An akkermansia muciniphila supplement has two human trials behind it, and both are small. In the bigger trial, 90 adults had just finished a diet. Over the next 24 weeks they put about 1.2 kg back on, while the placebo group put back about 3.2 kg (Trial). That result came from the heat-killed form. The best known US product sells the live form, which did not beat placebo in the earlier trial (Trial).
Both trials trace back to the research group behind the company that sells the product. The signal here is real enough. The evidence behind it is thin, and the label matters more than the price.
What Akkermansia Actually Is
Akkermansia muciniphila lives in the mucus layer that lines your gut wall. It feeds on mucin, the main protein in that mucus, and takes its carbon and nitrogen from it (Safety opinion). Eating the gut’s own lining sounds alarming. In a healthy gut it seems to be routine housekeeping, and the body keeps rebuilding the layer.
Most adults already carry it. A capsule tops up a resident rather than bringing in a stranger.
The commercial interest started with a pattern in the data. People with obesity and type 2 diabetes tend to carry less of it. One systematic review pooled 42 human studies of gut bacteria and type 2 diabetes (Systematic review). Akkermansia was one of five groups that came out lower in people who had the disease.
But that is a link, not proof. It is usually measured at a single moment in time. Low Akkermansia might help cause the problem. It might just as easily be a result of it.
The review’s own authors flag the gap. Most human studies in this field, they write, never try to work out which bacteria cause what (Systematic review).
That gap is exactly why someone put the bacterium in a capsule. If low levels do harm, topping them up should help. Only a trial can settle it.
The 2019 Proof-of-Concept Trial
The first human test was small, and the researchers said so themselves.
They signed up 40 overweight adults with early insulin problems, and 32 finished the three months (Trial). A coin flip decided who went into which group. Eleven took placebo, twelve took heat-killed Akkermansia, and nine took the live bacterium. Both active groups swallowed 10 billion cells a day.
The heat-killed group came out ahead on several blood test numbers. Their bodies handled insulin about a quarter better than the placebo group. Fasting insulin fell by about a third, and total cholesterol dropped by around 9% (Trial). Two liver enzymes also settled down.
The live group looked different. Against placebo it showed no clear gain on those numbers. It did improve one insulin score against its own starting point, which is a much weaker test. The liver and inflammation numbers did not move at all (Trial).
Weight barely shifted in anyone. The heat-killed group ended about 2.3 kg lighter than the placebo group. That gap was small enough that chance could explain it (Trial).
None of this was the trial’s main job. It was built to test safety and tolerance first, with the metabolism numbers riding alongside (Trial). The authors write that the study was never big enough to settle questions about metabolism. Treat the results as a lead worth chasing, not as a finding.
The trial registry lists 54 people signed up, and a fourth group on a lower live dose (Registry record). The published paper reports 40 signed up and 32 finishing (Trial).
Public grants and charity prizes paid for this one. Two authors had co-founded the company that grew out of the work. Five are named on patent applications covering the bacterium for obesity (Trial).
Why Heat-Killed Beat Live
That is the odd part, because heat normally ruins a probiotic.
The best explanation so far comes from mice. A protein on the surface of Akkermansia, called Amuc_1100, keeps its shape right up to the 70 C used for heating (Study). In lab dishes, a made-to-order copy of that protein switches on one immune sensor on gut cells, and leaves the neighbors alone. Killing the bacterium seems to leave that protein easier for the gut lining to reach.
The animal results matched. In obese mice, the heat-killed version beat the live one at holding down fat gain and insulin resistance (Study). That is what pushed the idea into human testing.
The researchers printed the caveat themselves. This explanation lives in mice and in cell cultures. How pasteurization boosts the effect, they wrote, still remains to be worked out (Study).
The 2026 human trial could not fill that gap either. Its authors list the missing comparison, a version with the active part taken out, as a limit (Trial).
There is a naming problem too. A probiotic is a live organism by definition, so a dead one is a postbiotic. The version with the human weight data behind it is not really a probiotic at all.
The 2026 Weight-Regain Trial
The second trial is the stronger one, and it asked a sharper question. Losing weight is the easy half. Keeping it off is where most people come unstuck.
Ninety adults with overweight or obesity spent 8 weeks on a low-energy diet, and each lost at least 8% of their body weight (Trial). Then they ate freely for 24 weeks. A coin flip decided who got what. One group stirred a daily dose of heat-killed Akkermansia into water before breakfast. The other took placebo.
The supplement group put about 1.2 kg back on. The placebo group put back about 3.2 kg. Measured from the very start, the supplement group finished about 3.1 kg lighter than the placebo group (Trial).
Two things make this more convincing than the pilot. The team registered weight regain as its question before the trial began, not a lucky result spotted afterwards (Trial). And 80 of the 90 people finished, according to trade reporting on the paper (Report).
Starting levels may matter. A study co-author who works for the sponsor described a second look at the data. People who began with less Akkermansia of their own seemed to gain the most, on weight and on blood pressure (Report). The paper itself says only that starting levels tracked the response (Trial).
Outsiders got a look too. The Science Media Centre asked three scientists with no part in the trial what they made of it (Expert reaction).
One called the roughly 2 kg gap a proof of concept, not a reason to change practice. He pointed at the small group, the short follow-up and the narrow set of volunteers. A second called the weight effect limited. He saw the supplement as a possible add-on to existing obesity treatments rather than a rival. A third was more enthusiastic. He called the observation important, though he noted the trial was short. He asked whether it might help people who regain weight after stopping GLP-1 drugs. None of the three told the public to go and buy it.
The funding matters as much as the result. The Akkermansia Company paid for the trial, and supplied the product and the placebo (Trial). Several authors are tied to that company, including its co-founder. Three are named on a patent application linked to the study.
That does not make the result wrong. Company-run trials can be clean, and this one was registered in advance and run blind. It does mean the finding needs repeating by a group with nothing riding on it.
Akkermansia Muciniphila Supplement Costs
The trial form is sold as Akkermansia Essential, by the company that funded the trial. In August 2026 it listed at $65 for 30 capsules, about a month’s worth (Product listing). Each capsule carries what the company counts as 30 billion pasteurized cells.
The best known American product is different. Pendulum Akkermansia listed at about $54 a month on the maker’s subscription option (Product listing). It holds live Akkermansia, at 100 million active cells per capsule.
Line those numbers up against the trials. The 2019 pilot gave people 10 billion cells a day (Trial). The live product sits roughly 100 times below that count. And live is the form that did not beat placebo.
A third trial sharpens the point. In China, 58 adults with type 2 diabetes and extra weight took a live strain or placebo for 12 weeks (Trial). Both groups lost some weight and improved their blood sugar, with no clear difference between them. Only the people who started low responded, and the bacterium only settled in for them. That trial had no heat-killed group at all (Registry record).
Run two checks on the label before you pay. First, look for the word pasteurized or heat-killed, because that is the form with the weight data. Second, compare the cell count against the 10 billion a day that was tested.
Then there is the subscription problem. Neither trial banked the effect. People took a dose every day: three months in the first trial, six in the second (Trial). Nobody followed them after they stopped. Stop paying and you stop taking it.
Safety and Who Should Wait
Safety is the strongest part of this record. Neither trial reported a serious side effect linked to the treatment (Trial). In 2019, 10 billion cells a day for three months was described as safe and well tolerated, in both the live and heat-killed form (Trial).
European regulators have assessed the pasteurized form twice. In 2021 a safety panel set a ceiling of 34 billion cells a day for adults (Safety opinion). Pregnant and breastfeeding women were left out, and the cells have to be properly dead. The European Commission then cleared pasteurized Akkermansia for sale from March 2022, at that same limit (Regulation).
A 2025 opinion from the same panel covered teenagers. It judged 21 billion cells a day safe from age 12, and 30 billion from 14 (Safety opinion). It also said safety in pregnancy and breastfeeding is not established.
The American route was different. A pasteurized Akkermansia ingredient reached the market through a new dietary ingredient notification, filed in late 2024 (FDA letter). The agency’s letter says accepting a notification for filing is a paperwork step. It is not a finding that the ingredient is safe.
Clearance is a safety and marketing decision. No regulator anywhere has said the product works.
Some people should wait regardless. Anyone with a weak immune system or a serious illness should ask a clinician first, because both trials deliberately left people like them out. The European panel set its dead-cell rule partly to head off possible effects on the gut barrier in more vulnerable people (Safety opinion).
One loose end deserves a mention. Akkermansia runs high, not low, in some brain and nerve conditions. The largest study of its kind found more of it in people with multiple sclerosis than in people from their own households (Study). The gap was clearest in those not yet on treatment. A pooled analysis of Parkinson’s disease studies pointed the same way (Meta-analysis). Those authors raise it as a reason for care, while stressing that the picture is unresolved.
How to Raise Your Own Levels
There is a cheaper route, and its evidence is older and broader.
Food can move Akkermansia, but not on command. A systematic review gathered 29 human trials covering 1,444 people, and 11 of them measured Akkermansia (Systematic review). Only a handful of things raised it: calorie cutting, pomegranate extract, resveratrol, inulin, polydextrose, sodium butyrate and a yeast product. Kiwifruit capsules and resistant starch did nothing. A low FODMAP diet lowered it. The reviewers judged the studies poor in quality, and said their own conclusions were not final.
Inulin has the clearest single trial. Sixty adults with type 2 diabetes took 10 g of inulin a day, sodium butyrate, both together, or placebo for 45 days (Trial). Inulin on its own raised Akkermansia against placebo, and so did sodium butyrate. Oddly, the two combined did not. Each group held only about 15 people. Levels drifted up in the placebo group too, and people responded very differently. Treat it as encouraging, not settled.
Inulin comes from chicory root, onions, garlic and leeks. It is one reason dietary fiber keeps turning up in gut research.
Polyphenols are the other lever, and the results are very personal. Orange juice, resveratrol and Schisandra juice have each raised Akkermansia in a human study (Review). The resveratrol result showed up in only part of the group. Pomegranate is a different case. Some people can turn its polyphenols into a compound called urolithin A, and they carry more Akkermansia than the people who cannot. So the bacterium tracks that ability, rather than the juice itself. For berries, cocoa and green tea, most of the Akkermansia evidence is still in mice.
Calorie restriction is the most interesting case, and the most misread. Among 49 overweight adults on six weeks of calorie cutting, those who started with more Akkermansia improved the most (Study). But in that same high-starting group, Akkermansia fell during the diet, even though it stayed above everyone else’s. Dieting shifted the bacterium around. It did not simply raise it.
So food has a real limit here. No meal reliably guarantees one particular species. What food does well is feed the whole community at once, and the same goes for fermented foods.
Fiber and polyphenols feed the microbes you already have, at grocery prices. A single strain in a capsule costs roughly $54 to $65 a month, and targets one name on a very long list.
Frequently Asked Questions
Does an akkermansia muciniphila supplement actually help you lose weight?
Not by itself, on the evidence so far. The one trial that measured weight properly tested people who had already dieted, and it was about holding the loss (Trial). Those taking it put back about 2 kg less over 24 weeks than the placebo group. In the earlier trial, the weight difference was small enough that chance could explain it (Trial).
Should I buy live or pasteurized Akkermansia?
The human evidence sits with the pasteurized, heat-killed form. In the 2019 trial, the heat-killed group improved on insulin and cholesterol numbers, while the live group did not beat placebo (Trial). A separate trial of a live strain in 58 adults found no overall difference against placebo (Trial). The best known US product sells the live form, so read the label.
How long do you have to take it?
Both trials dosed daily, with no break: three months in 2019, and six months in 2026 (Trial). Neither followed people after they stopped taking it (Trial). So nobody knows whether any benefit holds once you stop.
Can I raise Akkermansia with food instead?
Sometimes, and not predictably. Across those 29 human trials, only a few things raised it, including inulin, calorie cutting and some polyphenol foods (Systematic review). One 45-day trial found that 10 g of inulin a day raised it against placebo (Trial). People respond very differently, so treat food as a way to feed the whole gut community.
Key Takeaways
- Two human trials, both small. The heat-killed form has one pilot with 32 finishers, and one proper trial in 90 people (Trial).
- Heat-killed beat live. In 2019 the pasteurized group improved on insulin and cholesterol numbers, and the live group did not beat placebo (Trial).
- About 2 kg less regain. After an 8-week diet, the supplement group put back 1.2 kg over 24 weeks, against 3.2 kg on placebo (Trial).
- The company connection is real. The Akkermansia Company funded the 2026 trial and supplied the product, and its people are among the authors (Trial).
- The label trap. The best known US product sells live Akkermansia, the form that did not beat placebo in 2019 (Product listing).
- Fiber is the cheap route. Inulin raised Akkermansia in a 45-day trial, at grocery prices (Trial).
Feed the Gut You Already Have
Akkermansia is a genuine signal, and it is not yet a metabolic upgrade you can buy.
If one group has a case for trying it, it is people coming off a weight-loss phase who want to hold what they lost. That is the only situation the stronger trial tested. It is not a stand-in for the diet, and no scientist who reviewed the trial suggested otherwise (Expert reaction).
For everyone else, the cheap levers still carry the broader evidence. Eat more fermentable fiber. Eat more colorful plants. Feed the lining you already have, rather than buying one bacterium to put in it.
The evidence is still building, and this one is worth watching.
This article is for educational purposes and is not medical advice. Talk to a qualified clinician before changing your health regimen.
Bio-Hacking.Blog 
Leave a comment