Association of Early Pregnancy Perfluoroalkyl and Polyfluoroalkyl Substance Exposure With Birth Outcomes

Abstract

Importance: Prenatal perfluoroalkyl and polyfluoroalkyl substances (PFAS) have been linked to adverse birth outcomes. Previous research showed that higher folate concentrations are associated with lower blood PFAS concentrations in adolescents and adults. Further studies are needed to explore whether prenatal folate status mitigates PFAS-related adverse birth outcomes.

Objective: To examine whether prenatal folate status modifies the negative associations between pregnancy PFAS concentrations, birth weight, and gestational age previously observed in a US cohort.

Design, setting, and participants: In a prospective design, a prebirth cohort of mothers or pregnant women was recruited between April 1999 and November 2002, in Project Viva, a study conducted in eastern Massachusetts. Statistical analyses were performed from May 24 and October 25, 2022.

Exposure: Plasma concentrations of 6 PFAS compounds were measured in early pregnancy (median gestational week, 9.6). Folate status was assessed through a food frequency questionnaire and measured in plasma samples collected in early pregnancy.

Main outcomes and measures: Birth weight and gestational age, abstracted from delivery records; birth weight z score, standardized by gestational age and infant sex; low birth weight, defined as birth weight less than 2500 g; and preterm birth, defined as birth at less than 37 completed gestational weeks.

Results: The cohort included a total of 1400 mother-singleton pairs. The mean (SD) age of the mothers was 32.21 (4.89) years. Most of the mothers were White (73.2%) and had a college degree or higher (69.1%). Early pregnancy plasma perfluorooctanoic acid concentration was associated with lower birth weight and birth weight z score only among mothers whose dietary folate intake (birth weight: β, -89.13 g; 95% CI, -166.84 to -11.42 g; birth weight z score: -0.13; 95% CI, -0.26 to -0.003) or plasma folate concentration (birth weight: -87.03 g; 95% CI, -180.11 to 6.05 g; birth weight z score: -0.14; 95% CI, -0.30 to 0.02) were below the 25th percentile (dietary: 660 μg/d, plasma: 14 ng/mL). No associations were found among mothers in the higher folate level groups, although the tests for heterogeneity did not reject the null. Associations between plasma perfluorooctane sulfonic acid and perfluorononanoate (PFNA) concentrations and lower birth weight, and between PFNA and earlier gestational age were noted only among mothers whose prenatal dietary folate intake or plasma folate concentration was in the lowest quartile range. No associations were found among mothers in higher folate status quartile groups.

Conclusions and relevance: In this large, US prebirth cohort, early pregnancy exposure to select PFAS compounds was associated with adverse birth outcomes only among mothers below the 25th percentile of prenatal dietary or plasma folate levels.

Figure 1.. Associations of Early Pregnancy Plasma Polyfluoroalkyl and Polyfluoroalkyl Substances Concentrations With Birth Weight by Quartiles of Early Pregnancy Dietary Folate Intake (DFE) or Plasma Folate Concentrations Among Mother-Singleton Pairs in Project Viva

Models were adjusted for maternal age (continuous), educational level (≥college graduate vs not college graduate), race and ethnicity (Asian, Black, Hispanic, White, other), prepregnancy body mass index (continuous), smoking history (never, former, early pregnancy), nulliparous (yes vs no), breastfeeding history (yes vs no), paternal educational level (≥college graduate vs not college graduate), annual household income (>$70 000 vs ≤$70 000), infertility (yes vs no), planned pregnancy (yes vs no), infant sex (male vs female), gestational age at recruitment (continuous), and Alternative Healthy Eating Index in Pregnancy score in early pregnancy (continuous). PFNA indicates perfluorononanoate; PFOA, perfluorooctanoic acid; and PFOS, perfluorooctane sulfonic acid.

Figure 2.. Associations of Early Pregnancy Plasma Perfluorooctanoic Acid Concentrations With Birth Weight z Score (Standardized by Gestational Age and Sex) by Quartiles of Early Pregnancy Dietary Folate Equivalent (DFE) Intake or Plasma Folate Concentrations Among Mother-Singleton Pairs in Project Viva

Models were adjusted for maternal age (continuous), educational level (≥college graduate vs not college graduate), race and ethnicity (Asian, Black, Hispanic, White, other), prepregnancy body mass index (continuous), smoking history (never, former, early pregnancy), nulliparous (yes vs no), breastfeeding history (yes vs no), paternal educational level (≥college graduate vs not college graduate), annual household income (>$70 000 vs ≤$70 000), infertility (yes vs no), planned pregnancy (yes vs no), infant sex (male vs female), gestational age at recruitment (continuous), and Alternative Healthy Eating Index in Pregnancy score in early pregnancy (continuous).

Figure 3.. Associations of Early Pregnancy Plasma Perfluorononanoate Concentrations With Gestational Age by Quartiles of Early Pregnancy Dietary Folate Equivalent (DFE) Intake or Plasma Folate Concentrations Among Mother-Singleton Pairs in Project Viva

Models were adjusted for maternal age (continuous), educational level (≥college graduate vs not college graduate), race and ethnicity (Asian, Black, Hispanic, White, other), prepregnancy body mass index (continuous), smoking history (never, former, early pregnancy), nulliparous (yes vs no), breastfeeding history (yes vs no), paternal educational level (≥college graduate vs not college graduate), annual household income (>$70 000 vs ≤$70 000), infertility (yes vs no), planned pregnancy (yes vs no), infant sex (male vs female), gestational age at recruitment (continuous), and Alternative Healthy Eating Index in Pregnancy score in early pregnancy (continuous).