Antiviral therapies for treatment of COVID-19 may be associated with significant proarrhythmic potential. In the present study, the potential cardiotoxic side effects of these therapies were evaluated using a Langendorff model of the isolated rabbit heart. 51 hearts of female rabbits were retrogradely perfused, employing a Langendorff-setup. Eight catheters were placed endo- and epicardially to perform an electrophysiology study, thus obtaining cycle length-dependent action potential duration at 90% of repolarization (APD90), QT intervals and dispersion of repolarization. After generating baseline data, the hearts were assigned to four groups: In group 1 (HXC), hearts were treated with 1 µM hydroxychloroquine. Thereafter, 3 µM hydroxychloroquine were infused additionally. Group 2 (HXC + AZI) was perfused with 3 µM hydroxychloroquine followed by 150 µM azithromycin. In group 3 (LOP) the hearts were perfused with 3 µM lopinavir followed by 5 µM and 10 µM lopinavir. Group 4 (REM) was perfused with 1 µM remdesivir followed by 5 µM and 10 µM remdesivir. Hydroxychloroquine- and azithromycin-based therapies have a significant proarrhythmic potential mediated by action potential prolongation and an increase in dispersion. Lopinavir and remdesivir showed overall significantly less pronounced changes in electrophysiology. In accordance with the reported bradycardic events under remdesivir, it significantly reduced the rate of the ventricular escape rhythm.