Where You Store Body Fat Determines How Fast Your Heart Ages
A 21,000-person imaging study reveals visceral fat, liver fat, and muscle fat infiltration accelerate cardiovascular ageing differently by sex.
Summary
Researchers analyzed 21,241 UK Biobank participants using machine learning to predict cardiovascular age from 126 cardiac MRI traits, then linked those predictions to whole-body fat imaging. Visceral fat, liver fat fraction, and muscle fat infiltration accelerated cardiovascular ageing in both sexes. Crucially, abdominal subcutaneous fat and android (upper-body) fat mass raised cardiovascular age only in males. Gynoid fat distribution showed a genetically predicted protective association in women. Mendelian randomization supported causal relationships. The findings underscore that where fat is stored—not just how much—shapes heart ageing trajectories differently in men and women, pointing to fat distribution as a tractable intervention target for extending cardiovascular healthspan.
Detailed Summary
Obesity's impact on the heart is well-established, but total body weight or BMI misses a critical dimension: where fat accumulates. Sex hormones shape fat distribution—men tend toward visceral and upper-body depots, women toward lower-body (gynoid) depots—yet how these sex-specific patterns influence the pace of cardiovascular ageing has been largely unexplored. This study addresses that gap with one of the largest cardiac imaging datasets assembled for this purpose.
The research team analyzed 21,241 participants from the UK Biobank, a large prospective cohort of community-dwelling adults aged 40–69. Using a pre-trained CatBoost machine learning model, cardiovascular age was predicted from 126 image-derived phenotypes capturing vascular distensibility, cardiac chamber volumes, myocardial strain rates, and native T1 mapping (a marker of diffuse myocardial fibrosis). A cardiovascular 'age-delta'—the gap between predicted biological age and chronological age—served as the primary outcome. Fat phenotypes were extracted from whole-body Dixon MRI and DXA scans, covering visceral adipose tissue (VAT), abdominal subcutaneous adipose tissue (ASAT), muscle adipose tissue infiltration (MATI), liver proton density fat fraction (PDFF), android fat mass, gynoid fat mass, and total trunk and whole-body fat.
Multivariable linear regression stratified by sex revealed that VAT (β = 0.656), liver fat fraction (β = 1.066), and MATI (β = 0.183) were the strongest predictors of an older cardiovascular age-delta across both sexes. Strikingly, ASAT (β = 0.432) and android fat mass (β = 0.983) were significantly associated with accelerated ageing only in males, not females. Two-sample Mendelian randomization provided genetic evidence that these associations are likely causal rather than merely correlational. Genetically predicted gynoid fat distribution was associated with a decreased (more youthful) cardiovascular age-delta, supporting the hypothesis that lower-body fat depots in women may be hormonally protected and metabolically distinct.
The study also examined the role of oestradiol, noting that pre-menopausal women appear to benefit from its protective cardiovascular effects, which may partly explain why sex-specific fat patterns translate differently into ageing outcomes. After menopause, this protection diminishes, and female fat redistribution toward visceral depots may contribute to convergence of cardiovascular risk between sexes.
These results reframe obesity as a heterogeneous condition whose cardiovascular consequences depend heavily on fat depot identity and hormonal context. Therapeutic strategies—whether pharmacological (GLP-1 agonists, visceral fat-targeting agents) or lifestyle-based—may need to be sex-stratified and depot-specific to maximize cardiovascular benefit and extend healthy lifespan.
Key Findings
- Visceral fat, liver fat, and muscle fat infiltration accelerated cardiovascular ageing in both men and women.
- Abdominal subcutaneous and android fat mass increased cardiovascular age-delta in males only, not females.
- Genetically predicted gynoid fat distribution was associated with a younger cardiovascular biological age.
- Mendelian randomization supported causal links between specific fat depots and cardiovascular ageing.
- The pre-trained ML model used 126 cardiac MRI traits covering structure, motion, and myocardial fibrosis.
Methodology
Cross-sectional analysis of 21,241 UK Biobank participants using a pre-trained CatBoost gradient-boosted model to derive cardiovascular age from 126 cardiac MRI phenotypes. Fat phenotypes were extracted from whole-body Dixon MRI and DXA. Causal inference was assessed via two-sample Mendelian randomization.
Study Limitations
The UK Biobank is a volunteer cohort with known healthy-participant bias, limiting generalizability to broader or more diverse populations. Cross-sectional design cannot fully capture longitudinal changes in fat distribution and cardiovascular ageing. Mendelian randomization assumptions may be violated by pleiotropy.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.