Predictive ability of Guy's stone score in pediatric patients undergoing percutaneous nephrolithotomy.

ABSTRACT

BACKGROUND: Several authors have evaluated, in a number of external validation and predominantly adult studies, Guy's stone score (GSS) as a predictive tool for the assessment of stone clearance after percutaneous nephrolithotomy (PCNL). However, there are limited and conflicting data investigating whether GSS could be a potential independent factor associated with residual stone rates and complications of PCNL for children. OBJECTIVE: The aim of the current study was to evaluate the ability of GSS, as an independent factor, to predict residual stone rates and complications in pediatric patients undergoing PCNL. STUDY DESIGN: The records of the pediatric patients who had undergone PCNL for stone disease at the current department were retrospectively reviewed. Variables included patient, stone, and treatment parameters. The GSS was used for assessment of stone complexity. Univariate and multivariate analyses were performed to evaluate factors associated with residual stone rates and complications of pediatric PCNL. RESULTS: The study group consisted of 114 children (63 boys and 51 girls) with a median (interquartile range) age of 6 (3-11) years and who had undergone 122 PCNLs (eight bilateral). Multivariate logistic regression analysis (Summary Table) demonstrated that the independent risk factors for increased residual stone rate following pediatric PCNL monotherapy were GSS (OR 3.14, 95% CI 1.31-7.49, P = 0.01) and number of calyces involved (OR 4.02, 95% CI 1.11-14.53, P = 0.033). Multivariate logistic regression analysis demonstrated that the independent risk factors for complications following pediatric PCNL were surgical experience (OR 2.72, 95% CI 1.09-6.82, P = 0.032), and operation time (OR 1.01, 95% CI 1.00-1.03, P = 0.040). DISCUSSION: A prediction model is needed to pre-operatively counsel patients and evaluate outcomes for measuring treatment success of PCNL. The GSS provides a simple, reliable and reproducible tool for describing the complexity of PCNL while predicting postoperative stone-free status, despite its limitations. However, little research has been conducted on the applicability of GSS in pediatric patients undergoing PCNL. In addition, conflicting results have been reported about its association with complications of pediatric PCNL. CONCLUSIONS: The current study found that although GSS had a significant predictive ability for residual stone rates after pediatric PCNL revealed by multivariate logistic regression, the results did not conclude that complexity of GSS was directly proportional to complications of pediatric PCNL. The factors, including calyceal involvement and surgeon experience, may need to be incorporated in GSS to further improve its ability to predict outcomes of pediatric PCNL. IRB APPROVED PROTOCOL NUMBER 1422.