pharmacological and chemical classes with clinical torsadogenic propensity, as well as drugs without the latter effect in clinical settings. Overall, the Screenit system and its associated analysis classified the drugs based on their effects on AP morphology and conduction and additionally identified, in a qualitative manner, drugs clinically associated with TdP. Such an identification is based on the triangulation, reverse-use dependency, and instability of the AP, as well as on the direct indexes of proarrhythmia such as early afterdepolarization (EADs), ventricular tachycardia (VT), and ventricular fibrillation (VF). Overall, drugs that readily induce arrhythmia and/or EADs and/or causes triangulation, reverse-use dependency, and/or instability and/or a chaotic Poincaré plot in a range of concentrations likely to be achieved in man is likely to cause TdP in man, eventually. Only if none of these elements is present, at concentrations well exceeding the free therapeutic plasma concentration, can one expect that the drug will probably be devoid of torsadonenicity. Therefore, this in vitro model provides detailed information on the overall profile of drug-induced electrophysiological effects. In combination with other in vitro and in vivo repolarization assays and with pharmacokinetic data in man, it is a valuable tool to establish an integrated cardiovascular risk assessment of pharmaceutical compounds.