Ultrasound Prediction of Segmental Bowel Resection in Women with Rectovaginal Endometriosis: A Single-Center Experience

Abstract

Study objective: To retrospectively evaluate the ability of routinely collected preoperative ultrasound data to predict bowel resection during surgery for rectovaginal endometriosis.

Design and setting: Patients at the University College London Hospital who underwent surgery for rectovaginal endometriosis during a 6-year period were identified from the prospectively generated hospital (British Society for Gynaecological Endoscopy) database. Imaging data were collected and analyzed to determine associations with the requirement for bowel resection.

Patients: We evaluated 228 consecutive women undergoing bowel surgery (shave, disc resection, or segmental resection) for rectovaginal endometriosis.

Interventions: The patients in our study underwent surgical resection of rectovaginal endometriosis and interventions included shave, disc resection, and segmental resection of the bowel. All patients underwent a preoperative transvaginal ultrasound to assess the extent of endometriosis.

Measurements and main results: There were 206 rectal shaves (90.4%), 2 disc resections (0.9%), and 20 segmental bowel resections (8.8%). A multivariable analysis demonstrated an association between bowel resection and ≥2 nodules located in the rectovaginal space (odds ratio [OR] 6.85; 95% confidence interval [CI], 1.37-34.2), nodules in the vesicouterine pouch (OR 5.87; 95% CI, 1.03-33.3), and increasing nodule size (OR 2.39 per 1 cm increase per 1 cm diameter increase; 95% CI, 1.56-3.64).

Conclusion: Ultrasound findings of endometriotic nodule location, number of nodules, and increasing size are independent predictors of segmental bowel resection at the time of surgery for rectovaginal endometriosis. This highlights the importance of accurate diagnostic evaluation to aid counseling and surgical planning in the preoperative setting for women with rectovaginal endometriosis.

Keywords: Bowel resection; Endometriosis; Endometriotic nodules; Risk prediction; Ultrasound.