10 Long-Term Carnivore Dieters Had a Surprisingly Intact Gut Microbiome

A small but striking study published in Microbiota and Host in 2026 challenges one of the most repeated assumptions in nutrition science: that eliminating dietary fiber inevitably degrades gut microbiome health. The research compared 10 people who had followed a strict all-animal diet for an average of three years against 874 matched omnivore controls, and the results were neither a clean vindication nor a straightforward condemnation of the carnivore approach.

Carnivore dieters showed unexpected microbiome richness

The most surprising result was what did not happen. Standard thinking holds that diet is a primary driver of gut composition, and that slashing fiber should reduce microbial diversity dramatically. Yet the carnivore group showed no significant difference from omnivores on Shannon entropy, a widely used index of diversity. In fact, Chao1 richness, a metric that captures low-abundance bacterial species, was significantly higher in carnivore subjects. More rare bacteria were present in the fiber-free gut, not fewer.

Beta diversity told a different story: the overall composition of the carnivore microbiome was clearly distinct from that of omnivores (P = 0.001), with the phyla Synergistetes and Desulfobacterota appearing 90 and 9 times more abundant, respectively, in carnivore guts. Species associated with proteolytic fermentation, meaning bacteria that break down protein rather than carbohydrate, dominated. These included Alistipes onderdonkii, Fusobacterium varium, Alistipes putredinis, and Desulfovibrio piger, all of which are enriched in high-protein, low-fiber environments.

Notably absent, or sharply reduced, were Faecalibacterium prausnitzii, multiple Bifidobacterium species, and Roseburia hominis, all organisms commonly linked to fiber fermentation and colon health.

What a fiber-free gut means for metabolic function

The functional picture was equally mixed. The study identified 317 metabolic pathways and 13 functional modules that differed significantly between groups, all of them enriched in carnivores. Several were encouraging: modules for gut barrier function, acetate production, propionate production, and vitamin B12 synthesis were elevated. The carnivore group's microbiome appeared to be fermenting host-derived glycans, compounds the body itself secretes, rather than dietary fiber, a potential compensatory mechanism the authors flag for future investigation.

At the same time, adverse modules were also elevated: cytotoxicity markers, inflammation indicators, constipation indicators, and fructose intolerance signals all showed higher predicted activity in carnivore guts. Researchers examining carnivore diet risks and long-term health outcomes have previously flagged concerns about cardiovascular risk and nutrient deficiency; this study adds a functional microbiome dimension to that picture without resolving it.

The carnivore subjects themselves reported excellent self-assessed health. Their average BMI was 23.28 kg/m2, they averaged about seven hours of quality sleep per night, reported regular physical activity and low stress, and scored 8.81 out of 10 on a self-reported health scale, significantly higher than omnivore controls (P = 0.004). Whether the diet, the lifestyle, or the selection effect explains that gap is a question this study cannot answer.

How the study design both strengthens and limits its conclusions

The 10 carnivore subjects had maintained their diet for 22 to 60 months, averaging 36 months, making this one of the longer-term observational snapshots of true carnivore adherence published to date. Researchers used 16S rRNA gene sequencing alongside multivariate regression modeling that adjusted for age, BMI, sleep, and place of residence. The carnivore diet emerged as the dominant independent predictor of both taxonomic composition and predicted metabolic function, with host variables playing a secondary role.

Still, cross-sectional designs carry real methodological constraints in diet-microbiome research, and this study is no exception. The 874 omnivore controls came from a proprietary data bank, not from a contemporaneously recruited cohort. The n of 10 for carnivore participants is small enough that a single outlier can shift results materially. No blood draws, anthropometric measures, or dietary diaries were collected; health status rested entirely on online questionnaires. A single stool sample per participant cannot capture the natural week-to-week variability in microbiome composition.

Specific open questions about fiber, mucins, and long-term risk

One of the paper's most thought-provoking observations involves carbohydrate degradation. Despite eating no plant-derived carbohydrates, carnivore subjects had enriched carbohydrate-degrading microbial machinery. The likely explanation, the authors propose, is that bile-tolerant taxa and mucin-degrading bacteria are filling the niche left by absent dietary fiber by feeding on the gut's own mucous lining. Whether that sustained mucosal consumption is neutral, adaptive, or gradually erosive is genuinely unknown.

The dramatic reduction in Faecalibacterium prausnitzii also merits scrutiny. This species is one of the most studied anti-inflammatory commensals in the human gut, and its near-disappearance in carnivore subjects is a signal that warrants longitudinal follow-up. The study also cannot exclude the possibility that carnivore dieters represent a self-selected population with unusually robust health to begin with, and that the apparent benefits would not replicate in a broader, randomized context.

Where carnivore microbiome research needs to go next

The authors of this study call their results preliminary and explicitly frame them as hypothesis-generating rather than conclusive. The next logical steps align with what researchers in the field have argued more broadly: future diet-microbiome studies need richer data collection, including metabolomics, detailed dietary records, and longitudinal sampling across months or years.

For now, the study's main contribution is disruptive in the productive sense. It demonstrates that long-term carnivore dieters do not show the collapsed microbiome diversity that fiber-centric models would predict, while simultaneously showing a functional microbiome profile that carries both potentially beneficial and potentially harmful signals. Fiber, the paper suggests, may not be the only substrate capable of sustaining a complex microbial community. What that means for long-term health remains, for the moment, an open question.

Evidence rating

INVESTIGATIONAL — 67/100

Caveats:

• The carnivore group comprised only 10 individuals, making all findings statistically underpowered and not generalizable.
• Two co-authors (Paul Hammer and Carsten Krumbiegel) are co-founders and equity holders of Biomes NGS Ltd, the company that performed all sequencing, owns the proprietary omnivore control dataset, and retains the right to deny third-party data access — a direct commercial conflict of interest.
• The lead author (Andrija Karačić) is founder and owner of The Gut Microbiome Center, which offers commercial microbiome testing services.
• The omnivore control dataset is proprietary to Biomes NGS Ltd and cannot be independently verified or reproduced, which prevents external validation.
• Functional pathway data are computationally inferred via PICRUSt2 from 16S rRNA amplicon data, not measured directly — these are predictions, not confirmed metabolic activities.
• No laboratory biomarkers, anthropometric measurements, or dietary recalls were collected from carnivore participants; health assessment relied entirely on self-report.
• Cross-sectional design cannot establish causation; healthy-user bias is likely given participants were recruited via a social media influencer and self-selected for a health-oriented diet.

Cross-sectional study comparing gut microbiota profiles of 10 long-term carnivore diet adherents against 874 omnivore controls drawn from a proprietary commercial database. The extreme group size imbalance and very small carnivore sample (n=10) severely limit statistical power and generalizability. Functional pathway data are computationally predicted from 16S amplicon sequencing rather than directly measured. The study demonstrates associations only and cannot establish causation; health assessment in the carnivore group relied solely on self-reported questionnaires with no clinical or laboratory validation. The proprietary nature of the control dataset precludes independent replication or verification.

References

1. Long-term adherence to the carnivore diet and its impact on the gut microbiota: a cross-sectional study in: Microbiota and Host Volume 4 Issue 1 (2026). https://doi.org/10.1530/mah-25-0015
2. Ping Zhang (2022). Influence of Foods and Nutrition on the Gut Microbiome and Implications for Intestinal Health. International Journal of Molecular Sciences. https://doi.org/10.3390/ijms23179588
3. Almiera Lietz, Janina Dapprich, Tobias Fischer (2026). Carnivore Diet: A Scoping Review of the Current Evidence, Potential Benefits and Risks. Nutrients. https://doi.org/10.3390/nu18020348
4. Mariantonia Maglio, Riccardo Troncone (2020). Intestinal Anti-tissue Transglutaminase2 Autoantibodies: Pathogenic and Clinical Implications for Celiac Disease. Frontiers in Nutrition. https://doi.org/10.3389/fnut.2020.00073
5. Emily R. Leeming, Panayiotis Louca, Rachel Gibson, Cristina Menni, Tim D. Spector, Caroline I. Le Roy (2021). The complexities of the diet-microbiome relationship: advances and perspectives. Genome Medicine. https://doi.org/10.1186/s13073-020-00813-7