Oscillations of cardiac wave length and proarrhythmia

Drug-induced action potential duration (APD) prolongation was first proposed to be antiarrhythmic, but is now widely presumed to be torsadogenic. To elucidate this paradox, we tested the effect of APD upon liability for torsade de pointes. In addition, torsadogenicity is commonly associated with disturbances of repolarization, but at least in theory, it could also result from disturbances of conduction. These possibilities were tested in female rabbit hearts. Dofetilide, ATX II, and sodium channel blockers that did not prolong the action potential duration were used to modulate the APD and induce disturbances of conduction and disturbances of repolarization. Torsadogenicity could be induced by dofetilide and ATX II starting at normal APD (210 ms), reaching a peak incidence around a doubling of APD (400 to 450 ms), to then sharply decline with further APD prolongation, until torsade de pointes disappeared above 725 ms. Early afterdepolarizations (EAD) were regular triggers for torsade de pointes; while most of the EADs occurred in the plateau range, their incidence declined with repolarization but their potential for torsadogenicity increased. Sodium channel blockers that shorten the APD, even when devoid of hERG blocking properties, can yield torsade de pointes. Torsade de pointes can occur at normal, prolonged, and shortened APD, so that QT prolongation is an incomplete predictor of torsadogenicity. Torsade de pointes can result not only from disturbances of repolarization but also from disturbances of conduction.