Introduction: Reliable detection of drug-induced proarrhythmia, especially the potential for polymorphic ventricular tachycardia, is of great importance in the development of new compounds that are safe for the heart and was evaluated in a blinded study.
Methods and results: In 142 female rabbits, the monophasic action potential was used to determine intraventricular conduction, action potential duration (APD), triangulation (APD30 to APD90), reverse use-dependence, instability and presence of chaotic behavior, early afterdepolarizations, torsades de pointes (TdP), and ventricular fibrillation. In addition, 31 coded drugs were tested in a blinded fashion in another 150 hearts. Prototype cardiovascular agents [quinidine (IA), lidocaine (IB), flecainide (IC), propranolol (II), sotalol (IIIB), amiodarone (IIIAB) and verapamil (IV)] were correctly characterized in terms of their effects upon conduction and APD. Agents documented in clinical practice to have proarrhythmic potential (droperidol, sotalol, mibefradil, bepridil, lidoflazine, ketanserin, sertindole, terfenadine, haloperidol, astemizole, cisapride, ziprasidone, lubeluzole, dofetilide, quinidine, ibutilide) were identified as such. Pimozide is reported to rarely produce TdP and was also found to elicit Class III action with few adverse effects. Equally important, agents believed not to be proarrhythmic (two solvents, atenolol, propranolol, fenoximone, cetirizine, verapamil, sildenafil, lidocaine, diltiazem) were identified as having no proarrhythmic activity.
Conclusion: The SCREENIT method properly characterized and quantified prototype cardiovascular drugs and correctly identified proarrhythmic noncardiovascular agents of various mechanisms, but it did not produce false-positive results.