Prenatal exposure to polyfluoroalkyl substances and visual acuity in preschool children: Findings from the Guangxi Zhuang birth cohort in China.

BACKGROUND

Prenatal exposure to Polyfluoroalkyl Substances (PFASs) had been associated with adverse effects on multiple systems in offspring. However, the effects on visual system had not been previously explored.

OBJECTIVE

The study aimed to assess the association between prenatal exposure to PFASs and visual acuity and visual impairment in preschool children.

METHODS

A total of 1008 mother-infant pairs were extracted from the Guangxi Zhuang Birth Cohort in China. Maternal serum concentrations of nine PFASs were measured using Ultra-Performance Liquid Chromatography-Mass Spectrometry (UPLC-MS). The preschool visual acuity (VA) in the offspring was obtained through the local maternal and child health information system. Multiple linear regression models and logistic regression models were performed to study the single effects of individual PFAS on VA and visual impairment (VI), respectively. Additionally, Bayesian kernel machine regression (BKMR) and quantile g-computation (Qgcomp) models were utilized to assess the joint effects of the nine PFASs.

RESULTS

Prenatal exposure to PFOS (β = 0.074, 95 % CI: 0.033, 0.116), PFHxS (β = 0.062, 95 % CI: 0.012, 0.111), and PFBS (β = 0.023, 95 % CI: 0.001, 0.044) exhibited significant positive associations with enhanced VA. Conversely, PFOS (OR = 0.481, 95 % CI: 0.245, 0.946) and PFHxS (OR = 0.450, 95 % CI: 0.244, 0.830) were significantly associated with reduced odds of VI. BKMR and quantile g-computation (Qgcomp) modeling consistently displayed positive joint-effects for nine PFASs mixtures on VA. Stratified analysis by sex indicated that certain significant associations were observed exclusively in boys or girls, and these associations were positively correlated with VA.

CONCLUSIONS

The evidence does not substantiate the hypothesis that prenatal exposure to PFASs adversely impacts visual development. Nonetheless, inverse associations reported here should not be interpreted as protective, as these associations might be driven by some unresolved confounding factors and biases. Their relationship still needs to be elucidated in future studies, using larger samples and accounting for both prenatal and childhood periods.