TMAO and Heart Disease. What the Gut-Heart Research Actually Shows.

The gut microbiome does not stop at the intestinal wall. A decade of research from the Cleveland Clinic and elsewhere has established a biologically coherent pathway from the bacteria in your colon to atherosclerotic plaque in your coronary arteries, running through a small molecule most physicians have never ordered: trimethylamine N-oxide, or TMAO. The association data are consistent and the mechanism is plausible, but the field has not yet crossed from association to proven causation in humans, and clinical practice has not changed as a result.

The Mechanism

The pathway begins in the gut lumen. When you eat foods rich in choline, phosphatidylcholine (lecithin), or L-carnitine, specifically red meat, eggs, dairy, and certain seafood, gut bacteria metabolize these compounds to trimethylamine (TMA). This step is entirely microbial: without the relevant bacterial species, the conversion does not happen at all, or happens at very low rates. The bacterial genera most implicated include Clostridium, Prevotella, Klebsiella, Citrobacter, and several species within Ruminococcaceae. The specific species profile varies considerably between individuals, which explains why two people eating the same meal can produce dramatically different plasma TMAO concentrations hours later.

TMA is absorbed from the gut and travels to the liver via portal circulation. In the liver, the enzyme flavin-containing monooxygenase 3 (FMO3) oxidizes TMA to TMAO. This oxidation step is the rate-limiting biochemical conversion in the pathway: individuals with severely reduced FMO3 activity accumulate TMA rather than TMAO, producing the fish-odor syndrome (trimethylaminuria), a rare metabolic disorder. In people with normal FMO3 activity, TMAO is released into systemic circulation and eventually cleared renally. Kidney impairment therefore raises TMAO independently of dietary intake or microbiome composition, a confounding variable that careful studies must account for.

The downstream cardiovascular effects of TMAO appear to operate through at least three mechanisms. First, TMAO promotes foam cell formation by upregulating scavenger receptors (SR-A and CD36) on macrophages while simultaneously suppressing reverse cholesterol transport, the process by which cholesterol is moved from peripheral tissues back to the liver for excretion. This dual effect means more cholesterol enters plaque and less is removed. Second, TMAO appears to increase platelet aggregability and thrombus formation in animal models, raising the theoretical risk of acute coronary events independent of plaque burden. Third, elevated TMAO is associated with suppression of bile acid synthesis via farnesoid X receptor (FXR) signaling, reducing the liver’s ability to clear cholesterol through the bile acid pathway.

A related metabolite, gamma-butyrobetaine, deserves mention. It is produced from L-carnitine by gut bacteria as a precursor to TMA and is itself converted to TMA by certain microbes. Koeth et al. identified it as the dominant intermediate in the carnitine-to-TMAO pathway, which means TMAO is the end-product of a multi-step microbial process, not a simple direct conversion. This matters because it illustrates how the depth of the microbiome’s enzymatic repertoire, not merely the presence of a single species, shapes the actual metabolic output from any given meal. 3 / Early

What the Evidence Shows

The foundational observation came from Wang et al. (2011, Nature), from the Hazen laboratory at the Cleveland Clinic. That study used untargeted metabolomic profiling of plasma from patients undergoing elective cardiac evaluation. The investigators identified TMAO, along with choline and betaine, as compounds that distinguished patients who went on to have major cardiovascular events (myocardial infarction, stroke, or death) from those who did not over three years of follow-up. The TMAO association held after adjustment for traditional risk factors including LDL, blood pressure, diabetes status, and smoking. This was the discovery paper that placed TMAO on the research map. (Wang et al., Nature 2011)

Tang et al. (2013, Nature Medicine) formalized the clinical association in a prospective study of 4,007 patients from the same Cleveland Clinic cardiac evaluation registry. Patients were followed for three years after initial evaluation. Across quartiles of plasma TMAO, the highest quartile showed a 2.54-fold increase in risk of major cardiovascular events compared to the lowest quartile (95% CI: 1.96 to 3.28). After full multivariate adjustment, including eGFR to account for renal clearance effects, the hazard ratio for the top quartile versus the bottom quartile was 1.80 (95% CI: 1.32 to 2.46). The inclusion of eGFR adjustment matters: because TMAO is renally cleared, patients with worse kidney function accumulate more TMAO regardless of diet, and kidney disease itself predicts cardiovascular events. The Hazen group’s careful handling of this confounder strengthens the association finding, though it does not resolve the causation question. (Tang et al., Nature Medicine 2013) 3 / Early

Koeth et al. (2013, Nature Medicine) conducted the L-carnitine intervention study that connected the dietary-microbiome-TMAO chain in human subjects. Healthy volunteers consumed oral L-carnitine supplements or a controlled red meat meal. Omnivores produced substantially more plasma TMAO than vegans given the same carnitine load, demonstrating that microbiome composition, shaped by habitual dietary pattern, determines TMAO production capacity. When omnivore participants were pretreated with a broad-spectrum antibiotic course to suppress gut bacteria, the TMAO rise after carnitine ingestion was eliminated and restored only after the microbiome reconstituted. This was mechanistic human evidence that the gut bacteria, not the dietary substrate alone, drive TMAO production. (Koeth et al., Nature Medicine 2013)

The PREDIMED trial, a large Spanish randomized trial comparing Mediterranean diet versus a low-fat control diet in adults at high cardiovascular risk, provided relevant indirect data. Participants assigned to Mediterranean diet had lower circulating TMAO at follow-up compared to the control group. The Mediterranean diet is rich in plant protein, fiber, and polyphenols while low in red meat. These characteristics select for a gut microbiome composition associated with lower TMAO-producing capacity. The TMAO analysis was not PREDIMED’s primary endpoint, but it aligned the cardiovascular benefit of plant-predominant dietary patterns with the TMAO mechanism in a randomized context. 3 / Early

What the evidence does not yet show is equally important. No randomized clinical trial has specifically reduced TMAO levels through a targeted intervention and then measured cardiovascular events as the primary outcome. All of the cardiovascular association data comes from observational cohorts. Observational data on TMAO are subject to confounding: people who eat more red meat and produce more TMAO also differ from low-TMAO individuals in many other ways that influence cardiovascular risk. The Hazen group adjusted carefully for measured confounders, but residual confounding in observational cardiovascular research cannot be fully excluded. The field requires an intervention that specifically reduces TMAO in a double-blind trial and shows event reduction. That trial has not yet been published. 3 / Early

A second complication is that not all dietary TMAO sources behave identically in epidemiological studies. Saltwater fish, including cod, haddock, and certain shellfish, contain high amounts of preformed TMAO and produce substantial circulating TMAO after ingestion. Yet fish consumption is associated with reduced cardiovascular risk in most prospective population studies, not increased risk. This inconsistency suggests several possibilities: TMAO from fish may carry a different accompanying metabolite profile that is cardioprotective, omega-3 fatty acids in fish may offset any TMAO-mediated risk, or the gut-microbiome-derived TMAO from red meat carries additional atherogenic compounds not present in fish-derived TMAO. The fish paradox is a real unresolved tension in the current evidence base, not a minor caveat to be set aside.

What Actually Shapes Microbiome Composition: Beyond the Dietary Headlines

If gut microbiome composition determines how much TMAO a person produces from a given food, the clinically relevant question is: what determines microbiome composition, and can it be meaningfully changed? The answer is more nuanced than “take a probiotic,” and the honest evidence differs substantially from the marketing around gut health.

Dietary fiber is the most consistently supported intervention. The gut microbiome is nourished primarily by dietary fiber, specifically, the fermentable polysaccharides and resistant starches that reach the colon undigested. Sonnenburg and colleagues at Stanford have documented in multiple publications (including a 2022 Cell paper comparing high-fiber and high-fermented-food dietary interventions) that increased dietary fiber increases microbiome gene diversity, a marker of microbiome health, and shifts the community composition toward taxa associated with lower TMAO production. The intervention is not exotic: legumes, whole grains, vegetables, and fruit collectively provide the fermentable substrate that sustains the fiber-adapted microbiome configurations associated with lower TMAO output. A man who goes from essentially no legumes to two half-cup servings per week produces a measurable shift in microbiome community structure within two to three weeks. 3 / Early

Fermented foods have an acute immune effect, not just a microbiome stocking effect. The same Sonnenburg 2022 Cell study found that high-fermented-food intake (yogurt, kefir, kimchi, sauerkraut) reduced 19 inflammatory protein markers including IL-6 and IL-12p70 over 10 weeks, independently of microbiome gene diversity changes. The mechanism appears to involve direct interaction between live microorganisms in fermented foods and the mucosal immune system, rather than durable colonization. Fermented foods do not “repopulate” the gut microbiome in the way probiotic marketing implies, the transient microbes do not persist, but their immunomodulatory effects are measurable in short-term randomized trials.

Probiotic supplements have more limited evidence. Commercially available probiotic supplements contain specific strains (commonly Lactobacillus and Bifidobacterium species) that are selected for manufacturing feasibility and commercial viability rather than TMAO-reducing or microbiome-diversifying properties specifically. A systematic review by McFarland and colleagues found that probiotics are effective for specific indications (antibiotic-associated diarrhea, certain IBS subtypes) but do not reliably alter the established microbiome community composition in healthy adults. The evidence that commercial probiotic supplements reduce TMAO specifically is not available at a level that warrants clinical recommendation.

Intermittent fasting creates temporal windows of microbiome restructuring. Animal data and preliminary human data show that time-restricted eating alters circadian oscillations in microbiome composition and reduces the abundance of TMA-producing bacteria during fasting windows. Zarrinpar and colleagues documented these circadian microbiome shifts in a 2014 Cell paper. Whether this translates to a measurable reduction in circulating TMAO in free-living humans is not yet established in adequately powered trials.

Antibiotics reduce TMAO acutely, and then it returns. The Koeth 2013 study demonstrated that pretreatment with broad-spectrum antibiotics essentially eliminated TMAO production from L-carnitine. After the antibiotic course, as the microbiome reconstituted, TMAO production returned. This establishes that the microbiome is necessary for TMAO production but also that antibiotic-mediated TMAO suppression is temporary and comes with the significant cost of broad microbiome disruption, including loss of protective species. Antibiotics for TMAO reduction is not a clinical strategy.

The practical summary: dietary fiber and food diversity are the most evidence-backed approaches to sustaining a microbiome configuration associated with lower TMAO production. Fermented foods have immunomodulatory benefits that are real and measurable. Commercial probiotics are not specifically indicated for cardiovascular TMAO management. The gut health supplement space overstates what any single intervention can do to the established microbiome of an adult who has had their current dietary pattern for decades.

What to Do This Week

1. Shift toward plant protein at two meals per week. Replacing red meat with legumes, lentils, or tofu reduces TMAO production by removing the primary substrate (L-carnitine and choline from meat) and, over weeks, by gradually selecting for a gut microbiome that is less efficient at TMA production. This shift does not require eliminating red meat permanently to have a measurable effect.

2. Increase dietary fiber. Fiber feeds microbiome species that compete with TMAO-producing bacteria for the same gut niche. Soluble fiber from oats, beans, and fruit is associated with greater microbiome diversity in multiple cohort studies, and higher microbiome diversity correlates with lower TMAO production at any given level of red meat intake.

3. If red meat is a regular fixture in your diet, consider the frequency and portion size. The Koeth et al. carnitine study demonstrated that omnivores produce more TMAO per unit of carnitine than vegans, and this is a microbiome effect, not merely a dose effect. Habitual heavy red meat consumption actively selects for a microbiome configured toward higher TMAO-producing capacity. Moderate sustained reduction compounds: less substrate and a less capable microbiome.

4. Do not interpret fish intake through a TMAO-risk lens. Current evidence does not support restricting fish consumption because of TMAO concerns. Fish produces high circulating TMAO, but fish consumption is associated with reduced cardiovascular mortality in prospective studies. The net effect of fish on cardiovascular risk is protective by current evidence. Keep eating fish if you eat it.

5. Do not purchase TMAO testing from direct-to-consumer health companies. There are no established clinical reference ranges for cardiovascular risk stratification, the interlaboratory variability is substantial, and no management algorithm exists that is tied to the result. A number without an evidence-based response protocol is not useful clinical information.

The TMAO story is one of the more compelling developments in cardiovascular research of the past fifteen years, not because it changes clinical practice today, but because it offers a mechanism connecting gut bacteria, dietary pattern, and atherosclerosis in ways that explain observations from decades of dietary epidemiology. The translation from association to clinical tool is not complete. What is already supported by strong evidence, eating more plants, reducing red meat, maintaining microbiome diversity through fiber, arrives at the same conclusion from multiple independent directions. That convergence is worth acting on regardless of whether TMAO ever becomes a routinely ordered test.

The Signal Check is fifteen questions mapping the cardiovascular risk pattern across the physiological domains most commonly missed in standard screenings. It produces a specific starting point for your next clinical conversation.