The extent of cardiac injury incurred during reperfusion as opposed to that occurring during ischemia is unclear. This study tested the hypothesis that simulated ischemia followed by simulated reperfusion causes significant "reperfusion injury" in isolated chick cardiomyocytes. Cells were exposed to hypoxia, hypercarbic acidosis, hyperkalemia, and substrate deprivation for 1 h followed by 3 h of reperfusion. Irreversible cell membrane injury, measured by propidium iodide uptake, increased from 4% of cells at the end of ischemia to 73% after reperfusion; death occurred in only 17% of cells kept ischemic for 4 h. Lactate dehydrogenase release was consistent with these changes. Lengthening ischemia from 30 to 90 min increased cell injury as expected, but of the total cell death, > 90% occurred during reperfusion. "Chemical hypoxia" composed of cyanide (2.5 mM) plus 2-deoxyglucose augmented injury before reperfusion compared with simulated ischemia. Inhibition of oxygen radical generation by use of metal chelator 1,10-phenanthroline reduced cell death from 73% to 40% after reperfusion (P = 0.001). We conclude that simulated reperfusion significantly augments the cellular membrane damage elicited by simulated ischemia in isolated cardiomyocytes devoid of other factors and suggest that reactive oxygen species, perhaps from the mitochondria, participate in this injury.