The MISSED score, a new scoring system to predict Mortality In Severe Sepsis in the Emergency Department

Abstract

To derive and validate a new scoring system to predict in-hospital mortality in septic patients in the emergency department (ED). Septic patients admitted to the ICU and those in whom early goal-directed therapy (EGDT) was carried out in the ED were identified from the ED record. Univariate and multivariate regression analyses identified independent variables associated with mortality. The variables were given a score weighted by the odds ratio, the sum of which yielded the Mortality In Severe Sepsis in the Emergency Department (MISSED) score. The performance of the MISSED score in predicting mortality was compared with that of the Acute Physiology and Chronic Health Evaluation II (APACHE II) score, the EGDT criteria and the severe sepsis criteria. The Hosmer-Lemeshow test was performed to calibrate the model. Independent variables identified were age at least 65 years, albumin level up to 27 g/l and international normalized ratio of 1.2 or more. The MISSED score ranged from 0-9; cut-off point 5.5. Mortality rates associated with a score of 0, less than 5.5 and 5.5 or more were 7.4, 17.7 and 40.6%, respectively. The sensitivity of the score was 96.8% (95% confidence interval 87.8-99.4%). The mortality rate and specificity associated with a score of 9 were 62.9 and 91.6% respectively. The area under the curve for the MISSED score and the APACHE II score were equal. The performance of the MISSED score of 5.5 or more in predicting mortality was similar to that of the EGDT criteria. The sensitivity of the score was equal to that for the severe sepsis criteria. The Hosmer-Lemeshow test confirmed good calibration. The MISSED score should be used in the ED.