Impacts of long-term ambient particulate matter and gaseous pollutants on circulating biomarkers of inflammation in male and female health professionals

Background: Systemic inflammation may serve as a biological mechanism linking air pollution to poor health but supporting evidence from studies of long-term pollutant exposure and inflammatory cytokines is inconsistent. Objective: We studied associations between multiple particulate matter (PM) and gaseous air pollutants and pro- and anti-inflammatory cytokines within two nationwide cohorts of men and women. Methods: Data were obtained from 16,151 women in the Nurses' Health Study and 7,930 men in the Health Professionals' Follow-up Study with at least one measure of circulating adiponectin, C-Reactive Protein (CRP), Interleukin-6 (IL-6) or soluble tumor necrosis-factor receptor-2 (sTNFR-2). Exposure to PM with aerodynamic diameter <= 2.5, 2.5-10, and <= 10 mu m (PM2.5, PM2.5-(10), PM10) and nitrogen dioxide (NO2) was estimated using spatio-temporal models and were linked to participants' addresses at the time of blood draw. Averages of the 1-, 3-, and 12-months prior to blood draw were examined. Associations between each biomarker and pollutant were estimated from linear regression models adjusted for individual and contextual covariates. Results: In adjusted models, we observed a 2.72% (95% CI: 0.43%, 5.95%), 3.11% ( 0.12%, 6.45%), and 3.67% (0.19%, 7.26%) increase in CRP associated with a 10 mu g/m(3) increase in 1-, 3-, and 12- month averaged NO2 in women. Among men, there was a statistically significant 5.96% (95% CI: 0.07%, 12.20%), 6.99% (95% CI: 0.29%, 14.15%), and 8.33% (95% CI: 0.35%, 16.94%) increase in CRP associated with a 10 mu g/m(3) increase in 1-, 3-, and 12-month averaged PM2.5-10, respectively. Increasing PM2.5-10 was associated with increasing IL-6 and sTNFR-2 among men over shorter exposure durations. There were no associations with exposures to PM2.5 or PM10, or with adiponectin. Findings were robust to sensitivity analyses restricting to disease-free controls and non-movers. Conclusions: Across multiple long-term pollutant exposures and inflammatory markers, associations were generally weak. Focusing on specific pollutant-inflammatory mechanisms may clarify pathways.