Association between air pollution and coronary artery calcification within six metropolitan areas in the USA (the Multi-Ethnic Study of Atherosclerosis and Air Pollution): a longitudinal cohort study

Abstract

Background: Long-term exposure to fine particulate matter less than 2.5 μm in diameter (PM2.5) and traffic-related air pollutant concentrations are associated with cardiovascular risk. The disease process underlying these associations remains uncertain. We aim to assess association between long-term exposure to ambient air pollution and progression of coronary artery calcium and common carotid artery intima-media thickness.

Methods: In this prospective 10-year cohort study, we repeatedly measured coronary artery calcium by CT in 6795 participants aged 45-84 years enrolled in the Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air) in six metropolitan areas in the USA. Repeated scans were done for nearly all participants between 2002 and 2005, for a subset of participants between 2005 and 2007, and for half of all participants between 2010 and 2012. Common carotid artery intima-media thickness was measured by ultrasound in all participants at baseline and in 2010-12 for 3459 participants. Residence-specific spatio-temporal pollution concentration models, incorporating community-specific measurements, agency monitoring data, and geographical predictors, estimated concentrations of PM2.5 and nitrogen oxides (NOX) between 1999 and 2012. The primary aim was to examine the association between both progression of coronary artery calcium and mean carotid artery intima-media thickness and long-term exposure to ambient air pollutant concentrations (PM2.5, NOX, and black carbon) between examinations and within the six metropolitan areas, adjusting for baseline age, sex, ethnicity, socioeconomic characteristics, cardiovascular risk factors, site, and CT scanner technology.

Findings: In this population, coronary calcium increased on average by 24 Agatston units per year (SD 58), and intima-media thickness by 12 μm per year (10), before adjusting for risk factors or air pollutant exposures. Participant-specific pollutant concentrations averaged over the years 2000-10 ranged from 9.2-22.6 μg PM2.5/m(3) and 7.2-139.2 parts per billion (ppb) NOX. For each 5 μg PM2.5/m(3) increase, coronary calcium progressed by 4.1 Agatston units per year (95% CI 1.4-6.8) and for each 40 ppb NOX coronary calcium progressed by 4.8 Agatston units per year (0.9-8.7). Pollutant exposures were not associated with intima-media thickness change. The estimate for the effect of a 5 μg/m(3) higher long-term exposure to PM2.5 in intima-media thickness was -0.9 μm per year (95% CI -3.0 to 1.3). For 40 ppb higher NOX, the estimate was 0.2 μm per year (-1.9 to 2.4).

Interpretation: Increased concentrations of PM2.5 and traffic-related air pollution within metropolitan areas, in ranges commonly encountered worldwide, are associated with progression in coronary calcification, consistent with acceleration of atherosclerosis. This study supports the case for global efforts of pollution reduction in prevention of cardiovascular diseases.

Funding: US Environmental Protection Agency and US National Institutes of Health.

Figure 1. Long-term outdoor air pollutant concentrations by area

Pollutant concentrations are averaged over the time between the first and last CT scan, rounded to the nearest full year. Boxes cover the 25–75^th percentile (IQR) with a centre line for the median concentration. Whiskers extend to the highest observation within 3 IQR of the box, with more extreme observations shown as points. Black carbon was measured by light absorption coeff cient, where 0·5 × 10⁻⁵ m⁻¹ was approximately equivalent to 0·5 μg/m³ of elemental carbon. PM_2·5=fine particulate matter less than 2·5 μm in diameter. NO_X=nitrogen oxides. NO₂=nitrogen dioxide. ppb=parts per billion.

Figure 2. Maps of predicted PM_2·5 and NO_X concentrations for 2005 in the MESA Air communities, by the city of the clinical site

Contrasts are smoothed in these figures, which might attenuate the visibility of fine-scale variation. Participant residential locations are shown by black dots, jittered to protect their privacy. PM_2·5=fine particulate matter less than 2·5 μm in diameter. NO_X=nitrogen oxides. ppb=parts per billion.

Figure 3. Long-term average air pollutant concentrations and coronary artery calcium progression

(A) The linear longitudinal association of fine particulate matter (PM_2·5), nitrogen oxides (NOx), nitrogen dioxide (NO₂), and black carbon, with coronary artery calcium progression (Agatston units per year), from linear mixed models adjusted for age, sex, ethnicity, city, income, employment outside the home, smoking status, second-hand smoke exposure, physical activity, adiposity, cholesterol, statin use, neighbourhood socioeconomic index, income, education and scanner type. (B) The concentration-response curve with 95% CIs for the overall change in and coronary artery calcium (progression rate associated with long-term average PM_2·5 concentrations). The curve is based on a mixed model that includes a thin plate regression spline with five degrees of freedom to more flexibly assess the potentially non-linear association. The relationship at the extremes is less certain and might rely on concentrations that are recorded only in one geographical region; the highest and lowest 5% of overall concentrations have been trimmed for visualisation. The histogram at the bottom of the right panel shows the relative overall distribution of long-term PM_2·5 concentrations in the cohort, with different colours representing each clinic location. *Black carbon as measured by light absorption coefficient, where 0·5 × 10⁻⁵ m⁻¹ is approximately equivalent to 0·5 μg/m³.

Figure 4. Association between pollutant concentrations and coronary artery calcium progression, by baseline participant characteristics

The shaded region indicates the 95% confidence interval for the overall association per 5 μg/m³ of PM_2·5.*Linear adjustments for age were replaced with categories. †Linear adjustments for adiposity were replaced with indicator for body-mass index >30 kg/m². ‡Linear adjustments for cholesterol were replaced with categories. §Not adjusted for pack-years or second-hand smoke.