The results of several recent studies have demonstrated that reactive oxygen metabolites are responsible for a major portion of ischemia/reperfusion (I/R) injury in skeletal muscle. Presumably, the cytotoxic oxidants are produced during reperfusion when molecular oxygen (the source of the reactive oxygen metabolites) is reintroduced to the tissues. The purpose of this study was to test the hypothesis that molecular oxygen must be provided at reperfusion to produce I/R injury in skeletal muscle. Isolated, maximally vasodilated (papaverine) canine gracilis muscles were reperfused, after 4 h of inflow occlusion, from reservoirs containing autologous blood equilibrated with either 95% O2-5% CO2 or 95% N2-5% CO2 gas mixtures. Arterial PO2 fell from approximately 120 mmHg to less than 3-5 mmHg, during the use of nitrogen. The solvent drag reflection coefficient for total plasma proteins (sigma f) and total vascular resistance was determined for the following conditions: control (no ischemia), reperfusion with oxygenated blood after 4 h ischemia; and reperfusion (after 4 h ischemia), first with anoxic blood and then oxygenated blood. Reperfusion with oxygenated blood, after 4 h of ischemia, significantly reduced solvent drag reflexion coefficient (sigma f) from 0.93 +/- 0.02 to 0.63 +/- 0.02, indicating a dramatic increase in vascular permeability. Total vascular resistance increased from 6.1 +/- 1.1 mmHg.ml-1.min.100 g during the preischemic period to 12.9 +/- 3.0 mmHg.ml-1.min.100 g during normoxic reperfusion. In muscles reperfused with anoxic blood, sigma f averaged 0.82 +/- 0.06, whereas vascular resistance increased by 56 +/- 13%.(ABSTRACT TRUNCATED AT 250 WORDS)