The effect of milrinone in the 16 postoperative shock patients of cardiovascular surgery was studied. The preoperative hemodynamic status were 12 of cardiogenic shock, 2 cases of chronic heart failure and 2 cases of unstable angina pectoris. The operative procedure were 8 cases of coronary artery bypass grafting, 4 cases of valvular surgery, 2 cases of closure of ventricular septal perforation, 2 cases of Bentall operation and 1 case of ascending aortic replacement. The postoperative hemodynamic status were 15 cases of cardiogenic shock, 10 cases of hemorrhagic shock and 1 case of septic shock. Continuous intravenous infusion of 0.5 microgram/kg/min without initial bolus loading was administered immediately after the entrance of the intensive care unit. Significant increase in the maximum blood pressure 3 hours after the infusion were observed (84 +/- 17 mmHg vs 94 +/- 12, p = 0.033). The maximum blood pressure was increased gradually until 24 hours after the infusion. Significant increase in the peripheral body temperature 3 hours after the infusion were observed (32.5 +/- 2.0 degrees C vs 35.9 +/- 1.1 degrees C, p = 0.001). The difference between the peripheral temperature and the central body temperature diminished until 24 hours after the infusion. No significant change in the central venous pressure, pulmonary arterial pressure, pulmonary and cardiac index wedge pressure were observed. No significant change in the platelet number was observed until 3 days after the infusion. Twenty patients (75%) were discharged. Four hospital deaths included 1 cardiac and 3 septic cause were seen. These data suggest that the administration of milrinone for the shock patients after cardiac surgery showed safe and that the continuous intravenous infusion of 0.5 microgram/kg/min without bolus loading showed effective for the recovery of the peripheral circulation.