Validation of the Emergency Surgery Acuity Score in Patients Who Underwent Emergency General Surgery in South Korea: A Multi-Centre, Retrospective, Post Hoc Analysis.

OBJECTIVES: This study aimed to validate the Emergency Surgery Acuity Score (ESAS) in Korean patients who underwent emergency general surgery (EGS) and evaluate its effectiveness in predicting various outcomes. DESIGN: Retrospective cohort study. SETTING: Two tertiary hospitals in South Korea. PARTICIPANTS: This study included 1629 adults who underwent EGS between July 2014 and December 2019. PRIMARY AND SECONDARY OUTCOME MEASURES: The ESAS and its abbreviated version, the Physiological Emergency Surgery Acuity Score (P-ESAS), were calculated for each patient. Predictive performance was assessed for 28-day mortality, post-operative complications, intensive care unit (ICU) admission, and hospital length of stay (LOS). C-statistics were used to compare ESAS with P-ESAS and other scoring systems, including the Charlson Comorbidity Index (CCI). Logistic regression models were employed to identify independent predictors of 28-day mortality, incorporating clinically relevant covariates such as demographic characteristics, comorbidities, physiological parameters, and surgical factors. RESULTS: The ESAS and P-ESAS demonstrated superior predictive abilities compared to the Charlson Comorbidity Index (CCI) across clinical outcomes. For 28-day mortality, the c-statistics were 0.917 for ESAS, 0.897 for P-ESAS, and 0.691 for CCI. Both ESAS and P-ESAS outperformed CCI in predicting post-operative complications (c-statistics: 0.701, 0.677, and 0.637, respectively) and ICU admission (c-statistics: 0.777, 0.737, and 0.645, respectively). Linear regression analysis revealed a significant relationship between ESAS and hospital length of stay (LOS = 2.96 × ESAS; p < 0.001, adjR2 = 0.430). Both ESAS and P-ESAS were identified as significant independent predictors of 28-day mortality in multivariable analyses. CONCLUSION: The ESAS is validated as an effective tool for predicting 28-day mortality, post-operative complications, ICU admission, and hospital LOS in Korean patients undergoing emergency general surgery. Its predictive performance compared to other scoring systems highlights its potential to improve risk stratification and resource management in emergency surgery settings.