Abstract
BACKGROUND: We compared the prognostic accuracy of in-hospital mortality of the initial Sequential Organ Failure Assessment (SOFA(ini)) score at the time of sepsis recognition and resuscitation and the maximum SOFA score (SOFA(max)) using the worst variables in the 24 h after the initial score measurement in emergency department (ED) patients with septic shock. METHODS: This was a retrospective observational study using a multicenter prospective registry of septic shock patients in the ED between October 2015 and December 2019. The primary outcome was in-hospital mortality. The prognostic accuracies of SOFA(ini) and SOFA(max) were evaluated using the area under the receiver operating characteristic (AUC) curve. RESULTS: A total of 4860 patients was included, and the in-hospital mortality was 22.1%. In 59.7% of patients, SOFA(max) increased compared with SOFA(ini), and the mean change of total SOFA score was 2.0 (standard deviation, 2.3). There was a significant difference in in-hospital mortality according to total SOFA score and the SOFA component scores, except cardiovascular SOFA score. The AUC of SOFA(max) (0.71; 95% confidence interval [CI], 0.69-0.72) was significantly higher than that of SOFA(ini) (AUC, 0.67; 95% CI, 0.66-0.69) in predicting in-hospital mortality. The AUCs of all scores of the six components were higher for the maximum values. CONCLUSION: The prognostic accuracy of the initial SOFA score at the time of sepsis recognition was lower than the 24-h maximal SOFA score in ED patients with septic shock. Follow-up assessments of organ failure may improve discrimination of the SOFA score for predicting mortality.