Eggs and TMAO: What Controlled Trials Actually Show

The egg–TMAO story has a missing step

The popular warning sounds tidy: eggs contain choline; gut microbes can turn choline into trimethylamine; the liver converts that into trimethylamine N-oxide, or TMAO; TMAO links to atherosclerosis; therefore eggs endanger the heart. Each arrow has biological plausibility. The full chain, however, needs to survive a more basic test: do people who eat more eggs actually show higher circulating TMAO under controlled conditions?

The best current answer is no, on average. A 2025 systematic review and meta-analysis of randomized controlled trials found that egg consumption did not significantly change circulating TMAO in adults: weighted mean difference, −0.08; p = 0.692; with low heterogeneity across trials. Aryaeian et al. registered the review in PROSPERO and searched through April 2025. This is a biomarker result, not a trial of heart attacks or strokes, but it directly tests the first dietary claim in the fear narrative.

Controlled trials do not show that eating eggs raises circulating TMAO in the general adult population.

The point estimate was slightly below zero, and the result was statistically non-significant. Low heterogeneity matters here: the pooled null result did not simply average large positive and negative effects that happened to cancel each other out. It does not prove that eggs can never raise TMAO in any person. It does mean that a general rule telling adults to avoid eggs for TMAO reduction runs ahead of the controlled evidence.

Choline intake does not dictate TMAO the way the slogan implies

Choline is a required nutrient and an input into several metabolic pathways. Gut microbes can convert some choline into trimethylamine, but that pathway competes with the body's own use of choline. In the egg meta-analysis, meta-regression found an inverse dose relationship between egg intake and circulating choline and betaine: higher egg intake tracked with lower, rather than higher, levels of those metabolites.

The authors discuss a plausible explanation: at higher choline exposure, the liver may route more choline into phosphatidylcholine production through the CDP-choline, or Kennedy, pathway. Betaine may also be consumed during homocysteine remethylation through betaine-homocysteine methyltransferase. These are mechanistic interpretations, not direct proof that either pathway caused the meta-regression result.

The form of choline also matters. In a randomized crossover trial in adult volunteers, Böckmann et al. found that different choline supplements produced distinct plasma metabolite profiles. Eggs supply choline largely in a phosphatidylcholine-rich food matrix, which is not metabolically interchangeable with choline chloride or other isolated precursors. A result from a supplement challenge therefore cannot be pasted onto a breakfast plate without some very large assumptions.

That does not make eggs metabolically invisible. It means precursor content alone does not predict a person's TMAO concentration.

Why two people can eat the same eggs and have different TMAO

A 2024 population-level analysis by Andreu-Sánchez et al. found that host genetics, gut microbiome composition, and kidney function dominate interindividual variation in TMAO and its precursors, while dietary source contributes less. FMO3 variants can affect how the liver oxidizes trimethylamine, and microbial communities differ in their capacity to produce trimethylamine in the first place.

Kidney function complicates the picture further because kidneys clear TMAO. In a 2025 cross-lagged cohort study, Peng et al. found a bidirectional relationship between TMAO and estimated glomerular filtration rate: lower kidney function predicted higher TMAO, and higher TMAO predicted later decline in estimated kidney function. A cross-lagged design improves the temporal analysis, but it remains cohort evidence rather than a randomized test of TMAO lowering.

This matters when reading studies that link TMAO with disease. High TMAO can reflect dietary precursor exposure, microbial metabolism, reduced renal clearance, metabolic illness, or some combination. A single blood level cannot assign blame to an omelet.

TMAO still has concerning associations

The egg-specific result should not become a claim that TMAO is harmless. In the multiethnic primary-prevention Dallas Heart Study, Talmor-Barkan et al. reported observational associations between TMAO-related metabolites and adverse cardiovascular events. Observational associations establish real-world correlations, but they cannot fully separate TMAO from renal function, overall dietary patterns, and cardiometabolic disease.

Prospective cancer data also warrant attention. Within the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial cohort, Byrd et al. measured TMAO and precursor metabolites in baseline serum from 761 people who later developed colorectal cancer and 761 individually matched controls. Higher serum TMAO and some precursor metabolites were associated with higher colorectal cancer risk. Because the blood samples preceded diagnosis, reverse causation is less persuasive than in a cross-sectional study. Residual confounding by diet, dysbiosis, and metabolic health remains possible.

Pregnancy adds another setting that the egg trials do not address. A 2025 case-control study included 201 women with gestational diabetes and 201 matched controls; Yao et al. also combined epidemiological evidence in a meta-analysis and used Mendelian randomization to examine causality. Elevated gut-derived metabolites, including TMAO, were associated with gestational diabetes risk, and the Mendelian-randomization component supported a causal association for TMAO metabolites. This evidence concerns maternal metabolites and gestational diabetes, not egg consumption. It should not be converted into an egg-specific verdict.

Some food and fiber findings do not generalize to eggs

Animal foods differ in their relevant precursors and food matrices. A randomized, non-inferiority feeding trial by Dhakal et al. found that lean pork altered TMAO levels and gut microbiota when compared with Dietary Guidelines for Americans–based comparators. Pork contains L-carnitine, which may behave differently from egg phosphatidylcholine. The trial did not include an egg arm, so it cannot quantify a pork-versus-egg difference, but it is a useful warning against treating every animal food as one exposure.

Nor is changing the microbiome an established TMAO fix. In a double-blind randomized trial in hemodialysis patients, 16 g per day of resistant starch type 2 for four weeks did not significantly lower plasma TMAO. Kemp et al. tested a population with severe kidney disease and high TMAO burden. A randomized crossover trial of inulin-type fructans in peritoneal dialysis patients likewise failed to significantly reduce plasma TMAO. Xiong et al. These results do not prove that fiber has no health value; they show that these specific prebiotic interventions did not lower this biomarker in dialysis populations.

Who should be more cautious?

In the egg meta-analysis, adults older than 45 years and people with metabolic syndrome showed significant increases in circulating choline. The pooled downstream TMAO result remained non-significant. The published information available for this review does not provide the subgroup effect estimates here, so the size and clinical importance of those choline changes remain uncertain.

Those groups deserve more individualized attention because insulin resistance and reduced kidney function can alter choline handling and TMAO clearance. People with known chronic kidney disease, dialysis treatment, metabolic syndrome, or pregnancy-related metabolic concerns should discuss dietary changes with a clinician who can account for their full diet, kidney status, medications, and cardiometabolic risk. An isolated TMAO result is not a diagnosis, and avoiding eggs is not an established treatment.

The practical verdict

For most adults, randomized trial evidence does not support cutting eggs solely to reduce circulating TMAO. The central pooled estimate found no significant effect, despite the mechanistically plausible choline-to-TMAO pathway.

The larger TMAO question remains unsettled. Elevated TMAO associates with cardiovascular events and colorectal cancer in observational research, and maternal-metabolite research raises concern about gestational diabetes. Yet no Mendelian-randomization study has tested genetically predicted egg intake or dietary choline against hard cardiovascular outcomes such as myocardial infarction, stroke, or cardiovascular death. No large prospective cohort in the available evidence directly links egg intake itself to incident cardiovascular disease.

That gap leaves a narrow, defensible conclusion: treat eggs and TMAO as a question of individual metabolism and kidney function, not a universal food scare. The controlled trials have tested the claimed egg-to-TMAO link. So far, it has not shown up.