Comparison of CROES, S.T.O.N.E, and Guy's scoring systems for the prediction of stone-free status and complication rates following percutaneous nephrolithotomy in patients with chronic kidney disease.

PURPOSE: To evaluate and compare the accuracy of S.T.O.N.E, Clinical Research Office of the Endourological Society (CROES), and Guy's stone score in patients with chronic kidney disease (CKD) following percutaneous nephrolithotomy (PNL).

METHODS: The charts of patients who had undergone a prone percutaneous nephrolithotomy from June 2006 to June 2016 were retrospectively analyzed. Patients accepted as stage 3 and higher according to chronic kidney disease epidemiology collaboration formula, were enrolled into the study. Calculation of the CROES, S.T.O.N.E, and Guy's scoring system (SS) was made as defined in original papers. Patients were categorized into four scores according to CROES, into nine scores according to S.T.O.N.E, and into four scores according to Guy's SS.

RESULTS: A total of 303 patients fulfilled the study inclusion criteria. The mean preoperative eGFR and creatinine levels were 47 mL/min and 1.55 mg/dL, respectively. In patients who were stone free and those with residual stones, the mean CROES SS was 179 and 137 (p < 0.001), the mean S.T.O.N.E score was 8.8 and 9.9 (p < 0.001), and the mean Guy's SS was 1.8 and 2.4 (p < 0.001), respectively. Multivariate regression analysis revealed CROES SS was the only scoring system, which had a predictive value for PNL outcome in patients with CKD (p = 0.011) and any of three SS were not useful for predicting PNL complications in patients with CKD.

CONCLUSION: Our study demonstrated the CROES SS was the only independent factor in the prediction of PNL outcome in CKD patients. Furthermore, three of the NSSs were not useful for predicting PNL complications in patients with CKD.