Exposure of Norwegian toddlers to perfluoroalkyl substances (PFAS): The association with breastfeeding and maternal PFAS concentrations.

High exposure to perfluoroalkyl substances (PFASs) has been associated with adverse health effects in children. PFASs exposure pathways of toddlers might differ from those of infants and adults, and the investigations on determinants of PFASs exposure in early childhood are scarce. Our aims were to examine the PFAS blood concentrations in Norwegian toddlers and to assess their relationship with maternal PFAS concentrations in pregnancy and breastfeeding duration. We determined PFAS concentrations in 112 plasma samples of 3-year-old children collected at 2010-2011 and 99 maternal serum samples collected around delivery at 2007-2008. PFAS concentrations in children were regressed on duration of breastfeeding, and the effect modification by maternal prenatal PFAS concentrations was examined in 55 mother-child pairs. Six PFASs were quantifiable in >50% of both maternal and children samples. Positive and significant correlations ranging between 0.50 and 0.66 were found between maternal and child concentrations of the same PFAS congeners. Nevertheless, toddlers had higher total PFAS blood concentrations than their mothers, due to higher concentrations of PFOA, PFNA and PFHxS. Every month of breastfeeding was associated with an increase of 3.3% (95% Confidence Intervals (CI): 0.8-5.8) for PFOS, 4.7% (95%CI: 2.8-6.6) for PFOA and 6.1% (95% CI: 2.6-9.7) for PFHpS in toddlers' plasma and a dose-response association was found, after adjustment for confounders. However, PFNA and PFUnDA concentrations in children were not associated with either maternal concentrations or breastfeeding duration. Our findings suggest that transplacental transfer, prenatally, and breastfeeding, postanatally, are among the main determinants of PFOS, PFOA, PFHxS and PFHpS concentrations in toddlers, while that was not the case for PFNA and PFUnDA. Nevertheless, due to the small number of mother child-pairs in our study, our results should be interpreted with caution.