The hemodynamic effects of amrinone were assessed in seven children following cardiac surgery. Amrinone was administered as a bolus of 1 mg kg-1 body wt., followed by continuous infusion at 10 micrograms kg-1 min-1 for 1 h and two stepwise increases to 20 and 40 micrograms kg-1 min-1 for 30 min each. Hemodynamic data were obtained and plasma concentrations of amrinone measured 1 h after the bolus dose and immediately before each increment of the infusion rate. Amrinone levels ranged from 0.7 to 2.3 mg l-1. Administration of amrinone lowered systemic vascular resistance from 20.0 +/- 4.3 to 16.5 +/- 4.6 mmHg l-1 min-1 m-2 (p less than 0.05) and reduced mean arterial pressure from 71.7 +/- 9.5 to 62.6 +/- 13.5 mmHg (p less than 0.05) at the highest infusion rate, confirming the known vasodilative effect of the drug. However, these effects did not result in a statistically significant increase in stroke volume (35.0 +/- 7.5 to 35.5 +/- 7.0 ml m-2, NS) or cardiac index (3.10 +/- 0.50 to 3.20 +/- 0.40 l min-1 m-2). One additional patient, in whom a higher loading dose was tried in order to achieve a higher plasma concentration, developed systemic hypotension. A correlation was established between the plasma concentrations of amrinone and the percentage decrease in systemic resistance (r = 0.70, p less than 0.05). These results suggest that in children after open heart surgery, amrinone acts primarily as a systemic vasodilator, with questionable inotropic effect. Accordingly, its use should be restricted to children with severe cardiac failure and documented highly elevated afterload.