In 4,365 post-myocardial infarction patients followed for a median 14 years in the Dutch Alpha Omega Cohort, each 10-percentage-point increase in ultra-processed food (UPF) intake — measured as grams per day via a 203-item food-frequency questionnaire — was associated with a 21% higher risk of stroke mortality (HR 1.21; 95% CI 1.02–1.42). Notably, no significant association emerged for overall CVD or coronary heart disease mortality. High UPF consumers (top quartile, ~29% of diet by weight) also ate fewer fruits and vegetables and scored lower on a validated CVD-adapted diet quality index, suggesting UPFs partly displace cardioprotective foods. The stroke mortality signal was amplified in older adults, women, those with higher BMI, lower physical activity, and lower socioeconomic status.
This finding adds meaningful granularity to the UPF-cardiovascular disease literature, which has primarily focused on incidence rather than cause-specific mortality in already-compromised patients. The divergence between stroke and CHD mortality associations is biologically plausible — stroke risk is particularly sensitive to blood pressure, endothelial inflammation, and sodium load, all of which UPFs disproportionately deliver. However, as an observational prospective cohort, residual confounding cannot be excluded; dietary data were collected at a single baseline point, and the post-MI population limits generalizability. The relatively modest mean UPF contribution (18% by weight) compared to other populations also suggests the effect could be larger in societies with higher UPF dependence. For post-MI patients and their clinicians, this reinforces the case for aggressively reducing UPF intake as a modifiable stroke-prevention lever.