Background: Safety and efficacy of antiarrhythmic drugs are still a great challenge in cardiology. Drug-induced EP changes in both Purkinje and ventricular cardiomyocytes have been implicated in arrhythmogenesis in vivo, and can be investigated through human-based computer models.
Objectives: To perform in silico drug trials to quantify the response of human ventricular (VC) and Purkinje (PC) virtual cells to 10 anti-arrhythmic drugs and its comparison to in vivo arrhythmic risk.
Methods: Two virtual populations of human VC (n=1213) and PC (n=668) were constructed using ex vivo EP recordings, accounting for inter-subject variability. Multichannel blocking action of 10 antiarrhythmic drugs (classes I, III, IV) at 4 concentrations was simulated using experimental IC50 values. Drug-induced changes in AP biomarkers in both human cell types were quantified and compared to reported in vivo arrhythmic risk.
Results: Both PC and VC display reduced upstroke velocity for class I (A) and prolonged AP duration (APD) for class III drugs, as well as EADs occurrence at high concentrations of strong hERG blockers (B). In Control, PC display longer APD than VC and this remains the same when applying safe drugs. However, drugs with clinical history of arrhythmias lead to smaller APD increase in PC than in VC, due to higher Na/K-pump activity. This inverts the APD gradient between VC and PC compared to control (C).
Conclusions: Simulations identify early repolarization in PC compared to VC following drug administration as a potential indicator of increased arrhythmic risk in vivo.
Abstract P511 Figure.
