Effectiveness of magnetic resonance imaging and MDCT- enteroclysis in the diagnosis of bowel endometriosis

Introduction: Although various techniques have been proposed for the diagnosis of bowel endometriosis, no gold standard is currently available. The objective of this study is to compare the effectiveness of magnetic resonance imaging (MRI) and multidetector computerized tomography enteroclysis (MDCTe) in determining the presence of bowel endometriotic nodules and the depth of infiltration of the nodules in the bowel wall. Materials and methods: This prospective study included 26 women (median age, 32 years; range, 19 – 38) with pain and gastrointestinal symptoms suggestive of colorectal endometriosis (diarrhea, constipation, painful bowel move- ments, dyschezia, rectorrhagia). Patients underwent MRI (1T magnet, phased array coil, multiplanar FSET1, T1 fat sat, T2, T1 post-Gado sequences) and MDCTe (16-row MDCT scanner). MDCTe was performed after intestinal preparation, hypotonisation, and retrograde colon distension (obtained introducing 2000 ml of water). After the injection of iodinated contrast medium, the patient was scanned from the dome of the diaphragm to the pubic symphysis. The exams were reviewed independently and blindly by two radiologists. All women underwent laparoscopy within 2 weeks from imaging. After ade- quate adhesiolysis, last part of the ileum, caecum, colon, and rectum were systematically examined; all endometriotic nodules were excised by either nodulectomy (partial or full thickness) or bowel resection. Radiological findings were compared with surgical and histological data. Statistical analysis was performed by using SPSS 13.0; sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated by using the CATmaker software. Results: Bowel endometriosis was detected by MRI in 11 (42.3%) women and by MDCTe in 12 (46.2%) women. Surgery confirmed the presence of bowel endometriosis in the 12 patients identified by MDCTe. In the diagnosis of patients with bowel endometriosis, sensibility, specificity, PPV, and NPV were 91.7%, 100%, 100%, 93.3% for MRI and 100%, 100%, 100%, 100% for MDCTe. 21 nodules were identified by MRI and 22 by MDCTe; surgery identified a total of 25 nodules, they were located on the rectum (n 1⁄4 13), the sigmoid colon (n 1⁄4 11), caecum (n 1⁄4 1). All nodules missed by MRI andMDCTe were located on the rectum. One false positive nodule was observed at MDCTe and it was judged to reach the serosa. Among the nodules correctly identified at MRI, the depth of infiltration in the bowel wall was estimated to reach the serosa in 8 cases and the muscularis in 13 cases. At MDCTe, 4 nodules were judged to infiltrate the serosa, 16 nodules to reach the muscularis propria, and 1 reached to reach the mucosa. MDCTe correctly estimated the depth of infiltration of the nodules significantly more frequently than MRI (P 1⁄4 0.048). Conclusions: Both MRI and MDCTe can reliably detect the presence of bowel endometriotic nodules; however, MDCTe is more accurate in estimating the depth of infiltration of the nodules in the bowel wall.