A frequently cited study did not prove that ordinary sourdough is safe for celiac patients. Researchers used selected bacteria, added enzymes and controlled processing to create very special sourdough that wasn’t reliably gluten free.
Every time sourdough comes up, someone says some version of the same thing:
“My friend has celiac disease and can eat real sourdough.”
“The long fermentation breaks down the gluten.”
“People in Europe eat sourdough and they are fine.”
Or my personal favorite:
“There was a study proving people with celiac disease can eat sourdough.”
There was a study.
That is not what the study proved.
Researchers did not take ordinary wheat flour, add a traditional sourdough starter, leave it on the counter for 48 hours and discover that time had magically turned wheat bread into gluten-free bread.
They used selected strains of sourdough bacteria plus fungal protease enzymes specifically chosen to break gluten proteins apart.
That distinction is not a minor technicality. It is the entire point.
This was not Aunt Susie’s beloved starter, created with wheat flour and maintained on her kitchen counter.
It was not the artisan loaf at the farmers market advertising a 48-hour fermentation.
It was not imported European wheat.
It was not ordinary commercial sourdough from the grocery store.
It was an experimental process involving selected microorganisms, added protein-degrading enzymes and tightly controlled fermentation conditions. Earlier laboratory work by the same research group specifically described using selected sourdough lactobacilli together with fungal proteases to degrade wheat gluten.
What did the sourdough study actually test?
In the small human trial, researchers gave people with celiac disease baked goods made with wheat flour that had undergone different degrees of hydrolysis.
The untreated baked goods contained approximately 80,127 parts per million of gluten.
A partially hydrolyzed version produced a result of approximately 2,480 ppm.
That sounds like a dramatic reduction.
It was.
It was also nowhere close to gluten-free.
The two participants who ate those partially hydrolyzed baked goods reported no clinical complaints, but both developed subtotal villous atrophy.
They felt fine.
Their small intestines were not fine.
That result matters because people regularly use symptoms as proof that sourdough must be safe:
“I ate it and nothing happened.”
“I can tolerate it.”
“It doesn’t bother my stomach.”
Celiac disease does not require immediate symptoms to cause intestinal injury. I discuss that more fully in my article about silent celiac disease:
The study itself showed the problem clearly. The people eating the partially hydrolyzed product did not report symptoms, yet their biopsies showed substantial damage.
Where did the sourdough myth come from?
The most extensively hydrolyzed product in the study produced a reported result of approximately 8 ppm gluten.
Five participants completed that arm of the study. After eating the experimental baked goods for 60 days, researchers did not find significant deterioration in their celiac blood tests or intestinal biopsy measurements.
That result is scientifically interesting.
It is not proof that ordinary sourdough bread is safe for people with celiac disease.
The product was made using:
- specially selected sourdough bacteria;
- added fungal proteases;
- controlled fermentation;
- specially processed wheat flour;
- and an experimental manufacturing protocol.
And the human evidence came from five people.
Not 105.
Five.
A 60-day study involving five participants and a specialized experimental product cannot establish that ordinary wheat sourdough is safe for the broader celiac population.
The reported 8 ppm result has another complication
Hydrolyzed gluten is unusually difficult to measure.
Standard sandwich ELISA tests work best when gluten proteins remain sufficiently intact for antibodies to recognize multiple binding sites. Fermentation and hydrolysis break those proteins into smaller peptides.
Once that happens, the assay may no longer recognize or accurately quantify everything that remains.
In other words, a lower test result may reflect genuine gluten degradation, but it may also reflect the test’s reduced ability to measure the altered protein fragments.
That is why it is more accurate to say the experimental bread produced an 8 ppm test result than to confidently declare that it contained only 8 ppm of every potentially relevant gluten fragment.
The FDA specifically states that conventional analytical methods may not accurately detect and quantify gluten in fermented foods. It also says sandwich ELISA methods do not adequately detect gluten in fermented and hydrolyzed foods.
The FDA’s final rule goes further: because gluten proteins are no longer intact after fermentation or hydrolysis, they currently cannot be adequately detected and quantified through finished-product testing.
The full FDA guidance is here:
The FDA final-rule summary is here:
The full Federal Register rule is here:
The Federal Register states that there is currently no scientifically valid analytical method capable of precisely detecting and quantifying gluten in fermented or hydrolyzed foods as an equivalent amount of intact gluten.
Why ELISA testing is difficult after gluten is hydrolyzed
Imagine an ELISA test as looking for a recognizable section of a gluten protein.
An intact protein may contain the repeated structures the test needs in order to detect and measure it properly.
Hydrolysis chops that protein into smaller fragments.
Some fragments may still be recognizable.
Some may have only one antibody-binding site instead of the two needed by a sandwich assay.
Others may be altered enough that the test does not measure them reliably.
That does not automatically mean those fragments are harmless to someone with celiac disease.
A 2019 review of gluten testing in fermented and hydrolyzed products describes accurate quantification as a major challenge because the degree of protein breakdown varies and appropriate reference materials are lacking.
Full research article:
https://pmc.ncbi.nlm.nih.gov/articles/PMC6611335/
This is why the FDA does not simply tell manufacturers to test a fermented finished product and rely on the resulting ppm number.
For a fermented or hydrolyzed food to carry a gluten-free claim, manufacturers must maintain records showing that the food or its ingredients met the gluten-free requirements before fermentation or hydrolysis, and that gluten cross-contact was evaluated and controlled.
That matters enormously.
Fermentation is not recognized by the FDA as a magic process that can take ordinary wheat flour and convert it into a legally gluten-free food.
Under FDA rules, a food made from a gluten-containing grain cannot simply become eligible for a gluten-free claim because fermentation later produced a low or uncertain test result.
Fermentation may reduce gluten
Here is where the myth gains traction.
Sourdough fermentation can alter and reduce gluten proteins.
That part is true.
But reduced gluten and gluten-free are not the same thing.
The partially hydrolyzed product in the human study contained far less measurable gluten than the untreated bread.
It still caused substantial intestinal damage.
A reduction of unknown size is not an adequate safety standard for celiac disease.
The same problem arises with some beers made from barley and then treated to break down gluten. They may be described as “gluten-reduced,” but hydrolyzed gluten cannot be reliably quantified in the finished beverage using conventional testing.
I discussed the same testing problem in this article:
The lesson is consistent:
Breaking gluten apart is not the same as proving that every celiac-triggering fragment has been removed.
Does longer fermentation make sourdough safer?
Longer fermentation may produce more protein breakdown than a short fermentation.
That still does not establish that the bread is gluten-free or safe for celiac disease.
Fermentation results depend on much more than time:
- the bacterial strains present;
- the activity of those strains;
- the flour;
- temperature;
- acidity;
- enzyme activity;
- hydration;
- and the overall production process.
A bakery cannot prove that its wheat bread is safe for celiac disease merely by advertising that it fermented for 24, 48 or 72 hours.
- silent celiac disease
https://fatceliac.net/2025/04/16/understanding-silent-celiac-disease-symptoms-and-risks/ - how I approach gluten-free labels
https://fatceliac.net/2026/02/11/gluten-free-labels-believe-them/ - why the absence of a gluten-free claim matters
https://fatceliac.net/2026/06/18/if-they-wont-call-it-gluten-free-why-should-i/ - gluten testing in fermented alcoholic drinks
https://fatceliac.net/2021/05/04/gluten-free-alcohol/ - Read Before You Eat
https://fatceliac.net/resources/
“Long fermented” is a description of process.
It is not a gluten-free certification.
It does not tell you how much intact gluten remains.
It does not tell you which peptides remain.
It does not tell you whether the product can trigger a celiac immune response.
And it does not recreate the research protocol involving selected bacteria and added fungal proteases.
SEO title:
Is Sourdough Safe for Celiac Disease? What the Study Really Found
But what about people who say sourdough does not bother them?
Some of those people may have non-celiac gluten sensitivity, non-celiac wheat sensitivity or irritable bowel syndrome rather than celiac disease.
That distinction matters.
Some people who believe they react to gluten may instead be reacting to fructans or other fermentable carbohydrates in wheat. Sourdough fermentation may lower fructan content, making certain breads easier for some people to tolerate.
Other people may react to another wheat component.
And some people may simply tolerate smaller amounts of gluten without obvious symptoms.
A study of partial sourdough fermentation found that fermentation altered the digestion profile of immunogenic gluten peptides, but did not necessarily reduce the total concentration of immunogenic peptides.
Full article:
https://pmc.ncbi.nlm.nih.gov/articles/PMC8229354/
For a person without celiac disease who is managing symptoms, individual tolerance may be part of the decision.
They may reasonably ask:
Can I eat this bread without feeling sick?
Celiac disease asks a different question:
Can I eat this bread without triggering an autoimmune response and damaging my small intestine?
Those are not the same standard.
A person with non-celiac gluten or wheat sensitivity may tolerate a traditionally fermented sourdough loaf.
That does not prove the bread is gluten-free.
It means that person may tolerate that particular bread.
People with celiac disease cannot safely use symptoms alone as the test.
Ordinary wheat sourdough is still wheat bread
I understand why people want this myth to be true.
Good sourdough is delicious.
The idea that time, tradition or European technique transforms wheat bread into something safe feels hopeful.
But hope is not the same thing as evidence.
Ordinary sourdough made with wheat flour contains wheat gluten.
Long fermentation may change that gluten.
It may reduce some of it.
It may break portions of it into smaller fragments.
It does not automatically turn the bread into a gluten-free food.
If a packaged food carries a gluten-free claim, that claim must meet specific FDA ingredient and manufacturing requirements. I explain how I approach gluten-free labeling here:
I also explain why the absence of a gluten-free claim matters here:
And for people who are tired of sorting through contradictory food advice, my free Read Before You Eat guide provides a practical system for evaluating gluten-free food decisions:
The bottom line
Long fermentation does not automatically make wheat sourdough safe for celiac disease.
The often-cited study did not test ordinary bakery sourdough.
Researchers used specially selected sourdough bacteria, added fungal protease enzymes and tightly controlled processing.
The partially hydrolyzed bread still damaged the intestines of participants who reported no symptoms.
The most extensively hydrolyzed product was eaten by only five people for 60 days.
And the reported 8 ppm result must be understood in the context of serious limitations in testing hydrolyzed gluten.
Some people with non-celiac gluten or wheat sensitivity may tolerate long-fermented sourdough.
People with celiac disease should not eat ordinary wheat sourdough.
Not because everyone will immediately feel sick.
Because feeling fine is not the same as being safe.
Scientific research and regulatory sources
Greco et al., 2011
Safety for patients with celiac disease of baked goods made of wheat flour hydrolyzed during food processing
https://pubmed.ncbi.nlm.nih.gov/20951830/
Rizzello et al., 2007
Highly efficient gluten degradation by lactobacilli and fungal proteases during food processing: new perspectives for celiac disease
https://pubmed.ncbi.nlm.nih.gov/17513580/
Open-access full text:
https://pmc.ncbi.nlm.nih.gov/articles/PMC1932817/
Panda et al., 2019
Detection and Quantitation of Gluten in Fermented-Hydrolyzed Foods by Antibody-Based Methods: Challenges, Progress, and a Potential Path Forward
https://pmc.ncbi.nlm.nih.gov/articles/PMC6611335/
Ogilvie et al., 2021
A Case Study of the Response of Immunogenic Gluten Peptides to Sourdough Fermentation
https://pmc.ncbi.nlm.nih.gov/articles/PMC8229354/
FDA questions and answers on gluten-free labeling
FDA final rule for fermented and hydrolyzed foods
Full Federal Register rule
FAQ
Is traditional sourdough gluten-free?
No. Traditional sourdough made with wheat, barley or rye is not gluten-free merely because it has undergone a long fermentation.
Does sourdough fermentation remove gluten?
Fermentation may break down or reduce some gluten proteins, but ordinary fermentation does not reliably remove enough gluten to make wheat sourdough safe for people with celiac disease.
What did the sourdough celiac study find?
The study tested wheat flour processed with selected sourdough bacteria and added fungal protease enzymes. A partially hydrolyzed product caused intestinal damage despite participants reporting no symptoms. Only five people completed the extensively hydrolyzed arm of the 60-day study.
Why is hydrolyzed gluten difficult to test?
Hydrolysis breaks intact gluten proteins into smaller peptides. Standard sandwich ELISA methods may not recognize or accurately quantify all of those fragments, which is why the FDA does not rely solely on finished-product testing for fermented and hydrolyzed foods.
Can someone with non-celiac gluten sensitivity eat sourdough?
Some people with non-celiac gluten or wheat sensitivity may tolerate certain long-fermented breads better, possibly because fermentation alters fructans or other wheat components. Individual tolerance does not make the bread gluten-free or safe for celiac disease.
Can someone with celiac disease eat sourdough if it causes no symptoms?
No symptoms do not prove that no intestinal damage occurred. People with celiac disease may experience silent intestinal injury after gluten exposure.
Leave a Reply