ABSTRACT
BACKGROUND: Three-phase CT urography (CTU) is the gold standard for evaluating the upper urinary tract in patients with hematuria. We aimed to evaluate the accuracy of CTU for detecting upper urothelial cell carcinomas (UCC) in patients with hematuria and negative cystoscopy. Secondly, we aimed to determine the tumor visibility on each CTU phase. MATERIAL AND METHODS: This retrospective study included all patients with hematuria referred to CTU after a negative cystoscopy during 2016 and 2017. The original CTU reports were dichotomized as negative or positive. All patient charts were reviewed after a minimum of 18-month follow-up in order to register missed cancers. The results of biopsies and clinical follow-up were used as the reference standard. Two reviewers retrospectively evaluated the tumor visibility of each CT sequence in all true-positive CTUs. RESULTS: We included 376 patients with hematuria who underwent CTU after a negative cystoscopy. Macroscopic and microscopic hematuria occurred in 87% (327) and 13% (49), respectively. The incidence of upper urothelial cell carcinoma was 1.9% (7), and the sensitivity of CTU was 100% (95% CI, 59-100), specificity was 99% (95% CI, 98-100), positive predictive value was 88% (95% CI, 47-99), and negative predictive value was 100% (95% CI, 99-100). The accuracy was 99% (95% CI, 90-100). All UCCs were visible on the nephrographic phase for both reviewers. CONCLUSION: CTU is highly accurate for detecting upper UCCs. All cases were seen on the nephrographic phase. This suggests that the CTU protocol can be simplified. KEY POINTS: ⢠CT urography is highly accurate for detecting upper urothelial cell carcinomas. ⢠All cancers were seen on the nephrographic phase. ⢠All cancers were detected in patients with macroscopic hematuria.