Efficacy, Safety, and Recurrence Prediction of Minimally Invasive Percutaneous Nephrolithotomy in Uyghur and Han Pediatric Patients with Upper Urinary Tract Calculi

Introduction: The aim of the study was to evaluate the efficacy, safety, and recurrence risk of minimally invasive percutaneous nephrolithotomy (PCNL) in Uyghur and Han pediatric patients with upper urinary tract calculi and develop predictive models with nomograms for postoperative recurrence. Methods: Clinical data from 231 children (159 Uyghur, 72 Han) treated with Mini-PCNL or Microperc between June 2019 and June 2024 were retrospectively reviewed. Recurrence-free survival was analyzed using Kaplan-Meier estimates. Cox regression was applied to construct two models: Model 1 (Outpatient Clinical Predictive Model, OCPM) based on clinical variables, and Model 2 (Comprehensive Inpatient Predictive Model, CIPM) incorporating imaging and urine culture. Nomograms were established, and model performance was assessed using time-dependent ROC curves, calibration curves, and 10-fold cross-validation. Clinical utility was evaluated by decision curve analysis and clinical impact curves. Results: Stone-free rates (SFRs) and complication rates (CRs) were comparable between groups, but Uyghur children had a higher recurrence rate (25.2% vs. 13.9%, HR = 2.31, 95% CI: 1.30–4.10). Model 1 (OCPM, age, ethnicity, urine pH, serum creatinine) showed good discrimination for outpatient screening (AUC 0.79–0.85), while Model 2 (CIPM, adding stone size, laterality, multiplicity, and hydronephrosis) achieved superior accuracy (AUC >0.90) and calibration at higher risk levels. Conclusion: PCNL is effective and safe in both ethnic groups, but Uyghur patients have a higher recurrence risk. The predictive models provide valuable tools for optimizing postoperative management and follow-up strategies.