A growing body of experimental data indicates that the "no-reflow" phenomenon is a type of reperfusion injury in skeletal muscle which may, in part, be mediated by oxygen free radicals, and thus may be attenuated by using agents that scavenge or inhibit formation of these reactive oxygen metabolites. This study was undertaken to assess the efficacy of recombinant human manganese superoxide dismutase (rhMnSOD) in reducing reperfusion injury in skeletal muscle. The specific advantage of this agent over other SOD types is a much longer plasma half-life (5 to 7 hr), allowing better equilibration between extra- and intracellular compartments. The rat cremaster model was used to study "no-reflow" in skeletal muscle. Reperfusion injury in the muscle was assessed by fluorescein dye perfusion, myocyte creatine phosphokinase (CPK) release, and contractile function in response to electrical field stimulation. Compared with untreated saline control animals, those treated with rhMnSOD after 5 hr of cremasteric ischemia, had a significantly higher percentage area of blood reflow (78 percent +/- 6 percent of normal), a greater percentage tetanic (66 percent +/- 9 percent of normal) and twitch (56 percent +/- 9 percent of normal) contractile strength, and less CPK release (21.5 percent higher than pre-reperfusion baseline CPK levels) (p < 0.05). Untreated saline control CPK release (21.5 percent higher than the prereperfusion level. Animals treated with allopurinol also had a significantly higher percentage twitch contraction (47 percent +/- 14 percent of normal) and a lower CPK release (11.1 percent of the prereperfusion value) 45 min after reperfusion than untreated saline controls.(ABSTRACT TRUNCATED AT 250 WORDS)