Pelvic floor imaging: comparison between magnetic resonance and conventional defecography

Purpose: to evaluate the diagnostic capabilities of Magnetic Resonance Imaging (MRI) in the study of morphologic and functional pelvic floor disorders and to compare them with conventional defecography (CD). Materials and methods: 18 consecutive patients with clinical symptoms of pelvic floor dysfunction underwent pelvic examination with a closed-configuration MR system. Rectal distension was obtained with ultrasound gel. MRI protocol included static high resolution T2-weighted fast spin-echo (FSE) sequences, performed in the coronal, sagittal and transverse planes, and dynamic midsagittal FIESTA-cine acquisitions (1 image/sec) obtained, with the patient maximally contracting the sphincter (squeezing), at rest, straining and defecating. MR images were blindly evaluated by two radiologists and compared with CD and clinical examination available in all patients and considered as the gold standard. Results: disease prevalence was 72.2%. MRI showed sensitivity of 92.3% (95% CI: 77.8%-106.8%), specificity of 80% (95% CI: 44.9%-115.1%), accuracy of 88.9%, predictive positive value (PPV) of 92.3%, predictive negative value (NPV) of 80%. Static images provided information about levator ani muscle morphology showing thickening of pubo-rectal fascicles in spastic pelvic floor syndrome. Dynamic phases detected 9/10 rectoceles, 5/5 spastic pelvic floor syndrome, 7/7 descending perineum syndrome, 4/4 enterocele. MRI identified 2 cystoceles, 1 hysteroptosis and 2 peritoneocele undetected on CD. Conclusion: MRI is a non-invasive imaging technique that allows a complete morphologic and dynamic evaluation of pelvic structures. It can replace CD in the diagnosis of the pelvic floor disorders, providing useful information for the therapeutic planning.