Objectives: Cardiopulmonary bypass is associated with the release of proinflammatory cytokines (tumor necrosis factor alpha, interleukin 1beta, interleukin 6, and interleukin 8) and anti-inflammatory cytokines (interleukin 10 and transforming growth factor beta(1)). On the one hand this cytokine release is related to the postoperative systemic inflammatory response syndrome, and on the other hand it is related to deterioration of the immune system, for example in monocyte or polymorphonuclear neutrophil function, leading to an increased susceptibility to infections. To gain further insight into the alterations of immune cell reactivity and possible regulatory mechanisms, we studied lipopolysaccharide-induced tumor necrosis factor alpha synthesis in whole blood from cardiac surgical patients. Methods: Fifteen patients undergoing elective heart surgery with cardiopulmonary bypass were included in the study. Ex vivo lipopolysaccharide-induced tumor necrosis factor alpha synthesis was measured in a whole blood assay before, during, and after bypass. Corresponding tumor necrosis factor alpha messenger RNA levels were determined by semiquantitative reverse transcriptase-polymerase chain reaction. In addition, the influence of patient serum on whole blood responsiveness and its relationship to anti-inflammatory cytokines were evaluated in vitro. Results: Tumor necrosis factor alpha synthesis was significantly reduced after 30 minutes of cardiopulmonary bypass and showed the lowest values at the end of bypass (mean +/- SD 0.109 +/- 0.105 ng/10(6) white blood cells after 30 minutes of bypass and 0.050 +/- 0.065 ng/10(6) white blood cells at the end of bypass, vs 0.450 +/- 0.159 ng/10(6) white blood cells preoperatively, P < .001). As a further indication of reduced cytokine biosynthesis, diminished messenger RNA levels for tumor necrosis factor α were detected. Serum withdrawn from patients at the end of cardiopulmonary bypass reduced tumor necrosis factor α synthesis in heterologous blood from healthy volunteers highly significantly to 39.93% +/- 23.18% relative to control serum (P = .005) and preoperatively drawn serum (P = .024). This effect was dose dependent and was not specific for lipopolysaccharide-induced tumor necrosis factor α synthesis. Anesthesia and heparin administration did not influence tumor necrosis factor α production significantly. Ex vivo tumor necrosis factor α synthesis was negatively related to interleukin 10 serum levels, positively but weakly related to interleukin 4, and was not related to transforming growth factor β(1) (Spearman correlation coefficients -0.565, P < .001, 0.362, P < .001, and -0.062, P = .460, respectively). However, interleukin 10 levels in patient serum after cardiopulmonary bypass were 300-fold below the quantities needed for half-maximal inhibition of tumor necrosis factor α synthesis in vitro. Moreover, the inhibitory activity could not be removed by immune absorption of interleukin 10. Conclusions: These results suggest that during cardiac operations cytokine-inhibitory serum activities are released or newly formed. These activities could not be explained by the actions of interleukins 4 and 10 or transforming growth factor β(1). Although their exact nature remains undetermined, these substances may contribute to the diminished immune cell functions after cardiopulmonary bypass and thus need further characterization.
