Abstract / Summary
Abstract Background Rectosigmoid deep infiltrating endometriosis (DIE) is a complex condition that may require a range of surgical treatments, from rectal shaving to disc excision or segmental bowel resection. Because the choice of procedure depends partly on the depth and extent of bowel-wall involvement, accurate preoperative assessment is crucial for surgical planning. Endoscopic ultrasonography (EUS) can visualize the individual layers of the rectosigmoid wall and may provide useful information regarding the extent of bowel infiltration. However, whether EUS wall-layer grading is associated with the type of subsequent bowel surgery remains uncertain. This study aimed to evaluate this association in patients with rectosigmoid DIE. Methods We retrospectively studied 25 patients with DIE who underwent lower gastrointestinal EUS before surgery at one center between 2018 and 2024. EUS wall-layer involvement was graded 0–3 by the deepest suspected layer, and bowel surgical management from no therapeutic bowel-wall treatment to segmental resection. The primary analysis used Spearman correlation; perioperative and paired EUS–magnetic resonance imaging (MRI) analyses were exploratory. Results EUS grades 0–3 were assigned to 5, 4, 11, and 5 patients, and surgical-management grades 0–3 occurred in 8, 3, 4, and 10. Higher EUS grade was associated with more extensive bowel surgical management (Spearman's rho = 0.77, P < 0.001). Segmental resection occurred in 10/16 patients (62.5%) with EUS grades 2–3 and 0/9 with grades 0–1 (Fisher's exact P = 0.003). Grades 2–3 were associated with greater blood loss, longer operative time, and longer hospital stay (all P ≤ 0.032). Among 21 operative-positive paired examinations, EUS identified involvement in 17 (81.0%) and MRI in 11 (52.4%; exact McNemar P = 0.109). Conclusion Higher preoperative EUS wall-layer grades were associated with more extensive bowel surgical management in rectosigmoid DIE. EUS may complement counselling and multidisciplinary preparation for possible bowel intervention, but should not independently indicate segmental resection. Clinical trial number: not applicable