Abstract
BACKGROUND: This retrospective study aimed to assess the suitability of POSSUM and its modified versions, E-PASS and its modified score, SRS, and SORT scores for predicting postoperative complications and mortality in patients undergoing laparoscopic radical gastrectomy for gastric cancer. MATERIALS AND METHODS: Data analysis was performed on 349 patients who underwent laparoscopic radical gastrectomy at Tianjin Medical University General Hospital between January 2016 and December 2021. The discriminative ability of the scoring systems was evaluated using the area under the receiver operating characteristic curve (AUC). The primary endpoint focused on the prediction of postoperative complications, while the secondary endpoint assessed the prediction of postoperative mortality. RESULTS: Among the scoring systems evaluated, the modified E-PASS (mE-PASS) score exhibited the highest AUC (0.846) and demonstrated the highest sensitivity (81%) and specificity (79%) for predicting postoperative complications. All other scores, except for POSSUM, showed moderate discriminative ability in predicting complications. In terms of predicting postoperative mortality, the E-PASS score had the highest AUC (0.978), while the mE-PASS score displayed the highest sensitivity (76%) and specificity (90%). Notably, both E-PASS and mE-PASS scores exhibited excellent discriminative ability. CONCLUSIONS: The P-POSSUM, O-POSSUM, E-PASS, mE-PASS, SRS, and SORT scoring systems are useful tools for predicting postoperative outcomes in laparoscopic radical gastrectomy. Among them, the mE-PASS score demonstrated the best predictive power. However, the POSSUM system could only be applicable to predict postoperative mortality.