We critically review potential involvement of trimethylamine N-oxide (TMAO) as a link between diet, the gut microbiota and CVD. Generated primarily from dietary choline and carnitine by gut bacteria and hepatic flavin-containing mono-oxygenase (FMO) activity, TMAO could promote cardiometabolic disease when chronically elevated. However, control of circulating TMAO is poorly understood, and diet, age, body mass, sex hormones, renal clearance, FMO3 expression and genetic background may explain as little as 25 % of TMAO variance. The basis of elevations with obesity, diabetes, atherosclerosis or CHD is similarly ill-defined, although gut microbiota profiles/remodelling appear critical. Elevated TMAO could promote CVD via inflammation, oxidative stress, scavenger receptor up-regulation, reverse cholesterol transport (RCT) inhibition, and cardiovascular dysfunction. However, concentrations influencing inflammation, scavenger receptors and RCT (≥100 µm) are only achieved in advanced heart failure or chronic kidney disease (CKD), and greatly exceed pathogenicity of <1–5 µm levels implied in some TMAO–CVD associations. There is also evidence that CVD risk is insensitive to TMAO variance beyond these levels in omnivores and vegetarians, and that major TMAO sources are cardioprotective. Assessing available evidence suggests that modest elevations in TMAO (≤10 µm) are a non-pathogenic consequence of diverse risk factors (ageing, obesity, dyslipidaemia, insulin resistance/diabetes, renal dysfunction), indirectly reflecting CVD risk without participating mechanistically. Nonetheless, TMAO may surpass a pathogenic threshold as a consequence of CVD/CKD, secondarily promoting disease progression. TMAO might thus reflect early CVD risk while providing a prognostic biomarker or secondary target in established disease, although mechanistic contributions to CVD await confirmation.