Evaluate MRI criteria predictive of the type of surgery to be performed for an "in sano" resection of endometriotic involvement of the recto-sigmoid by a resection without opening of the rectal mucosa ("shaving") versus a segmental resection.

PURPOSE: To retrospectively determine the accuracy of MRI rectal and pararectal signs in predicting a segmental resection of lesions located in the rectum until the recto-sigmoid junction.

MATERIALS AND METHODS: Institutional review board approval for this study was obtained and waived written informed consent. The MR images of 61 patients treated for rectal endometriosis over a 6 years period were reviewed. Two blinded readers performed in consensus a systematic analysis of 7 rectal (length, transverse axis, thickness and circumference of the lesion, and presence of a convex base, submucosal edema and hyperintense cystic areas) and 4 pararectal (posterior vaginal fornix, parametrial, ureteral and sacro-recto-genital septum involvements) signs for each lesion. MR imaging results were compared with surgical procedure (shaving versus segmental resection). Efficacy parameters were calculated with 95 % confidence intervals (CIs).

Inclusion Criteria:

• operated patient for a pelvic endometriosis with resection without opening of the rectal mucosa ("shaving") or with segmental resection of the recto-sigmoidienne.
• Performing a preoperative pelvic MRI with at least 2 perpendicular T2 planes

Exclusion Criteria:

• Refusal to use clinical data and exploration acquired
• Absence of interpretable preoperative pelvic MRI

Patients treated in the gynecological surgery departments of the University Hospital Center of the Civil Hospitals of Lyon: Lyon Sud.