Correlation between S.T.O.N.E score and channel size selection in percutaneous nephrolithotomy

To evaluate the predictive value and influence of the S.T.O.N.E. nephrolithiasis scoring system on access tract size selection during percutaneous nephrolithotomy (PCNL).

A retrospective analysis was performed on the clinical data of 323 patients who underwent PCNL. Preoperative stones were assessed using the S.T.O.N.E. score, which was categorized into a high-score group (9-13) and a low-score group (5-8). Postoperative complications were graded using the Clavien classification system and divided into a mild group (no or Grade I complications) and a severe group (Grade II or higher complications). Stratified univariate analysis was conducted to investigate the relationship between access tract size selection during PCNL and postoperative outcomes, including the stone-free rate (SFR) and Clavien complication grades, in both high- and low-score groups.

Based on preoperative S.T.O.N.E. scores, 323 patients were classified into a high-score group (n = 110) and a low-score group (n = 213). In the high-score group, the SFR of the standard-tract (24F) PCNL (SPCNL) [72.6% (45/62)] was significantly higher than that of the mini-tract (18F) PCNL (MPCNL) [33.3% (16/48)]. The mean operative time for SPCNL [112.5 (104.5-123.0) min] was shorter than for MPCNL [134.0 (118.8-137.0) min] (P < 0.05). However, the proportion of severe complications was comparable between SPCNL [56.5% (35/62)] and MPCNL [52.1% (25/48)] (P = 0.648). In the low-score group, the SFR and operative time did not differ significantly between SPCNL [92.0% (80/87), 90.0 (70.0-118.0) min] and MPCNL [86.5% (109/126), 85.0 (64.0-109.3) min] (P > 0.05). However, SPCNL had a higher rate of severe complications [28.7% (25/87)] than MPCNL [12.7% (16/126)] (P < 0.05). Stepwise and multivariate logistic regression analyses identified the S.T.O.N.E. score (OR = 3.15, 95% CI 1.60-6.21, P < 0.001) and operative time (OR = 1.04, 95% CI 1.02-1.06, P < 0.001) as independent risk factors for stone residuals, whereas tract size (OR = 2.88, 95% CI 1.52-5.48, P = 0.001) served as an independent protective factor. The ROC curve demonstrated that the predictive model performed significantly better than any single independent factor (AUC = 0.836).

The S.T.O.N.E. score was significantly associated with tract size selection in PCNL. Among patients with high S.T.O.N.E. scores, SPCNL yielded higher stone-free rates and shorter operative times with similar safety compared to MPCNL, supporting the use of standard tracts in this group. In low-score patients, SPCNL and MPCNL demonstrated comparable efficacy, whereas MPCNL resulted in fewer complications. Multivariable analysis identified the S.T.O.N.E. score and operative time as independent risk factors for residual stones, while larger tract size was an independent protective factor. These findings suggest that the S.T.O.N.E. score may guide personalized tract size selection in PCNL.