Source category-specific PM2.5 and urinary levels of Clara cell protein CC16. The ULTRA study

Introduction: We have previously reported that outdoor levels of fine particles (PM2.5, diameter <2.5 mu m) are associated with urinary CC16, a marker for lung damage, in Helsinki, Finland, but not in the other two ULTRA cities (Amsterdam, The Netherlands, and Erfurt, Germany). We here evaluated whether PM2.5 from specific source categories would be more strongly associated with CC16 than (total) PM2.5. In addition, we compared two source apportionment methods. Methods: We collected biweekly spot urinary samples over 6 months from 121 subjects with coronary heart disease for the determination of CC16 (n = 1251). Principal component analysis (PCA) was used to apportion daily outdoor PM2.5 between different source categories. In addition, the multilinear engine (ME) was used for the source apportionment in Amsterdam and Helsinki. We analyzed associations of source category-specific PM2.5 and PM2.5 absorbance, an indicator for combustion originating particles, with logarithmized values of CC16 adjusting for urinary creatinine using multivariate mixed models in STATA. Results: In the pooled analyses, CC16 was increased by 0.6% (standard error 0.3%) per 1 x 10(-5) m(-1) increase in the same-day levels of PM2.5 absorbance. Source category-specific PM2.5 concentrations were not consistently associated with the levels of CC16 in the three cities. Correlations between source category-specific PM2.5 determined using either PCA or ME were in general high. Associations of source category-specific PM2.5 with CC16 in Amsterdam and Helsinki were statistically less significant when ME was used. Conclusions: The present results suggest that PM2.5 from combustion sources increases epithelial barrier permeability in lungs.