Development of a scoring system from noncontrast computerized tomography measurements to improve the selection of upper ureteral stone for extracorporeal shock wave lithotripsy.

PURPOSE: We investigated the role of noncontrast computerized tomography in predicting the treatment outcome of shock wave lithotripsy on upper ureteral stones to formulate a clinical algorithm to facilitate clinical management.

MATERIALS AND METHODS: Adult patients with upper ureteral stones confirmed by noncontrast computerized tomography and scheduled for primary in situ shock wave lithotripsy were prospectively recruited. Standardized treatment was performed on each patient. The primary end point was stone-free status at 3 months. Pretreatment noncontrast computerized tomography was assessed by a single radiologist blinded to the clinical parameters. Predictive values of computerized tomography measurements on the treatment outcome were then assessed.

RESULTS: Between October 2004 and July 2007 a total of 94 patients (60 male and 34 female) were recruited for the study. Logistic regression showed that stone volume, mean stone density and skin-to-stone distance were potential predictors of successful treatment. From ROC curves the optimum cutoff for predicting treatment outcomes for stone volume, mean stone density and skin-to-stone distance was 0.2 cc, 593 HU and 9.2 cm, respectively. A simple scoring system was constructed based on the 3 factors of stone volume less than 0.2 cc, mean stone density less than 593 HU or skin-to-stone distance less than 9.2 cm. The stone-free rate for patients having 0, 1, 2 and 3 factors was 17.9%, 48.4%, 73.3% and 100%, respectively (linear-by-linear association test 22.83, p <0.001).

CONCLUSIONS: Stone volume, mean stone density and skin-to-stone distance were potential predictors of the successful treatment of upper ureteral stones with shock wave lithotripsy. A scoring system based on these 3 factors helps separate patients into outcome groups and facilitates treatment planning.