Abstract

In the present study the electrophysiological characteristics and the proarrhythmic potential of cisapride and a structurally related drug, mosapride, were compared. In the anesthetized guinea pig, cisapride and d-sotalol (0.01-10 micromol/kg i.v., n = 6) dose-dependently prolonged the duration of the monophasic action potential recorded from the left ventricle. The maximal lengthening was 18 +/- 3.2% at 1.0 micromol/kg (mean +/- S.E.M., P < .01 vs. base line) and 19 +/- 2.5% at 10 micromol/kg (P < .001) for cisapride and d-sotalol, respectively. In contrast, mosapride did not increase this variable. In a rabbit model of the acquired long QT syndrome, infusion of cisapride (0.3 micromol/kg/min for 10 min maximum, n = 6), but not mosapride or vehicle, was associated with a significant lengthening of the QTU interval (43 +/- 3.8 ms, P < .01). Furthermore, torsades de pointes appeared in two of the six rabbits given cisapride. In isolated rabbit Purkinje fibers (PF), cisapride increased the action potential duration (48 +/- 5.6% at 0.1 micromol/l, P < .01 vs. control, n = 4). Mosapride did not significantly influence the action potential duration (3 +/- 2.0% increase at 1.0 micromol/l, n = 6). However, after mosapride was washed out, the addition of cisapride (0.1 micromol/l) caused a 46 +/- 3.2% lengthening of the action potential duration (P < .01 vs. 1.0 micromol/l mosapride). Early afterdepolarizations and triggered activity appeared in four of eight cisapride-superfused PF stimulated at a very low frequency (0.1 Hz). In isolated rabbit cardiomyocytes, cisapride concentration-dependently blocked (IC50 = 9 nmol/l) the rapid component of the delayed rectifying K+ current (I(Kr)). Mosapride was approximately 1000-fold less potent in blocking I(Kr) (IC50 = 4 micromol/l). It is concluded that the electrophysiological characteristics of cisapride may explain the recently reported propensity to prolong the QT interval and to induce torsades de pointes in susceptible patients, although a structurally related benzamide, mosapride, did not appear to have electrophysiological features of relevance for induction of torsades de pointes in common with cisapride.