Novel predictors of sepsis outperform the american burn association sepsis criteria in the burn intensive care unit patient

The purpose of this study was to determine whether systemic inflammatory response syndrome (SIRS) and American Burn Association (ABA) criteria predict sepsis in the burn patient and develop a model representing the best combination of novel clinical sepsis predictors. A retrospective, case-controlled, within-patient comparison of burn patients admitted to a single intensive care unit from January 2005 to September 2010 was made. Blood culture results were paired with documented sepsis: positive-sick, negative-sick (collectively defined as sick), and negative-not sick. Data for all predictors were collected for the 72 hours before blood culture. Variables were evaluated using regression and area under the curve (AUC) analyses. Fifty-nine subjects represented 177 culture periods. SIRS criteria were not discriminative: 98% of the subjects met criteria. ABA sepsis criteria were different on the day before (P = .004). The six best-fit variables identified for the model included heart rate > 130 beats per min, mean arterial pressure < 60 mm Hg, base deficit <-6 mEq/L, temperature < 36 C, use of vasoactive medications, and glucose > 150 mg/dl. The model was significant in predicting positive-sick and sick, with an AUC of 0.775 (P < .001) and 0.714 (P < .001), respectively; comparatively, the ABA criteria AUC was 0.619 (P = .028) and 0.597 (P = .035), respectively. Usefulness of the ABA criteria to predict sepsis is limited to the day before blood culture is obtained. A significant contribution of this research is the identification of six novel sepsis predictors for the burn patient.