Air pollution and fetal brain morphological development: a prospective cohort study

Abstract

Background: There is a scarcity of evidence of the influence of exposure to air pollution during pregnancy on the human fetal brain characterised prenatally. We aimed to evaluate the association of exposure to air pollution with fetal brain morphology.

Methods: In this prospective cohort study, we used data from the Barcelona Life Study Cohort, Spain, which recruited 1080 pregnant women at 8-14 weeks of gestation between Oct 16, 2018, and April 14, 2021, from three major university hospitals in Barcelona. Eligible participants were aged 18-45 years, had a singleton pregnancy, and had a fetus without major congenital anomalies. Third-trimester transvaginal neurosonography was applied to evaluate fetal brain morphological development. We integrated comprehensive data on time-activity patterns with land use regression, dispersion, and hybrid models to estimate exposure to NO₂, PM_2·5, and black carbon at home, workplace, and commuting routes during pregnancy until the neurosonography date. Single-pollutant linear mixed regression models and multipollutant ridge regression models were applied to estimate the associations between air pollutants and fetal brain outcomes, controlled for confounders. Distributed lag linear models were used to identify the vulnerable windows.

Findings: Among 1080 participants recruited at baseline, 954 attended the follow-up for the neurosonographic examination, 754 of whom were included in this study. In single-pollutant models, we found that prenatal exposure to NO₂, PM_2·5, and black carbon was associated with a wider anterior horn of lateral ventricles, wider cisterna magna, and larger cerebellar vermis. We also observed that higher exposure to black carbon was related to a shallower Sylvian fissure. No clear pattern or associations were observed between air pollution and other structures of brain morphology. Multipollutant models showed that these associations with black carbon remained significant, whereas associations with PM_2·5 and NO₂ lost significance for some indicators. A potential vulnerability window in mid-to-late pregnancy was identified for these associations.

Interpretation: Exposure to air pollution might affect brain morphological development as early as the fetal stage. Our findings could have important policy implications as they highlight the need to mitigate exposure of pregnant individuals to air pollution in urban areas to protect fetal brain development.

Funding: European Research Council.

Figure 1

Adjusted percentage difference in brain morphological structures associated with one IQR increase in exposure to ambient pollutants (A) Exposure to NO₂. (B) Exposure to PM_2·5. (C) Exposure to black carbon. Percentage differences are adjusted for fetal sex (female vs male), parity (multiparous vs nulliparous), maternal education (with university degree vs without university degree), ethnicity (White European vs other), active smoking during pregnancy (no vs yes), passive smoking during pregnancy (no vs yes), alcohol consumption during pregnancy (no vs yes), and gestational age at ultrasound (days), and the rater and hospital as random effects. All outcomes were measured in mm and natural log-transformed for analysis.

Figure 2

Adjusted percentage difference in brain morphological structures associated with one IQR increase in exposure to NO₂, PM_2·5, and black carbon in the multipollutant models Percentage difference are adjusted for fetal sex (female vs male), parity (multiparous vs nulliparous), maternal education (with university degree vs without university degree), ethnicity (White European vs other), active smoking during pregnancy (no vs yes), passive smoking during pregnancy (no vs yes), alcohol consumption during pregnancy (no vs yes), and gestational age at ultrasound (days), and the rater and hospital as random effects.