Cytokine balance and immunosuppressive changes at cardiac surgery: contrasting response between patients and isolated...

In vitro work suggests that IL-10 plays a pivotal role in controlling the balance of pro- and anti-inflammatory cytokines and monocyte HLA-DR expression. In 20 patients undergoing cardiac surgery, we investigated elaboration of interleukin 10 (IL-10) and its relationship to pro- and anti-inflammatory cytokines and leucocyte expression of HLA-DR and adhesion molecules. There were small increases in pro-inflammatory cytokines (IL-1, IL-8 and tumour necrosis factor (TNF) after induction, returning to baseline on induction of cardiopulmonary bypass (CPB). After CPB another transient increase in IL-8 occurred (P < 0.05). The anti-inflammatory response began with elevated IL-10 during CPB (P < 0.001), which peaked early in recovery (P < 0.001), by which time IL-1 receptor antagonist (IL-1ra) and the TNF soluble receptors (TNFsr) had also increased (P < 0.01). The next day IL-10 and IL-1ra were decreasing but TNFsr continued to increase. Induction of anaesthesia caused HLA-DR downregulation. The IL-10 peak was associated with further monocyte HLA-DR downregulation (P < 0.001) and return towards baseline of granulocyte adhesion molecule expression which transiently increased during CPB (P < 0.001). To determine which aspects of the immune response arose from the interaction of blood with the CPB apparatus, the above variables were studied within an isolated CPB circuit and the influence of fentanyl on the magnitude of any such changes determined. Five healthy volunteers donated two, 250-ml samples of blood to which was added either fentanyl 175 micrograms with heparin 1050 u. or heparin alone 1050 u. These were used to prime two identical isolated CPB circuits and circulation was conducted under identical conditions for 90 min. Of the pro-inflammatory cytokines, only IL-8 was elevated at 90 min CPB (P < 0.05). There was no increase in anti-inflammatory cytokines and TNFsr decreased (P < 0.001). Granulocyte adhesion molecules were increased during CPB. In the fentanyl group, the CD11b increase was greater and preceded CPB. The reduction in lymphocyte HLA-DR expression, observed throughout the study period (P < 0.01), was greater with fentanyl (P < 0.05). Monocyte HLA-DR expression increased (P < 0.05), but to a lesser extent with fentanyl (P > 0.05). In contrast with the in vivo response where there was a phased anti-inflammatory response beginning with IL-10, in the isolated CPB model no anti-inflammatory cytokine response occurred.