The Gut Is Not a Sealed Wall: Rethinking “Leaky Gut”

A healthy gut barrier is selectively permeable, not sealed. Learn what intestinal permeability can show, what it cannot prove, and why “leaky gut” oversimplifies the biology.

What would happen if your intestine became perfectly sealed?

It would not become a stronger barrier. It would stop doing its job.

Water, electrolytes, and nutrients have to move from the intestinal lumen into the body, often through regulated transport across epithelial cells. The interface also permits controlled movement between cells and specialized immune sampling. These routes are not equivalent, but they share a basic principle: healthy barrier function depends on selectivity, not impermeability (Camilleri, 2019; Vanuytsel et al., 2021).

The word “leaky” sets up the wrong contrast. It suggests that a healthy intestine is watertight and a damaged one develops holes. In reality, the gut is a living, selective interface. It absorbs, senses, tolerates, defends, and repairs. It succeeds not by blocking all passage, but by regulating what crosses, where, by which route, and in what amount.

Intestinal barrier dysfunction is not imaginary. Altered permeability is real, measurable in humans, and clinically meaningful in several defined diseases. But a change in permeability is only the beginning of an interpretation. On its own, it does not tell us that the lining is injured, that microbes have entered the blood, that systemic inflammation followed, or that the gut caused a symptom or disease.

The useful opposite of “leaky” is not sealed. It is well regulated.

A Healthy Gut Cannot Be Sealed

A living interface, not a wall

The intestinal barrier is often pictured as a single layer of tiles held together by tight junctions. That layer matters, but the real system is broader. Mucus helps control contact with the epithelial surface. Polarized epithelial cells transport selected substances in specific directions. Junctions regulate movement between cells. Antimicrobial defenses, rapid cell turnover, tissue repair, and mucosal immune responses all help determine what reaches the tissue beneath.

There is also more than one route across the epithelium. Paracellular movement occurs between cells and is influenced by junctional complexes. Transcellular movement occurs through cells by channels, transporters, endocytosis, or transcytosis. The pathways are not interchangeable, and tight junctions are not switches that simply flip from “closed” to “open.” They take part in dynamic, molecule-specific regulation (Camilleri, 2019; Vanuytsel et al., 2021).

Nor is the immune system meant to remain completely blind to the intestinal lumen. Specialized mucosal structures help sample dietary and microbial antigens as part of immune surveillance. Ex vivo human ileal tissue has demonstrated antigen and bacterial transport across follicle-associated epithelium, the specialized surface overlying Peyer’s patches. The distinction matters: a specialized route of uptake is not the same as generalized epithelial rupture (Keita et al., 2006).

Selection depends on cargo, route, place, and time

“Intestinal permeability” is incomplete unless it refers to something specific. Permeability to which molecule? Through which route? In the stomach, small intestine, or colon? During fasting, after a meal, under physiological stress, or during active disease? Over what collection window, and measured by which method?

A barrier can handle water, ions, nutrients, small experimental probes, peptides, and microbial material differently. Its behavior can also change with conditions. A result observed after a lipid challenge may not describe a fixed trait at rest (Genser et al., 2018). A colonic abnormality does not automatically establish a whole-gut defect.

This is the first correction to the sealed-wall model. Permeability is not one number permanently attached to a person. It is a relationship between a defined barrier and a defined cargo under defined conditions. Once that is clear, “leaky gut” begins to split into several biological questions.

One Phrase, Several Different Biological Events

Popular explanations often glide from increased permeability to barrier damage, microbial translocation, inflammation, symptoms, and chronic disease. The sequence sounds plausible because each step can occur somewhere in biology. The problem is that demonstrating one step does not demonstrate the next.

Altered transport is not the same as injury

Permeability describes passage across an epithelium under specified conditions. Barrier dysfunction is broader. It can involve mucus, junctional regulation, epithelial continuity, repair, antimicrobial defense, and immune coordination. Epithelial injury is more specific again: cells are damaged or lost.

These states can overlap, but they can also separate. A sugar-probe test may show altered flux without identifying structural injury. Conversely, plasma intestinal fatty acid-binding protein, or I-FABP, can rise when enterocytes are injured. In a controlled human intestinal ischemia-reperfusion model, I-FABP tracked morphological epithelial damage. That makes it useful for injury biology, but not a direct measure of paracellular permeability (Schellekens et al., 2014).

Passage is not the same as translocation

Even when transport changes, the result does not identify what crossed. A small inert probe does not stand in for a food peptide, lipopolysaccharide, bacterial DNA, or a viable bacterium. Detecting movement across the epithelium is also different from demonstrating that microbial material reached tissue, lymph, or blood beyond its normal compartment.

Translocation is another biological event. It requires evidence about both the cargo and its destination. A local mucosal exposure may be sampled, neutralized, or contained. If microbial material reaches the portal circulation, binding proteins and the liver can further shape what enters the systemic circulation. How far the material travels matters as much as the first crossing.

A systemic signal is not the same as disease causation

Systemic inflammation sits farther downstream. It depends on the identity and dose of the material, the route it took, immune recognition, host clearance, and the condition of the person exposed. A local barrier change can occur without a measurable distant effect. Inflammation can also alter the barrier, creating feedback in the opposite direction.

Symptoms and diseases add another layer. When altered permeability and a clinical condition occur together, the barrier change might be an early contributor, a consequence, an amplifier, a shared effect of another process, or a marker of disease activity. Establishing causal contribution takes more than a plausible pathway. Timing, mediation, intervention, dose-response, and other discriminating evidence become important.

Each arrow needs its own evidence.

This ladder does not claim that every disease unfolds in this order. Biology often runs in loops. It is an evidentiary map. Every time an explanation moves down one row, it takes on a new burden of proof.

What Crosses Matters More Than the Word “Leaky”

A sugar probe is not a peptide, bacterial product, or bacterium

Many human permeability studies use orally administered sugars. Their recovery in urine can reveal how selected probes moved through the gastrointestinal tract under a defined protocol. These tools are valuable precisely because the probes are known and measurable. That strength is also their limit: they report the behavior of those probes, not every molecule that might encounter the intestine.

Human celiac disease makes the difference visible. In biopsies from active disease, intact gliadin peptides crossed through pathways that were not adequately represented by mannitol flux or electrical conductance. Small-sugar permeability could not fully represent the transport of a biologically important antigen (Ménard et al., 2012).

The same principle works in the other direction. Specialized intestinal tissue can take up antigens and bacteria through organized mucosal pathways. Without context, the fact that material crossed an epithelial interface cannot tell us whether the event was specialized sampling, abnormal transport, tissue injury, or clinically important translocation.

Different routes carry different meaning

This is why the popular phrase “undigested food enters the bloodstream” is so misleading. Intact peptides and antigens can cross in particular physiological or disease-specific settings. That is not the same as showing that generalized quantities of incompletely digested food routinely circulate and cause nonspecific symptoms.

The word “toxins” hides even more. LPS is a defined component of Gram-negative bacteria. Bacterial DNA is a different signal. A viable bacterium is different again. Each has its own route, detectability, biological activity, and relationship to host defense. Lumping them together erases the information needed to judge the consequence.

“More permeable” is not a complete biological conclusion. More permeable to what? Through which route? In which region? Under what conditions? With what demonstrated destination?

Those questions also explain why no single test can capture the entire barrier.

What a Permeability Test Can and Cannot Tell You

Different tools answer different questions

Intestinal barrier research does not lack useful methods. The trouble begins when different methods are treated as interchangeable versions of a universal “leaky gut test.” A method becomes informative when its biological target and limits are clear (Vanuytsel et al., 2021; Vieujean et al., 2026).

• Sugar probes: Lactulose/mannitol, lactulose/rhamnose, sucrose, sucralose, and multi-sugar approaches provide functional information about the recovery of defined probes. Results depend on dose, timing, collection window, intestinal transit, absorptive surface, renal handling, diet, and analytical protocol. A ratio can also change because its numerator, denominator, or both have changed.
• I-FABP: This is primarily an enterocyte-injury signal. It can help identify cellular damage, but it does not directly quantify transepithelial flux.
• LBP and low-level plasma LPS: Lipopolysaccharide-binding protein is a host-response marker influenced by bacterial-product exposure, inflammation, and liver biology. It is downstream and not intestine-specific. Plasma LPS sounds more direct, but measuring very low concentrations is analytically difficult because binding, masking, interference, sample preparation, and contamination can affect the result.
• Ussing chambers, confocal endomicroscopy, and histology: These methods can reveal ex vivo molecular flux, in vivo microscopic leakage or epithelial gaps, and structural injury. They offer valuable mechanistic or specialist information, but they sample particular tissue, require invasive procedures, or depend on specialized interpretation.
• Stool microbiome profiles: These describe aspects of microbial composition or function. They do not measure movement across the epithelium. An international consensus statement published in 2025 also concluded that the clinical usefulness of many direct-to-consumer microbiome tests remains insufficiently established (Porcari et al., 2025).

Why no single test captures the whole barrier

Different markers can disagree for good biological reasons. Sugar flux, epithelial injury, microbial material, host response, inflammation, and symptoms occupy different points in the chain. They should not be expected to rise and fall together. Discordance is not necessarily noise. Sometimes it is evidence that the layers really are distinct.

Commercial “zonulin” testing shows why analytical identity also matters. Commercial ELISA kits studied in the cited literature did not reliably identify the expected pre-haptoglobin-2 analyte. Investigators reported signals from other proteins, and circulating kit-derived concentrations did not validly mirror independently measured colonic permeability (Scheffler et al., 2018; Ajamian et al., 2019; Meira de-Faria et al., 2021). This does not make junction-regulating biology fictional. It means that a commercial signal labelled zonulin cannot be assumed to be a validated standalone diagnosis of pathological permeability.

No current method compresses functional transport, structural integrity, epithelial injury, cargo identity, microbial translocation, systemic response, and clinical cause into one answer. A test can be scientifically useful without carrying all of those meanings (Vieujean et al., 2026).

Where Barrier Changes Carry Real Clinical Meaning

The strongest human evidence does not produce a long list of diseases “caused by leaky gut.” It shows that barrier findings answer different clinical questions in different contexts.

Evidence of timing and prognosis

Crohn’s disease provides the most important evidence that altered permeability can precede diagnosis. In the prospective GEM cohort, 1,420 asymptomatic first-degree relatives of people with Crohn’s disease were followed for a median of 7.8 years. Fifty developed the disease. An abnormal baseline lactulose/mannitol ratio was associated with later Crohn’s disease, including when the measurement occurred more than three years before diagnosis (Turpin et al., 2020).

That is stronger than a cross-sectional association because the barrier finding came first. It supports the possibility that altered permeability participates in the preclinical process. It does not establish permeability as the first or sufficient cause. Genetics, microbes, environmental exposures, and subclinical inflammation may already be interacting in that high-risk population, and the result is not a general-population screening test.

In established inflammatory bowel disease, barrier state can carry a different kind of meaning. In the prospective ERIca study, 181 patients in clinical remission underwent confocal laser endomicroscopy. Barrier healing predicted a lower risk of later adverse outcomes and provided prognostic information beyond endoscopic or histological remission alone (Rath et al., 2023).

That finding shows that epithelial barrier behavior can matter clinically even after symptoms and visible inflammation have improved. It does not prove that a treatment acting only on permeability would change the disease course by itself.

Evidence of cargo-specific and systemic consequences

Celiac disease answers a cargo question. Active disease is associated with abnormal sugar permeability, but human tissue studies also show that intact gliadin peptides can use transport routes that small probes do not adequately represent. The clinical meaning comes from the specific antigen, immune context, and disease, not from a generic porous-wall model (Ménard et al., 2012).

Advanced cirrhosis answers a translocation question. In a prospective cohort of 294 people with cirrhosis who had no bacterial infection at enrollment, circulating bacterial DNA, used by the investigators as a marker of translocation, was detected in 36.1 percent. Detectable bacterial DNA was associated with overt hepatic encephalopathy and higher LBP and IL-6 levels. It was also independently associated with six-month mortality (Thanapirom et al., 2024).

This shows that a systemic translocation marker can be clinically relevant in advanced disease. It does not establish viable bacteremia or prove that translocation caused encephalopathy or death. Cirrhosis also involves portal hypertension, immune dysfunction, severe disease, and altered hepatic clearance. It is not a model for assuming the same process in an otherwise healthy person with fatigue and a commercial panel result.

Evidence of heterogeneity

Irritable bowel syndrome does not have one universal barrier phenotype. More consistent findings occur in selected groups, including post-infectious and diarrhea-predominant IBS (Vanuytsel et al., 2021). A 2025 study of 51 participants with post-Campylobacter IBS and 39 healthy volunteers found higher 2-to-24-hour lactulose excretion, interpreted as greater colonic permeability. It also linked the phenotype with transcriptomic and metabolomic patterns, including metabolites associated with permeability and diarrhea (Caceres Lessa et al., 2025). Because the measurements were made after IBS was established, it remains unclear whether the barrier change was a mediator, consequence, or amplifier of the post-infectious process.

The metabolic literature also resists a simple equation. In a mechanistic study of 122 people with and without severe obesity, severe obesity was associated with several jejunal barrier abnormalities at baseline. Permeability to larger molecules became especially apparent after a lipid challenge and was linked with inflammation and type 2 diabetes (Genser et al., 2018). This supports an endpoint- and context-dependent phenotype in severe obesity. It does not establish a universal trait in all people with obesity or prove that intestinal permeability caused their metabolic disease.

These cases do not converge on one story. They show timing in one setting, prognosis in another, cargo specificity in a third, a translocation marker in advanced disease, and subgroup- or challenge-dependent vulnerability elsewhere. Each case earns a different inference.

Cause, Consequence, Amplifier, or Marker?

The familiar warning that association is not causation is correct, but incomplete. An association can still carry useful biological information. The real task is to say what that information adds.

If a barrier abnormality precedes diagnosis, as it can in a high-risk Crohn’s cohort, it may be an early contributor or a marker of a broader preclinical process. If barrier healing predicts later outcomes in established IBD, it can be a prognostic marker even when the mechanism of benefit is not established. If altered permeability appears after an intestinal infection or during active inflammation, it may be a consequence, a mediator, an amplifier, or all three at different points in time.

Feedback loops are often more faithful than one-way stories. Inflammation can change epithelial transport, damage cells, alter repair, and reshape the mucosal environment. Greater exposure to luminal antigens or microbial products can then sustain local immune signaling. The barrier can be downstream of inflammation and still help maintain it.

The same measurement can therefore mean different things before disease, during active disease, and in remission. Timing and disease stage change the causal question. So do the endpoint and population. A prognostic signal in IBD, a cargo-specific transport pathway in celiac disease, and circulating bacterial DNA in cirrhosis are all meaningful, but they do not assign the same role to intestinal permeability.

This plural model is less dramatic than a single root cause. It is also more biologically useful. It explains why real barrier pathology can matter without becoming a universal syndrome, and why treating inflammation may improve barrier function even when the barrier was not the initial trigger.

How to Evaluate a “Leaky Gut” Claim Without Turning It Into a Diagnosis

Six questions that keep the biology honest

Before accepting a headline, test interpretation, or repair claim, ask six questions.

1. What was measured? Was it functional flux, epithelial injury, tissue structure, microbial material, host response, inflammation, symptoms, or a clinical outcome?
2. Where and when? Which intestinal region, collection window, challenge, disease phase, and physiological state did the result describe?
3. What crossed? Was it a small probe, intact peptide, LPS-related signal, bacterial DNA, or a viable organism?
4. How far did it go? Was movement shown only across the epithelium, into local tissue or lymph, or into the systemic circulation with a measurable distant effect?
5. Which direction is plausible? Could the finding be a cause, consequence, shared effect of another driver, contributor, amplifier, feedback component, or biomarker?
6. Did changing it matter clinically? Did the intervention alter the barrier endpoint, and was symptom relief or disease benefit shown to depend on that change?

These questions do not dismiss permeability science. They stop a real observation from being asked to carry conclusions it was not designed to support.

A changed marker is not the same as clinical recovery

Some probiotics, prebiotics, synbiotics, dietary strategies, and disease-specific interventions can change selected barrier-related endpoints in selected populations. A 2025 meta-analysis reported effects on some markers, but the trials differed substantially in population, intervention, dose, duration, and measurement. Because the pooled endpoints had different biological meanings and analytical limitations, the combined result cannot be read as one uniform demonstration of barrier repair (Ghorbani et al., 2025).

Clinical outcomes can move separately from permeability markers. In a meta-analysis of four randomized larazotide trials in celiac disease, some symptom effects during gluten challenge did not correspond to a significant pooled improvement in the lactulose/mannitol ratio (Hoilat et al., 2022). A person can feel better through mechanisms a particular barrier test does not capture. A marker can also improve without proving that the person has recovered or that the marker change caused the benefit.

There is therefore no evidence-based universal protocol to “seal” or “repair” the gut. A stool microbiome panel, a commercial zonulin value, or a collection of indirect markers should not substitute for an established diagnostic pathway. Nonspecific symptoms do not establish pathological permeability.

Persistent gastrointestinal symptoms deserve attention on their own terms. Bleeding, unexplained weight loss, anemia, fever, severe pain, or chronic diarrhea warrant conventional clinical assessment rather than self-diagnosis through a catch-all “leaky gut” explanation.

The Biology Is Real. The Shortcut Is the Problem.

A healthy intestine is permeable because life requires controlled exchange. Pathology begins not with passage itself, but when the regulation, cargo, route, tissue response, or clinical context changes in a consequential way.

The evidence is strong that altered intestinal transport can occur and that barrier dysfunction matters in defined diseases. It becomes more conditional as the claim travels toward microbial translocation, systemic inflammation, nonspecific symptoms, and broad disease causation. Those later steps can be real, but they cannot be assumed.

That leaves a better model than either “leaky gut explains everything” or “the biology is fake.” Ask what was measured, what crossed, how far it travelled, and what the evidence actually connects to a clinical outcome. Each arrow must earn its own evidence.

The gut barrier is selective, not sealed. The biology is real. The shortcut is the problem.

Related Reading

Frequently Asked Questions

Can a healthy gut be permeable?

Yes. A healthy intestinal barrier must allow regulated movement of water, electrolytes, nutrients, and selected biological material. The healthy state is selective permeability, not complete sealing.

Is intestinal permeability the same as “leaky gut”?

Not exactly. Intestinal permeability is a measurable biological property. “Leaky gut” often bundles permeability, tissue injury, microbial translocation, inflammation, symptoms, and disease causation into one story, even though each step requires separate evidence.

Does increased intestinal permeability mean bacteria are entering the bloodstream?

No. A permeability result does not identify the cargo or prove that microbial material reached the blood. Microbial translocation is a separate biological event that requires its own evidence.

Can zonulin or a stool microbiome test diagnose intestinal permeability?

Commercial zonulin assays studied in the cited literature have important analytical limitations, and a stool microbiome profile does not measure movement across the intestinal epithelium. Neither is a validated standalone diagnosis of a universal “leaky gut” syndrome.

Does intestinal permeability cause systemic inflammation or chronic disease?

It can contribute to disease biology in defined contexts, but the relationship is not automatic or universal. The cargo, destination, host response, disease state, and causal direction all matter.

Can probiotics or diet repair a “leaky gut”?

Some interventions can change selected barrier-related markers in selected populations. That does not establish a universal repair protocol, and a changed marker is not the same as symptom recovery or disease modification.

References

Ajamian, M., Steer, D., Rosella, G., & Gibson, P. R. (2019). Serum zonulin as a marker of intestinal mucosal barrier function: May not be what it seems. PLoS ONE, 14(1), e0210728. https://doi.org/10.1371/journal.pone.0210728

Caceres Lessa, A. Y., Edwinson, A., Sato, H., et al. (2025). Transcriptomic and metabolomic correlates of increased colonic permeability in postinfection irritable bowel syndrome. Clinical Gastroenterology and Hepatology, 23(4), 632-643.e13. https://doi.org/10.1016/j.cgh.2024.06.028

Camilleri, M. (2019). Leaky gut: Mechanisms, measurement and clinical implications in humans. Gut, 68(8), 1516-1526. https://doi.org/10.1136/gutjnl-2019-318427

Genser, L., Aguanno, D., Soula, H. A., et al. (2018). Increased jejunal permeability in human obesity is revealed by a lipid challenge and is linked to inflammation and type 2 diabetes. Journal of Pathology, 246(2), 217-230. https://doi.org/10.1002/path.5134

Ghorbani, Z., Shoaibinobarian, N., Noormohammadi, M., et al. (2025). Reinforcing gut integrity: A systematic review and meta-analysis of clinical trials assessing probiotics, synbiotics, and prebiotics on intestinal permeability markers. Pharmacological Research, 216, 107780. https://doi.org/10.1016/j.phrs.2025.107780

Hoilat, G. J., Altowairqi, A. K., Ayas, M. F., et al. (2022). Larazotide acetate for treatment of celiac disease: A systematic review and meta-analysis of randomized controlled trials. Clinics and Research in Hepatology and Gastroenterology, 46(1), 101782. https://doi.org/10.1016/j.clinre.2021.101782

Keita, Å. V., Gullberg, E., Ericson, A.-C., et al. (2006). Characterization of antigen and bacterial transport in the follicle-associated epithelium of human ileum. Laboratory Investigation, 86(5), 504-516. https://doi.org/10.1038/labinvest.3700397

Meira de-Faria, F., Bednarska, O., Ström, M., et al. (2021). Colonic paracellular permeability and circulating zonulin-related proteins. Scandinavian Journal of Gastroenterology, 56(4), 424-431. https://doi.org/10.1080/00365521.2021.1879247

Ménard, S., Lebreton, C., Schumann, M., et al. (2012). Paracellular versus transcellular intestinal permeability to gliadin peptides in active celiac disease. American Journal of Pathology, 180(2), 608-615. https://doi.org/10.1016/j.ajpath.2011.10.019

Porcari, S., Mullish, B. H., Asnicar, F., et al. (2025). International consensus statement on microbiome testing in clinical practice. The Lancet Gastroenterology & Hepatology, 10(2), 154-167. https://doi.org/10.1016/S2468-1253(24)00311-X

Rath, T., Atreya, R., Bodenschatz, J., et al. (2023). Intestinal barrier healing is superior to endoscopic and histologic remission for predicting major adverse outcomes in inflammatory bowel disease: The prospective ERIca trial. Gastroenterology, 164(2), 241-255. https://doi.org/10.1053/j.gastro.2022.10.014

Scheffler, L., Crane, A., Heyne, H., et al. (2018). Widely used commercial ELISA does not detect precursor of haptoglobin2, but recognizes properdin as a potential second member of the zonulin family. Frontiers in Endocrinology, 9, 22. https://doi.org/10.3389/fendo.2018.00022

Schellekens, D. H. S. M., Grootjans, J., Dello, S. A. W. G., et al. (2014). Plasma intestinal fatty acid-binding protein levels correlate with morphologic epithelial intestinal damage in a human translational ischemia-reperfusion model. Journal of Clinical Gastroenterology, 48(3), 253-260. https://doi.org/10.1097/MCG.0b013e3182a87e3e

Thanapirom, K., Suksawatamnuay, S., Wejnaruemarn, S., et al. (2024). Bacterial DNA translocation is associated with overt hepatic encephalopathy and mortality in patients with cirrhosis. Clinical and Translational Gastroenterology, 15(5), e00697. https://doi.org/10.14309/ctg.0000000000000697

Turpin, W., Lee, S.-H., Raygoza Garay, J. A., et al. (2020). Increased intestinal permeability is associated with later development of Crohn’s disease. Gastroenterology, 159(6), 2092-2100.e5. https://doi.org/10.1053/j.gastro.2020.08.005

Vanuytsel, T., Tack, J., & Farré, R. (2021). The role of intestinal permeability in gastrointestinal disorders and current methods of evaluation. Frontiers in Nutrition, 8, 717925. https://doi.org/10.3389/fnut.2021.717925

Vieujean, S., Atreya, R., Buda, A., et al. (2026). Tools and advanced imaging technologies for assessing intestinal epithelial barrier integrity: A systematic review. American Journal of Physiology-Gastrointestinal and Liver Physiology, 330(4), G424-G458. https://doi.org/10.1152/ajpgi.00447.2025