Persistent organic pollutants in pregnancy, pre- and postnatal growth, and childhood obesity

Abstract

Humans have been exposed to an increasing amount of persistent organic pollutants (POPs) during the 20th century. Perfluoroalkyl substances (PFASs) and organochlorines (OCs) are two important groups of POPs. Concentrations of PFASs increased extensively until the 1990s and decreased from around 2000, while OCs have decreased since the 1980s due to world-wide bans and restrictions on use. POPs have persistent properties, they bioaccumulate in the food chain, and diet is the main exposure route for humans. They are transferred from the mother to the foetus via the placenta during pregnancy and via breastmilk to infants after birth. These POPs are thought to act as “endocrine disrupting chemicals” (EDCs), and it has been hypothesized that prenatal exposure to EDCs could potentially lead to harmful developmental consequences in the offspring, including foetal growth restriction (FGR) and/or later childhood obesity, but the evidence is sparse and inconclusive. This doctoral thesis uses data from the Scandinavian Successive Small-for-Gestational Age (SGA) Births Study. The SGA Births Study is a large, prospective, population-based multicentre study, conducted in Trondheim and Bergen (Norway) and Uppsala (Sweden) from 1986 to 1994. The women were followed from early pregnancy and screened at their study centre in gestational weeks 17, 25, 33, 37, and at delivery. After birth, a follow-up study examined selected children through their first year of life up to five years of age. Maternal serum samples, donated during the second trimester, were analysed for several PFASs and OCs. Three peer-reviewed and published papers make up this doctoral thesis. Paper I reports maternal serum PFAS and OC concentrations and examines associated maternal factors. Overall, Swedish women had higher serum PFAS and polychlorinated biphenyl (PCB) 153 concentrations compared to Norwegian women. Previous breastfeeding duration, sampling date, maternal age and time since last breastfeeding period influenced maternal PFAS and OC concentrations collected from 1986 to 1988. All maternal serum concentrations of PFASs and OCs were inversely associated with previous breastfeeding duration. Throughout the recruitment period maternal serum PFAS concentrations increased while OC concentrations declined. Maternal age and time since last breastfeeding period were positively associated with maternal serum concentrations of OCs and PFASs, respectively. Maternal pre-pregnancy BMI, smoking status, education level and alcohol consumption were also associated with some maternal serum PFAS and OC concentrations. A full description of the results can be found in the Results section and the appended published manuscripts. In Paper II, maternal serum concentrations of perfluorooctanoic acid (PFOA), PCB 153 and HCB were associated with indices of impaired foetal growth (including higher odds for SGA birth), but only among Swedish mother-child pairs. Some indications of stronger associations among Swedish male offspring were found. In Paper III, maternal serum concentrations of PFOS and PFOA were positively associated with offspring BMI and triceps skinfold z-scores and higher odds for offspring overweight/obesity at five-year follow-up. In country-stratified analyses, these associations only remained among Norwegian mother-pairs. Some evidence of non-linearity implied possible involvement of non-monotonic dose-response relationships. This work highlights challenges in establishing causal effects of POPs on pre- and postnatal growth patterns due to the complexity of developmental processes, as well as the intricacy of correlated exposures, lack of monotonicity, and multiple possible exposures routes and concentrations of POPs between different regions. However, the results indicate a possible influence of background exposures to POPs on pre- and postnatal growth. Welldesigned prospective studies with contemporary data on current environmental concentrations and mixtures of POPs, as well as longer follow-up are needed to replicate and confirm these findings.