Evaluation of definitions and parameters for sepsis assessment in patients after cardiac surgery.

In 110 patients undergoing elective heart surgery on extracorporeal circulation, various parameters were studied regarding the early assessment of septic complications. In a first step, the Elebute score definition for postoperative sepsis validated in general surgery patients (score > or = 12) could be confirmed in an extended form (> or = 12 on > or = 2 days) for cardiac surgery patients. According to this definition (overall classification accuracy for clinically defined sepsis-related mortality: 94%), septic complications occurred in 16 patients and were associated with a significantly worse prognosis than in non-septic patients (mortality 69% vs. 1%, p < 0.0001). In contrast, SIRS (best classification criterion: positive on > or = 3 days) displayed a lower specificity for clinically defined sepsis-related mortality, at least during the early postoperative course (accuracy: 67%). Based on the Elebute score classification, other more practicable parameters were investigated regarding their usefulness for an early sepsis risk assessment in post cardiac surgical patients. Five additional severity scores (APACHE II, MOF-Goris, HIS, SAPS, SSS) were comparable (ROC area: 0.94 to 0.96) and superior to plasma PMN-elastase and neopterin, haemodynamics and clinical parameters in predicting the risk for septic complications as early as by the first postoperative day.