Traffic-borne pollution as a risk factor for acute vascular events: a population study in Milan

Abstract

Abstract Background Traffic-borne noise and air pollution have both been associated with cardiovascular and cerebrovascular diseases, albeit with inconsistent findings and issues of collinearity/mutual confounding. The present study aims at evaluating the role of long-term exposure to traffic-borne pollution as a risk factor for acute vascular events in a highly urbanized setting. Methods This is a population-based retrospective dynamic cohort study including all residents aged >35 years in the municipality of Milan over the years 2011-2018 (N = 1087110). A noise predictive model and a NO2 land-use regression model were used to assign mean values of traffic noise at the day-evening-night level (Lden, dB) and NO2 concentration (µg/m3) to the residential address of each subject. Cox proportional hazards models were performed to assess the incidence of acute vascular events, with adjustment for potential confounders (age, sex, nationality, a socio-economic deprivation index) and sub-analyses for different outcomes (acute myocardial infarction, ischemic stroke, hemorrhagic stroke). Results A total of 27282 subjects (2.5%) had an acute vascular event. Models using NO2 yielded inconsistent results. When using Lden as a proxy of traffic intensity, there was a positive trend in risk with increasing levels of exposure, with an optimal cut-off for dichotomization set at 70 dB (HR 1.025, 95% C.I. 1.000-1.050). The association was observed specifically for ischemic stroke (HR 1.043, 95% C.I. 1.003-1.085) and hemorrhagic stroke (HR 1.036, 95% C.I. 0.969-1.107). When stratifying by age group and sex, a remarkable effect was found for hemorrhagic stroke in men aged <60 (HR 1.439, 95% C.I. 1.156-1.792). Conclusions Living close to high-traffic roads was found to exert a small but tangible effect on the risk of stroke. The varying effects observed for specific outcomes and in different age and sex groups are likely due to different pathogenetic mechanisms at play, which warrant further investigation. Key messages Residential proximity to roads with high traffic intensity (mean traffic noise level over 70 dB) is a risk factor for stroke, especially for hemorrhagic stroke in middle-aged men. Further interventions aimed at reducing traffic intensity in highly urbanized cities may be justified in order to reduce morbidity and mortality from stroke.