Comparison of supine magnetic resonance imaging with and without rectal contrast to fluoroscopic cystocolpoproctography for the diagnosis of pelvic organ prolapse.

PURPOSE: To compare supine magnetic resonance imaging (MRI), with and without rectal contrast, with fluoroscopic cystocolpoproctography (CCP) for the diagnosis of pelvic organ prolapse.

MATERIALS AND

METHODS

Supine MRI and CCP studies were reviewed in 82 patients. All patients were women with an average age of 58.8 years, and the studies were done a mean of 25 days apart. Magnetic resonance imaging was performed with rectal contrast (n = 35) and without rectal contrast (n = 47). Fluoroscopic cystocolpoproctography was performed with rectal (n = 82), vaginal (n = 82), small bowel (n = 81), and bladder (n = 78) contrast, and images were corrected for magnification. Each study was independently reviewed by 2 readers, and outcome variables were presence/absence of cystocele, vaginal prolapse, enterocele, sigmoidocele, and anterior rectocele. Sigmoidoceles were included with enteroceles for data analysis.

RESULTS

For the entire patient group, the prevalence of cystoceles was 89% on CCP and 80% on MRI; vaginal prolapse was 81% on CCP and 56% on MRI; enteroceles, 38% on CCP and 24% on MRI; and anterior rectoceles, 45% on CCP and 37% on MRI. There were significantly more cystoceles (odds ratio [OR] 4.7, P = 0.003), vaginal prolapses (OR 5.2, P < 0.0005), and enteroceles (OR 3.8, P< 0.0005) on CCP than on MRI. For MRI with rectal contrast versus CCP, the prevalence of cystoceles was 94% on CCP and 91% on MRI; vaginal prolapse, 74% on CCP and 70% on MRI; enteroceles, 36% on CCP and 19% on MRI; and anterior rectoceles, 51% on CCP and 59% on MRI. There was statistical significance only for enteroceles, more of which were found on CCP (OR 7.4, P = 0.003). For MRI without rectal contrast versus CCP, the prevalence of cystoceles was 85% on CCP and 72% on MRI; vaginal prolapse, 86% on CCP and 46% on MRI; enteroceles, 40% on CCP and 28% on MRI; and anterior rectoceles, 39% on CCP and 21% on MRI. There were significantly more cystoceles (OR 6.6, P = 0.003), vaginal prolapses (OR 20.8, P < 0.0005), enteroceles (OR 2.9, P = 0.015), and rectoceles (OR 4.9, P = 0.001) on CCP than on noncontrast MRI.

CONCLUSIONS

Magnetic resonance imaging without rectal contrast showed statistically fewer pelvic floor abnormalities than CCP. Except for enteroceles, MRI with rectal contrast showed statistically similar frequency of pelvic organ prolapse as CCP.