Long-term PM2.5 exposure and incident heart failure in the UK Biobank: relative and absolute risk estimates

Abstract

Background To investigate the association between long-term air pollution exposure and incident heart failure in the UK Biobank and further explore potential effect modification by cardiometabolic vulnerability. Methods We conducted a prospective cohort study among 149,410 UK Biobank participants free of heart failure and ischemic stroke at baseline. Annual mean concentrations of fine particulate matter (PM2.5), nitrogen dioxide (NO2), and particulate matter ≤10 μm (PM10) were assigned using land-use regression models. Incident heart failure and ischemic stroke were ascertained through linkage to hospital and death records. Cox proportional hazards models were used to estimate hazard ratios (HRs) per interquartile range (IQR) increase in pollutants with sequential adjustment for demographic, socioeconomic, lifestyle, and cardiometabolic factors. Effect modification by diabetes, chronic kidney disease, obesity, multimorbidity, and C-reactive protein was assessed using interaction terms. Results During a median follow-up of 17.6 years (interquartile range 14.4–20.3), 4,964 incident heart failure events and 3,795 ischemic stroke events occurred. PM2.5 was associated with a higher risk of incident heart failure after full adjustment (HR per IQR 1.045, 95% CI 1.005–1.086; p = 0.027). Participants in the highest quartile of PM2.5 exposure also had a higher risk of heart failure than those in the lowest quartile (HR 1.133, 95% CI 1.037–1.238; P-trend = 0.005). Dose–response analyses suggested an approximately linear relationship between PM2.5 and heart failure risk, with no strong evidence of non-linearity. No statistically significant interaction was observed between PM2.5 and cardiometabolic or inflammatory vulnerability factors (all P for interaction >0.10). In contrast, no significant associations were observed between air pollution metrics and ischemic stroke. Standardized absolute risks of heart failure were low, with a 20-year absolute risk difference of approximately 0.2 per 100 persons between the highest and lowest PM2.5 exposure groups. Conclusion In this large population-based cohort, long-term exposure to PM2.5 was associated with a modestly increased risk of incident heart failure. We found no strong evidence that this association differed across major cardiometabolic or inflammatory vulnerability groups. Although the absolute risk difference was small at the individual level, these findings add to the evidence linking long-term PM2.5 exposure to incident heart failure.