Per- and polyfluoroalkyl substances and adolescent bone mineral density: assessing periods of susceptibility.

CONTEXT: Per- and polyfluoroalkyl substances (PFAS) may affect adolescent bone mineral density (BMD), but few studies have examined periods of susceptibility.

OBJECTIVE

To assess associations of repeated serum PFAS concentrations with early adolescent BMD.

METHODS

Among 218 children in a prospective pregnancy and birth cohort, we quantified serum concentrations of perfluorohexane sulfonic acid (PFHxS), perfluoroctane sulfonic acid (PFOS), perfluorooctanoic acid (PFOA), and perfluorononanoic acid (PFNA) at delivery (cord) and ages 3, 8, and 12 years. We measured BMD at 6 skeletal sites using dual-energy X-ray absorptiometry at age 12 years. We estimated covariate-adjusted differences and 95% confidence intervals (CIs) in BMD Z-scores per interquartile range (IQR) increase in each PFAS at each time point using multiple informant linear regression models.

RESULTS

Higher PFOA concentrations were associated with lower 1/3 distal radius BMD at every time point: delivery (β: -0.39; 95% CI: -0.69, -0.10), 3 years (β: -0.36; 95% CI: -0.86, 0.14), 8 years (β: -0.54; 95% CI: -0.91, -0.16), and 12 years (β: -0.40; 95% CI: -0.75, -0.04). Higher 3-year PFHxS and PFOS concentrations were significantly associated with higher whole body, total hip, and femoral neck BMD Z-scores, while higher 12-year PFNA concentrations were associated with lower BMD Z-scores. Associations of 12-year PFAS with lower 1/3 distal radius BMD were stronger among females than males.

CONCLUSION

Serum PFOA concentrations from delivery to age 12 years were associated with lower 1/3 distal radius BMD in early adolescence, while associations of other PFAS with BMD varied by exposure assessment time point, skeletal site, and sex.