The Langendorff-perfused rat heart with regional ischemia is increasingly used for evaluating drugs for prevention of phase-1, ischemia-induced ventricular fibrillation (VF). Surprisingly, the effectiveness of Class I antiarrhythmics has not been characterized in this model. One lower and one higher concentration of quinidine (0.79 and 7.90 microM), lidocaine (3.88 and 12.93 microM), and flecainide (0.74 and 1.48 microM), representing the peak unbound plasma and total blood concentrations, respectively, at "therapeutic" dosage, were evaluated. The left main coronary artery was occluded for 30 min to elicit phase-1 VF, after which reperfusion-induced VF was examined. In hearts perfused with Krebs' solution containing 3 mM K(+), the higher concentrations of quinidine and lidocaine reduced the incidence of phase-1 VF from 92% to 0% and 17% respectively, (each p < 0.05). The lower drug concentrations were ineffective. Flecainide was equi-effective at low and high concentrations, with VF incidence reduced from 92% to 17% (p < 0.05). Neither low nor high concentrations of any of the drugs affected the incidence of reperfusion-induced VF. Using hearts perfused with Krebs' containing 5 mM K(+), sufficient to substantially reduce control phase-1 VF incidence, the experiment was repeated to test for possible proarrhythmic activity. None of the three drugs increased arrhythmia incidence. In this model, it was not possible to suppress ischemia-induced and reperfusion-induced VF with flecainide, lidocaine, or quinidine at concentrations equivalent to peak unbound plasma levels after clinical administration. This may explain the lack of clinical benefit with these drugs against sudden cardiac death. Because none of the drugs were proarrhythmic in ischemic hearts in which arrhythmia susceptibility had been lowered by high K(+), it would seem that clinical proarrhythmia seen with these drugs may not be related to exacerbation of phase-1, ischemia-induced VF.