A novel nomogram for predicting stone-free status after retrograde intrarenal surgery in patients with kidney stones.

BACKGROUND: This study aimed to develop and validate a novel preoperative nomogram to predict stone-free status (SFS) in patients undergoing retrograde intrarenal surgery (RIRS) for kidney stones. METHODS: A total of 220 patients who underwent RIRS were retrospectively analyzed. Feature selection was performed using Boruta and LASSO algorithms, identifying six key preoperative predictors: inferior pole stone (classified by RIPA), calyx pelvic height (CPH), number of stones, maximum stone diameter, stone volume, and mean stone density. A nomogram was constructed using multivariable logistic regression and evaluated by receiver operating characteristic (ROC) analysis, calibration curves, decision curve analysis (DCA), and clinical impact curves (CIC). Internal validation was performed with 1,000 bootstrap resamples. RESULTS: The nomogram demonstrated strong discriminative ability with an AUC of 0.873 in the training cohort and 0.862 in the validation cohort. Calibration plots showed good agreement between predicted and observed outcomes. DCA and CIC confirmed its superior clinical utility across a range of threshold probabilities. Notably, inferior pole stones with RIPA ≤ 30° and higher CPH were strongly associated with SFS failure. Compared to existing scoring systems, the new model achieved better predictive performance. Including both stone volume and maximum diameter offered a more accurate assessment of stone burden than using either metric alone. CONCLUSION: We developed and internally validated a nomogram that outperformed existing tools in predicting SFS after RIRS. It may assist clinicians in individualized risk assessment and preoperative planning. As this predictive model was developed based on a single-centre study, it will be necessary to conduct further multicentre or prospective studies in the future.