Repeated measurements of per- and polyfluoroalkyl substances (PFASs) from 1979 to 2007 in males from Northern Norway: assessing time trends, compound correlations and relations to age/birth cohort

Abstract

Background: Longitudinal biomonitoring studies can provide unique information on how human concentrations change over time, but have so far not been conducted for per- and polyfluoroalkyl substances (PFASs) in a background exposed population.

Objectives: The objectives of this study were to determine: i) serum PFAS time trends on an individual level; ii) relative compositions and correlations between different PFASs; and iii) assess selected PFAS concentrations with respect to periodic (calendar year), age and birth cohort (APC) effects.

Methods: Serum was sampled from the same 53 men in 1979, 1986, 1994, 2001 and 2007 in Northern Norway and analysed for 10 PFASs. APC effects were assessed by graphical and mixed effect analyses.

Results: The median concentrations of perfluorooctane sulphonic acid (PFOS) and perfluorooctanoic acid (PFOA) increased five-fold from 1979 to 2001 and decreased by 26% and 23%, respectively, from 2001 to 2007. The concentrations of PFOS and PFOA peaked during 1994-2001 and 2001, respectively, whereas perfluorohexane sulphonic acid (PFHxS) increased to 2001, but did not demonstrate a decrease between 2001 and 2007. Perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), and perfluoroundecanoic acid (PFUnDA) displayed increasing trends throughout the entire study period (1979-2007). Although PFOS comprised dominating and stable proportions of PFAS burdens during these years, the contributions from PFOA and PFHxS were considerable. The evaluation of APC effects demonstrated that calendar year was the dominating influence on concentrations of PFOA, PFUnDA, and PFOS, although time-variant and weaker associations with age/birth cohort were indicated.

Conclusions: The concentration changes of 10 PFASs in the repeated measurements from 1979 to 2007 demonstrated divergent time trends between the different PFASs. The temporal trends of PFASs in human serum during these 30years reflect the overall trends in historic production and use, although global transport mechanisms and bioaccumulation potential of the different PFASs together with a varying extent of consumer exposure influenced the observed trends. Sampling year was the strongest descriptor of PFOA, PFUnDA and PFOS concentrations, and the calendar-year trends were apparent for all birth year quartiles. Discrepancies between the trends in this current longitudinal study and previous cross-sectional studies were observed and presumably reflect the different study designs and population characteristics.

Keywords: Age–period–cohort effects; Per- and polyfluoroalkyl substances; Repeated measurements; Serum; Time trends.