In vivo experimental approach for the risk assessment of fluoroquinolone antibacterial agents-induced long QT syndrome.

The proarrhythmic effects of fluoroquinolone antibacterial agents, sitafloxacin, gatifloxacin and moxifloxacin, were compared using three in vivo models. In the halothane-anesthetized dogs (n=5), intravenous 10-min infusion of gatifloxacin and moxifloxacin (1-3 mg/kg) prolonged the ventricular effective refractory period and the repolarization period to a similar extent, whereas sitafloxacin (1-3 mg/kg) prolonged the former only. No significant change was detected in other cardiovascular parameters. In the chronic complete atrioventricular block dogs (n=4), oral administration of 100 mg/kg of gatifloxacin (2 of 4) and moxifloxacin (3 of 4) induced torsades de pointes, which was not observed by sitafloxacin. In the alpha-chloralose-anesthetized rabbits (n=5), intravenous 20-min infusion of 60 mg/kg of gatifloxacin induced torsades de pointes (1 of 5) in the presence of methoxamine infusion, which was not observed by sitafloxacin or moxifloxacin. Thus, the halothane-anesthetized model is suitable for assessing QT prolongation, whereas the chronic complete atrioventricular block model is sensitive for detecting torsadogenic action of drugs. The alpha-chloralose-anesthetized model is the simplest and least expensive method, but its sensitivity to detect proarrhythmic action may be less great.