Contents
- Summary
- Bacteria in the Large Intestine
- Bacteria in the Small Intestine
- Iranian Villagers
- Conclusion
- Bibliography
Summary
Given that many otherwise healthy vegans develop B12 deficiency when not supplementing their diets with B12, intestinal bacteria cannot be relied upon to prevent B12 deficiency in vegans.
Are raw foodists or people who eat fermented foods exceptions? No. See the section B12 Status of Raw Foodist Vegans.
Bacteria in the Large Intestine
It’s long been assumed that B12 is produced by bacteria in the large intestine (aka the colon) of humans, but because such B12 is produced below the ileum where most B12 is absorbed, it isn’t usable. This theory is reinforced by the fact that many species of totally or primarily vegetarian, non-ruminant animals eat their feces, allowing them to obtain B12 from the B12 produced by bacteria in their lower digestive tracts.
The most convincing evidence I’ve found, for B12 being produced in the large intestine where it isn’t bioavailable, is reported in a paper by Herbert (1988). In the paper, Herbert discusses a study from the 1950s in England where vegan volunteers with B12 deficiency (as shown by megaloblastic anemia) were fed B12 extractions made from their own stools which cured their deficiency. Herbert says this proves that the colonic bacteria of vegans produce enough B12 to cure deficiency, but the B12 is excreted rather than absorbed.
However, the study Herbert cites as the source, “Callender ST, Spray GH. Latent pernicious anemia. Br J Haematol. 1962;8:230-40,” does not mention this experiment.
There is another study by Callender and Spray that sounds like it could be the one Herbert is describing, “Preparation of hematopoietically active extracts from faeces. Lancet 1951(June 30):1391-2.” This study was not performed on vegans, but rather on people with pernicious anemia who cannot properly absorb B12. The B12 was isolated from the stool samples and given to the subjects intravenously. Because these people were ingesting B12, the B12 in their stool could have been from the B12 they were eating.
On the other hand, according to Lactobacillus lactis Dorner and Lactobacillus leichmannii assays, there were substantial amounts of B12 analogue found in the feces (e.g., 5 µg per 10 ml (2 teaspoons)). This seems like too much to have been provided by only the diet and enterohepatic circulation. Apparently, some of this B12 analogue was active, and there was enough to counteract any inactive B12 analogue in their stools. Thus, this study provides good evidence that there is active B12 produced by bacteria in the colon of at least some humans.
A variable to consider is that there are over 400-500 species of bacteria in the average human’s colon and these bacteria have not all been delineated. It is plausible that some humans have B12-producing bacteria in significant amounts while other humans do not. Some bacteria in the digestive tract absorb B12 for their own use, further complicating this situation.
Allen and Stabler (2008) found that more than 98% of B12 analogue in the human stool is inactive. This was in people who had a consistent intake of vitamin B12. They determined that 81% of non-absorbed, ingested B12 was destroyed or degraded into inactive analogue. This may or may not be the case in people with much lower, or no, vitamin B12 intakes.
There is evidence that some B12 might be absorbed from the colon. Kurpad et al. (2023) found that if B12 is injected into the colon some of it will be absorbed.
Bacteria in the Small Intestine
B12 deficiency has been found with relatively high frequency among vegetarian Indian immigrants in England, while at one time it was thought to be uncommon among native Indians with identical dietary patterns, possibly because healthy Indian subjects have a more extensive amount of bacteria in their small intestine than people in the West (Albert, 1980).
Albert et al. (1980) measured B12 production of bacteria in the small intestines of people in India using a Euglena gracilis Z assay. Results were confirmed by an Ochromonas malhamensis assay, which is thought to be specific for active B12. They determined that some active B12 was produced by members of the bacteria genera Klebsiella and Pseudomonas. Further confirmation using chromatography and bioautography showed a molecule with similar properties to cyanocobalamin. Albert et al. speculated that when Indians migrate to the West, their digestive tracts become like those characteristic of people in Western countries: with little or no bacteria in their upper small intestines. An article in Nutrition Reviews (1980) suggests some alternative causes of Indian immigrants to Britain having more B12 deficiency than Indian natives:
- In India, water is contaminated with various bacteria, including those from human and animal feces.
- The practice of defecating in open fields and lack of proper sewage.
- The mode of toilet hygiene where water is used instead of toilet paper.
It should also be noted that B12 deficiency is fairly common in India (see the table below), especially in lower income, lacto-ovo-vegetarians (Sarode 1989).
| Table 1. B12 Status of a Set of Healthy Native Indians age 27–55 | |||||
|---|---|---|---|---|---|
| Number | average serum B12 (pg/ml) |
serum B12 < 203 pg/ml | MMA > .26 µmol/l | HCY > 15 µmol/l | |
| Non-VegA Lacto-OvoB |
36 27 |
216 | 46% | 70% | 81%C |
| Source: Refsum, 2001, Table 1 A. Ate small amounts of animal products B. 1 person was vegan. C. Low folate status could have contributed to the high HCY levels (Antony, 2001) HCY – homocysteine • MMA – methylmalonic acid |
|||||
Iranian Villagers
Halstead et al. (1960) reported that some Iranian villagers with very little animal product intake (dairy once a week, meat once a month) had normal B12 levels. None had megaloblastic anemia. Their average B12 level was 411 pg/ml which was quite high considering their diet. The authors speculated this could be because their diets, which were very low in protein, allowed for B12-producing bacteria to ascend into the ileum where the B12 could be absorbed. They also speculated that because they lived among their farm animals and their living areas were littered with feces, they picked up enough B12 through contamination.
Halstead et al.’s 1960 report was in contrast to Wokes et al.’s 1955 report in which numerous British vegans were found to have neurological symptoms of B12 deficiency.
Conclusion
It’s possible that some vegans can ward off overt vitamin B12 deficiency, and even mild B12 deficiency, through B12 production by bacteria in the small intestine. However, this is an unusual condition, especially in Western countries, and should not be relied upon, including by raw foodists.
Bibliography
Herbert V. Vitamin B-12: plant sources, requirements, and assay. Am J Clin Nutr 1988;48:852-8.
4 thoughts on “Intestinal Bacteria as a Vitamin B12 Source”
Jack, hello. Terrific article. It turns out that SARS-COVID long haulers almost all have loss/damage to intestinal bacteria, and that it’s contributing to our acute B12 defeciences, many of which are not being caught by our docs. I’ve been researching how to repair and restore the bacteria that make active B12, and came across your piece here.
I’m dropping into the comments almost exactly eight years after the last one, because I just learned that the B12 tests in common use now are nearly useless for some people, and that there really is no one perfect test to determine deficiency of active B12, or its cause. It seems I was one of the post-SARS patients with markers like macrocytosis, lymphopenia, etc. that was deficient.
Fascinating writeup on the tests here: https://www.sciencedirect.com/topics/biochemistry-genetics-and-molecular-biology/schilling-test
This means, that a person can continue to get normal B12 tests in their labwork while they might be seriously B12 deficient. This can go on long-term, and be missed by doctors. Patients may develop macrocytosis/megaloblastic anemia, lymphopenia, nerve and muscle weakness and/or pain, nerve damage, brain damage, dementia.. it’s really an awful list.
Several things place people who fared poorly after SARS-COVID-19 infections (especially multiple infections) in danger of acute B12 deficiency. Most “Long COVID” patients have significant damage to their digestive tracts, stomachs and intestines. “Long COVID” cases are in the hundreds of millions now worldwide, and there could be a large population being left behind by the faulty modern assays, and because B12 deficiency isn’t front and center in our doctors’ minds. The days of only missing people with pernicious anemia is over, and now how many hundred million Long COVID and ME/CFS patients are suffering from what appears to be autoimmune injury, cognitive damage (“Brain Fog”), dementia, SARS/AIDS, leukemia, Addisons, immune damage, recurring cancers… another terrible list?
SARS-COVID leaves viral segments, active virus and antibodies in the blood, triggers latent other viruses (especially herpesviri like Epstein-Barr, herpes simplex, chickenpox/shingles, etc.) and antibodies alone can interfere with modern B12 tests. COVID also destroys stomach and gut bacteria, rendering people incapable of accessing the active B12 they normally would produce in our ileum, intestines and colons. People with SARS damage could be dangerously ill from long-term deficiency. I know that I was, and after one month of active B12 supplementation, I am strongly recovering. My labs are showing repair. None of my doctors were aware that the tests were not working for me (or people like me)- nor did they know about why, or how much of the population it affects.
Many papers suggest that the number of people left behind by modern B12 tests could be 25% or higher (which is insanely high for our doctors not to be on this) but if one adds in the number of people harmed by multiple SARS infections, that number could be staggering- certainly more than 50%. After all, most people have had COVID more than twice, and do not mask in public (I do). Absolutely great article here, from 2008:
https://pmc.ncbi.nlm.nih.gov/articles/PMC2696961/
While there are no good tests to offer the huge numbers of people who could be deficient, there are active B12 patches and liquids out there to trial- look for METHYLATED B12- this will show as cyanocobalamin or methylcobalamin on the labels.
If people are malabsorbing B12, though, liquids, pills or patches probably won’t help repair damage, and they could need B12 injections to improve. A doctor’s scrip is needed for that (in the US) and a course of injections could be needed. So fellow LC/ME/CFS patients, please ask your nutritionist, trainer, or doctor to help (I see a great link above for plant-based dieticians) or consider trialing active B12 patches.
Supplementing with active Vitamin B12 while our doctors catch up could save our lives.
Kate,
I appreciate you sharing your experience and concern. A few clarifications that may help readers:
Gut bacteria and B12: Humans don’t rely on B12 made in the colon; it’s produced too far downstream to be absorbed. Dietary/supplemental B12 is absorbed in the terminal ileum via intrinsic factor; a small fraction (about 1–2%) is absorbed anywhere along the gut by passive diffusion from high‑dose supplements.
Testing: Total serum B12 alone can miss functional deficiency, but modern practice uses better markers rather than abandoning testing. Methylmalonic acid (MMA) is the most specific functional marker; holotranscobalamin (holoTC) can help; homocysteine is supportive. The Schilling test is obsolete.
Long COVID: It’s linked to gut dysbiosis and barrier issues, but there’s no solid evidence that long COVID uniquely blocks passive diffusion, and only case reports (not population data) suggest it increases pernicious anemia. If deficiency is present without dietary cause, intrinsic factor and parietal cell antibodies, gastrin/pepsinogen, and sometimes endoscopy can confirm autoimmune gastritis/pernicious anemia (source).
Treatment: High‑dose oral B12 (e.g., 1,000 mcg/day cyanocobalamin) is usually effective even with malabsorption because of passive diffusion. Injections are appropriate for severe neurologic symptoms, very low levels, intolerance, or adherence concerns. Evidence for transdermal B12 patches is limited.
Forms: There is robust evidence for cyanocobalamin (via supplements) and hydroxocobalamin (via injections) correcting deficiency.
Practical advice for readers with symptoms (fatigue, macrocytosis, neuropathy, glossitis):
Test: serum B12 + MMA (or holoTC), consider homocysteine.
If low/indeterminate and not explained by intake: test intrinsic factor antibodies; if negative but suspicion remains, parietal cell antibodies and supportive gastric markers; consider GI evaluation.
Treat promptly (oral high‑dose or IM) and recheck labs/clinical response.
Hi,
I had a few questions for you based on the information you have presented here:
1. You said: “B12 deficiency has been found with relatively high frequency among vegetarian Indian immigrants in England, while it is supposedly uncommon among native Indians with identical dietary patterns (3, 4).”
Question: Your fourth citation, Refsum et al., indicates that B12 deficiency is common in India but here you state that it is “uncommon among native Indians”. Is this just a typo? Should it not read that it is common? Or did you mistakenly include Refsum et al. there when in reality you just meant to include Albert et al. (the third citation) there?
2. You indicate that Refsum et al. used 27 lacto-ovo subjects and 36 non-veg subjects. I feel this is also a mistake because it states, in their article, that the subjects on a vegetarian diet involved 27% of the 63 healthy subjects. Thus, I feel as if the correct number of lacto-ovo healthy subjects should have been listed at 17 rather than 27 lacto-ovo healthy subjects, as 17 represents 27 percent of the 63 healthy subjects. This would then mean that 47 healthy subjects would have been non-veg. Or am I missing something here?
Additionally, for the lacto-ovo and non-veg healthy subjects you list their combined average serum B12 at 216. I was just wondering if you could explain to me how it was you arrived at this figure? In Refsum et al.’s study, on table 1 of page 235, it states that this figure should be 160.
Lastly, it would be greatly appreciated if you could please explain to me how you concluded that 46% of the healthy subjects had a serum B12 < than 203.
Thanks in advance for your response. Look forward to hearing from you.
Hi Dustin–
Thank you for your comments and paying such close attention to this article.
1. I’m not sure why I cited Refsum et al. for the idea that B12 deficiency is rare in India—the study certainly doesn’t suggest that and I’ve removed that citation and modified the statement. Thanks for catching that.
2. I was reporting the percentages of healthy subjects with elevated homocysteine and MMA as listed in Table 1 of Refsum et al. Go to the bottom of the table and you’ll see I copied the percentages just as they list them. Regarding the average B12 level, I converted 160 pmol/l to 216 pg/ml (conversion factor of 1.35). I list B12 values in pg/ml throughout the VeganHealth.org site. I’ve now added the measurement to table header to prevent confusion.