The POSSUM: a good scoring system for predicting mortality in elderly patients undergoing emergency surgery?

• Published in: The Pan African medical journal Vol. 24; p. 166
• Main Authors: Mzoughi, Zeineb, Bayar, Rached, Djebbi, Achref, Talbi, Ghofrane, Romdhane, Hayfa, Aloui, Wafa, Lassaad, Gharbi, Khalfallah, Mohamed Taher
• Format: Journal Article
• Language: French
• Published: Uganda 2016
• Subjects: Aged; Case-Control Studies; Digestive System Surgical Procedures - mortality; Emergencies; Female; Humans; Male; Postoperative Complications - mortality; Reproducibility of Results; Retrospective Studies; ROC Curve; Sensitivity and Specificity; mortality; score; emergency; POSSUM; elderly; surgery
• ISSN: 1937-8688
• DOI: 10.11604/pamj.2016.24.166.9528

POSSUM (Physiologic and Operative Severity Score for the enumeration of Mortality and morbidity) is a scoring system for predicting mortality which is largely used in elective aortic and abdominal surgery. The aim of our study was to validate POSSUM with regards to elderly patients (> 70 years) undergoing digestive emergency surgery. We wanted to determine the optimal threshold for POSSUM, consisting of a physiological score and of an operative score for predicting mortality in this population. This is a case-control analytic retrospective study of 291 patients aged ≥70 years undergoing digestive emergency surgery. These patients were divided into two groups each comprising 50 patients. Group "DC": patients died in the immediate postoperative period and Control group "SURV" chosen by lot. We analyzed the reliability of POSSUM in predicting mortality and morbidity. Subsequently, we created ROC curve to find the thresholds with the best sensitivity/specificity couple. The physiological score, operative score and mortality and morbidity rates predicted by POSSUM and mortality predicted by P-POSSUM are predictors of mortality (P < 0.0001). The threshold values for the physiological and operative score with the best sensitivity/specificity couple were 23 and 15 respectively. Predicting mortality in patients allows to target care management programs and to inform the patient and his family of the risks.