Contrast-enhanced gray-scale and color Doppler voiding urosonography versus voiding cystourethrography in the diagnosis and grading of vesicoureteralreflux

Purpose. The purpose of this study was to compare contrast-enhanced gray-sc ale voiding urosonography (CE-VUS) and contrast-enhanced color Doppler void ing urosonography (CE-CDVUS) with voiding cystourethrography (VCUG) to veri fy whether the use of color Doppler imaging improves the diagnosis and grad ing of vesicoureteral reflux (VUR). Methods. In 74 patients, CE-VUS and CE-CDVUS were compared with VCUG, which was used as the gold standard. SHU 508 A (Levovist) was used as the echo-e nhancing contrast agent. VUR was diagnosed if hyperechoic dots or color sig nals were visualized in the ureter on sonograms. VUR grading was based on m orphologic and dynamic findings on CE-VUS and morphologic and color finding s on CE-CDVUS. VCUG was performed conventionally, and grading by VCUG was i n accordance with the international system of radiographic VUR grading. Pat ients who voided during 1 examination only (either CE-VUS and CE-CDVUS or V CUG) were excluded from the study. Agreement between the results of CE-VUS and VCUG and between those of CE-CDVUS and VCUG in diagnosing VUR was calcu lated by kappa statistics. CE-VUS and CE-CDVUS were compared for diagnostic accuracy by the McNemar test. Results. The agreement between CE-VUS and VCUG in predicting VUR was 90% (k appa score, 0.77; p < 0.001). The agreement between CE-CDVUS and VCUG was 9 6% (<kappa> score, 0.91; p < 0.001). CE-CDVUS showed a significantly higher diagnostic accuracy than did CE-VUS (96% Versus 90% of cases correctly cla ssified; McNemar <chi>(2) = 4; P < 0.05). This was mainly related to the lo wer number of false-negative results for grade I and grade II VUR when CE-C DVUS was used. Conclusions. The use of color Doppler imaging significantly improves the ac curacy of contrast voiding urosonography in the detection and grading of VU R. (C) 2001 John Wiley & Sons, Inc. J Clin Ultrasound 29:65-71, 2001.