Objective: The purpose of this study was to compare the accuracy of CT urography and excretory urography for the detection and localization of upper urinary tract urothelial carcinoma. Materials and Methods: Of 128 patients at high risk for upper tract urothelial carcinoma who were examined with both CT urography and excretory urography between 2002 and 2007, 24 were undiagnosed and excluded. CT urography and excretory urography results of the remaining 104 patients and 552 urinary tract segments were compared with histopathologic examination or follow-up imaging at 1 year. Two readers independently scored the confidence levels for the presence or absence of upper urinary tract urothelial carcinoma in each of six upper urinary tract segments on both CT urography and excretory urography; differences were resolved by consensus. Results: Upper urinary tract urothelial carcinoma was diagnosed in 77 (14%) segments of 46 (44%) patients. Per-patient sensitivity, specificity, overall accuracy, and area under the receiver operating characteristic curves for detecting carcinomas with CT urography (93.5% [43/46], 94.8% [55/58], 94.2% [98/104], and 0.963, respectively) were significantly greater than those for excretory urography (80.4% [37/46], 81.0% [47/58], 80.8% [84/104], and 0.831, respectively) (p = 0.041, p = 0.027, p = 0.001, and p < 0.001, respectively). Persegment sensitivity and overall accuracy for the localization of upper urinary tract urothelial carcinoma were significantly greater with CT urography (87.0% [67/77] and 97.8% [540/552]) than with excretory urography (41.6% [32/77] and 91.5% [505/552]) (p < 0.0001). Conclusion: CT urography was more accurate than excretory urography in the detection and localization of upper urinary tract urothelial carcinoma and should be considered as the initial examination for the evaluation of patients at high risk for upper urinary tract urothelial carcinoma.
ASJC Scopus subject areas