External Validation of CROES Nephrolithometry as a Preoperative Predictive System for Percutaneous Nephrolithotomy Outcomes

Abstract

PURPOSE: We externally validated CROES (Clinical Research Office of the Endourological Society) nephrolithometry and evaluated the predictive accuracy of the nomogram. MATERIALS AND METHODS: Data were collected on patients who underwent percutaneous nephrolithotomy between January 2012 and February 2015. The CROES nomogram was applied to all patients and externally validated. The AUC and calibration plot were used for discrimination and clinical validity assessment. RESULTS: A total of 176 patients were included in study. Mean ± SD patient age was 55.2 ± 13.9 years and the mean stone burden was 640.0 ± 911.4 mm(2). The CROES nomogram was significantly associated with stone number, location and burden, and the number of implicated calyces, punctures and tracts. The postoperative treatment success rate was 85.8%. The number of stones, number of implicated calyces and CROES score were independent predictors of treatment success. The estimated AUC was 0.715 and the model provided good calibration. CONCLUSIONS: The CROES nomogram is an accurate tool to estimate renal stone complexity. CROES nephrolithometry provides great accuracy to predict postoperative efficacy.