Physical activity modifies the association between prenatal perfluorooctanoic acid exposure and adolescent cardiometabolic risk

Abstract

BACKGROUND AND AIM: No studies have examined whether previously observed associations between higher gestational PFAS exposure and worse offspring cardiometabolic outcomes might be ameliorated through more physical activity and healthier diet in childhood. METHODS: In 164 mother-child pairs from a prospective cohort in Cincinnati, OH (HOME Study), we measured concentrations of four PFAS in maternal serum collected at 16-weeks gestation. When children were age 12 years, we assessed visceral adiposity using dual x-ray absorptiometry, blood pressure, and fasting serum biomarkers. Using these variables, we calculated age/sex-standardized cardiometabolic risk z-scores. At this same visit, we assessed general physical activity levels and Healthy Eating Index (HEI) scores using the Physical Activity Questionnaire and three 24-hour diet recalls, respectively. We determined whether covariate-adjusted associations of PFAS with cardiometabolic risk z-scores and its components were modified by physical activity or HEI using multivariable linear regression. RESULTS:Physical activity modified the association between perfluorooctanoic acid (PFOA) and cardiometabolic risk z-scores (interaction p-value0.01). Each doubling of PFOA was associated with worse cardiometabolic risk z-scores among children with physical activity levels below the median (β:1.2; 95%CI:0.4, 2.0, n=80), but not among those with physical activity levels greater than the median (β:-0.3; 95%CI:-1.1, 0.6, n=84). Adverse associations among children with lower physical activity were most apparent for insulin resistance, adiponectin-leptin ratio, and visceral fat area (interaction p-values=0.01-0.06), but not triglyceride-high density lipoprotein ratio or systolic blood pressure (interaction p-values=0.41-0.65). The pattern of modification by physical activity was similar, but not statistically significant, for perfluorosulfonic acid and perfluorononanoic acid (interaction p-values=0.44 and 0.23). Diet quality (HEI scores vs. 50) did not modify any associations (interaction p-values0.28). CONCLUSIONS:Childhood physical activity modified the association between gestational PFOA concentrations and cardiometabolic risk among adolescents in this cohort, suggesting that behavioral and lifestyle interventions may be an avenue to ameliorate the adverse effects of PFOA exposure. . KEYWORDS: PFAS, Endocrine disrupting chemicals, Children's environmental health, Obesity and metabolic disorders