The effects of the potassium channel openers (KCO), cromakalim or pinacidil, were evaluated in an anesthetized porcine model of pacing- and ischemia-induced ventricular fibrillation (VF). Hearts were paced at 180 bpm and the left anterior descending coronary artery was occluded until VF was induced. Reproducible times to VF (in seconds) were obtained allowing at least 20 min recovery following defibrillation. Cromakalim (0.3 mg/kg) or pinacidil (3 mg/kg) produced equivalent drops in mean arterial blood pressure. At these doses, cromakalim reduced monophasic action potential duration measured at 90% repolarization (APD90). Although time to VF in the cromakalim group was significantly greater than the vehicle treated group, it was not significantly different from its predrug value. In contrast, pinacidil reduced APD90, and significantly increased time to VF from 134 +/- 5 to 322 +/- 62 s (p < 0.05). Neither cromakalim nor pinacidil affected whole-cell calcium currents recorded in guinea pig myocytes. During ischemia, cromakalim or pinacidil further reduced APD90; however, pinacidil had a two-fold greater effect than did cromakalim. The Class III antiarrhythmic agent, dofetilide, prolonged APD90, but did not increase time to VF. In conclusion, the increased time to VF observed with pinacidil coincides with its ability to shorten APD, and is consistent with activation of ATP-sensitive K+ channels (K+ ATP). It is suggested that indirect reduction of calcium influx through K+ ATP activation and APD shortening is sufficient to increase time to VF in this model. However, the inability of dofetilide to be effective suggests that this model would not be useful to test for Class III antiarrhythmic agents.