Particulate matter properties and health effects:: consistency of epidemiological and toxicological studies

Identifying the ambient particulate matter (PM) fractions or constituents, critically involved in eliciting adverse health effects, is crucial to the implementation of more cost-efficient abatement strategies to improve air quality. This review focuses on the importance of different particle properties for PM-induced effects, and whether there is consistency in the results from epidemiological and experimental studies. An evident problem for such comparisons is that epidemiological and experimental data on the effects of specific components of ambient PM are limited. Despite this, some conclusions can be drawn. With respect to the importance of the PM size-fractions experimental and epidemiological studies are somewhat conflicting, but there seems to be a certain consistency in that the coarse fraction (PM10-2.5) has an effect that should not be neglected. Better exposure characterization may improve the consistency between the results from experimental and epidemiological studies, in particular for ultrafine particles. Experimental data indicate that surface area is an important metric, but composition may play an even greater role in eliciting effects. The consistency between epidemiological and experimental findings for specific PM-components appears most convincing for metals, which seem to be important for the development of both pulmonary and cardiovascular disease. Metals may also be involved in PM-induced allergic sensitization, but the epidemiological evidence for this is scarce. Soluble organic compounds appear to be implicated in PM-induced allergy and cancer, but the data from epidemiological studies are insufficient for any conclusions. The present review suggests that there may be a need for improvements in research designs. In particular, there is a need for better exposure assessments in epidemiological investigations, whereas experimental data would benefit from an improved comparability of studies. Combined experimental and epidemiological investigations may also help answer some of the unresolved issues.