Retrograde intrarenal surgery (RIRS) is a widely used treatment for renal calculi. However, prolonged operative time (≥90 minutes) is associated with increased complications and healthcare resource utilization. While several predictive models exist, including STONE Score, Guy’s Stone Score, RUSS Score, and Modified STONE Score, their complexity limits clinical utility. This study aims to identify key predictors of prolonged operative time and determine whether a simplified model can provide comparable predictive performance while improving clinical practicality.

This retrospective study analyzed RIRS patients, with prolonged operative time (≥90 minutes) as the primary outcome. Statistical analyses included: 1. Univariate Analysis: Comparison of patient demographics, stone characteristics, and anatomical factors between operative time groups (<90 min vs. ≥90 min). 2. Multivariable Logistic Regression: Identification of independent predictors of prolonged operative time. 3. Receiver Operating Characteristic (ROC) Curve Analysis: Evaluation of the predictive performance of individual parameters and established scoring systems. 4. Cut-off Determination: Using Youden’s J index to define optimal cut-off values for significant predictors. 5. Development of a Simplified Risk Score: Constructing a clinically relevant predictive model based on key variables.

Among 142 patients, univariate analysis showed significant differences in stone size (p < 0.001), stone number (p = 0.003), and Inferior Pole Angle (IPA) (p < 0.001) between groups. Multivariable analysis confirmed stone size (OR = 1.34, p = 0.029) and IPA (OR = 0.91, p < 0.001) as independent predictors. Model Development A simplified RIRS risk score was constructed using two key variables: • Stone size (≥16mm) • IPA (≤60°) Patients were classified into three risk groups: • Low risk (0-2 points): Suitable for RIRS. • Moderate risk (3-4 points): RIRS feasible with experienced surgeons. • High risk (5-6 points): High likelihood of prolonged operative time; consider PCNL. Predictive Performance The simplified model (stone size & IPA) achieved an AUC of 0.792, with sensitivity 69.0%, specificity 84.5%, PPV 81.7%, and NPV 73.2%. It performed comparably to STONE Score (AUC = 0.809) and Modified STONE Score (AUC = 0.807), and slightly lower than RUSS Score (AUC = 0.868). Validation The simplified RIRS risk score exhibited a significant correlation with operative time (p < 0.001). The mean operative time for each risk group was: • Low-risk group (0-2 points): 72.5 ± 18.3 minutes • Moderate-risk group (3-4 points): 98.4 ± 22.1 minutes • High-risk group (5-6 points): 124.2 ± 30.7 minutes

This simplified model provides a practical and effective tool for predicting prolonged RIRS operative time, which can guide surgical decision between RIRS and PCNL. Further validation across diverse populations is recommended.