Investigation of Multisequence Magnetic Resonance Imaging for Detection of Recurrent Tumor after Transurethral Resection for Bladder Cancer

Andrew B. Rosenkrantz, Islamiat O. Ego-Osuala, Victoria Khalef, Fang Ming Deng, Samir S. Taneja, William C. Huang

Research output: Contribution to journalArticle

Abstract

Purpose The aim of this study was to evaluate multisequence magnetic resonance imaging (MRI) in detecting local recurrence after transurethral resection for bladder cancer. Methods Thirty-six patients with bladder cancer with previous transurethral resection underwent bladder MRI incorporating T2-weighted imaging, diffusion-weighted imaging, and delayed contrast-enhanced T1-weighted imaging, followed by cystoscopy. Two radiologists (R1 and R2) evaluated examinations for suspicious findings. Results Forty-seven percent of patients had recurrent tumor at cystoscopy and biopsy. Using multisequence MRI, sensitivity and specificity were 67% and 81% for R1 and 73% and 62% for R2. Both readers missed 1 high-grade pathologic stage T1 recurrent tumor; otherwise, all missed tumors were low-grade pathologic stage Ta lesions. All false positives for R1 and 7 of 9 false positives for R2 were in patients receiving previous bacillus Calmette-Guerin therapy. Furthermore, 40% to 50% of solitary abnormalities and 83% to 100% of multifocal abnormalities were tumor recurrences; 12% to 20% of smooth wall thickening, 50% to 75% of irregular wall thickening, and 88% to 100% of papillary masses were tumor recurrences. Conclusions Although multisequence MRI exhibited moderate performance for detecting recurrent tumor, nearly all missed tumors were low grade and noninvasive.

Original languageEnglish (US)
Pages (from-to)201-205
Number of pages5
JournalJournal of computer assisted tomography
Volume40
Issue number2
DOIs
StatePublished - Mar 1 2016

    Fingerprint

Keywords

  • MRI
  • bladder cancer
  • contrast-enhanced imaging
  • cystoscopy
  • diffusion-weighted imaging

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this