Prenatal exposure to endocrine disrupting chemicals and risk of being born small for gestational age: Pooled analysis of seven European birth cohorts

Abstract

Background and aims: There is evidence that endocrine disrupting chemicals (EDCs) have developmental effects at environmental concentrations. We investigated whether some EDCs are associated with the adverse birth outcome Small for Gestational Age (SGA). Methods: We used PCB 153, p,p’-DDE, HCB, PFOS and PFOA measured in maternal, cord blood or breast milk samples of 5446 mother-child pairs (subset of 693 for the perfluorinated compounds) from seven European birth cohorts (1997–2012). SGA infants were those with birth weight below the 10th percentile for the norms defined by gestational age, country and infant's sex. We modelled the association between measured or estimated cord serum EDC concentrations and SGA using multiple logistic regression analyses. We explored effect modification by child's sex and maternal smoking during pregnancy. Results: Among the 5446 newborns, 570 (10.5%) were SGA. An interquartile range (IQR) increase in PCB 153 was associated with a modestly increased risk of SGA (odds ratio (OR) of 1.05 [95% CI: 1.04–1.07]) that was stronger in girls (OR of 1.09 [95% CI: 1.04–1.14]) than in boys (OR of 1.03 [95% CI: 1.03–1.04]) (p-interaction = 0.025). For HCB, we found a modestly increased odds of SGA in girls (OR of 1.04 [95% CI: 1.01–1.07] per IQR increase), and an inverse association in boys (OR of 0.90 [95% CI: 0.85–0.95]) (p-interaction = 0.0003). Assessment of the HCB-sex-smoking interaction suggested that the increased odds of SGA associated with HCB exposure was only in girls of smoking mothers (OR of 1.18 [95% CI: 1.11–1.25]) (p-interaction = 0.055). Higher concentrations of PFOA were associated with greater risk of SGA (OR of 1.64 [95% CI: 0.97–2.76]). Elevated PFOS levels were associated with increased odds of SGA in newborns of mothers who smoked during pregnancy (OR of 1.63 [95% CI: 1.02–2.59]), while an inverse association was found in those of non-smoking mothers (OR of 0.66 [95% CI: 0.61–0.72]) (p-interaction = 0.0004). No significant associations were found for p,p’-DDE. Conclusions: Prenatal environmental exposure to organochlorine and perfluorinated compounds with endocrine disrupting properties may contribute to the prevalence of SGA. We found indication of effect modification by child's sex and smoking during pregnancy. The direction of the associations differed by chemical and these effect modifiers, suggesting diverse mechanisms of action and biological pathways.