&NA;Available data demonstrate that oxygen free radicals and derived reactive species of oxygen are produced during myocardial ischemia as well as upon reperfusion of the ischemic tissue. The present study was designed to determine if polyethylene glycol‐conjugated superoxide dismutase (PEG‐SOD), with its extended plasma half‐life in excess of 30 hours in contrast to the native form of the enzyme (Native‐SOD), could provide protection to the ischemic myocardium subjected to a 6‐hour regional ischemia followed by reperfusion for 24 hours. We hypothesized that myocardial injury due to an ischemic interval is a dynamic process involving the sustained production of cytotoxic oxygen radicals that may continue beyond the ischemic interval. The ability to demonstrate a protective effect of the free radical scavenger enzyme superoxide dismutase would require the continued presence of the antioxidant during the ischemic interval and especially during reperfusion. To test this hypothesis, 22 anesthetized, open‐chest dogs underwent 6 hours of circumflex coronary artery occlusion followed by reperfusion for 24 hours. Rapid administration of either Native‐SOD (1,000 U/kg), PEG‐SOD (1,000 U/kg), PEG‐albumin (PEG‐ALB), or 0.9% sodium chloride solution for injection (saline) was administered via the left atrium 15 minutes before occlusion of the vessel. A continuous infusion of an additional 1,000 U/kg of the respective enzyme interventions or an equivalent volume of PEG‐ALB or saline was given during the 6‐hour coronary artery occlusion and terminated 15 minutes after reperfusion. The animals were euthanized 24 hours after reperfusion, and the myocardial region at risk and the infarct region were quantitated by the tetrazolium method. The area of myocardium at risk of infarction, expressed as a percent of the left ventricle, did not differ among the groups: Native‐SOD (n=8), 46.2±1.8%; PEG‐SOD (n=6), 45.7±2.1%; PEG‐ALB, 38.4±2.3% (n=4); and saline 46.0±2.1% (n=4). Hemodynamic parameters, the calculated rate‐pressure‐product, as well as regional myocardial blood flow (radiolabeled microsphere method) in the endocardial, midmyocardial, and epicardial segments of the risk and the nonrisk regions were comparable for all groups. Mean infarct size, determined 24 hours after reperfusion, in the group treated with PEG‐SOD was 47.1±2.9% of the area at risk (n=6), significantly smaller than that observed in each of the other treatment groups: Native‐SOD, 63.5±2.2% (n=8); PEG‐ALB, 64.6±2.4% (n=4); saline, 70.8±2.2% (n=4). The present studies provide support for the concept that superoxide dismutase can prevent myocardial necrosis due to oxygen radicals produced during the ischemic interval as well as the period of reperfusion. Because PEG‐SOD was more effective than Native‐SOD, the results suggest that the sustained presence of oxygen radical scavenger activity is necessary to prevent rather than delay myocardial necrosis. (Circulation Research1989;64:665‐675)