Can Phase-2 Early Afterdepolarizations Propagate in Cardiac Tissue? Insights from Multiple Action Potential Models

Early afterdepolarizations (EADs) have been widely observed in cardiac myocytes and tissue, and are considered as arrhythmia triggers in the form of premature ventricular complexes (PVCs). The mechanism is that EADs overcome the source-sink effect to propagate into PVCs. An early experimental study in Purkinje fibers by Damiano and Rosen (1984) showed that phase-2 EADs could not propagate into PVCs while phase-3 EADs could. However, EADs observed in isolated myocytes are almost exclusively phase-2 EADs, raising a fundamental question on whether phase-2 EADs are indeed responsible for the PVCs. Here we use computer simulations of multiple action potential models to address this question. We show that in all the action potential models used in this study, phase-2 EADs are either suppressed by the repolarization gradient or confined to long action potential regions of heterogeneous tissue due to the source-sink effect. However, PVCs can be generated in these models with a substantially enhanced L-type calcium current (at least two-fold increase), which is much higher than that required for the genesis of phase-2 EADs. Moreover, PVCs are enhanced by increasing the sink effect (or increased repolarization gradient), contrary to the sour-sink requirement for EAD propagation. We show that PVCs are not a result of the propagation of phase-2 EADs. Instead, they are caused by the combination of enhanced L-type calcium current, and tissue-scale dynamical instabilities originated from the steep repolarization gradient regions.