Neonates undergoing heart surgery are exposed to high levels of circulating catecholamines. Our objective was to determine to what extent administration of magnesium counters epinephrine-induced cardiotoxicity. We assessed left ventricular function (pressure-volume data obtained by the conductance catheter/micromanometer technique) and ultrastructure in newborn pigs 3 to 5 days of age before and after administration of epinephrine alone (2 micrograms/kg/min, group A, n = 6) and simultaneously with magnesium sulfate (8 mmol/L, 5 ml/hr, group B, n = 6). Plasma levels of magnesium were maintained at 200% to 250% of control, and ionized calcium was maintained at normal levels. During administration of epinephrine, there was a significant increase in end-systolic elastance from 8.9 +/- 2 to 15 +/- 3 mm Hg/ml in group A and from 7.8 +/- 2 to 16 +/- 3 mm Hg/ml in group B (p < 0.05). This increase was accompanied by an increase in chamber stiffness index (p < 0.05) and shortening of the time constant of isovolumic relaxation (p < 0.05; group A, 19 +/- 3 to 13 +/- 3 msec; Group B, 20 +/- 2 to 15 +/- 2 msec). After epinephrine was discontinued, however, systolic and diastolic indexes returned to baseline levels in group B, whereas group A exhibited a significant reduction in end-systolic elastance (5 +/- 1 mm Hg/ml; p < 0.05) and an increase in chamber stiffness index (0.7 +/- 0.08 versus 0.4 +/- 0.1 ml-1; p < 0.05) and time constant (25 +/- 1 versus 19 +/- 3 msec). Left ventricular dysfunction in group A was associated with focal sarcolemmal rupture and mitochondrial swelling, whereas only minor reversible changes (microvesicular lipid accumulation) were seen in group B. We conclude that magnesium has a protective effect against epinephrine-induced cardiotoxicity because of its blocking action on calcium influx of ionized calcium and could be of therapeutic benefit in the perioperative period.