Voiding dysfunction after surgery for colorectal deep infiltrating endometriosis: an updated systematic review and meta-analysis.

Abstract

To define the risk factors of post-operative voiding dysfunction according to the type of surgical procedure performed. A systematic review through PubMed, the Cochrane Library, and Web of Science databases was performed. The Medical Subject Headings terms aimed for English articles about colorectal endometriosis surgery and voiding dysfunction published until December 26, 2022 were used. The primary outcome was the occurrence of post-operative voiding dysfunction. Secondary outcome was the presence of a persistent voiding dysfunction at 1 month. MeSH terms included ''deep endometriosis'', ''surgery'', or ''voiding dysfunction''. Two reviewers (AM, PE) assessed the quality of each article independently. A Study Quality Assessment Tool was used to provide rating of the quality of the included studies. 22 studies were included in the final analysis. Rectal shaving was associated with less voiding dysfunction than segmental resection (OR 0.33; 95%CI [0.20: 0.54]; I² = 0%; p < 10^-3). No difference was found between rectal shaving and discoid excision (OR 0.44; 95%CI [0.07: 2.84]; I² = 55%; p = 0.39), nor between discoid excision and segmental resection (OR 0.44; 95%CI [0.18: 1.09]; I² = 49%; p = 0.08). Conservative surgery (i.e., shaving and discoid) was associated with less voiding dysfunction than radical surgery (i.e., segmental resection) (OR 0.37; 95%CI [0.25: 0.55]; I² = 0%; p < 10^-3). Regarding persistent voiding dysfunction, rectal shaving and discoid excision were less associated with voiding dysfunction than segmental resection (respectively, OR 0.30; 95%CI [0.14: 0.66]; I² = 0%; p = 0.003 and OR 0.13; 95%CI [0.03: 0.57]; I² = 0%; p = 0.007). Conservative bowel procedures are associated with lower rates of persistent post-operative voiding dysfunction and should be considered first when possible.Trial registration: Our meta-analysis was registered under the PROSPERO number: CRD42023395356.