External validation of Modified Seoul National University Renal Stone Complexity Score to predict outcome and complications of retrograde intrarenal surgery: a RIRSearch Group study

Abstract

Introduction The Modified Seoul National University Renal Stone Complexity Score (S-ReSC) is a simple model based solely on stone location regardless of stone burden. The aims of this study were to validate S-ReSC for outcomes and complications of retrograde intrarenal surgery (RIRS) and to evaluate its predictive power against the stone burden. Material and methods Data of 1007 patients with kidney stones who had undergone RIRS were collected from our RIRSearch database. Linear-by-linear association, logistic regression, ANOVA/post hoc analysis and ROC curve (with Hanley and McNeil's test) were used for evaluation. The main outcomes were stone-free status and complications of RIRS. Results The overall stone-free rate was 76.8% (773/1007). Higher S-ReSC scores were related to lower stone-free rates and higher total, perioperative and postoperative complication rates (p<.001, p<.001, p=.008 and p<.001, respectively). S-ReSC score (p=.02) and stone burden (p<.001) were independent predictors of stone-free status. But stone burden (AUC = 0.718) had a more powerful discriminating ability than the S-ReSC score (AUC = 0.618). Conclusions The S-ReSC score is able to predict not only stone-free status but also complications of RIRS. Although this location-only based scoring system has a fair discriminative ability, stone burden is a more powerful predictor of stone-free status after RIRS. An ideal scoring system aiming to predict outcomes of RIRS must include stone burden as a parameter.