Comparison of Scoring Systems in Predicting Success of Percutaneous Nephrolithotomy.

BACKGROUND: Scoring systems (SS) are useful to inform the patients about the success and complication rates of the operation prior the surgery.

AIMS: We aimed to determine the applicability of the popular SS (Guy's, S.T.O.N.E and CROES) by means of examining preoperative data of patients treated with percutaneous nephrolithotomy (PNL).

STUDY DESIGN: Cross sectional.

METHODS: We retrospectively reviewed files of the patients who had undergone PNL in our center between 2011 and 2015. Excluded from the study were patients aged < 18 years, and those who were not assessed preoperatively with computed tomography (CT). Preoperative CT images of all patients were assessed by a single observer (MT), and patients were graded based on three SS. Demographic data were analyzed along with perioperative data (operation, fluoroscopy, length of hospital stay, changes in hematocrit values, location, and number of access sites, stone-free and complication rates).

RESULTS: A total of 298 patients who had been treated with 300 procedures were enrolled into the study. Mean age, stone burden, number of stones, and density were 48.1±12.9 years, 663.5±442.8mm2, 1.8±1.1 and 888.3±273 HU respectively. Scores of the cases based on Guy's, S.T.O.N.E, and CROES SS were calculated as 2, 7.6, and 222,1 points respectively. 81.6% of the patients were stone-free. Complications were detected in 30 (9.9%) patients. Based on ROC curve analysis a positive correlation was detected between success rate and scoring systems, i.e., Guy's (p:<0.001, r:-0.309), S.T.O.N.E. (p:<0.001, r:-0.295), and CROES (p:<0.001, r:0.426). The CROES scoring system had the highest predictive value. The sensitivity rates rates for Guy's, Croes and Stone SS were as 78.78%, 80% and 82.34% respectively.

CONCLUSION: Our results show that all of SS predicted correctly the success of the PNL procedures. The CROES scoring system had the highest predictive value.