Predicting Postoperative Complications After Cholecystectomy for Acute Cholecystitis: Comparative Performance of...

Background: Although early laparoscopic cholecystectomy represents the standard treatment for acute cholecystitis [AC], reliable preoperative risk stratification remains challenging. This study compared the discriminative performance of the following five prognostic scores: two disease-specific tools (Chole-Risk and a locally modified variant, Chole-Risk mod) and three general indices (POSSUM Physiological Score, APACHE II, and Charlson Comorbidity Index [CCI]) for predicting postoperative complications [POCs] and prolonged hospital stay. Methods: This single-center retrospective study included 211 consecutive patients who underwent cholecystectomy for AC between 2015 and 2024. Primary endpoint: the occurrence of any POC. Secondary endpoint: prolonged length of stay (LOS), defined as postoperative hospitalization exceeding the 75th percentile (>6 days). Discrimination was assessed using the area under the receiver operating characteristic curve (AUC), with pairwise comparisons performed using the DeLong test. Calibration was evaluated graphically, and clinical utility was explored through decision curve analysis. Results: POC occurred in 55 patients (26.1%), and prolonged LOS in 31 (14.7%). Chole-Risk mod showed the best discrimination (AUC 0.925) and the strongest association per one-standard-deviation increase (OR 16.60; 95% CI 9.49-43.56). Other scores showed lower performance as follows: POSSUM PS (AUC 0.732), CCI (0.712), Chole-Risk (0.699), and APACHE II (0.695). For prolonged LOS, Chole-Risk mod demonstrated the highest discrimination (AUC 0.869). Decision curve analysis confirmed a net clinical benefit for Chole-Risk mod across a broad range of decision thresholds. Conclusions: The modified Chole-Risk score showed the highest discrimination among the evaluated scores for predicting adverse outcomes after cholecystectomy for acute cholecystitis in this single-center exploratory cohort. These findings suggest that incorporating disease-specific variables may improve preoperative risk stratification, although prospective multicenter validation is warranted.