Accuracy and factors affecting the outcome of multi-detector computerized tomography urography for bladder tumors in...

The objective of this study was to investigate the diagnostic accuracy of multi-detector computerized tomography urography (MDCTU) for the detection of bladder tumors.

We retrospectively reviewed the medical records of 143 patients who were scanned by use of 64-channel MDCTU and who underwent cystoscopy due to painless hematuria or a clinical suspicion of bladder tumor. We examined the accuracy of MDCTU for the detection of bladder tumors by comparing the results obtained by MDCTU with those obtained by cystoscopy. The associations between tumor characteristics, frequency of transurethral resection (TUR), and bladder volume and detectability of bladder tumors on MDCTU were also analyzed.

Of 143 patients, 50 patients had a history of urothelial carcinomas. In these patients, the sensitivity and specificity of MDCTU were 60.0% and 80.0%, respectively. In 93 patients without previous urothelial carcinomas, the sensitivity and specificity of MDCTU were 86.7% and 96.8%, respectively. Falsely diagnosed cases had a smaller distended bladder volume (p=0.014) and a smaller tumor size (p=0.022) than did true diagnosed cases. The false-negative rate increased when the bladder tumor was located at the bladder neck. In the univariate analysis, the tumor location, size, frequency of TUR, bladder volume, and initial hematuria were associated with detectability by MDCTU (p<0.05).

To improve the accuracy of MDCTU for diagnosing bladder tumors, bladder filling is recommended. Thus, cystoscopy should be considered as a standard diagnostic tool for bladder tumors even in patients with normal MDCTU results, especially in the evaluation of recurrent, bladder neck-located, small, or sessile bladder tumors.