The effectiveness of high-potassium cardioplegic solution in the neonatal heart remains controversial. Our previous study indicated that the protection afforded by a cardioplegic solution was inadequate in the neonatal heart. On the hypothesis that oxyradicals were responsible for the ineffectiveness of cardioplegic solution in neonatal heart, the effects of a cardioplegic solution (a modified St. Thomas' Hospital cardioplegic solution) with recombinant human superoxide dismutase on the isolated perfused neonatal guinea pig hearts (within 2 days after delivery, body weight of 60 to 120 g) were studied in comparison with those on the adult hearts (6 to 8 weeks after delivery, body weight of 300 to 500 g). After arrest induced by modified St. Thomas' Hospital cardioplegic solution, hearts were subjected to 120 min of ischemia at 20 degrees C, during which time the cardioplegic solution was injected every 30 minutes. Then the heart was reperfused for 60 minutes at 37 degrees C. Under this condition, the left ventricular developed pressure recovered to 84.4% +/- 4.0% of the preischemic value in the adult heart, whereas the recovery was only 68.1% +/- 3.1% in the neonatal heart. Thiobarbituric acid-reactive substance level, a parameter of lipid peroxidation by oxyradicals, significantly increased during ischemic arrest both in the adult and neonatal heart. However, the increase was much greater in the neonatal heart than in the adult. Cardioplegia with recombinant human superoxide dismutase (300 and 1,000 U/mL) significantly inhibited this accumulation of thiobarbituric acid-reactive substance in the neonatal heart; at 1,000 U/mL, the myocardial function of the reperfused neonatal heart recovered to the level of the adult heart.(ABSTRACT TRUNCATED AT 250 WORDS)