Comparative performance of five nephrolithometry scoring systems and a novel nomogram for predicting stone-free rate...

Percutaneous nephrolithotomy (PCNL) is the first-line treatment for complex pediatric renal stones, yet accurately predicting postoperative outcomes remains a clinical challenge. This study aimed to systematically compare the performance of five established nephrolithometry scoring systems-the S.T.O.N.E. (size, tract length, obstruction, number of involved calices, and essence) score, Guy's Stone Score (GSS), Seoul National University Renal Stone Complexity (S-ReSC) score, Clinical Research Office of the Endourological Society (CROES) nomogram, and Stone Kidney Size (SKS) score-and a newly developed Friendship nomogram in predicting the stone-free rate (SFR) and postoperative complications after pediatric PCNL.

We retrospectively analyzed the clinical data of 96 pediatric patients who underwent PCNL at Beijing Friendship Hospital, Capital Medical University, from July 1, 2020, to July 1, 2025. All five scores were calculated for each patient. Univariate and multivariate logistic regression analyses were used to identify factors associated with SFR and complications. A novel nomogram, termed the Friendship nomogram, was constructed based on independent predictors of SFR. The predictive performance of all models was evaluated and compared using the area under the receiver operating characteristic curve (AUC).

Among the 96 patients, the overall SFR was 77.08%, and the postoperative complication rate was 43.75%. For predicting SFR, all scoring systems demonstrated good predictive value, with AUCs from highest to lowest as follows: the CROES nomogram (0.882), the Friendship nomogram (0.878), the SKS score (0.843), the S-ReSC score (0.841), the GSS (0.822), and the S.T.O.N.E. score (0.775). Multivariate analysis confirmed that age [odds ratio (OR) =0.83, 95% confidence interval (CI): 0.71-0.97, P=0.02], stone burden (OR =0.99, 95% CI: 0.99-0.99, P=0.04), and the Stone-to-Kidney Index (SKI) (OR =0.15, 95% CI: 0.04-0.59, P=0.006) were independent predictors of SFR, forming the basis of the Friendship nomogram. For predicting complications, only the S.T.O.N.E. score (AUC =0.623, P=0.03) and the degree of hydronephrosis (AUC =0.672, P=0.02) showed weak predictive power, while all other systems had no significant predictive value.

All evaluated nephrolithometry scoring systems can effectively predict postoperative SFR in pediatric PCNL, with the CROES nomogram demonstrating the highest performance. Although our novel Friendship nomogram is slightly less accurate than the CROES model, its advantage lies in its simplicity, achieving comparable predictive accuracy with only three common clinical variables (age, stone burden, and SKI), making it more practical for rapid clinical assessment. However, none of the currently analyzed scoring systems can accurately predict postoperative complications.