Factors affecting operative time in ureteroscopic stone removal: A cross-sectional study for prediction model construction

Objective:​ To identify independent factors influencing operative time and develop a preoperative prediction model for patients achieving stone-free status after a single ureteroscopic stone removal procedure, supporting clinical decision-making and mitigating risks associated with prolonged surgery. Methods:​ In an observational cohort study, 495 patients undergoing ureteroscopic lithotripsy at Shandong Provincial Third Hospital (December 2023–December 2024) were included. Preoperative data included demographics, imaging features, and laboratory results. Variable selection used univariate analysis followed by LASSO regression. A Gamma regression model (Generalized Linear Model framework) served as the primary prediction tool, assessed via variance inflation factor (VIF) and residual analysis. Internal validation employed 1000 bootstrap samples; a nomogram was created for clinical use. A logistic regression model using dichotomized operative time was evaluated by ROC curve and Decision Curve Analysis (DCA). Results:​ Among 14 significant variables, three key predictors emerged: stone length and maximum stone density (risk factors), and simple ureteral stone (protective factor). The Gamma model showed no multicollinearity (VIFs < 2) and was robust on bootstrap validation (MSE: 20.2, RMSE: 26.9). Despite slight overestimation for very short procedures, predicted and observed operative times correlated significantly (p < 0.05). The logistic model demonstrated strong discrimination (AUC = 0.879) and favorable clinical utility on DCA. Conclusion:​ Stone length and density independently predict longer operative time, while simple ureteralstones shorten it. The Gamma and logistic regression models provide reliable, clinically applicable tools for preoperative planning, potentially improving resource allocation and patient safety.