Abstract:
Introduction: Long-term exposure to fine particulate matter air pollution (particulate matter less than 2.5 um in diameter, or PM2.5) has been associated with cardiovascular morbidity and mortality in several large epidemiologic studies, but the mechanisms remain unclear. One means by which PM2.5 may exert its negative cardiovascular effects is through changes in cardiac structure and function. We hypothesized that exposure to urban background PM2.5 and proximity to major roadways, a marker of finer-scale exposure to traffic-related air pollution, would be associated with increases in left-ventricular mass (LVM) and decreases in stroke volume (SV).
Methods: The residential addresses of 4,478 participants in without cardiovascular disease (CVD) in 6 large United States cities in the Multi-Ethnic Study of Atherosclerosis were geocoded. Urban background PM2.5 exposure was estimated by interpolating year 200 average PM2.5 monitor values to participant home address using ordinary kriging. Exposure to traffic-generated pollutants was estimated by measuring the distance from each participant's home to the nearest major roadway. LVM and SV were measured by cardiac magnetic resonance imaging (MRI). We performed multiple linear regression to determine the adjusted cross-sectional relationship between measures of proximity to major roadways, predicted residential fine PM concentration in 2000, and measures of cardiac structure and function (LVM and SV), adjusted for multiple potential confounding factors. We also examined for the presence of effect modification by demographic covariates, risk factors, and disease status.
Results: Among MESA participants with at least 5 years residence at their home address and compared to participants living >150 meters from a major roadway, living within 50 meters of a major roadway was associated with an adjusted 3.37 gram, (95% Cl: 0.33, 6.41) increase in LVM. Living between 50 and 150 meters was associated with no significant difference in LVM compared to participants living >150 meters from a major roadway. Larger adjusted point estimates of the 50 meter association were observed for smokers, persons with diabetes, males, African Americans, and Hispanics. SV was not associated with proximity to major roadways. City of residence was noted to be a significant confounding factor in all exposure-outcome relationships. Because adjustment for city was required, there was insufficient variability in urban background kriged PM2.5 within cities to detect effects of kriged PM2.5 exposure.
Conclusions: Close proximity to major roadways is associated with increased LVM, in a manner consistent with an effect of traffic-related air pollution. This association may be stronger in certain sensitive subpopulations.