The Breast Milk Perfluorome is Associated with Infant Growth in the First Year of Life

Abstract

BACKGROUND AND AIM: Early life exposure to poly- and perfluoroalkyl substances (PFAS) has significant obesogenic effects that are manifested in early life. However, PFAS examined do not include the wide range of PFAS used in commercial applications. METHODS: This pilot included 56 Hispanic mother-child pairs from Southern California. Maternal and infant characteristics were assessed, including infant weight, length, and fat mass. All mothers were breastfeeding, and breast milk was collected at 1-month postpartum. The perfluorome was characterized using untargeted high-resolution mass spectrometry. Organofluorine compounds were identified using the EPA CompTox Dashboard. Multivariable linear and logistic regression analyses were used to examine associations between PFAS with growth measures (12-months, change in first year, rapid growth) after adjusting for infant age, sex, pre-pregnancy BMI, gestational age, and formula feedings. Results are reported for associations with p<0.05 and were considered statistically significant using a false discovery rate (FDR) of 20%. RESULTS: 484 suspect PFAS compounds were detected in ≥75% of samples. PFAS were associated with 12-month weight [n(%): 10(2.1%) positive, 12(2.5%) negative], change in weight [10(2.1%) positive, 22(4.5%) negative], 12-month length [21(4.3%) positive, 17(3.5%) negative] change in length [22(4.5%) positive, 25(5.2%) negative], 12-month fat mass [9(1.9%) positive, 12(2.5%) negative], and change in fat mass [18(3.7%) positive, 21(4.3%) negative]. Additionally, 12(2.5%) and 13(2.7%) of PFAS compounds were associated with an increased (OR=2.0-3.8) and decreased odds (OR=0.28-0.47) of rapid growth in the first, respectively. After FDR, three compounds, each detected in >98% of samples, were associated with change in length (q=0.01; n=2 positive, n=1 negative). Tentative annotations of these compounds identified novel PFAS, including methyl 4,4,5,5,5-pentafluoropent-2-ynoate3,4,4,5,5-Pentafluoro-2-methoxycyclopent-2-en-1-one, ethyl [5-(heptafluoropropyl)-1-methyl-1H-pyrazol-3-yl]acetate, and ethyl 2-acetyl-4,4,5,5,5-pentafluoro-3-oxopentanoate. CONCLUSIONS: Results from this pilot suggest that a wide range of PFAS are present in human breast milk, which may impact infant growth in the first year of life. KEYWORDS: breast milk, PFAS, infants