Validation of the updated sequential organ failure assessment (SOFA)-2 score in emergency department septic shock.

objectivesThe original Sequential Organ Failure Assessment (SOFA-1) score is widely used to quantify organ dysfunction in critically ill patients, including those with septic shock. A recently updated, data-driven SOFA score (SOFA-2) revises component thresholds and incorporates contemporary organ support modalities. While SOFA-2 has been validated in intensive care unit populations, its performance in the emergency department (ED), particularly among patients with septic shock, is uncertain. We therefore sought to validate SOFA-2 in ED patients with septic shock.

methodsIn this observational study, we analyzed two prospective septic shock registries. Adult ED patients with septic shock were included. SOFA-1 and SOFA-2 scores were calculated using the worst physiologic and laboratory values recorded within the first 24 h in the ED. We compared score distributions and examined mortality across score ranges. Discrimination for in-hospital, 28-day, and 90-day mortality was evaluated using the area under the receiver operating characteristic curve (AUROC). Calibration for in-hospital mortality was assessed using the calibration curve and Brier scores. Reclassification performance was evaluated using continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI).

resultsThe study included 2669 patients in Cohort A and 1443 patients in Cohort B. SOFA-1 and SOFA-2 score distributions differed significantly. Mortality generally increased with higher SOFA-1 and SOFA-2 scores across all outcomes. SOFA-2 demonstrated significantly higher AUROCs than SOFA-1 for in-hospital, 28-day, and 90-day mortality in both cohorts (Cohort A: 0.734 vs. 0.702, 0.719 vs. 0.694, and 0.680 vs. 0.658; Cohort B: 0.719 vs. 0.671, 0.737 vs. 0.689, and 0.711 vs. 0.666; all p < 0.001). SOFA-2 showed generally closer agreement between predicted and observed in-hospital mortality and lower Brier scores. Continuous NRI (0.276-0.635) and IDI (0.024-0.062) also favored SOFA-2 across all mortality outcomes (all p < 0.001). In adjusted analyses, each 1-point increase in SOFA-2 was independently associated with higher mortality. Sensitivity analyses and pooled analyses showed consistent findings.

conclusionAmong ED patients with septic shock, SOFA-2 showed modestly improved discrimination, closer agreement between predicted and observed in-hospital mortality with lower overall prediction error, and more favorable continuous reclassification metrics compared with SOFA-1. These findings support the external validation of SOFA-2 in this population and warrant further evaluation in diverse ED settings.