Differentiating Calcium Oxalate and Hydroxyapatite Stones In Vivo Using Dual-Energy CT and Urine Supersaturation and pH Values

Abstract

Knowledge of urinary stone composition can guide therapeutic intervention for patients with calcium oxalate (CaOx) or hydroxyapatite (HA) stones. In this study, we determined the accuracy of noninvasive differentiation of these two stone types using dual-energy CT (DECT) and urine supersaturation (SS) and pH values. Patients who underwent clinically indicated DECT scanning for stone disease and subsequent surgical intervention were enrolled. Stone composition was determined using infrared spectroscopy. DECT images were processed using custom-developed software that evaluated the ratio of CT numbers between low- and high-energy images. Clinical information, including patient age, gender, and urine pH and supersaturation profile, was obtained from electronic medical records. Simple and multiple logistic regressions were used to determine if the ratio of CT numbers could discriminate CaOx from HA stones alone or in conjunction with urine supersaturation and pH. Urinary stones (CaOx n = 43, HA n = 18) from 61 patients were included in this study. In a univariate model, DECT data, urine SS-HA, and urine pH had an area under the receiver operating characteristic curve of 0.78 (95% confidence interval [CI] 0.66-0.91, P = .016), 0.76 (95% CI 0.61-0.91, P = .003), and 0.60 (95% CI 0.44-0.75, P = .20), respectively, for predicting stone composition. The combination of CT data and the urinary SS-HA had an area under the receiver operating characteristic curve of 0.79 (95% CI 0.66-0.92, P = .007) for correctly differentiating these two stone types. DECT differentiated between CaOx and HA stones similarly to SS-HA, whereas pH was a poor discriminator. The combination of DECT and urine SS or pH data did not improve this performance.