Parabiotically perfused, isolated cat hearts were subjected to global ischemia for 2 hours at 27 degrees C and were then reperfused for 2 hours. One group of hearts was simply perfused with hypothermic blood 5 minutes before ischemia and again before normothermic reperfusion. A second group received a crystalloid cardioplegic solution. A third group received cardioplegic solution supplemented with superoxide dismutase and catalase; at the start of reperfusion the anesthetized cats supplying blood to perfused hearts in this group received an i.v. infusion of these enzymes. Based on comparisons of postreperfusion ventricular pressure development, maximal ventricular dP/dt, rate-pressure products and coronary blood flow, we concluded that cardioplegic solution and enzyme treatment were superior to enzyme-free cardioplegia solution under otherwise similar experimental conditions. The results suggest that a component of "ischemic" cardiac damage may involve cytotoxicity from oxygen metabolites such as superoxide anion, hydrogen peroxide, or both, and that this component of damage can be reduced by enzyme supplements such as superoxide dismutase or catalase.