Serum trimethylamine N-oxide and choline, and risk of cardiovascular events and mortality in 14 countries: a case-cohort analysis from the Prospective Urban and Rural Epidemiology (PURE) study

Background Trimethylamine N-oxide (TMAO) is a gut microbiota–derived metabolite produced from dietary precursors such as betaine, carnitine, and choline, reportedly found in red meat. Elevated TMAO levels have been proposed to increase cardiovascular disease (CVD) risk through mechanisms involving lipid metabolism, inflammation, and thrombosis(1). Both TMAO and red meat intake have been controversially linked to CVD, but evidence remains inconsistent and largely from high-income, European-ancestry populations. We investigated associations of TMAO and its precursors with incident CVD and mortality in the Prospective Urban and Rural Epidemiology (PURE) study, a large, multiethnic international cohort(2). Purpose To identify demographic, clinical, and dietary predictors of serum TMAO, and assess associations of TMAO and its precursors (choline, carnitine, betaine) with incident major cardiovascular events (MACE), mortality, and type 2 diabetes (T2D). Methods A case-cohort analysis was conducted among 9,287 PURE participants from 14 countries spanning high-, middle-, and low-income settings, including all major ethnicities. Serum metabolites were quantified using multisegment injection–capillary electrophoresis–mass spectrometry, and 3,639 participants had paired creatinine-normalised urinary TMAO(3). Dietary intake was assessed using validated, country-specific food frequency questionnaires(4). Primary outcome was incident MACE (composite of cardiovascular death, myocardial infarction, stroke, and heart failure), secondary outcomes were incident T2D and death. Weighted Cox frailty models with random intercepts estimated hazard ratios (HRs) adjusted for demographic, clinical, and dietary covariates. Results Over a median of 15 years (interquartile range:12.3–15.6), 2,198 MACE and 2,281 deaths occurred as of April 30th 2025. Higher serum TMAO was associated with older age, male sex, baseline T2D, coronary heart disease, higher body mass index, and egg intake, and inversely with intake of carbohydrates, processed meat, and white meat, and rural residence (all q<0.05; Figure 1). Serum TMAO was not independently associated with MACE (HR per 1-standard deviation increase: 1.04, 95% Confidence Interval:0.97–1.11), all-cause mortality (HR 1.01, 0.96–1.07), or T2D (HR 0.96, 0.91–1.01). Subgroup analyses of urinary TMAO showed similar null findings. Among precursors, only serum choline was positively associated with MACE (HR 1.08, 1.01–1.15; p=0.03; Table 1). Conclusion In this large, multiethnic cohort, circulating and urinary TMAO were not independently associated with MACE, mortality, or T2D, while serum choline showed a modest positive association with MACE. TMAO was weakly correlated with red meat intake, likely reflecting metabolic or microbial processes rather than diet. Our findings do not support TMAO as a predictive biomarker of CVD risk. Future work should examine mechanistic pathways linking choline metabolism to CVD as potential targets.Figure 1.Multivariable RankingsFor image description, please refer to the figure legend and surrounding text. Table 1.Choline Hazard RatiosFor image description, please refer to the figure legend and surrounding text.