alcohol: the auto-brewery syndrome

 the auto-brewery syndrome (ABS) has hit the popular press https://www.nytimes.com/2026/04/03/well/autobrewery-syndrome.html); this syndrome results from the gut microbiome producing excess ethanol and exceeding the hepatic ability to metabolize it. at times it results in individuals being accused of DUI (driving under the influence of alcohol) despite no alcohol consumption, having the breathalyzer test at high a level as 0.18% (see NatMicrobiology2026 in dropbox or https://doi.org/10.1038/s41564-025-02225-y and https://www.ncbi.nlm.nih.gov/books/NBK513346/). In addition, per the NY Times, an individual left his job after being stigmatized for smelling of alcohol, some individuals had frank alcohol intoxication and were felt to be liars by their doctors and spouses, and some even having alcohol withdrawal symptoms when not eating sufficient carbohydrates to stimulate the gut to produce ethanol

---

Background (largely from the review article: auto-brewery syndrome review NatReviews2024 in dropbox, or https://doi.org/10.1038/s41575-024-00937-w), which has exquisite detail of our knowledge of the role and mechanisms of the gut microbiome in elaborating ethanol production, occasionally leading to significantly high blood alcohol levels and potentially severe clinical outcomes

    -- for another review article that includes some different studies and has a somewhat different approach, see https://www.ncbi.nlm.nih.gov/books/NBK513346/

-- as per several recent and remote blogs (https://gmodestmedblogs.blogspot.com/search?q=microbiome), the gut microbiome plays a leading role in much human physiologic functioning. the gut microbiome affects more than half of the variance in the plasma metabolome (metabolites found within cells, tissues, organs, and the whole body, which includes amino acids, lipids, sugars, vitamins, hormones, and signaling molecules). These metabolites can enter the blood circulation and be transported via the plasma throughout the body.

    -- the genome is a largely stable entity though there are some episodic epigenetic changes spurred by environmental factors that can turn genes on and off; in contrast, the gut microbiome is exquisitely sensitive to external influences (diet, exercise, stress, antibiotics,..). This liquidity of the microbiome includes destructive dysbiosis when the microbiome has a decrease in microbial diversity and unbalance. Changes that enhance negative microbial species can then interfere with normal physiologic function. fortunately, changes that improve the microbiome health (diet, exercise, decreased stress or other causes of chronic inflammation are reparative: https://pmc.ncbi.nlm.nih.gov/articles/PMC7213601/000) 

    -- as noted in the last blog, the microbiome functions extend well into the central nervous system (gut-brain axis), the liver (gut-liver axis), as well as changes in the blood supplying all organs: https://gmodestmedblogs.blogspot.com/2026/05/long-term-microbiome-effects-after.html

-- the issue of ethanol production in the microbiome is complex:

    -- ethanol is a major microbial metabolite that is produced in the intestines in nearly all individuals.

    -- there are clearly many different microbes responsible for this ethanol production

    -- there are some individuals in whom the gut microbiome ethanol production exceeds the ability of their livers to metabolize it (90-98% of ethanol consumed or produced is metabolized in the liver by alcohol dehydrogenase and some other enzymes)

    -- this ethanol production creates a variable effect on the human body, from no significant change, to causing metabolic dysfunction-associated steatotic liver disease (MASLD, the former NAFLD), to the more serious metabolic dysfunction-associated steatohepatitis (MASH), to the rarer auto-brewery syndrome (ABS), whereby patients may have symptoms of ethanol intoxication despite drinking no alcohol

        -- documentation of MASH in some individuals may well portend the future development of cirrhosis and hepatocellular carcinoma, though there is no direct evidence of these latter complications directly related to ABS

    -- interestingly, knowledge of this ABS phenomenon dates back to documentation in France, in 1894…, having the first recorded cases of patients with severe endogenous ethanol production

    -- the amount of ethanol produced is also elevated in patients with metabolic diseases such as type II diabetes and MASLD, as well as MASH (not surprisingly, since these latter problems themselves are associated with significant gut microbiome changes). And liver biopsies reveal that there is no significant difference in individuals having endogenous ethanol production (ie ABS) vs alcoholic liver disease in drinkers; both can have hepatic steatosis and increased Mallory bodies on biopsies. 

    -- excessive ethanol production is a manifestation of a disrupted gut microbiome

        -- there are cases of alcohol withdrawal syndrome with ABS, which may be related to abrupt decreases in carbohydrate ingestion (carbohydrates provide a necessary substrate for fermentation and ethanol production by multiple different bacteria and fungi)

        -- a low carbohydrate diet can help some of those with ABS, though this may need to be combined with antibiotics and/or antifungals: https://pubmed.ncbi.nlm.nih.gov/37060744/

-- there are several microbes identified that are associated the endogenous ethanol production

    -- E. coli is one of many bacteria that can produce ethanol

        --of note, in the 1950s it was established that E. coli can produce ethanol at a rate of 0.8 g per hour for every gram of bacteria, and low levels of ethanol production are associated with type II diabetes and liver disease

    -- Klebsiella pneumoniae can also produce ethanol, with effects similar to those found with E. coli, and studies have found that ethanol producing strains of Klebsiella pneumoniae can elicit steatotic liver disease in mice: Cell Metab. 30, 675–688.e7 (2019). Transplant of fecal ethanol-producing K. pneumoniae from a patient with ABS into mice resulted in steatohepatitis in the mice, further supporting the similar potential hepatotoxicity from either endogenous or exogenous systemic ethanol

    -- there may be a role of fungal species in the microbiome (eg, candida varieties) causing increased ethanol production, though the data is a bit less clear. however, antifungal meds are considered to be part of ABS treatment, as noted above

-- there are also some very important beneficial microbes in the gut that are diminished by alcohol (eg Akkermansia, which seems to play an important role itself in improving glucose intolerance and diabetes control: https://gmodestmedblogs.blogspot.com/2024/07/microbiome-in-cardiovascular-disease.html)

    -- antibiotics, a major microbiome antagonist (https://gmodestmedblogs.blogspot.com/2026/05/long-term-microbiome-effects-after.html), lead to depletion of short-chain fatty acid-producing (SCFAs) microbial strains such as acetate, butyrate, and propionate, which serve as a vital energy resource for colonic cells and enhance the antimicrobial actions of macrophages and regulatory T-cells. The SFCA depletion leads to impaired mitochondrial respiration in colonic epithelial cells, and then to decreased oxygen consumption of the cells. This change leads to a hypoxic environment further promoting the growth of facultative anaerobic bacteria associated with ethanol production

        -- proton pump inhibitors (PPIs) should be avoided for several reasons:

            -- they have been found to be a recognized ABS risk factor, likely from decreases in gastric acid which serves as a barrier to upper GI tract microbial colonization and elevated postprandial ethanol levels in plasma

            -- PPIs promote small intestinal bacterial overgrowth (SIBO) as well as fungal overgrowth (SIFO), both are catalysts for ABS; and this is a cumulative process, with each additional month of PPI use being associated with a 4.3 percentage point increase in SIBO prevalence

            -- PPIs are also directly associated with increases in endogenous ethanol production

            -- and PPIs are associated with growth of some ethanol-producing facultative anaerobes, including pathogenic E coli and Lactobacillales (eg Enterococcaceae, Streptococcaceae and Enterobacteriaceaea)

    -- one important concern with the intestine function is the role of the mucus layer which acts as a buffer between the luminal bacteria and intestinal epithelial cells. This barrier prevents translocation of microorganisms and their antigens to spaces outside the intestines and inside the cells and blood.

    -- this is also a very important dietary issue since emulsifiers are used that disturb the mucosal barrier, allowing gut bacteria to migrate closer to the underlying epithelial cells, triggering chronic inflammation and increasing the risk of contamination into the blood supply. Food emulsifiers, particularly predominant in ultraprocessed foods, lead to a significant increase in risk of metabolic syndrome and inflammatory diseases: https://pmc.ncbi.nlm.nih.gov/articles/PMC9331555/; alcohol itself increases the ability of harmful bacteria to bypass the gut barrier

-- a few other comments from the above reviews:

    -- the above reviews go into great detail on the very large array of bacteria and likely fungi that are involved in generating endogenous ethanol, as well as reviewing the many mechanisms by which this ethanol affects the gut itself and the human body

    -- there is some evidence that the brain itself may have a role in the metabolism of alcohol (and we do know that there is also a rather profound gut-brain axis with bidirectional effects of each on each other)

    -- and there are some studies suggesting there may be a therapeutic role of targeting K. pneumoniae by the use of phages to ameliorate ethanol-induced steatohepatitis in mice.

 

---

A recent small study of 22 patientswith auto-brewery syndrome (ABS) found that these individuals had enrichment of the adverse gut microbiome bacterial genes in metabolic pathways associated with ethanol production, with one patient having improvement after a microbiome transplant leading to improving the gut microbiome health: see  auto-brewery syndrome full article Nature microbiology2026 in dropbox or https://www.nature.com/articles/s41564-025-02225-y

Details:

-- 22 patients with documented ABS, 21 unaffected household partners, and 22 healthy controls were prospectively enrolled in the study

-- serial stool samples were collected for microbial analysis and assessment of physiological changes over time

-- microbial ethanol production was initially assessed in fecal samples using an anaerobic in vitro culturing system

-- megagenomic sequencing was done to identify taxonomic differences and functional metabolic pathways that distinguished the patients with ABS from their household partners and biologically-matched healthy controls

-- one patient had a fecal microbiome transplant with a healthy microbiome sample, with subsequent analyses to assess shifts in gut microbial composition with physiological measurements of ethanol production and symptom improvement, in order to test the biological relevance of their findings

Results:

-- microbiome analysis of those with ABS versus their asymptomatic partners and healthy controls:

    -- substantially more ethanol production

    -- the proteobacteria E. coli was a predominant ethanol-producing microbe in most patients, though Bacteroides was also elevated with smaller increases in Actinobacteria and lower amounts of Firmicutes (the above review paper goes into great detail of the known ethanol-producing bacteria in health and disease)

        -- ie, the microbes in those with ABS had overrepresentation of those with fermentation pathways capable of generating ethanol (the fermentation pathway is carbohydrate-derived pyruvate which is first decarboxylated into acetaldehyde, and this is then reduced to ethanol through metabolic activity of bacteria)

-- changes in the single patient who had a fecal microbiome transplant:

    -- improvement in symptoms, which correlated with reductions in ethanol-producing microbes and fermentation pathways in that patient

Commentary:

-- as per the comments above, ABS is a rare condition in which individuals have gut microbiomes that produce high levels of ethanol that enter the bloodstream and are associated with the diverse effects of direct ethanol ingestion

    -- individuals with ABS can have symptoms of ethanol intoxication, especially after consuming a high carbohydrate meal, with symptoms such as dizziness, slurred speech, impaired coordination, memory problems, and altered behavior such as lethargy, euphoria, or other behavioral shifts including aggression, mood swings, and decreased self-control that are not typical for the individual

-- the gold standard approach to diagnosis of ABS involves monitoring a patient in a clinical setting to assure no drinking alcohol for up to 24 hours to confirm a rise in blood alcohol levels after an oral glucose challenge

-- the current treatment for this condition which can have some benefit, as outlined above, involves decreased carbohydrate ingestion and taking antibiotics and antifungals, though the above study does suggest that a fecal microbiome transplant may also be beneficial (as is well-documented in those with severe and relapsing C. difficile infection)

    -- of course, it would be healthier and more generalizable to see if aggressive non-medical interventions (improved diet, increased exercise, decreased stress, decreased processed foods and sugars, avoiding unnecessary antibiotics) might improve the microbiome diversity and health, leading to less ethanol production. A healthy microbiome is so, so important for so many physiological functions

-- the overall evidence of this study strongly supports the conclusion that ABS is a real entity (though rare), that endogenous alcohol production is consistent with the finding that this alcohol can be associated with both profound social implications (mimicking intoxication and DUI issues, stigmatization, etc) as well as the array of clinical complications associated with alcohol use disorders

-- the study also confirms that gut-derived ethanol production may well be relevant to conditions such as diabetes and steatotic liver disease in which low-level endogenous ethanol production has been detected, as noted above

-- this brings up the following musing:

    -- endogenous alcohol production in the gut microbiome is near universal

    -- it is important to determine the level of endogenous alcohol production that is associated with adverse outcomes

        -- the clinical information we have strongly suggests that the goal of alcohol drinking to prevent adverse clinical effects is zero alcohol: https://gmodestmedblogs.blogspot.com/2023/04/alcohol-consumption-small-amounts-not.html

    -- we also know that fatty liver disease (steatosis) is a condition that is associated with increased cardiovascular and cancer morbidity/mortality as well as increased likelihood of developing steatosis and diabetes that can progress to liver failure and cancer

    -- and we know that about 25% of the world's population has fatty liver disease

    -- major risk factors include overweight/obesity, several clinical factors (hypertension, diabetes, metabolic syndrome), and genetic factors

    -- but we also know that about 15% (10-20%) of those who have are not overweight: https://pmc.ncbi.nlm.nih.gov/articles/PMC8533092/

    -- and, we do know, as per above, that restoring a healthy microbiome through a low carbohydrate diet and other lifestyle changes (exercise, decreasing stress and other inflammatory conditions), and avoiding antibiotics unless really necessary (along with preferentially prescribing those that some are better than others: https://gmodestmedblogs.blogspot.com/2026/05/long-term-microbiome-effects-after.html

so, should we be routinely testing people for their microbiome composition?

    -- as noted, a high level of gut-derived systemic alcohol is bad (specific level needs to be determined; also is there a sensitive blood test??)

    -- and we already do stool tests pretty often for FIT testing as a colon cancer prevention test, or H Pyori testing (more than 50% of the world is infected, with reasonable argument that we should be testing individuals broadly to prevent the myriad of bad associated outcomes: https://gmodestmedblogs.blogspot.com/2026/03/new-h-pylori-treatment-guidelines.html has a lot of information on this)

    -- the finding of higher levels of endogenous alcohol is actionable: reinforcing the lifestyle changes that improve microbiome health, decreasing dietary carbohydrates

        -- though we do need much more information: is the above approach to microbiome health durable and effective in decreasing the known adverse clinical outcomes of alcohol production?? are long-term complications (eg diabetes, steatosis) really avoided?

        -- as per usual, this study engenders the need for more studies.....

-- the most direct implication of the above studies is that we clinicians, public health people, etc should be wary of assuming that a person with apparent alcohol use disorder with the attendant social/medical complications is an alcohol consumer. Auto-brewery syndrome is real, though we do not have a practicable method to confirm it absolutely (are we able to admit patients to observation units for 24 hours???), though perhaps there are creative possibilities (observation in an alcohol detox program for 24 hours??, a home-based program??). this is important given the very different strategies of treatment for the profound alcohol-related problems associated with both ABS or alcohol consumption

    -- we do know that more than half of individuals developing liver cancer do not consume alcohol (https://pmc.ncbi.nlm.nih.gov/articles/PMC4239482/). do many of them have undetected ABS??? or even just higher ethanol levels that surpass the unknown appropriate cutpoint for elevated endogenous ethanol production???

    -- by the way, a recent survey did have the very positive conclusion that alcohol consumption in adults in the US has been declining rapidly in the past 2-3 years: https://onlinelibrary.wiley.com/doi/10.1111/add.70323?af=R

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

to get access to the dropbox: https://www.dropbox.com/scl/fo/pcf4vymtpo77f49wd2ljr/ANyJqB40CuBd-DMEr67XDoI?rlkey=5dsalmuhf8ekd0svsgvn93xm1&dl=0

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them