The authors tested whether allopurinol or oxypurinol could limit infarct size when treatment was started just prior to reperfusion and continued until sacrifice. In closed chest, anesthetized dogs, a branch of the left coronary artery was reversibly occluded for 90 mins followed by 24 h of reperfusion. Fifteen minutes prior to reperfusion, dogs received a bolus of either allopurinol (10 dogs), or oxypurinol (nine dogs), 10 mg/kg intravenously, followed by a 24 h infusion, 55 mg/kg/day. Eleven control dogs received equal volumes of saline. Investigators were blinded to the identity of the agent. Hearts were removed 24 h after reperfusion. Arrhythmias for 30 mins after reperfusion were quantitated. Oxypurinol caused 28% less of the risk zone to infarct for any level of collateral flow than that seen in the controls. This difference was significant. Allopurinol-treated hearts averaged only 18% less infarction and did not achieve significance. Ventricular arrhythmias and mortality did not differ among the three groups. Therefore, unlike allopurinol, oxypurinol with continued administration can limit tetrazolium-indicated necrosis in the dog heart in the absence of pretreatment.