MDCT urography of upper tract urothelial neoplasms.
OBJECTIVE: The purpose of our study was to review the MDCT urography appearance of pathologically proven transitional cell carcinomas of the renal collecting system and ureter and to correlate the MDCT urography findings with pathology findings. MATERIALS AND METHODS: Of 370 MDCT urography examinations performed over an 18-month period, 18 patients were diagnosed with 27 renal collecting system or ureteral urothelial neoplasms at endoscopic biopsy (n = 8) or surgery (n = 19). Initial MDCT reports were reviewed to determine the sensitivity of original reviewers in detecting these neoplasms. Two radiologists also retrospectively reviewed these scans and characterized the CT appearance of the neoplasms on both axial CT and 3D reformatted images. Findings at retrospective review were correlated with pathology results to determine whether any CT features could be used to predict tumor grade. RESULTS: Eighteen of 27 neoplasms were prospectively identified on MDCT urography, and an additional six neoplasms were detected on retrospective review. Three ureteral neoplasms could not be visualized. The 24 retrospectively detected neoplasms had three distinct MDCT appearances: circumferential urothelial wall thickening (n = 14), small masses (> 5 mm in maximal diameter) (n = 5), and large masses (> 5 mm in maximal diameter) (n = 5). All detected lesions could be seen on axial excretory phase images provided wide window settings were reviewed; however, only six were detected on 3D reconstructions. MDCT urography appearance did not correlate with tumor grade. CONCLUSION: MDCT urography is a promising technique for detecting upper urinary tract neoplasms. The static 3D reconstructions used in this study are insufficient for visualization. Axial image review remains essential for tumor identification.
Caoili, EM; Cohan, RH; Inampudi, P; Ellis, JH; Shah, RB; Faerber, GJ; Montie, JE
Volume / Issue
Start / End Page
Pubmed Central ID
International Standard Serial Number (ISSN)
Digital Object Identifier (DOI)