BACKGROUND Steroid use during cardiac operations may reduce the risk of postperfusion lung syndrome, but both cardiopulmonary bypass and steroids are immunosuppressive. The synergistic effects of the bypass and steroids on patients' immunologic activities, hemodynamics, and metabolisms during and after heart operations have not been clarified systematically. METHODS Twenty-four patients undergoing valve replacement were studied in a randomized, double-blind trial. Twelve of these patients (S group) received bolus methylprednisolone, 20 mg/kg body weight, and the remaining 12 patients (C group) received a placebo intravenously before and after bypass. Blood cell count, C-reactive protein, lymphocyte surface markers (CD3, CD4, CD8, CD16, and CD20), phytohemagglutinin response, interleukin-2 production, and natural killer cell activity were examined on admission through day 7. Cardiac output, blood gas, electrolyte, lactate, and serum glucose levels were examined perioperatively. RESULTS The peak white blood cell count in the S group was higher than that in the C group (analysis of variance: p [group] = 0.0436). The peak C-reactive protein level was higher in the C group than in the S group (p [group] < 0.0001). From the analysis of the surface markers, the steroid increased the natural killer cells before and soon after bypass (p [group] = 0.0117), and later tended to increase the CD4+ T and B cells during the postoperative recovery period. The phytohemagglutinin response in both groups decreased after bypass (p [time] < 0.0001), but the steroid caused exaggerated decreases before (p < 0.01 by Student's t test) and soon after (p < 0.001) bypass in the S group (analysis of variance: p [group] = 0.0127). The interleukin-2 production was suppressed by bypass alone after the bypass in the C group, but was further suppressed by the steroid before and after bypass in the S group (p [group] = 0.0446). The cardiac index, water balance, electrolytes, arterial oxygen tension, and timing of extubation were not different between the groups. In contrast, the glucose (p [group] = 0.0486) and lactate (p [group] = 0.0525) levels were higher in the S group than those in the C group. CONCLUSIONS T-cell functions are synergistically suppressed by cardiopulmonary bypass and high-dose methylprednisolone in heart operations. The hemodynamic benefits of the steroid are negligible, whereas glucose tolerance is worsened by the steroid during bypass.