Nine Circulating Gut Metabolites Predict Coronary Heart Disease Across Four Cohorts

Nine molecules produced by gut bacteria and measurable in a blood draw can predict coronary heart disease risk years before symptoms appear, according to a prospective multi-stage metabolomics study published in PLOS Medicine. The research, involving Black, White, and Asian adults across five cohorts in the United States and China, is one of the most rigorous attempts yet to move gut-microbiome science from the lab bench into the cardiovascular clinic.

Nine gut microbial metabolites tied to incident coronary heart disease

The team, led by Yanhong Zheng and colleagues at Vanderbilt University Medical Center, began with an untargeted plasma screen of 896 incident coronary heart disease (CHD) cases matched to 896 controls from the Southern Community Cohort Study and the Shanghai Women's and Men's Health Studies. That discovery stage flagged 73 microbiota-related metabolites at a false discovery rate below 0.10. Of the 61 that could be taken into a computational validation step using data from the Atherosclerosis Risk in Communities study (N=3,539; 663 cases) and the Multi-Ethnic Study of Atherosclerosis (N=3,860; 446 cases), 24 held up in the same direction. A final targeted quantitative assay of 864 new cases and 864 matched controls whittled the list to nine metabolites with consistent, validated associations across all three stages.

Those nine span three broad biochemical families: amino acid derivatives, lipid-related compounds, and xenobiotic metabolites. Imidazole-propionate, a histidine-derived compound generated by specific gut microbes, showed the most consistent signal across all stages: an odds ratio of 1.35 in discovery, a hazard ratio of 1.14 in the in-silico validation cohorts, and an odds ratio of 1.26 in the targeted validation. The mean time between blood draw and CHD diagnosis ranged from 5.6 to 15.0 years across the cohorts, which means these metabolite signatures were circulating well before clinical disease appeared.

Why a ketone body showing up as a CHD risk factor is surprising

One of the more eyebrow-raising findings involves 3-hydroxybutyrate, the primary ketone body that has attracted attention in discussions of ketogenic diets and metabolic health. In this study, higher circulating 3-hydroxybutyrate was positively associated with CHD risk, and the association was particularly pronounced in non-obese individuals. The result does not necessarily condemn ketosis in healthy people; the authors note that circulating ketone levels in fasting or metabolically stressed states may reflect impaired glucose metabolism rather than deliberate dietary manipulation. Still, the finding adds a note of caution to the conversation around ketone elevation as an unambiguous cardiac benefit.

Also on the list is 3-hydroxy-2-ethylpropionate, a marker of disrupted isoleucine catabolism. Branched-chain amino acid metabolism has been implicated in insulin resistance and cardiovascular risk before, and this metabolite's appearance here adds another data point to that picture. It is worth noting that research on TMAO as a cardiovascular risk factor established that gut-derived small molecules can influence atherosclerosis through multiple pathways, including inflammation and lipid oxidation, so the metabolite diversity found here is scientifically plausible even if the exact mechanisms remain to be worked out.

A multi-cohort design that resists the usual single-study criticisms

The study's credibility rests heavily on its staging architecture. Most metabolomics studies in cardiovascular disease have been limited by small samples, single cohorts, or reliance on prevalent rather than incident cases. This team separated discovery from validation and ran a confirmatory targeted assay with freshly selected cases, which is closer to the design standards used in large genetic studies. Metabolomics in cardiovascular research has long been recognized as a powerful tool for identifying disease mechanisms and therapeutic targets in large epidemiological settings, but the field has been hampered by reproducibility problems when findings are not validated across independent populations.

By drawing on cohorts that include substantial numbers of Black and Asian participants, the investigators also addressed a persistent criticism of biomarker research: that it overwhelmingly reflects European-ancestry populations. Associations were adjusted for age, sex, race, education, income, smoking, alcohol intake, physical activity, diet quality, BMI, diabetes, hypertension, and dyslipidemia, which limits the risk that the metabolite signals are simply proxies for known risk factors.

"Our findings underscore the importance of gut microbiota and microbial metabolism in cardiovascular disease development and highlight promising metabolites that may serve as novel biomarkers or therapeutic targets for future mechanistic and interventional studies." — Zheng et al.

Coronary heart disease remains the leading cause of death worldwide. The increasing global CVD burden documented by The Lancet reached 437 million disability-adjusted life years in 2023, a 1.4-fold increase since 1990, driven by aging populations and modifiable risk factors. Novel biomarkers that operate through pathways not captured by cholesterol panels or blood pressure readings could meaningfully extend the toolkit available for early risk stratification.

Causality remains unproven, and some associations varied by subgroup

The authors are careful to flag what their design cannot establish. Observational data, even from five well-characterized cohorts with long follow-up, can identify associations but cannot prove that any of these metabolites drives coronary heart disease rather than simply correlating with it. As a companion perspective in PLOS Medicine notes, gut metabolite causality in CHD remains a persistent gap between association and established mechanism, demanding stringent epidemiological and experimental scrutiny before clinical translation.

Stratified analyses also revealed some heterogeneity. Certain metabolite-CHD associations differed when the data were split by race or age group, suggesting that microbial metabolism may be modulated by factors that vary across populations, including diet, antibiotic exposure history, and host genetics. The study was not powered to definitively resolve those subgroup differences.

What comes next for gut microbiome therapy in heart disease

If even a subset of the nine identified metabolites proves to have a causal role, the therapeutic implications are substantial. Imidazole-propionate, for instance, is produced by specific bacterial taxa responding to dietary histidine, which means its circulating level is potentially modifiable through diet or targeted microbiome interventions. Research on modulating gut microbiome for CVD prevention through Mediterranean diets, high-fiber intake, probiotics, and pharmacological approaches is already underway, though none of those trials were designed with imidazole-propionate or 3-hydroxybutyrate as endpoints.

The immediate next steps the authors call for include mechanistic studies in model systems to test whether these metabolites actively promote atherosclerosis or cardiac dysfunction, biomarker validation in clinical settings, and eventually intervention trials that can manipulate specific metabolites and measure CHD outcomes. Given the 5-to-15-year lead time between metabolite measurement and clinical disease in this dataset, there is a plausible window for early intervention if the causal story holds up. Whether blood-based gut metabolite panels ever reach routine cardiology practice will depend on those trials, but this study sets a methodological standard that future work in the field will have to match.