Restitution of action potential duration during sequential activation: a simulation study

Restitution of action potential duration (APD) is widely considered to play a key role in initialization and maintenance of ventricular fibrillation (VF). Restitution properties are quantified by using protocols such as the standard and the dynamic protocol. However, these protocols eliminate the so-called 'memory' effects by pacing for several tens of beats at a fixed cycle length before acquiring the relationship between diastolic interval (DI) and APD. In order to investigate the effects of previous activation history on APD, i.e. the memory effects, we used a feedback based pacing protocol that permitted explicit control of DI. Activation was simulated using a Luo-Rudy dynamic model. Explicit control of DI allowed evaluation of cumulative memory effects which cannot be quantified by the use of contemporary protocols. We observed that for the same DI, the APD were quite different when the prior activation history was different. Furthermore, APD changes were significantly different than those predicted by the standard protocols. Our results show, therefore, that during sequential activation, DI is not the only dominant factor that determines next APD. These results may explain the lack of correlation between pro-fibrillatory activity and restitution relationship.