Contamination levels and distribution of persistent and non-persistent pollutants in paired hair, urine and serum samples from a Belgian non-occupationally exposed population.

Designing ideal human biomonitoring studies involves the selection of reliable markers of exposure in adequate biological matrix. Besides conventional matrices such as blood or urine, hair has been increasingly investigated as a promising noninvasive alternative. However, understanding the pollutant distribution between differing biological compartments is essential for reliable interpretation of data collected. Therefore, the contamination levels and the distribution of some persistent (8 perfluoroalkyl substances - PFAS - and 6 polychlorobiphenyls - PCBs) and non-persistent pollutants (2 bisphenols and 3 parabens) were investigated in paired serum and hair samples, or paired spot urine and hair samples obtained from 30 Belgian non-occupationally exposed individuals. The levels measured were close to those reported in recent larger-scale studies. PFAS, PCB and bisphenol distributions largely differed depending upon the matrix and within the same chemical family depending upon the congener. The correlation and agreement between pollutant levels in differing matrices demonstrated that the information provided is comparable only for highly chlorinated PCBs and parabens, while the classification of exposure for bisphenols was substantially different according to the matrix. The selection of the human matrix thus remains complex and might markedly bias the results obtained, especially when assessing the health risk related to chemical exposure.