Repletion with Ca2+ often leads to damage of previously Ca(2+)-depleted hearts (the calcium paradox). The behavior of the liver under similar conditions is not well understood. With a perfused rat liver model, we examined liver cell damage and lipid peroxidation during Ca2+ depletion and repletion and used lucigenin-enhanced chemiluminescence as a measure of oxygen radicals. During 30 minutes of Ca2+ depletion, release of lactate dehydrogenase and thiobarbituric acid-reactive substance did not change significantly. When Ca2+ depletion was extended to 150 minutes, release of lactic acid dehydrogenase and thiobarbituric acid-reactive substance and tissue oxygen radical levels all increased progressively, accompanied by decrease in oxygen uptake. Ca2+ repletion after 30 minutes of Ca2+ depletion caused small increases in release of lactic acid dehydrogenase and thiobarbituric acid-reactive substance but significantly suppressed the changes described, compared with expression in depleted livers without Ca2+ repletion. There were large releases of sinusoidal glutathione and glutathione disulfide at the onset of Ca2+ depletion, which declined within 15 minutes. On Ca2+ repletion, sinusoidal glutathione level decreased to its baseline but glutathione disulfide level did not change significantly. During long-term Ca2+ depletion, sinusoidal glutathione level was significantly higher than baseline but glutathione disulfide level remained low. These results indicate that long-term Ca2+ depletion causes oxidative stress and liver damage. Ca(2+)-dependent release of sinusoidal glutathione appears to result from causes other than oxidative stress. There is no evidence for the calcium paradox in the liver; in fact, reexposure to Ca2+ protects the liver from the injury caused by Ca2+ depletion.