Objective: To report on randomized controlled trials (RCTs) that examine the surgical treatment of endometriosis-associated pelvic pain and to highlight their strengths and weaknesses.
Data sources: We performed a systematic review of English-language, full-text articles addressing the surgical management of pain symptoms associated with endometriosis. The terms endometriosis, pain, surgery, laparoscopy, plasma, and laser were used for searches in Cochrane, MEDLINE, EMBASE, and clinical trial databases. Additional studies were identified from references in electronically located articles.
Methods of study selection: A literature search was conducted by 2 authors, and abstracts were independently screened for inclusion, with the resolution of any discrepancy by a third author. Randomized studies that reported pain before and after surgery were eligible for inclusion. Supporting data from nonrandomized trials were used for discussion. The Cochrane risk-of-bias assessment was performed on included studies.
Tabulation, integration, and results: Search results for available articles from 1996 to October 2019 revealed 594 potential studies, with 20 studies meeting the final inclusion criteria. Comparative studies of surgery vs no surgery for an effect on pain, surgical approach, the effect of different locations of disease on pain, nerve-dividing techniques for pain, and nerve-sparing effects for pain were studied. RCTs reported a substantial reduction in pain compared with no surgery in up to 80% of women; however, up to a third of women in these studies reported a placebo response. There was no evidence of a difference in pain reduction with the mode of surgery (laparoscopy, laparotomy, or robot-assisted laparoscopy). There is limited evidence stating that excision is superior to ablative surgery; however, there are confounders in the reporting of disease location and depth and the pain symptoms most affected. We need to reconsider the hypothesis that disc excision results in fewer complications and has superior outcomes to those of segmental resection in light of the first RCT on this subject. Nerve-dividing surgery for pain has been demonstrated to be of no value for uterosacral nerve ablation and/or division and of limited (if any) value for presacral neurectomy.
Conclusion: Although surgical RCTs have always been difficult to undertake, there are 16 RCTs on endometriosis-associated pain. Ethical considerations, the equipoise of surgeons and participants, and follow-up duration are important parameters in establishing RCTs. In addition, we must be willing to accept and adopt the evidence when it does demonstrate a particular outcome, such as the fact that surgical uterosacral nerve disruption does not improve pain or that disc excision does not substantially reduce complications compared with segmental resection for bowel disease, as suggested by previous nonrandomized studies. If we accept that a well-conducted RCT provides best-quality evidence, then we should at least be open to the possibility that our long-held views may be challenged and changed with new science in our practice.
Keywords: Endometriosis and pain; Randomized controlled trial; Surgical treatment.
Copyright © 2019 AAGL. All rights reserved.