A widely used leaky gut blood test was measuring the wrong protein

Key takeaways

• In a 2018 study by German and U.S. researchers, a widely used zonulin blood test didn’t detect zonulin; the likeliest protein it picked up was properdin, part of the immune system.
• Urine sugar tests are the research standard for gut permeability, but a 2022 review found they aren’t standardized and are unlikely to see wide clinical use as they stand.
• A leakier gut is well documented in untreated celiac disease and active Crohn’s disease, and celiac patients’ readings fell sharply with treatment.
• Among 1,420 healthy relatives of people with Crohn’s disease, a leaky reading tripled the later risk of Crohn’s, yet only 50 of them developed it during follow-up.
• Glutamine made no overall difference to gut permeability in a 2024 pooled analysis of trials in sick adults, and the only collagen study we found was uncontrolled and paid for by a collagen maker.

In 2018, a team of German and U.S. researchers took a hard look at a blood marker that many studies had used as a sign of “leaky gut”: zonulin¹. Zonulin is a protein that helps open the seals between the cells lining your gut¹⁴.

The team measured it in 376 people with a widely used commercial kit made by the German company Immundiagnostik¹. Then they ran a simple check. Zonulin is an early form of a protein called haptoglobin 2. People who inherit two copies of the other gene version, HP1, should make none of it¹.

Those people had zonulin readings anyway. Results didn’t differ by gene type¹. When the team fed the kit lab-made zonulin, it didn’t register. Of the purified proteins they tried, the kit reacted to properdin, an immune-system protein, though it read far below the true amount¹.

One co-author was Alessio Fasano, whose 1991 and 1995 papers on a cholera toxin helped launch the study of zonulin¹. The core finding was blunt: the kit doesn’t recognize the protein now identified as zonulin¹.

The authors did leave one door open. The proteins the kit catches may be structural, and possibly functional, cousins of zonulin, with properdin the likeliest candidate¹. Either way, the number on the report isn’t zonulin.

Our earlier article on gut-healing claims showed that a leakier gut is real and that most protocols sold to fix it are untested. This one looks at the tests, the diseases that truly involve a leak, and how to judge a diagnosis.

The leaky gut theory, as its supporters see it

Your gut lining is a single layer of cells, held together by protein seals called tight junctions. If those seals loosen, the theory goes, bits of food and bacterial toxins slip into the body. The immune system reacts, and over time that steady reaction feeds illness far from the gut.

In this view, the leak comes first and illness follows. Cleveland Clinic calls the resulting “syndrome” a hypothetical condition, and notes that doctors don’t recognize it as a diagnosis⁹.

The clinic allows that low-grade inflammation may play a part in obesity, diabetes, arthritis, chronic fatigue syndrome, asthma and fibromyalgia. Whether a leaky gut drives that inflammation, it says, is unclear⁹.

Parts of the theory stand on firm ground. The barrier is real, measurable and able to loosen. Inflammatory bowel disease and celiac disease are known to come with a leakier lining⁹.

The open question is whether the leak starts the trouble or follows it⁸⁹. That shapes what any test result can mean for you.

Zonulin kits failed check after check

In 2019, a team that included Peter Gibson tested kits from Immundiagnostik and the Chinese company CUSABIO². They used the kits’ antibodies to fish proteins out of blood, then identified the catch with mass spectrometry, a lab method that sorts molecules by mass².

The CUSABIO kit pulled out haptoglobin and complement C3, another immune protein. The Immundiagnostik kit pulled out C3². Neither, the team concluded, was detecting the real protein².

In 2021, Canadian GEM Project researchers, who study Crohn’s disease families, tried the Immundiagnostik kit in 39 healthy relatives of patients³. Its readings didn’t line up with a sugar-drink permeability test, and people with the HP1-1 gene type scored about the same as everyone else³.

How could a test drift so far from its target? A 2021 letter in the journal Gut, from a group that included two authors of the 2018 study, explained that the kits were built on an early protein sequence later found to be unrelated to zonulin¹⁴.

The letter flagged a surprise, too. Lab mice don’t make this form of zonulin, because the haptoglobin 2 gene type exists only in humans⁴. That casts doubt on mouse studies reporting blood zonulin levels.

The letter also noted work by Nicholas Talley and colleagues, who found that serum zonulin failed to pick out people with irritable bowel syndrome, functional dyspepsia or wheat sensitivity⁴.

None of this proves zonulin is unimportant; the letter’s authors said it may still help regulate the barrier⁴. The problem is practical. These kits don’t reflect real zonulin levels and track actual permeability poorly, so a result says little about whether your gut is leaky⁴.

The sugar-drink test is the research standard, with fine print

The test researchers trust more is simpler. You fast, drink a solution of two sugars your body barely breaks down, and collect your urine for several hours⁶.

Mannitol, the smaller sugar, passes through the lining cells fairly easily. Lactulose, a bigger one, mostly squeezes between cells where the seals are loose⁶. A higher ratio of lactulose to mannitol in your urine means a leakier small intestine⁶.

The test can catch real changes. In a 2014 study, 40 healthy women took it after 600 milligrams of aspirin and, a week apart, after a placebo⁵. Aspirin raised the lactulose coming through, while mannitol stayed the same⁵. A common painkiller loosened the barrier enough to measure.

But the fine print is long:

• There’s no single protocol. Studies use different sugar mixes, fasting times, urine windows and lab methods⁶. The 2014 team found results most consistent 2.5 to 4 hours after the drink; other studies collect for 2 to 24 hours⁵⁶.
• Colon bacteria interfere. Gut microbes break down lactulose in the colon, so the classic test mostly reflects the small intestine¹¹. Sucralose could in theory probe the whole gut, but that use is still experimental, and kidney function can sway the reading¹¹.
• Cutoffs aren’t agreed on. A 2025 review found that standard thresholds are lacking and that many methods have been checked only in lab samples, not large groups of patients¹¹.

A 2022 systematic review in BMC Gastroenterology pooled 34 studies of the sugar test in celiac and Crohn’s disease⁶. Its verdict: the test needs one agreed, optimized protocol before it can be widely used, including for functional gut disorders, a catch-all label for poorly understood gut problems. In its current form, the authors judged, it is unlikely to see wide clinical use⁶.

Even in celiac disease, where the leak is large, it’s a weak screen. In four studies, it caught about 83% of cases and gave a false alarm in about 3 of every 10 people who didn’t have the disease⁶. Our celiac guide covers the better tests.

A leaky gut test decoder

In this table, “validated” means checked against an accepted standard in enough people to guide care.

Test you may be offered	What it measures	Validated for leaky gut?	What a result can tell you	What it can’t tell you
Urine sugar test (lactulose-mannitol)	How much of two swallowed sugars reaches urine6	No; not standardized6	Whether small-bowel permeability looks higher than in research groups6	What caused it, or whether it causes your symptoms89
Blood zonulin	Whatever the kit’s antibodies bind; studies show it isn’t zonulin12	No123	Very little about your gut34	Whether your gut is leaky4
Stool zonulin	Proteins a zonulin kit detects in stool1011	No; research stage10	Tracked the sugar test in people with overweight or obesity in one small study10	A diagnosis for one person
LPS-binding protein (blood)	An inflammation alarm protein that rises when the barrier is impaired1011	No; promising for research10	Possible barrier stress in study groups10	The source, since infection and trauma raise it too11
I-FABP (blood or urine)	A protein released when small-bowel lining cells are injured11	Studied mainly for acute bowel injury11	Lining damage in emergencies such as blocked blood flow11	Everyday “leaky gut”
Antibody or toxin panels	Immune reactions to gut bacteria products9	No standard test exists9	Not established	Not established
Biopsy or advanced endoscopy	Tissue tested in a lab chamber, or a close-up view of the lining9	Research methods9	Direct evidence in a study	Not part of routine care9

Where the gut really does leak

In the 2022 review, healthy people averaged a lactulose-mannitol ratio of about 0.014⁶. People with untreated celiac disease averaged about 0.133, more than nine times higher. On treatment, their average fell to about 0.037⁶. Crohn’s disease followed its activity: about 0.093 during flares and 0.028 when quiet⁶.

Cleveland Clinic’s list of causes adds HIV, chemotherapy and radiation, food allergies, and chronic overuse of alcohol or of NSAID painkillers such as aspirin and ibuprofen⁹. Because the lining keeps renewing itself, the clinic says, wearing it down takes a heavy, sustained blow⁹. Arthritis, fibromyalgia and the other conditions tied to leaky gut in wellness marketing aren’t on that list⁹.

Chicken or egg? One Crohn’s study came closest

Celiac disease shows how a leak can follow a disease. Treat it, and the ratio drops⁶. Cleveland Clinic says treating the underlying condition is the only known fix; treatments aimed at the lining alone haven’t been shown to improve disease⁹.

Crohn’s disease offers the best test of the other direction. From 2008 to 2015, researchers recruited 1,420 healthy relatives of people with Crohn’s disease, ages 6 to 35⁷. Each took the sugar test at the start. Over a median of 7.8 years, 50 developed Crohn’s⁷.

A high ratio, above 0.03, came with about three times the risk of a later diagnosis⁷. The link held, more weakly, when the test was done more than three years before diagnosis⁷. The authors said this supports the idea that a faulty barrier helps set the disease in motion, and that it might someday serve as a risk marker⁷.

Read the numbers both ways. A leaky reading tripled the risk, yet 1,370 of the 1,420 relatives were still free of Crohn’s at the end of the study⁷. Even in a high-risk group, a leak was a warning light, not a diagnosis. Before this study, the CHEO Research Institute notes, it wasn’t clear whether barrier changes cause Crohn’s or result from it⁸.

So the answer is “it depends.” In celiac disease, the leak follows the disease and fades with treatment. In Crohn’s disease, it may come first in some people. For most other conditions, nobody knows.

Gut-lining supplements: what the trials tested

L-glutamine. This amino acid is a common ingredient in gut-healing kits. Our earlier article covered its most striking result, a trial in people whose IBS began after a gut infection.

The wider evidence is less exciting. A 2024 pooled analysis in the journal Amino Acids gathered randomized trials published from 1998 to 2014, in people with Crohn’s disease, HIV, cancer, burns, sepsis, pancreatitis or major surgery¹². Overall, glutamine made no significant difference to gut permeability¹².

Some slices of the data, such as trials of two weeks or less, did show an effect, so the authors called for more trials¹². None of the trials studied healthy people.

The paper is also sloppy in places. Its summary and results disagree on how many people took part (352 vs. 428), and it gives the dose in milligrams in some sections and grams in others¹².

Collagen. The only human study of collagen’s digestive effects we found appeared in 2022 in JMIR Formative Research¹³. Women took 20 grams of collagen peptides a day for eight weeks. Of 36 who started, 14 finished¹³.

Their bloating scores fell 31%. But there was no comparison group, no placebo and no blinding, and gut permeability was never measured¹³. When bowel habits shifted within days, the authors themselves wrote that gut healing was an unlikely reason, since healing would take weeks¹³.

Rousselot, which makes the collagen used, paid for the study. One author worked for the company, and another had been paid to consult for it¹³.

How strong is the evidence?

Claim	Evidence level	Why
Untreated celiac disease and active Crohn’s disease involve a leakier gut that improves with treatment	Strong evidence	Pooled sugar-test data from 34 studies6
Aspirin and other NSAIDs can loosen the barrier	Promising but early	One small randomized test of a single aspirin dose; Cleveland Clinic lists chronic NSAID overuse as a cause59
A leaky barrier can come before Crohn’s disease	Promising but early	One large family study; a risk marker, not proof of cause7
The sugar test can diagnose leaky gut in one person	Not enough evidence	Not standardized or validated outside research611
Leaky gut causes arthritis, fatigue syndromes or fibromyalgia	Not enough evidence	Links listed as unclear9
Glutamine tightens the barrier in everyday use	Not enough evidence	No overall effect in trials of sick adults; none tested healthy people12
Collagen heals the gut lining	Not enough evidence	One uncontrolled, company-funded study; permeability not measured13
Commercial blood zonulin tests show gut permeability	Evidence says no, or unsafe	Kits detect other proteins and don’t track sugar tests1234

How to judge a “leaky gut” diagnosis

Practitioners use “leaky gut” in different ways. Some mean a result on a research-grade test. Others mean a cluster of symptoms. Some sell both the test and the “fix.”

Natural health asks a good question: what’s the root of the problem? A careful practitioner should welcome the follow-up: how do we know? Asking for proof isn’t distrust. It’s the same habit that makes you read a label before you buy.

Seven questions to ask:

1. Which test, and which lab? If it was blood zonulin, ask how the lab answers studies showing the kits miss zonulin¹².
2. Where did the normal range come from? Sugar-test cutoffs vary with the protocol⁶¹¹.
3. What else could explain the result? Infection, injury, heavy drinking and painkillers can move these markers⁵⁹¹¹.
4. What disease does this point to? If it might be celiac or Crohn’s disease, ask for the standard tests for those conditions.
5. Will the result change what we do? If not, the test may not be worth paying for.
6. How will we know the plan worked? Agree on a symptom measure and an end date first.
7. Are you selling the treatment too? A test sold alongside its fix deserves a second opinion.

Questions readers ask

The bottom line

Your gut barrier is real, and its failures matter in diseases such as celiac and Crohn’s disease. That part of the leaky gut idea has earned its place in mainstream science. The testing and selling built around a vague “syndrome” have not.

The widely used zonulin blood test doesn’t measure what its name promises, the best research test isn’t ready to diagnose one person, and gut-lining supplements haven’t shown they tighten the barrier in daily life. Before you pay, ask what a test measures, what it will change and who profits from the answer.

Related reading

• Leaky gut can be measured, but most “heal your gut” protocols haven’t been tested
• When the company reading your labs also sells the fix
• What a medical test can and cannot tell you
• Same stool, different verdicts: what home gut microbiome tests can tell you
• Sick guts hold different microbes, and proving the microbes did it is harder

Sources

1. Scheffler L, Crane A, Heyne H, et al. “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, 2018. frontiersin.org
2. Ajamian M, Steer D, Rosella G, Gibson PR. “Serum Zonulin as a Marker of Intestinal Mucosal Barrier Function: May Not Be What It Seems.” PLoS ONE, 2019 (DOAJ record). pubmed.ncbi.nlm.nih.gov
3. Power N, Turpin W, Espin-Garcia O, et al., CCC GEM Project Research Consortium. “Serum Zonulin Measured by Commercial Kit Fails to Correlate With Physiologic Measures of Altered Gut Permeability in First Degree Relatives of Crohn’s Disease Patients.” Frontiers in Physiology, 2021. frontiersin.org
4. Massier L, Chakaroun R, Kovacs P, Heiker JT. “Blurring the Picture in Leaky Gut Research: How Shortcomings of Zonulin as a Biomarker Mislead the Field of Intestinal Permeability.” Gut (letter), 2021. pubmed.ncbi.nlm.nih.gov
5. Sequeira IR, et al. “Standardising the Lactulose Mannitol Test of Gut Permeability to Minimise Error and Promote Comparability.” PLoS ONE, 2014 (DOAJ record). pubmed.ncbi.nlm.nih.gov
6. Systematic review and meta-analysis of the lactulose:mannitol test in coeliac and Crohn’s disease (PROSPERO CRD42021259836). BMC Gastroenterology, 2022. link.springer.com
7. Turpin W, Lee SH, Raygoza Garay JA, et al. “Increased Intestinal Permeability Is Associated With Later Development of Crohn’s Disease.” Gastroenterology, 2020 (McMaster University record). experts.mcmaster.ca
8. CHEO Research Institute. “Increased Intestinal Permeability Is Associated With Later Development of Crohn’s Disease” (project summary). 2020. cheoresearch.ca
9. Cleveland Clinic. “Leaky Gut Syndrome.” Medically reviewed April 2022. my.clevelandclinic.org
10. Seethaler B, Basrai M, Neyrinck AM, et al. “Biomarkers for Assessment of Intestinal Permeability in Clinical Practice.” American Journal of Physiology: Gastrointestinal and Liver Physiology, 2021 (University College Cork record). research.ucc.ie
11. Zhang J, Wang T, Qi H, Zheng L. “Biomarkers in Body Fluids and Their Detection Techniques for Human Intestinal Permeability Assessment.” Clinical Chemistry and Laboratory Medicine, 2025. pubmed.ncbi.nlm.nih.gov
12. Abbasi F, Haghighat Lari MM, Khosravi GR, et al. “A Systematic Review and Meta-Analysis of Clinical Trials on the Effects of Glutamine Supplementation on Gut Permeability in Adults.” Amino Acids, 2024. link.springer.com
13. Abrahams M, O’Grady R, Prawitt J. “Effect of a Daily Collagen Peptide Supplement on Digestive Symptoms in Healthy Women: 2-Phase Mixed Methods Study.” JMIR Formative Research, 2022. pubmed.ncbi.nlm.nih.gov
14. American College of Gastroenterology. “Irritable Bowel Syndrome.” Patient information, accessed October 2026. gi.org