Comparison of severity of illness scoring systems in the prediction of hospital mortality in severe sepsis and septic shock

Background : New scoring systems, including the Rapid Emergency Medicine Score (REMS), the Mortality in Emergency Department Sepsis (MEDS) score, and the confusion, urea nitrogen, respiratory rate, blood pressure, 65 years and older (CURB-65) score, have been developed for emergency department (ED) use in various patient populations. Increasing use of early goal directed therapy (EGDT) for the emergent treatment of sepsis introduces a growing population of patients in which the accuracy of these scoring systems has not been widely examined. Objectives : To evaluate the ability of the REMS, MEDS score, and CURB-65 score to predict mortality in septic patients treated with modified EGDT. Materials and Methods : Secondary analysis of data from prospectively identified patients treated with modified EGDT in a large tertiary care suburban community hospital with over 85,000 ED visits annually and 700 inpatient beds, from May 2007 through May 2008. We included all patients with severe sepsis or septic shock, who were treated with our modified EGDT protocol. Our major outcome was in-hospital mortality. The performance of the scores was compared by area under the ROC curves (AUCs). Results : A total of 216 patients with severe sepsis or septic shock were treated with modified EGDT during the study period. Overall mortality was 32.9%. Calculated AUCs were 0.74 [95% confidence interval (CI): 0.67-0.81] for the MEDS score, 0.62 (95% CI: 0.54-0.69) for the REMS, and 0.59 (95% CI: 0.51-0.67) for the CURB-65 score. Conclusion : We found that all three ED-based systems for scoring severity of illness had low to moderate predictive capability. The MEDS score demonstrated the largest AUC of the studied scoring systems for the outcome of mortality, although the CIs on point estimates of the AUC of the REMS and CURB-65 scores all overlap.