Akkermansia muciniphila (probiotic strain)Probiotic

• Insulin sensitivity and cholesterolPossibly helps

  In the first human trial (2019, 32 completers), 3 months of pasteurized A. muciniphila improved insulin sensitivity by +28.6% versus placebo (P=0.002) [1].
  
  Fasting insulin fell 34.1% and total cholesterol 8.7% against placebo in the same trial [1].
  
  The authors call it a small exploratory pilot, not powered for firm conclusions [1]. ^[1]

• Body weight and fat massPossibly helps

  Body weight (-2.27 kg), fat mass (-1.37 kg) and hip circumference (-2.63 cm) all moved favourably against placebo [1].
  
  None of the three reached statistical significance (P≈0.09 for all), so weight loss is not what that trial established [1]. ^[1]

• Liver enzymes and gut-barrier markersPossibly helps

  Pasteurized cells lowered gamma-glutamyl transferase by about 24% against placebo (P=0.009), and lowered AST as well [1].
  
  Plasma lipopolysaccharide, a marker of gut-barrier leakiness, also dropped [1].
  
  Fasting glucose, HbA1c and CRP did not change significantly [1]. ^[1]

• Weight regain after dietingPossibly helps

  A 2026 controlled randomized trial in Nature Medicine gave pasteurized strain MucT during the maintenance phase after weight loss [5].
  
  It reported less weight regain, with the strongest cardiometabolic effect in people who started with low Akkermansia abundance [5]. ^[5]

• Type 2 diabetesPossibly helps

  A phase-2 RCT of strain AKK-WST01 in overweight and obese adults with type 2 diabetes reported lower BMI and HbA1c [4].
  
  That held only in participants whose baseline gut Akkermansia was low, not across the whole group [4]. ^[4]

• Who responded — an unresolved caveatPossibly helps

  ⚠ Two separate trials found the benefit concentrated in low-baseline responders, which complicates every average on this page [4][5].
  
  Nobody measures your baseline abundance before you buy a bottle. Read the pooled figures as an average across people who may not respond alike. ^[4]

• Muscle strength in older adultsPossibly helps

  A 12-week randomized trial of pasteurized strain HB05P in adults aged 60 and over reported improved muscle strength against placebo (P=0.0063) [3].
  
  Follistatin, a hormone that counteracts the muscle-wasting signal myostatin, rose in the same trial, with no notable safety concerns [3]. ^[3]

• Gut barrier and inflammatory bowel diseasePossibly helps

  Reviews describe A. muciniphila stimulating mucin production and strengthening intestinal tight-junction proteins, the basis for leaky-gut and IBD interest [6][7].
  
  That evidence is mechanistic and preclinical, plus the observation that abundance is lower in people with Crohn's disease and ulcerative colitis [7].
  
  A critical review concludes the human clinical evidence for supplementation stays limited and preliminary [8]. ^[7]

• Pasteurized (heat-inactivated)The form nearly all human data comes from.

  Elemental content

  Viable count capped below 10 CFU/g in EFSA's authorized specification — effectively no live cells [2]

  Absorption

  Supplied as a freeze-dried powder, water activity ≤0.43 and moisture ≤12% in the EFSA-cleared spec [2]

  Best for

  The form nearly all human data comes from. Insulin, cholesterol and liver-enzyme outcomes reached significance only in the pasteurized arm of the 2019 trial [1].

• Live (viable) cellsSold commercially, but it under-performed the pasteurized arm on several metabolic measures in the only human head-to-h…

  Elemental content

  Living cells of a Gram-negative, non-spore-forming, strictly anaerobic bacterium [1]

  Absorption

  Oxygen-sensitive and hard to culture; animal-derived growth media were replaced with a synthetic, human-compatible medium [1][9]

  Best for

  Sold commercially, but it under-performed the pasteurized arm on several metabolic measures in the only human head-to-head [1].

• Digestive tolerability

  The 2019 trial tracked nausea, flatulence, bloating, cramps, borborygmi and reflux every two weeks for 3 months at 10 billion cells/day [1].
  
  It found no adverse events attributable to supplementation, with over 99% compliance in every group [1].
  
  That is one small pilot of 32 completers, not a broad safety database [1]. ^[1]

• Pregnancy and breastfeeding

  EFSA's 2025 opinion states that no conclusions could be drawn from the three mouse reproductive and developmental toxicity studies submitted [2].
  
  Read that as an open question, not as clearance [2]. ^[2]

• Medication combinations nobody tested

  The 2019 trial excluded people taking glucose-lowering drugs, statins and immunosuppressants [1].
  
  Safety alongside those drugs has therefore not been tested — an evidence gap rather than a clean record [1]. ^[1]

• Conditions excluded from the research

  The same trial also excluded people with inflammatory bowel disease, irritable bowel syndrome, and gastroparesis or reduced gut motility [1].
  
  EFSA's opinions cover adults and adolescents from age 12, so nothing below that age has been assessed [2]. ^[1]

• Antibiotics

  The 2019 trial excluded antibiotic users, and 4 of its 7 dropouts were removed specifically for untimely antibiotic use during the study [1].
  
  A 2025 mouse study found penicillin exposure selected for A. muciniphila variants with a compromised ability to reduce obesity, and such variants are common in human microbiomes worldwide [13].
  
  That is a mouse-model mechanism, not a confirmed human effect [13]. ^[13]

• Berberine

  Berberine is not a competing supplement here. Multiple animal studies found it raises the gut's own Akkermansia abundance by stimulating colonic mucin secretion [14][15].
  
  No human trial comparing the two, or testing them together, was found in this research [15]. ^[15]

Nothing you eat contains it. A. muciniphila lives in your gut's mucus layer and is absent from all foods, fermented ones included [9].

Diet can still nudge the population you already have. Polyphenols — anthocyanins, flavan-3-ols, flavonols, flavanones, stilbenes such as resveratrol, and phenolic acids — raised its abundance across rodent and in-vitro studies [16].

Foods carrying those compounds include berries, pomegranate, red grapes, green tea, cocoa and olive oil [16].

⚠ Fiber is messier than the usual advice suggests. A review of fiber intervention studies found the effect on A. muciniphila inconsistent — sometimes up, sometimes down, depending on the disease context studied [17].

No trial has compared morning against evening dosing, so there is no evidence-backed best time [1].

The one concrete protocol is the 2019 trial's: every morning, on an empty stomach, from a product kept frozen at -20°C until it was swallowed [1].

That was chosen to keep the frozen, glycerol-suspended preparation intact, not because morning dosing outperformed anything [1].