Fermented foods: what sauerkraut, kefir and kimchi can do
A randomised study over 17 weeks showed both more microbial diversity and fewer inflammatory signals at the same time. There were 18 people per group. Both belong in the same sentence, and that is exactly what this article is about.
All articles in the gut cluster
Fermented food is not a medicine and not a substitute for a proper work up. It is, however, one of the few foods for which a randomised human study has described more microbial diversity and fewer inflammatory signals at the same time. That study is small, and its primary endpoint stayed unchanged. Both belong in the same sentence.
You are standing in front of the chilled aisle holding two jars of sauerkraut. One costs one euro and used to sit next to the tins. The other costs four euros, was refrigerated, and the lid carries a word you would not have expected: not pasteurised.
Someone told you the living bacteria are the decisive part. Someone else told you they do not survive the stomach anyway.
Many people know exactly this moment. It is typical for this topic. A great deal of enthusiasm, a great many rules, very few numbers.
So I do it the other way round. First the numbers, then the rules. And then the honest place where the numbers stop.
What awaits you here
- What happens in the jar when cabbage becomes sauerkraut
- The Stanford study with its real numbers and its real limits
- Kefir, yoghurt, sauerkraut, kimchi, miso, tempeh, kombucha checked one by one
- Whether the bacteria have to arrive alive, and what postbiotics have to do with it
- Which supermarket products actually contain living cultures
- Which amounts were given in the studies, and why that is not a recommendation
- For whom fermented food is not harmless, in detail and without glossing over
- Why the guidelines stay silent on it, and why that is not a rejection
Fermented food is not harmless for everyone. This paragraph deliberately sits right at the top and not as a footnote at the end.
- Histamine intolerance. Fermented foods are the one group that every published low histamine diet agrees on. More on that in the article on histamine intolerance and DAO.
- Relevant immunosuppression or serious underlying illness. Large amounts of living cultures should be discussed with a doctor beforehand.
- Acute flare of Crohn disease or ulcerative colitis. There are no robust data on this, neither for nor against. Details in the article on inflammatory bowel disease.
- High blood pressure and salt sensitive situations. Sauerkraut and kimchi are salt preserves, and salt counts in the daily balance.
- Irritable bowel with strong gas production. Fermented vegetables bring fermentable carbohydrates along. Small amounts and slow increases are not hesitation here, they are craft.
- Vitamin K antagonists such as phenprocoumon or warfarin. Natto is the food with the highest vitamin K2 content of all. Even a small portion can shift the anticoagulation setting. That belongs in a conversation with the practice that manages your INR beforehand.
- MAO inhibitors and the antibiotic linezolid. Fermentation produces not only histamine but also tyramine. Mature cheese, sauerkraut, kimchi, miso and soy sauce are among the classic tyramine sources. The combination can raise blood pressure sharply.
- Pregnancy and breastfeeding. Raw milk cheese, raw milk kefir and other unpasteurised dairy products can contain listeria, and listeriosis in pregnancy can endanger the child. Raw kombucha adds alcohol on top. In this period the pasteurised product is the right choice and not the second best one.
- Cow milk protein allergy. Casein and whey protein remain fully intact in kefir. Fermentation does not help here, and the reaction can be severe.
And one more thing: nothing in this article is a reason to change a prescribed medication or to postpone a recommended work up. You do not adjust any of it yourself, every adjustment belongs in medical hands. A jar of sauerkraut is no substitute for a colonoscopy or for laboratory diagnostics.
Red flags: this calls for an appointment, not a nutrition experiment
- Blood in the stool or black, tarry stool
- Unintended weight loss without explanation
- Fever together with abdominal complaints
- Complaints that wake you at night
- Repeated vomiting or difficulty swallowing
- A new, persistent change in bowel habit from around 45 to 50 years of age
- Anaemia in the laboratory, especially iron deficiency anaemia without an obvious reason
- Bowel cancer or inflammatory bowel disease in the family
These signs need medical assessment and should not be treated with food. If a colonoscopy, a gastroscopy or laboratory diagnostics has been recommended, getting there is the first task. Everything in this article comes afterwards.
What happens in the jar when cabbage becomes sauerkraut
Ask yourself something simple for a moment. Why does a jar of sauerkraut keep for months when fresh cabbage goes mushy after two weeks?
The answer is the whole trick. Fermentation is not an addition. It is a conversion.
When cabbage is shredded and salted, the salt draws water out of the cells. The vegetable lies in its own brine, the air is displaced, and a low oxygen environment forms. That is exactly where lactic acid bacteria feel at home, and they were already sitting on the leaves anyway. They need nobody to add them.
These bacteria do what they do best. They eat sugar and starch and give off lactic acid. The pH falls, often below 4. In that acidic environment putrefying microbes have a hard time, because most of them cannot handle it. The jar largely preserves itself.
Largely is the decisive word here. If there is too little salt, if air gets in or if the acidification does not get going, this protection does not hold. That is why a bulging lid on a home batch is not a sign of liveliness but a reason to throw it out, not least because of the rare but serious risk of botulism poisoning in badly run batches.
Think of a building site with strict access control. Salt and lack of oxygen are the fence. The lactic acid is the bouncer. And the bacteria that arrived first settle in.
Five things that change in the food during fermentation
- Sugar disappears. Lactic acid bacteria and yeasts metabolise part of the carbohydrates. That is why fermented vegetables do not taste sweet, and why mature cheese contains almost no lactose any more.
- New substances appear. Lactic acid and acetic acid, short chain fatty acids, bioactive peptides from the protein, exopolysaccharides and, in dairy products, conjugated linoleic acid. Chemically, the end product is no longer the starting product.
- Individual vitamins increase. Some microorganisms produce B vitamins, and under certain conditions folate as well. That is no reason to sell a fermented food as a vitamin source, but it explains part of the nutritional differences.
- Antinutrients are broken down. Phytate binds minerals such as zinc and iron. Fermentation can partly split it. With sourdough and with fermented soy products such as tempeh, that is the more interesting side effect.
- Fermentable carbohydrates can fall. With deliberately guided sourdough that is well documented. With kimchi and its generous garlic and onion, the opposite tends to apply, because new ones are added there.
The last point is the most important one for anyone with irritable bowel syndrome. Fermented does not automatically mean easier to digest. It depends on the product.
That leaves the question of what a fermented food actually is. There has been a clean definition for it since 2021, and it comes from an international expert panel.
A panel of the International Scientific Association for Probiotics and Prebiotics, ISAPP for short, set out in 2021 what fermented foods are: foods made through desired microbial growth and enzymatic conversion of food components.
The same paper explicitly separates fermented foods from probiotics and notes that they do not yet form a category of their own in national dietary recommendations.
For you that means: beer, wine, sourdough bread and coffee are fermented foods under this definition. They still contain no living microorganisms at the end, because they are heated, filtered or baked. Fermented and probiotic are two different things.
Marco ML, Sanders ME, Gänzle M et al. Nat Rev Gastroenterol Hepatol. 2021;18(3):196-208. PMID: 33398112 · DOI: 10.1038/s41575-020-00390-5 [Consensus paper, scientific association]The widespread idea goes: I eat sauerkraut so that the bacteria from the jar move into my gut.
That is the weakest justification of all. A sturdier one goes: I eat a food that microorganisms have already predigested, and I take their metabolic products along with it. Whether the bacteria themselves stay is an entirely different question. We come back to it later.
And one more boundary that frames this article. Here it is only about foods. Capsules, spore based products and colony forming units are covered in the article on probiotics in spore form and capsule. The feed for your own bacteria, meaning inulin, resistant starch and fibre, is covered in the article on prebiotics.
And now you know why the word fermented on its own says nothing at all about living cultures.
The Stanford study, read honestly
There is one piece of work in this field that everyone refers to. Usually without numbers, often without author names, almost always without the participant count.
Yet that is exactly what should be in there. So, one thing at a time.
A team around Hannah Wastyk and Gabriela Fragiadakis at Stanford University split 36 healthy adults into two groups of 18 people each. One group increased fermented foods, meaning yoghurt, kefir, fermented cottage cheese, fermented vegetables, vegetable brine drinks and kombucha. The other group increased plant fibre. Seventeen weeks in total, with deep profiling of microbiome and immune system.
In the arm with fermented foods, microbial diversity rose steadily across the intervention phase, and the authors report a decrease in 19 of the 93 measured inflammatory proteins. In the fibre arm, community diversity stayed stable, while microbially encoded glycan degrading enzymes increased. And here belongs the most important sentence: the primary endpoint, a cytokine response score, did not change in either arm. The rise in diversity and the drop in inflammatory proteins are secondary and exploratory endpoints. In the fibre arm, three different immune trajectories showed up instead, depending on how diverse the microbiome had been at baseline.
For you that means: this is the best randomised work so far showing that an eating pattern can raise microbial diversity and lower inflammatory signals. It is 18 people per group, all healthy, over four months, with no disease endpoint. Both belong in the same sentence.
Wastyk HC, Fragiadakis GK, Perelman D et al. Cell. 2021;184(16):4137-4153.e14. PMID: 34256014 · DOI: 10.1016/j.cell.2021.06.019 [RCT, n=36]Why do I stress the 18 so much? Because in the German language coverage this study turned into proof. It is not proof. It is a strong signal from a small group.
A strong signal from a small group is still worth something. Particularly when a second, entirely differently built investigation points in the same direction.
A group around Bryn Taylor analysed stool samples from 6,811 people in the American Gut Project, with sequencing and untargeted mass spectrometry. A further 115 people were recruited specifically by how often they ate fermented foods and followed over four weeks.
Between consumers and non consumers there were subtle but statistically robust differences in the composition of the microbiome. In the metabolic profile of the consumers, conjugated linoleic acid was enriched.
For you that means: people who regularly eat fermented foods have a measurably different microbiome. The effect is real and small. And it is an observation, not proof of cause and effect, because people who drink kefir usually differ in the rest of their lifestyle too. It should also be noted that three authors of this work come from the research arm of a dairy company. That does not devalue it, but it belongs in the picture.
Observational data, scientific context, not a health claim within the meaning of the European health claims regulation.
Taylor BC, Lejzerowicz F, Poirel M et al. mSystems. 2020;5(2):e00901-19. PMID: 32184365 · DOI: 10.1128/mSystems.00901-19 [Cross section with small longitudinal subgroup, n=6811]Why the fibre group is not an argument against fibre
People like to turn the Stanford work into: fermented beats fibre. That is not what it says.
What it does say: in ten weeks, more fibre did not raise community diversity. Something changed nonetheless, namely the enzyme equipment of the bacteria. They adapted their tools to the new feed before the composition shifted. And the immune trajectory depended on how diverse the microbiome had been beforehand.
So the obvious reading is not either or. It goes: fibre possibly needs more time and a suitable starting position. How much there is to the number 30 is covered in the article on fibre myths.
The usual sentence goes: I need more good bacteria.
The sturdier one goes: I need a more diverse community. In the Stanford work, diversity is the parameter that moved, not the amount of a single strain. And diversity builds over weeks through what lands on the plate regularly, not through one single ingredient.
And now you know why I cite this study and still do not turn it into proof.
Kefir, sauerkraut, kimchi, miso: what the data give per food
Now it gets uncomfortable. Because as soon as you take the family apart, the data get very unevenly thick.
For some of it there are reproducible effects in humans. For other parts there are twelve participants and a lot of marketing. I go through it one by one, and the format of the boxes shows you immediately how robust each piece is.
Kefir: the strongest hard finding in the whole field
Kefir is something different from yoghurt. It is set with kefir grains, a living community of lactic acid bacteria, acetic acid bacteria and yeasts. That is why it fizzes slightly and contains traces of carbon dioxide.
Steven Hertzler and Shannon Clancy gave 15 adults with lactose maldigestion a series of test meals containing 20 grams of lactose each: as milk, as plain kefir, as fruit kefir, as plain yoghurt and as fruit yoghurt. They then measured breath hydrogen and symptoms over eight hours.
The hydrogen curve was 224 ppm times hour for milk, 76 for plain yoghurt and 87 for plain kefir. Fruit kefir sat in between at 156. Perceived flatulence severity fell by 54 to 71 percent compared with milk.
For you that means: fermentation can make milk sugar measurably easier to digest, and yoghurt and kefir do that similarly well. The fruit addition worsens the result. How lactose intolerance is established in the first place is covered in the article on lactose, fructose and sorbitol.
Hertzler SR, Clancy SM. J Am Diet Assoc. 2003;103(5):582-587. PMID: 12728216 · DOI: 10.1053/jada.2003.50111 [RCT, n=15]A Turkish team around Ezgi Bellikci-Koyu gave 62 people with metabolic syndrome either 180 millilitres of kefir daily or 180 millilitres of unfermented milk for twelve weeks.
Apolipoprotein A1 rose by 3.4 percent under kefir and fell by 2.4 percent under milk, and the difference was statistically significant. TNF alpha, interleukin 6, interleukin 10, interferon gamma and homocysteine fell under kefir compared with baseline. In the subgroup with high LDL, LDL fell by 7.6 percent against the starting value, but not significantly against the milk group. Blood pressure fell in both groups.
For you that means: in the before and after comparison, kefir looks good. In the direct comparison against ordinary milk, little of the advantage remains. That is exactly how it needs writing down, otherwise it turns into an advertising slogan.
Bellikci-Koyu E, Sarer-Yürekli BP, Karagözlü C et al. Nutr Res. 2022;102:59-70. PMID: 35405603 · DOI: 10.1016/j.nutres.2022.02.006 [RCT, n=62]The same working group had looked at the microbiome earlier, in only 22 people in total. Under kefir, fasting insulin, HOMA-IR and blood pressure fell against baseline, but compared with the milk group no parameter stayed significant. In the microbiome, exactly one bacterial group rose measurably. Anyone selling kefir as a rebuild of the gut flora goes beyond these data. Bellikci-Koyu E et al. Nutrients. 2019;11(9):2089. PMID: 31487797 · DOI: 10.3390/nu11092089 [RCT, small]
Sauerkraut: a single study, and it is the most interesting one
On sauerkraut in irritable bowel syndrome there is exactly one controlled investigation. It is small, and it says something that contradicts the usual narrative.
A Danish group around Elsa Sandberg Nielsen gave 34 Norwegian people with irritable bowel syndrome sauerkraut daily for six weeks. Fifteen received pasteurised product, 19 received unpasteurised. Neither the participants nor the investigators knew who got what.
The symptom score IBS-SSS fell by an average of 38.57 points under pasteurised sauerkraut and by 56.99 points under unpasteurised. Both groups improved statistically significantly, the difference between the groups did not. The microbiota changed in both groups. Sauerkraut typical lactic acid bacteria were found more often in the stool only in the group with raw product.
For you that means: the heated sauerkraut helped in this study too. The authors attribute the effect more to the prebiotic components than to living microbes, and they call their own study underpowered. Even so, this is the sentence at which the story that only living bacteria count begins to wobble.
Nielsen ES, Garnås E, Jensen KJ et al. Food Funct. 2018;9(10):5323-5335. PMID: 30256365 · DOI: 10.1039/c8fo00968f [RCT, n=34]Kimchi: plenty of reputation, thin data
Kimchi is a fermented Chinese cabbage with chilli, garlic, ginger and fish sauce or its vegetarian version. There are studies on it, but they are small.
A Korean team around Kyungsun Han gave 24 women with obesity either fresh or fermented kimchi for eight weeks and examined stool and gene expression in blood.
The two versions differed in their effects on obesity related parameters. The links between clinical values, gene expression and microbiota were more pronounced in the fermented group.
For you that means: a hint that the fermentation itself and not just the vegetable makes the difference. With 24 participants that is a hint and not a basis for a recommendation.
Han K, Bose S, Wang JH et al. Mol Nutr Food Res. 2015;59(5):1004-1008. PMID: 25688926 · DOI: 10.1002/mnfr.201400780 [RCT, n=24]There is also a larger piece of work with 114 people in which a bacterial strain isolated from kimchi was tested as a capsule against placebo. Body fat mass differed by 0.8 kilograms in favour of the supplement at the end, waist circumference by 0.8 centimetres. That is a decent study, but it says nothing about kimchi as a food. It says something about a capsule. This exact mix up is standard online. It should be noted that two authors of this work are employed by the manufacturer of the tested preparation. That does not devalue it, but it belongs in the picture. Lim S, Moon JH, Shin CM et al. Endocrinol Metab (Seoul). 2020;35(2):425-434. PMID: 32615727 · DOI: 10.3803/EnM.2020.35.2.425 [RCT, n=114]
Miso, natto, tempeh: the largest number in the whole article
The Japanese JPHC cohort around Ryoko Katagiri followed 92,915 people between 45 and 74 years of age for an average of 14.8 years and recorded diet by questionnaire. During that time 13,303 people died.
Total soy intake had nothing to do with mortality. For fermented soy products, meaning natto and miso, the hazard ratio in the highest against the lowest fifth was 0.90 in men and 0.89 in women. In men the value sat exactly on the threshold of significance, and that belongs in the picture. Natto additionally showed an inverse association with cardiovascular mortality.
For you that means: in the largest investigation on this topic, it is precisely the fermented subgroup that stands out. It remains an observation. The authors themselves warn about unadjusted confounders, because people who eat a lot of natto usually eat differently in other ways too.
Observational data, scientific context, not a health claim within the meaning of the European health claims regulation.
Katagiri R, Sawada N, Goto A et al. BMJ. 2020;368:m34. PMID: 31996350 · DOI: 10.1136/bmj.m34 [Cohort, n=92915]Natto is the food with the highest vitamin K2 content of all. If you take a vitamin K antagonist such as phenprocoumon or warfarin, even a small portion can shift your anticoagulation setting. A systematic review of the acceptable vitamin K intake under these medications concludes that fermented soybeans are the one food consistently advised against, while green leafy vegetables remain possible in limited amounts.
Natto then does not belong on the plate without a prior conversation with the practice that manages your INR. For miso in culinary amounts the same does not apply to the same degree, for natto it very much does. This is no reason to panic, but it is a reason to ask before you start. And it is no reason to change anything about the medication itself.
Sato Y, Murata M, Chiba T, Umegaki K. Shokuhin Eiseigaku Zasshi. 2015;56(4):157-165. PMID: 26346860 · DOI: 10.3358/shokueishi.56.157 [Systematic Review]
On Korean doenjang, a fermented soy paste, there is a small randomised study in 56 women going through the menopause. What is remarkable about it is less the result than one detail: the symptom index improved in all groups, but most clearly in the group with the lower microbial count. More living bacteria did not bring more effect here. It should be noted that three authors of this work are based at the state fermentation institute that produces the tested product. Han AL, Ryu MS, Yang HJ et al. Nutrients. 2024;16(8):1194. PMID: 38674884 · DOI: 10.3390/nu16081194 [RCT, n=56]
Yoghurt and cheese: the unspectacular majority
The largest data sets on fermented foods come not from the health food shop but from the chilled aisle.
In the European EPIC-InterAct study with 340,234 people and almost four million person years, total dairy intake had no association with type 2 diabetes. Combined intake of fermented dairy products, by contrast, was inversely associated, with a hazard ratio of 0.88.
The Dutch Hoorn cohort with 2,262 adults found a 17 percent lower risk of prediabetes over 6.4 years for high fat fermented dairy, and 21 percent for high fat cheese. The multinational PURE study with 136,384 people from 21 countries found a hazard ratio of 0.86 for yoghurt alone for the combined endpoint.
For you that means: three independent large observations point in the same direction. In EPIC-InterAct the signal hangs on the fermentation rather than on the fat content, because total dairy showed no association there. In the Hoorn cohort it was different: there the association appeared precisely in the high fat fermented products and in high fat cheese, not in the low fat versions. In PURE total dairy intake was favourably associated as well, with a hazard ratio of 0.84, so the yoghurt signal cannot be cleanly separated from the other dairy products there. Observation remains observation. Three different populations pointing the same way makes the pattern more interesting, but it proves nothing.
Observational data, scientific context, not a health claim within the meaning of the European health claims regulation.
Sluijs I et al. Am J Clin Nutr. 2012;96(2):382-390. PMID: 22760573 · DOI: 10.3945/ajcn.111.021907 · Slurink IAL et al. Eur J Nutr. 2022;61(1):183-196. PMID: 34245355 · Dehghan M et al. Lancet. 2018;392(10161):2288-2297. PMID: 30217460 [Cohort, n=340234]The obvious explanation for that comes from a review by Arne Astrup and colleagues. Their thesis: a food is more than the sum of its nutrition table. Full fat and fermented dairy products do not show the unfavourable effects you would have predicted from their saturated fat and sodium content. The matrix counts, not the single nutrient. Astrup A, Geiker NRW, Magkos F. Adv Nutr. 2019;10(5):924S-930S. PMID: 31518411 · DOI: 10.1093/advances/nmz069 [Mechanism Review] Whether milk suits you in principle is a separate question and is covered in the article on dairy products.
Kombucha: twelve people, and one important side number
An American team around Chagai Mendelson gave twelve adults with type 2 diabetes either 240 millilitres of kombucha daily or a placebo for four weeks, then an eight week washout and then a switch.
Self measured fasting blood glucose fell under kombucha from 164 to 116 mg/dl, and under placebo from 162 to 141 mg/dl without statistical significance. In the product itself, lactic acid and acetic acid bacteria as well as yeasts were found. Lactic acid and acetic acid were each below one percent, ethanol at 1.5 percent.
For you that means: twelve people are twelve people. No recommendation can be built from that. The practically more important number is the alcohol content, and it is relevant for pregnant women, for people who are breastfeeding, for people in abstinence and for children.
Mendelson C, Sparkes S, Merenstein DJ et al. Front Nutr. 2023;10:1190248. PMID: 37588049 · DOI: 10.3389/fnut.2023.1190248 [RCT, n=12]The sentence I correct most often goes: fermented food is good for the gut.
Fermented is not a category with one effect. It is a family of very different foods with very different data. Kefir and lactose are well documented. Yoghurt and metabolism are well observed. Sauerkraut and irritable bowel are a pilot trial. Kombucha is twelve people. Anyone who does not separate those is selling a story instead of a tool.
And now you know why I phrase things differently for kefir than for kombucha.
Do the bacteria have to arrive alive?
This question decides whether you spend four euros or one euro on your sauerkraut. And it is answered less well than the price difference suggests.
First the journey. What you swallow lands in a stomach with a pH that on an empty stomach can be well below 3. After that comes bile, which attacks cell membranes. Then a transit lasting hours. For a bacterium that is a hike through the desert.
Some get through. How large that share is depends on the strain, on the food matrix and on whether you eat with a meal. Fat and protein can buffer the acid. Yoghurt at breakfast is therefore possibly something different from kefir on an empty stomach.
Incidentally, that is exactly one reason why I like to look at stomach acid early. Anyone who produces too little acid may not only stall on protein breakdown but possibly also have a weaker bouncer at the entrance. How strong that link is has not been settled scientifically. What I describe here is a consideration from practice and not a study result. Whether it plays a role in your case is settled by an examination and not by a blog article.
Betaine HCl, which is often mentioned in this context, is not a harmless household remedy. With a stomach ulcer, an active gastritis, under painkillers of the NSAID type or under an ongoing acid blocker therapy it can irritate the mucosa further. A self experiment on suspicion is therefore not part of this. How to assess it properly and when it comes into question at all is covered in the article on low stomach acid and betaine HCl.
And because this paragraph is easily misread, here is the addition that goes with it. Anyone taking an acid blocker does not stop it themselves because of this and does not reduce it on their own either. These medicines are prescribed for good reasons, and stopping abruptly can make complaints temporarily worse. Whether long term therapy still fits is a question for the consultation and not for a blog article. How acid blockers work and when a joint review can make sense is covered in the article on heartburn and acid blockers.
Now the second half of the answer. And that is the genuinely interesting part.
A second ISAPP panel around Seppo Salminen defined the term postbiotic in 2021: a preparation of inanimate microorganisms or their components that confers a health benefit on the host.
The decisive part of the definition is the word inanimate. Effective postbiotics have to contain inactivated microbial cells or cell components, with or without metabolic products.
For you that means: the field has created its own term for the fact that something can trigger a biological response without being alive. Cell wall components, peptides and acids do not disappear when you heat a jar.
Salminen S, Collado MC, Endo A et al. Nat Rev Gastroenterol Hepatol. 2021;18(9):649-667. PMID: 33948025 · DOI: 10.1038/s41575-021-00440-6 [Consensus paper, scientific association]That fits with what happened in the Norwegian sauerkraut study. Both groups improved, including the one with heated product. The authors themselves write that the effect is more likely attributable to the prebiotic components than to the living lactic acid bacteria.
And it fits with the Korean doenjang study, in which the group with the lower microbial count showed the larger effect.
The common question goes: are the bacteria still alive?
The better question goes: what arrives? What arrives in any case are the acids and the altered carbohydrates. Cell components and peptides can trigger biological effects, and how far that carries in detail is not yet settled. Living microbes can arrive on top of that. The first is the more reliable expectation, the second is a bonus.
Practical consequence: unpasteurised is the better choice when you can get it and when you do not belong to one of the groups for which that does not apply. In pregnancy it is the other way round. Raw milk cheese, raw milk kefir and other unpasteurised dairy products can contain listeria, and listeriosis in pregnancy can endanger the child. In this period the pasteurised product is the right choice and not the second best one. The same applies with relevant immunosuppression. Pasteurised is not a mistake otherwise either. The expensive mistake would be to eat none of it because the good version is too dear.
And now you know why I never talk anyone out of their tinned sauerkraut.
What is actually alive in the supermarket
Back to the shelf. Now you want to know how to tell.
The first rule is banal and carries far: anything that keeps for months unrefrigerated has been through a heat treatment. Living cultures need cooling. A jar that stands at room temperature for two years contains none.
The second rule is in the small print. Look for the words pasteurised, heat treated or preserved. And look, conversely, for raw, unpasteurised, with live cultures, or the note to keep refrigerated and use promptly after opening.
| Product | Living cultures likely | What to look for |
|---|---|---|
| Sauerkraut from the chilled aisle | yes, if declared raw | Check the label for raw or unpasteurised, cloudy brine is a good sign |
| Sauerkraut from a tin or unchilled jar | no | Heated and therefore shelf stable. The prebiotic components and the acids remain |
| Plain yoghurt | usually yes | The note heat treated after fermentation means: no living cultures left |
| Kefir | yes with real kefir | Labels such as kefir style product often mean a different culture without yeasts |
| Kimchi | yes with chilled product | Mind the salt content, garlic and onion matter with irritable bowel syndrome |
| Miso | yes with unpasteurised paste | Stir in only after cooking, do not boil it. Very salty |
| Tempeh | mostly no after frying | Eaten heated. The value lies in the broken down protein, not in living microbes |
| Kombucha | yes with raw product | Check residual sugar and alcohol content on the nutrition table |
| Sourdough bread | no, it is baked | A genuine long ferment against mere sourdough flavour in the ingredient list |
| Beer, wine, coffee | no | Fermented in the sense of the definition, without living microorganisms in the end product |
The special case of sourdough
Sourdough bread contains no living cultures at the end, because nothing survives the oven. Even so, many people report that genuine sourdough bread sits better. There is a plausible explanation for that from food chemistry.
A Viennese group around Vera Fraberger tested 13 sourdough typical yeast strains in a model system with a typical wheat carbohydrate profile and measured over four days how much fructan remained.
The strains differed clearly in their ability to break down fructans. One strain from Austrian traditional sourdough broke down the most.
For you that means: fermentation can lower fermentable carbohydrates, but it depends on the strain and on how the dough is handled. That is laboratory work and not a human study. A review of sourdough technology reaches the same conclusion and describes deliberately guided, low FODMAP rye sourdough breads.
Fraberger V, Call LM, Domig KJ, D'Amico S. Nutrients. 2018;10(9):1247. PMID: 30200589 · DOI: 10.3390/nu10091247 [In vitro] · Loponen J, Gänzle MG. Foods. 2018;7(7):96. PMID: 29932101 · DOI: 10.3390/foods7070096 [Mechanism Review]In practice that means: a bread with a twenty four hour ferment is something different from a yeast bread with sourdough extract for flavour. The difference is in the ingredient list. If you are working to FODMAP, that belongs in your plan. How the protocol is built is covered in the article on the FODMAP diet.
One thing absolutely needs saying here, because it otherwise gets lost. Easier to digest does not mean gluten free. Even a long guided sourdough does not break gluten down completely, and for people with coeliac disease wheat bread remains unsuitable, however good the ferment.
The timing matters even more. If coeliac disease is suspected, please do not drop gluten beforehand. A gluten free diet before the diagnostic work up can change antibodies and tissue findings so much that the diagnosis becomes impossible afterwards. First the work up, then the dietary change, and not the other way round. How that work up runs is covered in the article on recognising coeliac disease. If wheat causes complaints without coeliac disease being present, that is a separate question and is covered in the article on gluten and gliadin without coeliac disease.
Many people read the label like a value judgement. Raw equals good, pasteurised equals worthless.
The question of purpose makes more sense. If you want living cultures, you need the chilled aisle and raw product. If you want the acids, the peptides and the altered carbohydrates, you get those from the tin too. Both are legitimate, and cost is allowed to weigh in.
And now you know why the cheap jar is not the bad buy it is often taken for.
How much, how often, how to start without overwhelming your gut
Now comes the question I am asked most often in the consultation. How much, then?
The honest answer is unsatisfying and still the most important one in the whole article.
There is no official recommended daily amount for living microorganisms, neither in Germany nor anywhere else. In 2020 the expert panel itself asked openly whether there should be one. Every figure you read is a study amount and not a recommendation.
Marco ML, Hill C, Hutkins R et al. J Nutr. 2020;150(12):3061-3067. PMID: 33269394 · DOI: 10.1093/jn/nxaa323 [Mechanism Review]
So that you still get an idea, here are the amounts that were actually studied.
What was given in the cited work
- Stanford, 17 weeks
- Participants increased step by step to up to six servings of fermented foods per day. That is the upper end and was a supervised study protocol, not an everyday target.
- Kefir in metabolic syndrome
- 180 millilitres daily over twelve weeks, roughly one large glass.
- Sauerkraut in irritable bowel syndrome
- One serving daily over six weeks, in both study arms.
- Kombucha in type 2 diabetes
- 240 millilitres daily over four weeks, in a pilot study with twelve people.
These numbers describe study designs. They are not a protocol for you, and they take account of neither your salt balance nor your tolerance nor your existing conditions.
People often turn these numbers into a target. Six servings a day, starting tomorrow. That is exactly the route on which many people give up in frustration after a week.
The direction I consider sensible instead is more boring and lasts longer.
Directions for the start, not recipes
- Start small. A tablespoon of sauerkraut or half a glass of kefir a day is a start. Not three kinds at once.
- One kind after the other. Anyone who introduces four things at once and then gets complaints knows nothing afterwards.
- With a meal rather than on an empty stomach. Fat and protein can buffer the stomach acid, and the gut then often reacts more calmly.
- Read gas production as a signal, not as proof. A bit of rumbling at the start is common. Strong flatulence that persists over weeks is a reason to reduce the amount and look at the cause. Where the air comes from in the first place is covered in the article on bloating.
- Count the salt. Anyone eating sauerkraut and kimchi daily is eating salt preserves daily.
- Sort out the order first. If stomach acid is too low, an intolerance is in question or a work up is still open, that comes first. The jar of sauerkraut comes afterwards.
- Think in weeks. In the Stanford work, diversity rose over weeks, not over days. If you feel nothing after five days, you have done nothing wrong.
The most common thinking error: I need the right dose.
There is none. There is regularity. One spoonful every day over three months is very probably worth more than six servings on one Sunday and nothing after that. That is unspectacular and the only part you genuinely hold in your own hands.
And now you know why I do not give you a figure in grams.
For whom fermented food is not harmless
There is one sentence in this field that regularly annoys me. It goes: you can always eat fermented food, it is natural after all.
You cannot. And that has nothing to do with panic, it has to do with craft. You already read the short version right at the top. Here are the reasons.
Histamine intolerance
Fermentation means microbial activity, and part of that activity produces biogenic amines. Histamine is one of them. That is not a side note, it is the point on which the specialist literature agrees.
A Spanish group around Sònia Sánchez-Pérez compared ten low histamine diets published in the specialist literature and checked whether the excluded foods actually contain much histamine.
The diets differed strongly from one another. Only fermented foods were excluded in all ten. Part of the remaining exclusions could not be explained by the measured amine content at all.
For you that means: with known histamine intolerance, fermented food is not a good starting point but the one group that every concept agrees on. What lies behind it and how it is properly assessed is covered in the article on histamine intolerance and DAO deficiency. The same applies to mast cell topics, and there is the article on mast cell activation for that.
Sánchez-Pérez S, Comas-Basté O, Veciana-Nogués MT et al. Nutrients. 2021;13(5):1395. PMID: 33919293 · DOI: 10.3390/nu13051395 [Systematic Review]Tyramine, MAO inhibitors and linezolid
And one more group that explicitly belongs here. Fermentation produces not only histamine but also tyramine. Anyone taking an MAO inhibitor, for example tranylcypromine, or receiving the antibiotic linezolid has to avoid tyramine rich foods. Mature cheese, sauerkraut, kimchi, miso and soy sauce are among them. The combination can raise blood pressure sharply.
Both substances are prescription only, and this warning appears in their product information. If one of these medications is running in your case, this article is not your instruction manual but a reason to ask at the treating practice. And it is no reason to change anything about the medication itself.
Cow milk protein allergy, the other diagnosis behind dairy complaints
A distinction that belongs here. Lactose intolerance and cow milk protein allergy are two different things. With lactose intolerance it is about the milk sugar, and fermentation can help with that. With an allergy to milk protein it is about casein and whey protein, and those remain fully intact in kefir.
If you react to dairy products with skin reactions, swelling, breathlessness or circulatory problems, that is not a digestive topic and not a case for a self experiment with kefir. It needs an allergy work up. With breathlessness, swelling in the throat or circulatory problems the emergency number, 112 in Germany, is the right route, not an appointment at the practice.
Salt, blood pressure and an uncomfortable cohort figure
Sauerkraut and kimchi are salt preserves. Salt is not an accessory in them, it is part of the process, because without salt there is no controlled fermentation.
If you watch your salt because of high blood pressure, heart failure or kidney disease, fermented vegetables belong in the daily balance and not beside it. What matters is the total amount, not the single food.
Fermented is not automatically favourable
A Korean meta-analysis of 72 observational studies describes fewer stomach, bowel and breast cancer diagnoses for fruit and vegetables. For fermented soy products it describes the opposite, namely more frequent stomach cancer diagnoses. For other salt preserved foods the pooled estimates did not reach statistical significance.
I report these figures because they belong to the honest picture, and not as a statement about what a food can bring about in you. No prevention follows from observational data, and no recommendation follows from any of these figures.
Some context before anyone throws out their sauerkraut: this is Korean dietary reality with very high salt amounts over decades. It is an observation, not evidence of cause and effect, and it does not describe a spoonful of sauerkraut with dinner. Even so it needs saying, because salt preservation is a risk factor in its own right, independent of the fermentation.
Kim JH, Jun S, Kim J. Epidemiol Health. 2023;45:e2023102. PMID: 38037322 · DOI: 10.4178/epih.e2023102 [Meta-analysis, k=72]
Immunosuppression and serious underlying illness
Here it becomes medically relevant. Lactic acid bacteria are commensals, meaning ordinary housemates. With strongly weakened defences, housemates can become pathogens.
Leila Haghighat and Nancy Crum-Cianflone described a bloodstream infection with Lactobacillus acidophilus in a person with AIDS, temporally linked to very high consumption of probiotic enriched yoghurt, and reviewed the available literature on it.
Their conclusion: invasive infections with Lactobacillus species are increasingly reported in immunosuppressed people. The safety of lactobacillus containing products is unclear in this group.
For you that means: a single case is a single case, and nothing follows from it for healthy people. With relevant immunosuppression, after transplantation, under chemotherapy or with a central venous line, however, large amounts of living cultures should be discussed with a doctor beforehand.
Haghighat L, Crum-Cianflone NF. Int J STD AIDS. 2016;27(13):1223-1230. PMID: 26130690 · DOI: 10.1177/0956462415590725 [Case Series, n=1]The counter check comes from the large Cochrane review on probiotics for preventing Clostridioides difficile associated diarrhoea under antibiotic therapy. Thirty nine studies, 9,955 participants. The authors consider short term use safe and, at a high baseline risk, effective. In the post hoc subgroup analysis the effect appeared only in studies with a baseline risk above 5 percent, and at a low baseline risk there was no difference. And they add an important restriction: in patients who are not immunocompromised or severely debilitated. That inverse reading is exactly the point. Goldenberg JZ, Yap C, Lytvyn L et al. Cochrane Database Syst Rev. 2017;12:CD006095. PMID: 29257353 · DOI: 10.1002/14651858.CD006095.pub4 [Meta-analysis, k=39, n=9955] These studies concerned supplements, not sauerkraut. What that means for the time after antibiotic therapy is covered in the article on the gut after antibiotics.
Acute flare of Crohn disease or ulcerative colitis
Here I keep it short, because the data are short. On fermented foods in an acute flare there is no robust study, neither for nor against.
That is why I give no general recommendation on it. In a flare, nutrition belongs in the hands of the treating team, and immunosuppressive therapies are their own reason to ask first. In remission and with good tolerance, in my view little speaks against small amounts. That is a clinical assessment and not a study result. More on the overall picture in the article on Crohn disease and ulcerative colitis.
Irritable bowel with strong gas production
Fermented vegetables are not automatically low in FODMAPs. Sauerkraut in small amounts is well tolerated by many, in larger amounts often no longer. Kimchi brings additional fermentable carbohydrates with its garlic and onion, and those are a classic trigger in irritable bowel syndrome.
That is the two sidedness of this topic that few people ever say out loud. Fermentation can break down fermentable carbohydrates, as with deliberately guided sourdough. But it can also leave them standing or even raise them through the ingredients. Both are true, depending on the product. The mechanics behind it are covered in the article on irritable bowel syndrome and finding causes.
Kombucha in pregnancy and breastfeeding
Raw kombucha is an unpasteurised drink with living yeasts and a measurable alcohol content. In the only randomised investigation it was 1.5 percent ethanol in the product used.
In pregnancy that puts together both things you do not want there: alcohol and a raw, unpasteurised food. I therefore advise against raw kombucha in pregnancy, and because of the alcohol the same applies while breastfeeding. With relevant immunosuppression I advise against it as well. For people in abstinence the number is relevant too, even if it looks small.
What you should also know: there are individual case reports of liver damage and of acidification of the blood after high or home brewed kombucha consumption. Single cases prove no causal link, and for most people nothing follows from them. With a liver disease I would still advise against it. Gedela M, Potu KC, Gali VL et al. S D Med. 2016;69(1):26-28. PMID: 26882579 · SungHee Kole A, Jones HD, Christensen R, Gladstein J. J Intensive Care Med. 2009;24(3):205-207. PMID: 19460826 · DOI: 10.1177/0885066609332963 [Case reports]
And if you want to ferment yourself
Home fermenting is an old craft and completely normal in many kitchens. I still deliberately write no step by step instructions here, and I give no safety promise. What matters can be said in one paragraph: clean vessels, enough salt, the vegetables fully under the brine, kept cool and dark, and the batch looked at regularly. Anything that grows mould, turns slimy, smells putrid or bulges the lid should be thrown away, not skimmed, not least because of the rare but serious risk of botulism poisoning. Anyone who is pregnant, immunocompromised or seriously ill is better off avoiding raw home batches entirely. Anyone who wants to learn it properly learns it from someone who can do it, and not from a blog article.
None of this is a reason to change, reduce or stop a prescribed medication on your own. Every adjustment belongs in medical hands.
And none of this replaces a proper work up. If a colonoscopy, a gastroscopy or laboratory diagnostics has been recommended, getting there is the first task, regardless of how well the sauerkraut is tolerated. The red flags further up in the article continue to apply unchanged.
Caution is often read as hesitation. As if you did not dare.
I see it differently. Knowing who a tool is not suited to is part of the instruction manual. A food that causes complaints in histamine problems is not a bad food because of it. It is just not yours.
And now you know why this section does not sit at the end.
Why the guidelines stay silent on it, and why that is not a rejection
Perhaps while reading you thought: if it looks this good, why is it in no guideline?
A fair question. The answer is unspectacular and instructive at the same time.
The German S3 guideline on irritable bowel syndrome from DGVS and DGNM treats probiotics as a therapy option and stays reserved about them. It contains no statement of its own on fermented foods. Layer P, Andresen V, Allescher H et al. Z Gastroenterol. 2021;59(12):1323-1415. PMID: 34891206 · DOI: 10.1055/a-1591-4794 · AWMF 021/016 [Guideline]
The American ACG guideline on irritable bowel syndrome comes to the same result. It recommends a time limited trial of a low FODMAP diet and gut directed psychotherapy, and on probiotics it stays explicitly reserved. Lacy BE, Pimentel M, Brenner DM et al. Am J Gastroenterol. 2021;116(1):17-44. PMID: 33315591 · DOI: 10.14309/ajg.0000000000001036 [Guideline]
And the AGA technical review is the most differentiated. For individual indications such as pouchitis, the prevention of Clostridioides difficile associated diarrhoea and necrotising enterocolitis in preterm infants it speaks in favour. For irritable bowel syndrome and for inflammatory bowel disease it recommends use only within studies, because the data do not allow a statement. That is something other than proof that nothing happens. Preidis GA, Weizman AV, Kashyap PC, Morgan RL. Gastroenterology. 2020;159(2):708-738.e4. PMID: 32531292 · DOI: 10.1053/j.gastro.2020.05.060 [Guideline]
A guideline can only recommend what can be standardised and tested. A supplement is a defined strain in a defined dose. A jar of sauerkraut is not.
The reason for the reticence, in one sentenceThat is not a rejection, it is a methodological limit. And it is the point at which I explicitly understand my colleagues in gastroenterology. Anyone issuing a recommendation takes on responsibility for its reproducibility. With a food whose microbial count and composition vary from batch to batch, that is not given.
At the same time there is the other side of the same question, and it does not contradict this. The same expert community that stays silent as a guideline panel has acted as a scientific association. In 2021 ISAPP defined fermented foods as a category of their own, clarified the postbiotic term in the same year, and back in 2020 asked openly whether there should be a recommended daily intake of living microorganisms.
In 2026 another consensus paper followed, defining the term gut health: a state of normal gastrointestinal function without active disease and without gut related complaints that impair quality of life. It took until 2026 for that term to be cleanly established. Marco ML, Cunningham M, Bischoff SC et al. Nat Rev Gastroenterol Hepatol. 2026;23(5):432-448. PMID: 41709019 · DOI: 10.1038/s41575-026-01176-x [Consensus paper, scientific association]
How I handle this in the consultation
Documented by randomised human trials are essentially two things: that fermented dairy products can make lactose digestion measurably easier, and that a diet with fermented foods over weeks can raise microbial diversity and lower individual inflammatory proteins. The second comes from a study with 18 people per group and from secondary endpoints, not from the primary one.
Mechanistically plausible, with thin human evidence, is the rest: that the metabolites, the peptides and the acids formed have effects of their own at the mucosa. More on the interplay of butyrate and gut mucosa is in the article on L-glutamine and butyrate.
Clinical observation, not a study result: in my consultation I see people getting considerably further with small daily amounts than with big intentions. And I see tolerance deciding far more about success than the choice of variety. That is experience and not evidence, and I know the difference.
And the order stays the same with me as it always does. First the question of whether there is enough stomach acid up top and whether a work up is due. After that the jar of sauerkraut. Where that sits in the overall concept is shown in the article on the gut reset.
Silence in a guideline is often read as a verdict. Either as a warning or as evidence that something is being suppressed.
Neither is right. The guideline stays silent because a food is not a supplement and because the study base is not sufficient for a formal recommendation. No ban follows from that, and no free pass. What follows is that the decision sits with you, with realistic expectations and with an eye on the list of those it does not suit.
And now you know why there is no contradiction in my recommending fermented food while still considering the reticence of the professional bodies correct.
Frequent questions about fermented foods
What exactly are fermented foods, and does beer count?
The ISAPP consensus paper from 2021 defines them as foods made through desired microbial growth and enzymatic conversion of food components. Under that definition, beer, wine, sourdough bread and coffee are fermented foods. They still contain no living microorganisms at the end, because they are heated, filtered or baked. Fermented and probiotic are therefore two different things.
Which fermented foods are there, is there a list worth remembering?
Sorting them by raw material works well as a memory aid. Milk gives you yoghurt, kefir, buttermilk, quark and many cheeses. Vegetables give you sauerkraut, kimchi, pickles in brine and fermented beetroot. Soy gives you miso, natto, tempeh and soy sauce. Grain gives you sourdough bread, tea gives you kombucha. Only the uncooked and unfiltered versions of these contain living cultures.
How much sauerkraut or kefir a day makes sense?
There is no official daily amount for living microorganisms, neither in Germany nor anywhere else. In 2020 the expert panel itself asked openly whether there should be one. Every figure in circulation is a study amount. In the Stanford work participants increased up to six servings of fermented foods a day, in the kefir studies it was 180 millilitres daily over twelve weeks, in the sauerkraut study one serving daily over six weeks.
Tinned sauerkraut or sauerkraut from the chilled aisle, does it make a difference?
For the living bacteria yes, for the effect on irritable bowel complaints possibly less than many people assume. Tins and jars from the unchilled shelf have been heated, after which no living lactic acid bacteria remain. In the Norwegian irritable bowel study, though, the symptom score improved clearly in both groups, including under pasteurised sauerkraut. The authors attribute that more to the prebiotic components than to living microbes.
Kefir or yoghurt, which is better for the gut?
For lactose digestion both are good and roughly equally good. In a breath test study the hydrogen curve fell by about two thirds compared with milk for both plain yoghurt and plain kefir. Kefir additionally contains yeasts and a broader spectrum of microbes. For hard endpoints such as blood pressure or inflammatory values, kefir showed no clear advantage in the direct comparisons against unfermented milk. If you tolerate both, taste can decide.
Do the bacteria survive the stomach at all?
Some do, a large share do not. Stomach acid, bile salts and transit time make for a hard passage, and how much arrives depends on the strain, on the food matrix and on whether it is eaten with a meal. The second insight matters more: even killed cells and their components can have biological effects. That is exactly why the term postbiotic has existed since 2021.
How long does it take before something changes in the gut flora?
In the Stanford study microbial diversity rose steadily across the ten week ramp up and maintenance phase, not within a few days. So count in weeks, not in days. What happens when you stop again is unanswered, because the study ran 17 weeks in total and has no long follow up.
Sauerkraut gives me flatulence, do I have to stop?
Not necessarily, but it is a signal to reduce the amount and increase it more slowly. Fermented vegetables contain fermentable carbohydrates, and kimchi adds more with garlic and onion. If flatulence stays strong, does not settle after weeks or comes with pain, the cause needs looking at rather than eating through. Gas production is a signal, not proof of a benefit.
Fermented foods with histamine intolerance, is that possible at all?
Here the data are unusually clear. In an analysis of ten published low histamine diets, fermented foods were the only group excluded in all ten. With known histamine intolerance, fermented foods are therefore not a good starting point. Whether and what is individually possible is a question for medical or nutritional therapy guidance.
Can I eat fermented foods with Crohn disease or ulcerative colitis?
For the acute flare there is no robust study on fermented foods, neither for nor against. That is why I give no general recommendation on it. In remission and with good tolerance, in my view little speaks against small amounts, while in a flare nutrition belongs in the hands of the treating team. Immunosuppressive therapies are their own reason to ask first.
Is kombucha a problem because of the sugar and the alcohol?
Both belong on the table. In the only randomised pilot study, with twelve people, an ethanol content of 1.5 percent was measured in the kombucha used. For pregnant women, for people who are breastfeeding, for people in abstinence and for children that is relevant. Residual sugar varies widely with fermentation time and product, so a look at the nutrition table is worth it. Under relevant immunosuppression, raw kombucha should be discussed with a doctor first.
Do fermented foods replace a probiotic capsule?
No, they are something different. A supplement is a defined strain in a defined amount with its own body of studies. A jar of sauerkraut is a shifting community with no declared microbial count. The two can sit side by side, but studies on one do not automatically apply to the other. What holds for capsules and spore based products is covered in its own article.
Is sauerkraut a problem with high blood pressure because of the salt?
Fermented vegetables are a salt preserve, and that has to be counted in. With high blood pressure, heart failure or kidney disease, the relevant figure is the total daily salt intake, not the single food. A Korean meta-analysis also found an association between fermented soy products and a higher stomach cancer risk. That reflects Korean dietary reality with very high salt amounts and not a spoonful of sauerkraut, but it still needs saying.
Can I ferment at home without it becoming dangerous?
Home fermenting is an old craft and everyday practice in many kitchens. I still give no safety promises about it. What matters: clean vessels, enough salt, the vegetables fully under the brine, kept cool and dark, and the batch watched. Anything that grows mould, turns slimy, smells putrid or pushes on the lid should be thrown away, not skimmed. Anyone who is immunocompromised or pregnant should avoid raw home batches.
Where the jar of sauerkraut connects to the rest
Fermented food does not stand on its own. It hangs on what else is on the plate, on sleep quality, on the nervous system and on whether an intolerance is running in the background.
Fibre myths
What is behind the number 30 and what is not
Inflammatory foods
Why preparation often counts more than the ingredient
Sleep and the microbiome
The lever almost everyone underrates
Gut brain axis
Why tension arrives in the belly
Stool testing and dysbiosis
What a microbiome report can do and what it cannot
Mast cell activation
When histamine is not the only topic
Scientific sources
- Marco ML, Sanders ME, Gänzle M et al. The ISAPP consensus statement on fermented foods. Nat Rev Gastroenterol Hepatol. 2021;18(3):196-208. PMID: 33398112 · DOI: 10.1038/s41575-020-00390-5 [Consensus paper, scientific association]
- Salminen S, Collado MC, Endo A et al. The ISAPP consensus statement on the definition and scope of postbiotics. Nat Rev Gastroenterol Hepatol. 2021;18(9):649-667. PMID: 33948025 · DOI: 10.1038/s41575-021-00440-6 [Consensus paper, scientific association]
- Marco ML, Cunningham M, Bischoff SC et al. The ISAPP consensus statement on the definition and scope of gut health. Nat Rev Gastroenterol Hepatol. 2026;23(5):432-448. PMID: 41709019 · DOI: 10.1038/s41575-026-01176-x [Consensus paper, scientific association]
- Marco ML, Hill C, Hutkins R et al. Should There Be a Recommended Daily Intake of Microbes? J Nutr. 2020;150(12):3061-3067. PMID: 33269394 · DOI: 10.1093/jn/nxaa323 [Mechanism Review]
- Layer P, Andresen V, Allescher H et al. Update S3 guideline on irritable bowel syndrome, DGVS and DGNM. Z Gastroenterol. 2021;59(12):1323-1415. AWMF registry number 021/016. PMID: 34891206 · DOI: 10.1055/a-1591-4794 [Guideline]
- Lacy BE, Pimentel M, Brenner DM et al. ACG Clinical Guideline: Management of Irritable Bowel Syndrome. Am J Gastroenterol. 2021;116(1):17-44. PMID: 33315591 · DOI: 10.14309/ajg.0000000000001036 [Guideline]
- Preidis GA, Weizman AV, Kashyap PC, Morgan RL. AGA Technical Review on the Role of Probiotics in the Management of Gastrointestinal Disorders. Gastroenterology. 2020;159(2):708-738.e4. PMID: 32531292 · DOI: 10.1053/j.gastro.2020.05.060 [Guideline]
- Wastyk HC, Fragiadakis GK, Perelman D et al. Gut-microbiota-targeted diets modulate human immune status. Cell. 2021;184(16):4137-4153.e14. PMID: 34256014 · DOI: 10.1016/j.cell.2021.06.019 [RCT, n=36]
- Taylor BC, Lejzerowicz F, Poirel M et al. Consumption of Fermented Foods Is Associated with Systematic Differences in the Gut Microbiome and Metabolome. mSystems. 2020;5(2):e00901-19. PMID: 32184365 · DOI: 10.1128/mSystems.00901-19 [Cross section with small longitudinal subgroup, n=6811]
- Nielsen ES, Garnås E, Jensen KJ et al. Lacto-fermented sauerkraut improves symptoms in IBS patients independent of product pasteurisation. Food Funct. 2018;9(10):5323-5335. PMID: 30256365 · DOI: 10.1039/c8fo00968f [RCT, n=34]
- Hertzler SR, Clancy SM. Kefir improves lactose digestion and tolerance in adults with lactose maldigestion. J Am Diet Assoc. 2003;103(5):582-587. PMID: 12728216 · DOI: 10.1053/jada.2003.50111 [RCT, n=15]
- Bellikci-Koyu E, Sarer-Yürekli BP, Karagözlü C et al. Probiotic kefir consumption improves serum apolipoprotein A1 levels in metabolic syndrome patients: a randomized controlled clinical trial. Nutr Res. 2022;102:59-70. PMID: 35405603 · DOI: 10.1016/j.nutres.2022.02.006 [RCT, n=62]
- Bellikci-Koyu E, Sarer-Yürekli BP, Akyön Y et al. Effects of Regular Kefir Consumption on Gut Microbiota in Patients with Metabolic Syndrome. Nutrients. 2019;11(9):2089. PMID: 31487797 · DOI: 10.3390/nu11092089 [RCT, small]
- Han K, Bose S, Wang JH et al. Contrasting effects of fresh and fermented kimchi consumption on gut microbiota composition and gene expression related to metabolic syndrome in obese Korean women. Mol Nutr Food Res. 2015;59(5):1004-1008. PMID: 25688926 · DOI: 10.1002/mnfr.201400780 [RCT, n=24]
- Lim S, Moon JH, Shin CM, Jeong D, Kim B. Effect of Lactobacillus sakei, a Probiotic Derived from Kimchi, on Body Fat in Koreans with Obesity. Endocrinol Metab (Seoul). 2020;35(2):425-434. PMID: 32615727 · DOI: 10.3803/EnM.2020.35.2.425 [RCT, n=114]
- Mendelson C, Sparkes S, Merenstein DJ et al. Kombucha tea as an anti-hyperglycemic agent in humans with diabetes: a randomized controlled pilot investigation. Front Nutr. 2023;10:1190248. PMID: 37588049 · DOI: 10.3389/fnut.2023.1190248 [RCT, n=12]
- Han AL, Ryu MS, Yang HJ, Jeong DY, Choi KH. Evaluation of Menopausal Syndrome Relief and Anti-Obesity Efficacy of the Korean Fermented Food Doenjang. Nutrients. 2024;16(8):1194. PMID: 38674884 · DOI: 10.3390/nu16081194 [RCT, n=56]
- Katagiri R, Sawada N, Goto A et al. Association of soy and fermented soy product intake with total and cause specific mortality: prospective cohort study. BMJ. 2020;368:m34. PMID: 31996350 · DOI: 10.1136/bmj.m34 [Cohort, n=92915]
- Sluijs I, Forouhi NG, Beulens JWJ et al. The amount and type of dairy product intake and incident type 2 diabetes: results from the EPIC-InterAct Study. Am J Clin Nutr. 2012;96(2):382-390. PMID: 22760573 · DOI: 10.3945/ajcn.111.021907 [Cohort, n=340234]
- Slurink IAL, den Braver NR, Rutters F et al. Dairy product consumption and incident prediabetes in Dutch middle-aged adults: the Hoorn Studies. Eur J Nutr. 2022;61(1):183-196. PMID: 34245355 · DOI: 10.1007/s00394-021-02626-9 [Cohort, n=2262]
- Dehghan M, Mente A, Rangarajan S et al. Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE). Lancet. 2018;392(10161):2288-2297. PMID: 30217460 · DOI: 10.1016/S0140-6736(18)31812-9 [Cohort, n=136384]
- Astrup A, Geiker NRW, Magkos F. Effects of Full-Fat and Fermented Dairy Products on Cardiometabolic Disease: Food Is More Than the Sum of Its Parts. Adv Nutr. 2019;10(5):924S-930S. PMID: 31518411 · DOI: 10.1093/advances/nmz069 [Mechanism Review]
- Kim JH, Jun S, Kim J. Dietary intake and cancer incidence in Korean adults: a systematic review and meta-analysis of observational studies. Epidemiol Health. 2023;45:e2023102. PMID: 38037322 · DOI: 10.4178/epih.e2023102 [Meta-analysis, k=72]
- Goldenberg JZ, Yap C, Lytvyn L et al. Probiotics for the prevention of Clostridium difficile-associated diarrhea in adults and children. Cochrane Database Syst Rev. 2017;12:CD006095. PMID: 29257353 · DOI: 10.1002/14651858.CD006095.pub4 [Meta-analysis, k=39, n=9955]
- Sánchez-Pérez S, Comas-Basté O, Veciana-Nogués MT, Latorre-Moratalla ML, Vidal-Carou MC. Low-Histamine Diets: Is the Exclusion of Foods Justified by Their Histamine Content? Nutrients. 2021;13(5):1395. PMID: 33919293 · DOI: 10.3390/nu13051395 [Systematic Review]
- Haghighat L, Crum-Cianflone NF. The potential risks of probiotics among HIV-infected persons: Bacteraemia due to Lactobacillus acidophilus and review of the literature. Int J STD AIDS. 2016;27(13):1223-1230. PMID: 26130690 · DOI: 10.1177/0956462415590725 [Case Series, n=1]
- Fraberger V, Call LM, Domig KJ, D'Amico S. Applicability of Yeast Fermentation to Reduce Fructans and Other FODMAPs. Nutrients. 2018;10(9):1247. PMID: 30200589 · DOI: 10.3390/nu10091247 [In vitro]
- Loponen J, Gänzle MG. Use of Sourdough in Low FODMAP Baking. Foods. 2018;7(7):96. PMID: 29932101 · DOI: 10.3390/foods7070096 [Mechanism Review]
- Sato Y, Murata M, Chiba T, Umegaki K. A Systematic Review of the Acceptable Intake Level of Vitamin K among Warfarin Users. Shokuhin Eiseigaku Zasshi. 2015;56(4):157-165. PMID: 26346860 · DOI: 10.3358/shokueishi.56.157 [Systematic Review]
- Gedela M, Potu KC, Gali VL, Alyamany K, Jha LK. A Case of Hepatotoxicity Related to Kombucha Tea Consumption. S D Med. 2016;69(1):26-28. PMID: 26882579 [Case report]
- SungHee Kole A, Jones HD, Christensen R, Gladstein J. A case of Kombucha tea toxicity. J Intensive Care Med. 2009;24(3):205-207. PMID: 19460826 · DOI: 10.1177/0885066609332963 [Case report]
- The Stanford study is small and missed its primary endpoint. Eighteen people per group, all healthy, 17 weeks, no disease endpoint. The cytokine response score as the primary endpoint stayed unchanged in both arms, and the diversity and the 19 inflammatory proteins are secondary and exploratory analyses. It is the best work in this field and still a signal and not proof. Whether the rise in diversity persists after stopping is unanswered.
- Sauerkraut in irritable bowel syndrome is a pilot trial with 34 people. The authors call their own study underpowered. There is no second independent investigation confirming the result.
- On kimchi as a food there is no robust human study in bowel complaints. What exists is 24 participants in a nutrition study and a capsule with a single isolated strain. Both are something other than a jar of kimchi.
- Kombucha rests on twelve people. The blood glucose finding is an observation with question marks. What is reliable about this work is above all the measured ethanol figure of 1.5 percent.
- Miso, natto and tempeh rest on observational cohorts. The Japanese data are large, but they remain dietary questionnaires and associations. The authors themselves warn about unadjusted residual confounding. There is no randomised study on bowel symptoms.
- The kefir advantage disappears in the direct comparison. Against unfermented milk, no significant difference remained for most parameters in either Turkish study. Anyone citing only the before and after figures overstates it.
- There is no official daily amount for living microorganisms. All amounts in this article are study designs. They are explicitly not a recommendation and certainly not a personal dose.
- FODMAP breakdown through fermentation is best documented for sourdough, and there mostly from food chemistry and small clinical investigations. For sauerkraut or kimchi there is no comparable body of data. The yeast study cited is laboratory work and not a human study.
- On inflammatory bowel disease in an acute flare there is no robust study on fermented foods. The article says exactly that and issues neither a recommendation for nor against.
- The bacteraemia case under immunosuppression is a single case. It justifies asking a question, not a general warning. Nothing follows from it for healthy people.
- The stomach cancer finding comes from Korean observational data with very high salt amounts over decades. It cannot be transferred to a spoonful of sauerkraut and is mentioned here only with that context.
- All risk figures from cohorts and meta-analyses are observational data. They stand here as scientific context and not as a health claim about a food. Neither prevention nor a recommendation follows from them.
- Three of the cited works involved authors from the manufacturer or from the producing institute. This is noted in the text at the relevant place. It does not devalue those works, but it belongs in the picture.
- What deliberately does not appear here. No personal dosing recommendation, no eating protocol, no instructions for home fermenting and no advice to change, reduce or stop an existing medication. Every adjustment needs medical supervision. From no section does it follow that a recommended colonoscopy, endoscopy or laboratory work up should be postponed or replaced. What I describe from my consultation is marked as observation and is not a study result.