Air pollution: what matters most?: Physical, chemical and oxidative properties of air pollution components related to toxic effects

Abstract

Numerous studies have been published on the adverse health effects associated with both short- and long-term exposure to air pollution. Air pollution is a heterogeneous, complex mixture of gases, liquids, and particulate matter (PM). Up to now, PM mass concentration has been the metric of choice to quantify (particulate) air pollution exposure and has proven useful in demonstrating associations with cardiorespiratory morbidity and mortality outcomes. In recent years, it has become evident that PM mass per se may not be the causal factor that explains air pollution-induced effects, and therefore research has increasingly focused on identifying the influences of specific air pollution characteristics or sources. The research presented in this thesis was designed to establish which individual characteristics of the air pollution mixture (e.g. particle size fractions, chemical composition, oxidant properties etc.) are most closely related with adverse health effects seen after short-term air pollution exposure. It was hypothesized that air pollution-induced effects would be most strongly associated with particles oxidative potential (OP), since OP may be a more biologically meaningful metric for predicting human health effects than particle mass alone. Eight sampling sites across the Netherlands, representing different air pollution emission sources, were chosen to create high contrast and low correlations among different air pollutants. Measurement sites included an underground train station, three different road traffic sites, an animal farm, a sea harbor, a site located in the vicinity of steelworks, and an urban background site. Size-segregated PM was collected at each location for a detailed chemical characterization and subsequent in vitro and in vivo animal experiments. Furthermore, healthy volunteers were repeatedly exposed to air pollution at five of the eight sites. The research presented in this thesis focused on markers of pro-inflammatory (respiratory) effects (e.g. cytokines in the airways, differential cell counts in blood). Associations with markers related to lung function and cardiovascular effects were also investigated, but these results have been described elsewhere (thesis of M. Strak, 2012, Utrecht University). Despite differences in methodological design there was considerable coherence between the results of the different studies. PM exposure induced pro-inflammatory effects both in vitro and in vivo in animals and humans in a similar time response profile. In the human study, there were no consistent associations for either PM mass or particles’ OP with any of the physiological markers measured. Instead, there were significant associations with nitrogen dioxide, organic carbon, endotoxin, particle number concentration, nitrate and sulfate. These associations were robust and insensitive to adjustment for other pollutants. In summary, contrary to the hypothesis, particles’ OP did not predict acute effects due to short-term exposures better than PM mass or other air pollution characteristics. The results showed that there is no single exposure metric that is related to all physiological markers measured in this study, and thereby confirms that air quality legislations would benefit from a multi-component approach that combines gaseous pollutants and several PM characteristics instead of relying on individual gaseous pollutants and PM mass concentration only.