Exposure to perfluoroalkyl substances and longitudinal changes in bone mineral density in adolescents and young adults: A multi-cohort study

Abstract

Background: Per- and polyfluoroalkyl substances (PFAS) may impair bone development in adolescence, which impacts life-long bone health. No previous studies have examined prospective associations of individual PFAS and their mixture with bone mineral density (BMD) changes in Hispanic young persons, a population at high risk of osteoporosis in adulthood.

Objectives: To examine associations of individual PFAS and PFAS mixtures with longitudinal changes in BMD in an adolescent Hispanic cohort and examine generalizability of findings in a mixed-ethnicity young adult cohort (58.4% Hispanic).

Methods: Overweight/obese adolescents from the Study of Latino Adolescents at Risk of Type 2 Diabetes (SOLAR; n = 304; mean follow-up = 1.4 years) and young adults from the Southern California Children's Health Study (CHS; n = 137; mean follow-up = 4.1 years) were included in this study. Plasma PFAS were measured at baseline and dual x-ray absorptiometry scans were performed at baseline and follow-up to measure BMD. We estimated longitudinal associations between BMD and five PFAS via separate covariate-adjusted linear mixed effects models, and between BMD and the PFAS mixture via quantile g-computation.

Results: In SOLAR adolescents, baseline plasma perfluorooctanesulfonic acid (PFOS) was associated with longitudinal changes in BMD. Each doubling of PFOS was associated with an average -0.003 g/cm² difference in change in trunk BMD per year over follow-up (95% CI: -0.005, -0.0002). Associations with PFOS persisted in CHS young adults, where each doubling of plasma PFOS was associated with an average -0.032 g/cm² difference in total BMD at baseline (95% CI -0.062, -0.003), though longitudinal associations were non-significant. We did not find associations of other PFAS with BMD; associations of the PFAS mixture with BMD outcomes were primarily negative though non-significant.

Discussion: PFOS exposure was associated with lower BMD in adolescence and young adulthood, important periods for bone development, which may have implications on future bone health and risk of osteoporosis in adulthood.

Keywords: Adolescents; Bone mineral density; Children; Endocrine disruptors; Environmental epidemiology; Perfluoroalkyl substances.

Figure 1.

Association between a) plasma PFOS and trunk BMD at the median age for baseline and follow-up visits for n=226 Hispanic adolescents with overweight and obesity recruited from Sothern California between 2001 and 2012 (SOLAR cohort); and b) plasma PFOS and total BMD at the median age for baseline and follow-up visits for n=79 young adults recruited from the Children’s Health Study between 2014 and 2018 (CHS cohort). Figure represents point estimates and 95% confidence intervals for BMD at each age calculated using linear mixed effect models corrected for a) sex, parental education, Tanner stage, and study wave; and b) sex, ethnicity, and parental education. * denotes p-values for differences <0.05, calculated using t-statistics to evaluate whether the difference between individuals with low (10^th percentile) vs. high (90^th percentile) of PFOS was significantly different from 0.