Endometriosis and PCOS Co-occurrence
Endometriosis and polycystic ovary syndrome (PCOS) are the two gynaecological conditions patients most often ask about alongside ME/CFS. Their evidentiary status differs sharply: endometriosis has a robust, recent meta-analytic association with ME/CFS, whereas PCOS has essentially no direct evidence. The section treats them separately to avoid conflating a well-supported association with a merely suspected one.
1 Endometriosis: A Well-Supported Association
Endometriosis co-occurs with ME/CFS at a substantially elevated rate, and the association is supported by the strongest evidence in this chapter — a recent systematic review and meta-analysis.
A systematic review and meta-analysis (13 studies, sample sizes from 84 to 134,805) found endometriosis associated with an odds ratio of 2.79 (95% CI 2.00–3.89) for developing ME/CFS, with a pooled OR of 2.52 (95% CI 2.45–2.60) for the ME/CFS–endometriosis association (Compton et al. 2025). ME/CFS prevalence in endometriosis patients was ~17%; endometriosis prevalence in ME/CFS patients was ~13%. The association showed minimal heterogeneity (I² = 0.0%); the prevalence estimates showed extreme heterogeneity (I² > 98%), meaning the association is robust but the prevalence is context-dependent.
This is consistent with the earlier population-based finding that ME/CFS cases reported endometriosis at 36% versus 17% in controls (Boneva, Lin, and Unger 2011). A twofold-to-nearly-threefold association of this magnitude, from a meta-analysis with a large combined sample, is not plausibly explained by referral bias alone.
Certainty: 0.58 (discounted; direct ME/CFS, raw 0.58). Strong methodology (PRISMA 2020 meta-analysis), but cross-sectional designs predominate and 54% of cases were self-reported endometriosis.
Falsifiable prediction: If the association reflects shared biology (rather than only shared care-seeking), then in a prospective cohort, the incidence of ME/CFS in women with surgically-confirmed endometriosis should exceed that in matched controls, and ME/CFS severity should correlate with endometriosis disease activity — patterns that would not hold if the association were purely an ascertainment artifact.
Consequence: Endometriosis is a genuinely co-occurring condition, not a coincidence — so screening for endometriosis in ME/CFS women with pelvic symptoms (and considering ME/CFS in endometriosis patients with unexplained fatigue) is clinically warranted, and treating one condition may be relevant to managing the other.
The elevated co-occurrence of endometriosis and ME/CFS is plausibly explained by a shared mast-cell and neuroinflammatory axis. Endometriosis lesions produce IL-1β, TNF-α, and nerve growth factor, and endometriosis features mast cell hyperactivation, peripheral nerve sensitisation, and estrogen-driven inflammation — mechanisms also implicated in ME/CFS. On this account, the systemic low-grade inflammation of endometriosis could seed central sensitisation that contributes to ME/CFS; conversely, ME/CFS-associated immune dysregulation could permit retrograde-menstruation implants to escape clearance, producing endometriosis. Both directions are plausible and not mutually exclusive (Compton et al. 2025) (Zaitsu et al. 2007).
Certainty: 0.35 — the association is strong (see Endometriosis Is Associated with a ~2.8-Fold Higher Risk of ME/CFS) but the shared-mechanism account is an inference from overlapping features, not directly demonstrated in a combined cohort.
Falsifiable prediction: ME/CFS patients with comorbid endometriosis should show higher serum mast-cell and inflammatory markers (tryptase, IL-1β, NGF) than ME/CFS-only patients matched for severity; and surgical excision of endometriotic lesions should reduce ME/CFS symptom severity in a subset of dual-diagnosis patients. Falsified if dual-diagnosis patients do not show elevated inflammatory markers or if lesion excision has no effect on ME/CFS symptoms.
Consequence: If the shared axis is real, treating endometriosis (medical or surgical) could be a lever on the ME/CFS burden in affected women, and the estrogen-mast-cell mechanism (see HRT Cardiovascular and VTE Risk Is Route- and Regimen-Dependent) may connect the two conditions’ female predominance.
Severity applicability: Unknown — the meta-analysis and population studies are not ME/CFS-severity-stratified.
2 PCOS: A Frequently Asked-But Unsupported Comorbidity
Polycystic ovary syndrome is frequently raised as a possible comorbidity of ME/CFS, largely because both affect premenopausal women and both involve endocrine and metabolic dysregulation. The honest finding of the literature search is that there is no direct evidence of a PCOS–ME/CFS co-occurrence association. The section therefore records this as an open question, not an established comorbidity.
A targeted literature search found no direct, adequately powered study demonstrating a PCOS–ME/CFS co-occurrence association. The only relevant hit — a small fatigue study in PCOS (n=37, no control group, null fatigue correlation, low-tier journal) — fails quality-inclusion criteria and does not support an association. The association therefore remains unestablished.
This is not to say the question is uninteresting: PCOS is an estrogen/androgen-dominant, metabolically active endocrine condition in premenopausal women, so a mechanistic rationale for possible overlap exists (shared endocrine-metabolic dysregulation, possible mast cell/inflammatory involvement). But a rationale is not evidence, and integrating a claimed PCOS–ME/CFS association without evidence would mislead readers.
Certainty: n/a — absence of evidence; treated as a genuinely open research question.
Testable outcome: A properly powered prevalence study comparing PCOS rates (by Rotterdam criteria, not self-report) in a diagnosed ME/CFS cohort against matched controls would resolve the question — a significantly elevated PCOS prevalence in ME/CFS would support the association, while a null or comparable prevalence would settle it as unassociated.
Consequence: A clinician should not assert that PCOS and ME/CFS are linked — no evidence supports it. If a patient has both, they should be managed as two distinct conditions; and the field would benefit from a proper prevalence study before any claim of association is made.
Severity applicability: Not applicable — no established association, no severity data.