To determine if the addition of potassium enhances the myocardial protective effect of intracoronary perfusion hypothermia during aortic cross-clamping, 50 patients undergoing aortocoronary bypass grafting were studied in a randomized, prospective, double-blind fashion. Twenty-six patients received a cold crystalloid solution infused with a handheld syringe into the root of the cross-clamped aorta every 20 minutes, and 24 patients received the same solution but with 25 mEq/L of potassium chloride added, infused in a similar manner. Both groups were analyzed by mortality, rate of perioperative myocardial infarction (electrocardiographic changes, MB-CPK enzyme release, and preoperative and postoperative gated cardiac blood pool scans), intraoperative hemodynamic changes, intraoperative lactate determinations, postoperative arrhythmias, and requirement for pressor or intraaortic balloon pump support. One patient in the potassium cardioplegia group died (massive pulmonary embolism), and none in the hypothermic perfusion group died. Possible perioperative myocardial infarction was diagnosed by more than one marker in 4 of 26 patients in the hypothermic perfusion group and 5 of 24 patients in the potassium group (p = 0.61). There were no differences between the two groups in terms of hemodynamic changes, lactate production, postoperative arrhythmias, or the need for postoperative hemodynamic support. This study in human beings could not demonstrate a specific protective effect of potassium, beyond that afforded by myocardial perfusion hypothermia and wash-out. The data suggest that myocardial hypothermia, achieved through cold intracoronary arterial perfusion, may be the most important beneficial component of so-called cardioplegia for attaining effective intraoperative myocardial preservation in human beings.