Can MRI rule out bladder and rectal invasion in cervical cancer to help select patients for limited EUA?

• Published in: Gynecologic oncology 2006-05, Vol.101 (2), p.244-249
• Main Authors: Rockall, A.G., Ghosh, S., Alexander-Sefre, F., Babar, S., Younis, M.T.S., Naz, S., Jacobs, I.J., Reznek, R.H.
• Format: Article
• Language: English
• Subjects: Adult; Aged; Aged, 80 and over; Bladder invasion; Cervical cancer; Female; FIGO staging; Humans; Magnetic Resonance Imaging - methods; Middle Aged; Neoplasm Invasiveness; Neoplasm Staging; Rectal invasion; Rectal Neoplasms - diagnosis; Rectal Neoplasms - secondary; Retrospective Studies; Urinary Bladder Neoplasms - diagnosis; Urinary Bladder Neoplasms - secondary; Uterine Cervical Neoplasms - diagnosis; Uterine Cervical Neoplasms - pathology
• ISSN: 0090-8258; 1095-6859
• DOI: 10.1016/j.ygyno.2005.10.012

Although invasion of the bladder or rectum is rare in cervical carcinoma, endoscopic assessment of both organs is part of the standard FIGO clinical staging system, with associated increase in cost and risk of complications. Our objective was to evaluate whether MRI could be used to select patients who did not require invasive staging of the bladder or rectum. Two observers, blinded to the results of cystoscopy and endoscopic examination of the rectum, retrospectively reviewed the MR images of 112 patients with cervical carcinoma. A 5-point invasion score was used to determine bladder and rectal invasion (1 = no invasion, 5 = definite invasion). A confidence score of 3 or above was used to identify patients with possible bladder or rectal involvement. The results of cystoscopy and endoscopic examination of the rectum were recorded and correlated with the MR findings. MRI was negative for both bladder and rectal invasion in 94/112 patients. Cystoscopy and endoscopic examination of the rectum were confirmed to be normal in all 94 cases. MRI identified 12 patients with possible rectal invasion, 2 confirmed at endoscopy. MRI identified 14 patients with possible bladder invasion, one confirmed at cystoscopy. Using a low threshold cut-off score of > 3 to predict invasion resulted in a 100% negative predictive value (NPV) in detection of bladder and rectal invasion. The absence of bladder or rectal invasion can be diagnosed with sufficient confidence using an MRI scoring system to safely obviate the need for invasive cystoscopic or endoscopic staging in the majority of patients with cervical cancer. This could potentially lead to a reduction in staging costs and morbidity.