Prenatal exposure to perfluoroalkyl substances and birth outcomes in a Spanish birth cohort: assessing confounding by maternal glomerular filtration rate

摘要

Background: Prenatal exposure to some perfluoroalkyl substances (PFAS) has been associated with low birth weight (LBW) but this association may be confounded by maternal hemodynamics, such as glomerular filtration rate GFR during pregnancy. We evaluated the association between prenatal exposure to PFAS and birth outcomes, and the influence of maternal GFR in 1,206 mother-child pairs from a Spanish birth cohort study. Methods: We measured perfluorohexane sulfonate (PFHxS), perfluorooctane sulfonate (PFOS), perfluorooctanoate (PFOA), and perfluorononanoate (PFNA) in maternal plasma samples collected in the the 1st-trimester of pregnancy during the years 2003-2008. Birth outcomes included weight, LBW, LBW-at term, length, head circumference, ponderal index, gestational age, small-for-gestational-age (SGA), and preterm birth. Maternal GFR during pregnancy was estimated using the Cockcroft-Gault formula in 769 mothers. Results: We detected at least one PFAS in every maternal sample, being PFOS (mean: 5.78 ng/mL) and PFOA (mean: 2.32 ng/mL) the most abundant. Overall, we observed a strengthening of our associations after including GFR in our models. Specifically, all PFAS were associated with higher weight at birth, ranging from 38g (95% CI: 6.52, 68.71) up to 58g (16.86, 99.48) per doubling of PFAS concentrations. Also, higher PFOS concentration was associated with a decrease of 0.30 weeks (-0.53, -0.08) in gestational age, and 2.7 (1.15, 6.40) times higher odds of preterm birth only in boys (p-interactions = 0.06 and 0.07, respectively). Higher PFHxS, PFOS, and PFNA concentrations were associated with higher ponderal index at birth (betas from regression models ranged from 0.02 to 0.04). There was no association between PFAS and birth length, head circumference, LBW, LBW-at term, and SGA. Conclusions: In this study, prenatal PFAS exposure was associated with higher weight at birth in both sexes, and reduced gestational age and higher odds of being preterm only in boys. These results were especially strengthened after considering maternal GFR as a confounder. Studies using maternal PFAS concentrations would benefit from adjusting for maternal hemodynamics such as GFR, even if using maternal samples collected early in pregnancy.