Prospective prediction of the major component of urinary stone composition with dual-source dual-energy CT in vivo

• Chemical composition of urinary stone is important for management.
• Focusing on the major component of stones has more clinical significance.
• Prediction of stone composition by dual-source dual-energy CT (DSDECT) is feasible.
• DSDECT could predict the major component of stones with an accuracy of 97.5%.
• DSDECT could also accurately detect different types of stone composition.

AimTo prospectively evaluate the diagnostic accuracy of dual-source dual-energy computed tomography (DSDECT) for predicting the major component and determining the composition of urinary calculi in patients with urolithiasis, using postoperative in vitro Fourier transform infrared spectroscopy (FT-IR) analysis as the reference standard.Materials and methodsPatients with known urolithiasis underwent preoperative DSDECT evaluation, and subsequently, underwent surgical removal of the stones. All patients were examined using the dual-energy renal stone protocol. Material-specific chromatic images were made using dedicated post-processing software. The final determination of stone composition was made using FT-IR postoperatively. Diagnostic parameters of DSDECT for predicting the major component and detecting the presence of four composition types were calculated.ResultsA total of 81 urinary calculi were included in this study. Forty-three were pure stones and 38 were mixed stones according to FT-IR. DSDECT correctly identified the major component of all pure stones and 36 mixed stones. The major component of two mixed stones with uric acid as the major component was falsely interpreted as calcium oxalate. The overall accuracy of DSDECT for predicting the major component of stones was 97.5% (79/81). The accuracy of DSDECT for detecting the presence of four types of composition, uric acid, cysteine, hydroxyapatite, and calcium oxalate, was 97.5% (79/81), 93.8% (76/81), 80.2% (65/81), and 93.8% (76/81), respectively.ConclusionDSDECT could accurately predict the major component of urinary calculi and detect uric acid, cysteine, and calcium oxalate with a satisfactory accuracy.