Live bacteria
CFU means colony-forming units: a culture-based estimate of organisms able to grow under specified conditions. Strain, storage and shelf-life viability matter.
GUT MICROBE · FORMULATION MATTERS
Akkermansia.
Interesting human results.
Not all bacteria are alike.
Akkermansia research has moved beyond laboratory promise. Some pasteurized preparations have improved selected metabolic outcomes or limited weight regain. Larger trials also show important null results, and the product on the shelf may be a very different formulation.
See the measured resultsIngredient and formulation research, not proof of finished-product results. Reviewed 10 September 2026.
THE RESULTS, IN PLAIN ENGLISH
The early result is not the whole story.
Insulin sensitivity can be measured in several ways. A favorable small pilot and a later null primary result belong side by side. Low-baseline-microbe subgroup findings are interesting, but they do not erase the overall result.
TWO DIFFERENT ANSWERS TO TWO DIFFERENT TESTS
+28.62%
Insulin-sensitivity index
2019 pilot, pasteurized versus placebo.
± 7.02% SEM; p = 0.002.
12 pasteurized and 11 placebo completers. [1]
No clear benefit
Overall insulin sensitivity
2026 multicenter trial, four months.
30 billion pasteurized cells daily.
Primary Matsuda-index target not met. [2]
UNDERSTANDING THE INGREDIENT
What the name does, and does not, tell you.
Pasteurized bacteria
Heat-treated cells are not a live probiotic. Studies can quantify cells or fluorescent units even when organisms no longer form colonies. Pasteurized MucT cannot be matched to a CFU label by number alone.
AFU, TFU and mixtures
Active fluorescent units (AFU), total fluorescent units (TFU), total cells and CFU use different measurement methods. They are not universal conversion factors. A blend with several other actives needs its own evidence.
A CLOSER LOOK AT THE HUMAN RESEARCH
Read the result.
Keep the limits in view.
Positive findings, null results and reviews are shown together. Open any study for the population, dose, duration and limitations. Every summary is available without JavaScript.
DEPOMMIER · 2019 · EXPLORATORY TRIAL
The original pilot favored the pasteurized form
Insulin sensitivity improved in a small pasteurized-bacteria group. Weight loss was not statistically established. [1]
Study design, measured results and limitations
- Who and how
- Forty adults with excess weight and insulin resistance enrolled; 32 completed: 12 pasteurized, 9 live and 11 placebo. Participants received 10 billion A. muciniphila bacteria daily for three months, without a prescribed weight-loss program. The paper compares defined live and pasteurized preparations, not the multi-ingredient commercial mixture below.
- What changed
- Pasteurized treatment versus placebo: insulin sensitivity +28.62 ± 7.02% (SEM; p = 0.002), fasting insulin −34.08 ± 7.12% (p = 0.006), total cholesterol −8.68 ± 2.38% (p = 0.02). Weight difference was not significant (p = 0.091). The clearest signals were from pasteurized bacteria, not proof that all live preparations work similarly.
- What it means
- This was a small, exploratory proof-of-concept study with multiple metabolic endpoints and imbalanced attrition. The percentage refers to a calculated metabolic index, not percent better health or percent lower diabetes risk. Overall microbiome structure did not substantially change. Short-term tolerance and a promising index do not establish long-term clinical benefit.
Read the source · Primary full text
SUENAERT · 2026 · MULTICENTER TRIAL
A larger trial missed its main insulin-sensitivity target
The overall four-month result was null. Low-baseline-Akkermansia subgroups remain a research lead. [2]
Study design, measured results and limitations
- Who and how
- Daily capsules supplied 30 billion cells of pasteurized A. muciniphila MucT versus placebo for four months. The full report distinguishes 143 randomized participants, 142 exposed to treatment, and a 103-person prediabetes primary-analysis population (51 active, 52 control). Participants had metabolic syndrome.
- What changed
- Whole-body insulin sensitivity measured by Matsuda index did not significantly improve overall. Exploratory analyses suggested better responses among some people with low starting Akkermansia levels, including metabolic and glucose-stimulated GLP-1 signals.
- What it means
- The abstract’s 142 participants refers to the exposed population, not the randomized total. Secondary and subgroup p-values are exploratory. A GLP-1 response after a glucose test is not evidence of appetite suppression or equivalence to GLP-1 medicines. A stool test is not yet a validated instruction to take this product. Thirty billion pasteurized cells is not thirty billion CFU.
Read the source · Primary full text
ZHANG · 2025 · RANDOMIZED TRIAL
Live strain research also shows a mixed overall result
Body weight and HbA1c fell in both groups without a significant overall treatment difference. [3]
Study design, measured results and limitations
- Who and how
- Fifty-eight participants with type 2 diabetes and overweight or obesity received live A. muciniphila strain AKK-WST01 or placebo for 12 weeks, alongside routine lifestyle guidance. The accessed abstract did not specify the dose; no dose equivalence is assumed.
- What changed
- The overall between-group body-weight and HbA1c comparisons were not significant. Colonization and selected metabolic responses appeared stronger in the low-baseline-Akkermansia subgroup.
- What it means
- A subgroup signal is not the same as overall success. The strain identity matters, and the current product label does not establish a match to AKK-WST01. Mouse experiments within the same paper support mechanisms, not a second human trial.
Read the source · Primary abstract
MOUNT · 2026 · RANDOMIZED TRIAL
Less regain after weight loss, not weight loss from a capsule alone
A pasteurized MucT preparation helped limit regain after a structured low-energy diet. [4]
Study design, measured results and limitations
- Who and how
- The report describes 90 adults with overweight or obesity. An eight-week low-energy diet targeted at least 8% weight loss, followed by 24 weeks of weight maintenance with pasteurized MucT or placebo. The accessible publication preview did not specify the dose.
- What changed
- Reported regain during maintenance was 1.2 ± 0.7 kg with MucT and 3.2 ± 0.4 kg with placebo (p = 0.012). Both groups regained some weight. The preview did not define whether these ± values are SD or SEM, so they are reproduced as reported, not drawn as uncertainty bars.
- What it means
- The study is about maintenance after diet-induced loss, not starting a probiotic and losing weight without lifestyle change. Full endpoint denominators and complete statistical methods were not accessible in the preview. Authors reported company roles and patent interests. No serious treatment-related adverse events were reported; this is not a claim of no adverse events.
Read the source · Primary abstract and publisher preview
ROSHANRAVAN · ONLINE 2021 · SYSTEMATIC REVIEW
A systematic review, largely a foundation for later trials
Early evidence was promising, but human and rodent research must not be pooled into a consumer benefit claim. [5]
Study design, measured results and limitations
- Who and how
- The review searched through March 2020 and included 15 studies across human and rodent models. It examined supplementation and obesity-related mechanisms.
- What changed
- The authors described therapeutic potential while calling for further human clinical trials. This review predates the larger 2025 and 2026 trials summarized above.
- What it means
- Fifteen included studies does not mean fifteen human trials. The accessed abstract does not supply a human-only pooled dose, effect size or safety estimate. It is background context, not proof of weight loss from a finished probiotic blend.
Read the source · Systematic review, abstract
This is a selected evidence review, not an exhaustive systematic review of every possible use. A review can contain the trials shown above; those are not additional independent experiments. A p-value describes statistical evidence, not the size or importance of a benefit.
FROM THE STUDY TO THE LABEL
Compare the actual formulation.
Current catalog labels are product information, not independent proof of efficacy, identity, potency or bioequivalence. No finished Neurogan product below was the intervention in the cited trials.
Akkermansia Probiotic
Supplement Facts: One capsule: Akkermansia 150 mg (30 billion CFU), NMN 150 mg, berberine 100 mg, epigallocatechin 100 mg, micronized trans-resveratrol 50 mg and inulin 20 mg.
Directions: Directions: one or two capsules daily. The full listed blend is 570 mg per capsule; that is not 570 mg of Akkermansia.
Research comparison: The visible label does not specify strain, pasteurization status, culture/viability method or CFU guarantee at expiry. “Epigallocatechin” is not automatically EGCG. A 30-billion-CFU mixture is not the 30-billion-cell pasteurized trial preparation.
Current product listing · Original label image
View the label used for this comparison
This is the label image linked by the live product record when reviewed on 10 September 2026.
Serving size and daily directions are listed separately. A dose similarity is not evidence that a different formula produces the same outcomes.
Labels and listings can change. Check the package you receive and discuss individual use with a qualified clinician.
SAFETY AND REALISTIC EXPECTATIONS
Short trials cannot settle long-term safety.
The cited trials reported generally acceptable short-term tolerability in their selected populations. That does not establish safety in severely immunocompromised or critically ill people, or for every live strain. The catalog blend also contains NMN, berberine, epigallocatechin, resveratrol and inulin. Medication interactions, gastrointestinal tolerance and pregnancy precautions therefore cannot be assessed from an Akkermansia-only trial. Ask a qualified clinician before use, especially during pregnancy or breastfeeding, with glucose-lowering medicines, or with major illness. The label does not settle the live-versus-pasteurized question.
How to use this review responsibly
This page is educational, not a diagnosis, treatment plan or individualized dose recommendation. Do not stop prescribed treatment or replace established care on the basis of a biomarker result. “No significant difference” does not prove two treatments are equivalent, and “well tolerated” in a small trial does not rule out rare harms.
Bring the full ingredient list, medicines and health history to a clinician or pharmacist. In a blend, the safety question concerns every ingredient together, not only the name on the front label.
REFERENCES AND ACCESS
Follow the evidence.
Links lead to primary publications or systematic reviews. Access limitations are stated rather than filled with assumed methods or outcomes. Product comparisons use the live product records and linked label images above.
DEPOMMIER · 2019 · EXPLORATORY TRIAL
The original pilot favored the pasteurized form. Access reviewed: Primary full text.
Open publication · Return to study summarySUENAERT · 2026 · MULTICENTER TRIAL
A larger trial missed its main insulin-sensitivity target. Access reviewed: Primary full text.
Open publication · Return to study summaryZHANG · 2025 · RANDOMIZED TRIAL
Live strain research also shows a mixed overall result. Access reviewed: Primary abstract.
Open publication · Return to study summaryMOUNT · 2026 · RANDOMIZED TRIAL
Less regain after weight loss, not weight loss from a capsule alone. Access reviewed: Primary abstract and publisher preview.
Open publication · Return to study summaryROSHANRAVAN · ONLINE 2021 · SYSTEMATIC REVIEW
A systematic review, largely a foundation for later trials. Access reviewed: Systematic review, abstract.
Open publication · Return to study summary