Abstract / Summary
Endometriosis is associated with persistent pain and a broad symptom burden. We evaluated the effects of repeated structured exercise using published adjusted or longitudinal trial estimates where available and small-sample meta-analytic methods. This PRISMA 2020 systematic review was registered in PROSPERO (CRD420261400459). Six databases and ClinicalTrials.gov were searched through 20 August 2026 without language restriction; backward citation searching was also performed. Two reviewers independently screened records and full texts, extracted data, assessed result-level risk of bias and rated certainty; disagreements were resolved by discussion and, when required, a third reviewer. Random-effects meta-analyses used restricted maximum likelihood (REML) with Hartung–Knapp inference and alternative small-k models. GRADE used review-stage clinical decision thresholds. The searches identified 3,197 database records and 68 ClinicalTrials.gov records. Eight reports from six independent trials were included, and five trials contributed quantitative data. Overall endometriosis-related pain (4 trials, N = 271) yielded SMD − 0.36 (95% CI − 0.95 to 0.24; I²=57.2%; prediction interval − 1.74 to 1.03). Psychological distress (3 trials, N = 243; SMD − 0.29, 95% CI − 1.43 to 0.85) and sexual function (3 trials, N = 144; MD − 2.23, 95% CI − 5.96 to 1.51) were also very uncertain. Statistical significance differed across random-effects specifications for five syntheses. Harms reporting was inconsistent. Certainty was very low for all key outcomes. Current randomised evidence does not establish a clinically important benefit of structured exercise for endometriosis-associated pain or the wider symptom burden. Point estimates generally favoured exercise, but the evidence is very uncertain because trials were few and small, confidence intervals were wide, several syntheses were method-dependent, and risk-of-bias concerns were common. Decisions to use structured exercise should be individualised according to patient preference, general physical-activity guidance, comorbidity and tolerance rather than an assumed endometriosis-specific benefit; better powered trials and systematic harms reporting are needed. PROSPERO CRD420261400459 (registered 20 May 2026; post-registration amendments are reported in the manuscript and supplementary materials).