Outcome Prediction by Traditional and New Markers of Inflammation in Patients with Sepsis

Abstract

Patients (n = 242) admitted to intensive care unit for longer than 48 hours were categorised for sepsis according to American College of Chest Physicians/Society of Critical Care Medicine (ACCP/SCCM) Consensus Conference criteria. Body temperature, leukocyte count, C-reactive protein (CRP) and procalcitonin (PCT) as well as tumor necrosis factor-α (TNF-α), interleukin (IL)-6, IL-8, IL-10 and HLA-DR expression on monocytes were determined. Data of one randomly choosen day per patient entered analysis. Immunologic mediators contributing significantly to outcome were determined by logistic regression analysis. Area under the curves (AUC) of receiver operating characteristic curves of clinical markers of inflammation predicting prognosis were compared with AUC of relevant immunologic mediators. TNF-α, IL-6 and HLA-DR expression on monocytes were significantly associated with outcome; the AUC's were 0.835, 0.844 and 0.761 respectively. AUC's for clinical markers were 0.878, 0.811, 0.620 and 0.614 for PCT, CRP, leukocyte count and body temperature respectively. PCT had the highest AUC compared to other clinical markers. These data indicate that PCT might be a better marker than the classic criteria of inflammation, CRP, leukocyte count, and body temperature to identify patients endangered by severe infection or sepsis.