Multicenter External Validation and Comparison of Stone Scoring Systems in Predicting Outcomes After Percutaneous...

Area of Science:

• Urology
• Nephrology
• Surgical Outcomes

Background:

• Percutaneous nephrolithotomy (PCNL) is a key procedure for kidney stone removal.
• Predictive scoring systems are crucial for estimating outcomes like stone-free rates (SFR) and complications.
• Several systems, including Guy's stone score, S.T.O.N.E. nephrolithometry, and CROES nomogram, are used to predict PCNL results.

Purpose of the Study:

• To compare the predictive accuracy of commonly used scoring systems for PCNL outcomes.
• To assess the ability of Guy's stone score, S.T.O.N.E. nephrolithometry, and CROES nomogram to predict SFR, operative time (OT), complications, and length of stay (LOS).
• To develop risk group stratification based on these scoring systems for improved patient counseling.

Main Methods:

• Retrospective review of 586 patients undergoing PCNL at four academic institutions (2006-2013).
• Primary outcome: SFR within 3 weeks. Secondary outcomes: OT, complications, LOS.
• Statistical analysis included logistic, linear, and Poisson regressions, and ROC curve analysis with AUC calculation.

Main Results:

• Overall SFR was 67.4%. All three scoring systems (Guy's, S.T.O.N.E., CROES) predicted SFR with similar accuracy (AUCs: 0.629, 0.671, 0.646).
• S.T.O.N.E. nephrolithometry was the only independent predictor of longer operative time (p<0.001).
• None of the scores independently predicted postoperative complications or longer LOS. S.T.O.N.E. and CROES nomogram demonstrated distinct risk group stratification.

Conclusions:

• Guy's stone score, S.T.O.N.E. nephrolithometry, and CROES nomogram exhibit comparable predictive accuracy for SFR after PCNL.
• The S.T.O.N.E. score offers additional value by predicting operative time.
• Risk stratification using these scores can enhance patient counseling, though further research is needed to determine superiority in clinical utility.