Physiological Emergency Surgery Acuity Score Predicts Mortality and Anastomotic Leak in Emergency Laparotomy: An External Validation Study

Le résumé fourni par la source

BACKGROUND: The Physiological Emergency Surgery Acuity Score (PESAS) predicts 30-day mortality but has not been externally validated in emergency general surgery (EGS) patients. The study aim was to evaluate PESAS discrimination for predicting mortality and anastomotic leak in patients receiving emergency laparotomy. METHODS: We retrospectively evaluated PESAS in a single-center cohort of adults receiving exploratory laparotomy for emergent, nontraumatic indications. PESAS and 30-day mortality were calculated for all patients. Discrimination of PESAS for 30-day mortality was assessed using area under receiver operator curve (AUROC). Secondary analysis was conducted in similar fashion for the outcome of anastomotic leak requiring reoperation. RESULTS: In 1086 patients with 160 cases of 30-day mortality (14.7%), PESAS showed strong discrimination for the primary outcome (AUROC 0.86; 95% CI 0.83-0.88) with acceptable calibration-in-the-large. In an exploratory secondary analysis of 414 patients who received bowel anastomosis with 21 cases of leak (5.1%), PESAS also showed good discrimination for anastomotic leak (AUROC 0.82; 95% CI 0.73-0.90). Observed mortality rates showed clear inflection points at PESAS values of 5 and 10, respectively, corresponding to mortality exceeding 10% and 50%. CONCLUSIONS: PESAS demonstrates good discrimination, calibration-in-the-large, and net benefit as an ordinal risk stratification tool for prediction of 30-day mortality among patients undergoing emergency laparotomy, with exploratory analysis suggesting promise for predicting anastomotic leak. Because it is derived entirely from objective physiologic variables, it may be well-suited for automated risk stratification in EGS populations. Further patient-level calibration is needed to justify expanded use.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.