Differential associations of ambient air pollutants with acute cardiovascular and cerebrovascular admissions: a 10-year time-series analysis

Background Short-term exposure to ambient air pollution is associated with acute cardiovascular events, but effects may differ by pollutant type, clinical presentation, and exposure–lag structure. Purpose To evaluate pollutant-specific and lag-dependent associations between air pollution and acute cardiovascular and cerebrovascular admissions. Methods We analysed 50,537 consecutive admissions to a UK cardiac–stroke receiving unit (2015–2025). Daily counts of diagnoses (including ST-elevation myocardial infarction [STEMI], non-ST-elevation myocardial infarction, unstable angina, ischaemic stroke, atrial fibrillation [AF], acute heart failure, and total admissions) were linked to daily ambient pollutants (PM₂.₅, PM₁₀, nitrogen dioxide [NO₂], nitric oxide [NO], and ozone) and meteorology. Associations were assessed using single-pollutant quasi-Poisson time-series models adjusted for seasonality, long-term trends, and day-of-week effects. Effect estimates are presented per 10 µg/m³ increase. Distributed lag non-linear models (DLNM) for NO₂ and PM₂.₅ were fitted over 0–7 days. Results Associations differed by pollutant class and clinical phenotype. Traffic-related gaseous pollutants demonstrated consistent harmful effects on acute thrombotic events: NO₂ was associated with increased STEMI admissions (RR 1.09, 95% CI 1.01–1.18), and NO with both STEMI (RR 1.16, 1.02–1.32) and unstable angina (RR 1.15, 1.01–1.31). Similar directional increases were observed for ischaemic stroke, supporting a shared thrombo-inflammatory mechanism across arterial vascular beds. In contrast, associations for non-ST-elevation myocardial infarction and atrial fibrillation were smaller and less consistent, suggesting differential susceptibility of plaque rupture versus demand-mediated or arrhythmic presentations. Particulate pollutants showed divergent patterns. PM₂.₅ was associated with reduced admissions for acute heart failure (RR 0.90, 0.82–0.99) and total admissions (RR 0.92, 0.88–0.97), while ozone showed no significant associations. DLNM clarified these heterogeneous findings. Increased risk for thrombotic events (STEMI, unstable angina, and ischaemic stroke) was concentrated at short lags (0–2 days), consistent with acute triggering. In contrast, inverse associations for acute heart failure were delayed (lags 2–4 days), consistent with short-term displacement or behavioural avoidance rather than a protective effect. NSTEMI and AF showed attenuated or less temporally structured lag responses. Conclusion Ambient air pollution shows heterogeneous, diagnosis-specific effects with distinct lag structures. Traffic-related pollutants are associated with immediate increases in thrombotic events, while delayed inverse associations for heart failure likely reflect displacement or behavioural effects. Incorporating lag structure is essential to avoid misinterpretation of same-day associations.