We determined whether any of the antiischemic effects of nicorandil were due to direct cardioprotective effects such as potassium channel activation or to its peripheral hemodynamic effects. Nicorandil was administered either intravenously (i.v.) or directly into the ischemic coronary artery (i.c.) and compared with i.c. cromakalim (a potassium channel activator previously shown to improve reperfusion function directly in rat hearts) or vehicle for their ability to improve postischemic contractile function as measured by ultrasonic crystals in anesthetized dogs or in isolated perfused rat hearts. In a model of 25-min global ischemia and reperfusion in isolated perfused rat hearts, nicorandil (10-100 microM) did not improve reperfusion function or decrease LDH release, although 300 microM nicorandil did protect the hearts. Cromakalim (7 microM) significantly improved reperfusion function and reduced lactate dehydrogenase (LDH) release. In the dog studies, the left anterior descending coronary artery (LAD) was occluded for 15 min and was reperfused for 3 h. Nicorandil improved reperfusion function only when administered i.v., although i.c. cromakalim was efficacious in improving function. Neither nicorandil nor cromakalim improved collateral flow, although cromakalim significantly improved preischemic and reperfusion blood flows, particularly in the subepicardial region. Although i.c. treatment with cromakalim and nicorandil did not result in significant changes in peripheral hemodynamic status, i.v. nicorandil reduced both preload and afterload. Thus, at the dose used, nicorandil does not appear to have direct myocardial protective effects and the beneficial effects of nicorandil do not appear to be related to potassium channel activation in the myocardium. Potassium channel activation by cromakalim does result in direct cardioprotective effects whereas nicorandil appears to be dependent on peripheral actions for its efficacy.