Part one — how this normally works
The lining of the intestine is a single layer of epithelial cells joined by tight junctions, covered in mucus. It has to be thin enough to absorb nutrients and selective enough to keep bacteria and undigested protein out. Most of the body's immune tissue sits directly behind it.
The gut has its own nerve network, coordinated with the vagus nerve and the wider autonomic system. Transit speed, secretion and blood flow to the gut are all set by that traffic, not by the gut alone.
Resident bacteria ferment fiber into short-chain fatty acids that feed the lining itself, train immune tolerance, and influence how much histamine is present locally.
Part two — what varies between bodies
Part three — where it diverges
What moves this mechanism
Evidence brief · drafted 22 September 2026
Your gut is not inside you. The inside of your intestine is continuous with the outside world — a tube running through you — and everything you swallow stays outside your body until something lets it across.
What lets it across is a single layer of cells. One cell thick, along the entire length of your intestine. Everything you have ever absorbed came through it, and everything that should not get in was stopped by it.
Physiologists describe this as a gate and fence: selectively let nutrients in, keep everything else out. The gate is supposed to open. Permeability is not a malfunction — it is a regulated, moment-to-moment setting. 4
Here the honesty has to get sharper, because this is the single most oversold mechanism in the entire chronic illness space.
A meta-analysis of 11 studies found women with endometriosis carry an IBS diagnosis at roughly three times the rate of women without it — pooled odds ratio 3.26 (95% CI 1.97–5.39), and 2.72 (2.24–3.31) with one outlier removed. A new IBS diagnosis after an endometriosis diagnosis runs around two-fold. The number that matters most: a history of IBS before the endometriosis diagnosis, pooled OR 3.10 (95% CI 2.06–4.67). If you spent years being managed for IBS before anyone looked for endometriosis, that is the documented pattern, not an unlucky anecdote. 2
Eleven studies, consistent direction, no significant heterogeneity — but IBS has no diagnostic test, so the symptom overlap could inflate the association in both directions, and the authors state plainly that causality cannot be established. An IBS diagnosis alongside endometriosis is also not automatically a wrong one. The two genuinely co-occur. The failure is not that IBS was the wrong label — it is that the label closed the question.
Bowel endometriosis is real, visible at surgery and documented in pathology. Among people undergoing surgery for deep infiltrating endometriosis, roughly a quarter have bowel involvement, and of those the rectum and rectosigmoid junction account for up to three quarters. Nodules are also described in the appendix, terminal ileum and elsewhere in the colon. 5
Those figures come from retrospective surgical series. The population is people whose disease was severe enough to warrant surgery for deep disease — not everyone with endometriosis, and not the general population. Symptomatic disease is far more likely to reach a surgeon. Saying “a quarter of people with endometriosis have bowel disease” is wrong, and it is a very common error.
Most GI symptoms in endometriosis are not explained by a nodule on the bowel. The gut has its own dense nervous system and a continuous two-way conversation with the brain, and that pathway can amplify. It is the same wind-up process described in How the alarm is set, applied to a hollow organ, where it produces pain from normal distension. Visceral hypersensitivity is well described in functional GI disorders; its specific contribution in endometriosis is inferred rather than directly measured.
If several of those are familiar, the Start here tool will show you which other mechanisms they touch — most of these entries light up more than one.
This section is longer than on most pages, and it should be.
A 2025 scoping review in Human Reproduction Open covered 36 studies published 2016–2024 and concluded that sound evidence is lacking to support a specific gut dysbiosis profile in women with endometriosis. Most of those studies had fewer than 100 participants. The largest metagenome study, using shotgun sequencing, found no significant differences. Scattered signals may reflect confounders the studies did not control: menstrual cycle phase, diet, medication. Anyone selling you a protocol to fix your “endo gut” is selling you something that has not been shown to exist yet. 1
Researchers ran the two main commercial assays through mass spectrometry and immunoprecipitation: one detects haptoglobin and complement C3, the other complement C3. Neither reliably detected recombinant zonulin used as a positive control. The two assays did not agree with each other on identical samples. Two people genetically incapable of producing zonulin still tested positive. The authors advise caution until assay methodology is improved. If you have paid for a zonulin result, the number was measuring something — but it was not measuring zonulin. 3
Intestinal permeability is genuinely measurable in research settings and genuinely changes in disease. What is not established is that a marketed test can tell you yours, or that a supplement can change it in a way that changes your symptoms. And the direction of everything here is unresolved: whether gut changes contribute to endometriosis, result from it, result from the treatments, or simply travel alongside it.
Deliberately not on this page
The experience is real and widely described. It is not a defined diagnosis with criteria, and presenting it as one misleads.
The idea that gut bacteria drive endometriosis through estrogen recycling is a hypothesis, not a finding.
This figure circulates constantly and does not survive checking.
No source supports the versions in circulation.
Not used unless the denominator is stated with it: that number comes from a cohort of women with suspected endometriosis, not confirmed disease.
Red flags
These need assessment on their own terms. Do not file them under a chronic diagnosis you already have.
Draft · 22 September 2026
Full citation details are to be completed at build — each verified against the primary source before publication, per the standing rule on preprints and secondary listings.
