The predictive utility of the E-PASS score for postoperative complications in robot-assisted partial nephrectomy: a retrospective cohort study.

Abstract

Background: Robot-assisted partial nephrectomy (RAPN) is widely used for small renal tumors and provides favorable oncological and functional outcomes. However, a significant risk of postoperative complications remains a concern. Existing nephrometry scores focus on tumor anatomy but neglect patient-specific and intraoperative factors. The Estimation of Physiologic Ability and Surgical Stress (E-PASS) score, originally developed for gastrointestinal surgery, combines physiological and surgical parameters for the prediction of postoperative risk. This study evaluates the predictive performance of the E-PASS score in RAPN.

Methods: This observational study retrospectively analyzed 166 patients who underwent RAPN between March 2015 and September 2024. The preoperative risk score (PRS), surgical stress score (SSS), and comprehensive risk score (CRS) were calculated. Complications developing within 30 days were classified using the Clavien-Dindo system, with Grade 2 or higher events being considered significant. The performance of the CRS in predicting postoperative complications was evaluated via receiver operating characteristic (ROC) curve analysis, while model calibration was assessed by means of a calibration plot. Logistic regression was used to identify independent predictors of postoperative complication risk. Bootstrap resampling was used to estimate 95% confidence intervals.

Results: Grade 2 or higher complications occurred in 26 patients (15.7%). CRS was significantly greater in the complication group (p < 0.001). ROC analysis demonstrated good discrimination (AUC: 0.721; 95% CI: 0.629-0.812). A CRS cut-off of 0.083 yielded 76.8% sensitivity and 59.5% specificity. The calibration plot indicated good agreement between predicted and observed probabilities. Multivariate analysis identified CRS (OR: 1.537, p = 0.044), tumor size (OR: 1.048, p = 0.020), and off-clamp surgery (OR: 4.569, p = 0.003) as independent predictors.

Conclusions: E-PASS CRS reliably predicts postoperative complications in RAPN and enhances risk stratification by incorporating patient-specific surgical factors. Its integration may assist more personalized perioperative management in urological surgery.