The six scoring systems’ prognostic value in predicting 24-hour mortality in septic patients

M. Djikic, M. Milenkovic, M. Stojadinovic, T. Miladinovic, D. Gujanicic, I. Milicevic-Nesic, B. Uzelac, M. Laban, D. Markovic

Emergency Center, University Clinical Center of Serbia, Belgrade, Serbia. marija.milenkovic@med.bg.ac.rs

---

• Anesthesiology

OBJECTIVE: The use of scoring systems contributes to the faster identification of septic patients, especially those at a high risk of a fatal outcome. The best scoring system does not exist, so the search for the optimal one is always current. The aim of this study is to estimate the prognostic value of the six scoring systems in predicting 24-hour mortality among septic patients presented at the emergency department.

PATIENTS AND METHODS: An observational retrospective study was conducted in the Emergency Triage Room (ETR) of the Emergency Center (EC) at the University Clinical Center of Serbia (UCCS) in Belgrade. Consecutive septic patients, according to the Sepsis-3 definition, with or without shock, presented to the ETR and then hospitalized in Intensive Care Units were included in the study. Mortality data within 24 h and on the 28th day were extracted from the Hospital information system or the National mortality database. Scoring systems including sequential organ failure assessment (SOFA), quick sequential organ failure assessment (qSOFA), systemic inflammatory response syndrome (SIRS), National early warning score (NEWS), sepsis patient evaluation in the emergency department (SPEED), and mortality in emergency department sepsis (MEDS) were analyzed for all patients utilizing the available data. The primary outcome of this study was death within 24 hours of triage. Receiver operating characteristic (ROC) analysis was used to determine the most effective scoring system. Lactate was then added to this system to enhance its predictive accuracy.

RESULTS: Nineteen out of 120 patients included in the study (15.8%) experienced death within 24 hours of triage. The twenty-eight-day mortality rate was 55%. SOFA score demonstrated the highest predictive value for 24-hour mortality but was only moderately predictive overall, with an area under the receiver operating curve (AUC) of 0.755 (95% CI 0.625-0.885). SPEED, MEDS, and NEVS exhibited modest discriminatory power [0.673 (95% CI 0.543-0.803), 0.665 (95% CI 0.536-0.794), 0.630 (95% CI 0.528-0.724)], while SIRS and qSOFA remained insignificant in predicting 24-hour mortality. The predictive value of the SOFA score was increased by the addition of lactate (AUC 0.865, 95% CI 0.736-0.995; p=0.0081). All scores demonstrated better and satisfactory predictive power for 28-day mortality.

CONCLUSIONS: SOFA, with the addition of lactate, is a complex but reliable tool for the early stratification of septic patients who are presenting at an emergency department.

Free PDF Download