Associations of Air Pollution on the Brain in Children: A Brain Imaging Study

Monica Guxens, Malgorzata J Lubczynska, Laura Perez-Crespo, Ryan L Muetzel, Hanan El Marroun, Xavier Basagana, Gerard Hoek, Henning Tiemeier

Research output: Book/Report/Inaugural speech/Farewell speechReportAcademic


INTRODUCTION: Epidemiological studies are highlighting the negative effects of the exposure to air pollution on children's neurodevelopment. However, most studies assessed children's neurodevelopment using neuropsychological tests or questionnaires. Using magnetic resonance imaging (MRI) to precisely measure global and region-specific brain development would provide details of brain morphology and connectivity. This would help us understand the observed cognitive and behavioral changes related to air pollution exposure. Moreover, most studies assessed only a few air pollutants. This project investigates whether air pollution exposure to many pollutants during pregnancy and childhood is associated with the morphology and connectivity of the brain in school-age children and pre-adolescents.

METHODS: We used data from the Generation R Study, a population-based birth cohort set up in Rotterdam, the Netherlands in 2002-2006 (n = 9,610). We used land-use regression (LUR) models to estimate the levels of 14 air pollutants at participant's homes during pregnancy and childhood: nitrogen oxides (NOx), nitrogen dioxide (NO2), particulate matter with aerodynamic diameter ≤10 μm (PM10) or ≤2.5 μm (PM2.5), PM between 10 μm and 2.5 μm (PMCOARSE), absorbance of the PM2.5 fraction - a measure of soot (PM2.5absorbance), the composition of PM2.5 such as polycyclic aromatic hydrocarbons (PAHs), organic carbon (OC), copper (Cu), iron (Fe), silicon (Si), zinc (Zn), and the oxidative potential of PM2.5 evaluated using two acellular methods: dithiothreitol (OPDTT) and electron spin resonance (OPESR). We performed MRI measurements of structural morphology (i.e., brain volumes, cortical thickness, and cortical surface area) using T1-weighted images in 6- to 10-year-old school-age children and 9- to 12-year-old pre-adolescents, structural connectivity (i.e., white matter microstructure) using diffusion tensor imaging (DTI) in pre-adolescents, and functional connectivity (i.e., connectivity score between brain areas) using resting-state functional MRI (rs-fMRI) in pre-adolescents. We assessed cognitive function using the Developmental Neuropsychological Assessment test (NEPSY-II) in school-age children. For each outcome, we ran regression analysis adjusted for several socioeconomic and lifestyle characteristics. We performed single-pollutant analyses followed by multipollutant analyses using the deletion/substitution/addition (DSA) approach.

RESULTS: The project has air pollution and brain MRI data for 783 school-age children and 3,857 pre-adolescents. First, exposure to air pollution during pregnancy or childhood was not associated with global brain volumes (e.g., total brain, cortical gray matter, and cortical white matter) in school-age children or pre-adolescents. However, higher pregnancy or childhood exposure to several air pollutants was associated with a smaller corpus callosum and hippocampus, and a larger amygdala, nucleus accumbens, and cerebellum in pre-adolescents, but not in school-age children. Second, higher exposure to several air pollutants during pregnancy was associated with a thinner cortex in various regions of the brain in both school-age children and pre-adolescents. Higher exposure to air pollution during childhood was also associated with a thinner cortex in a single region in pre-adolescents. A thinner cortex in two regions mediated the association between higher exposure to air pollution during pregnancy and an impaired inhibitory control in school-age children. Third, higher exposure to air pollution during childhood was associated with smaller cortical surface areas in various regions of the brain except in a region where we observed a larger cortical surface area in pre-adolescents. In relation to brain structural connectivity, higher exposure to air pollution during pregnancy and childhood was associated with an alteration in white matter microstructure in pre-adolescents. In relation to brain functional connectivity, a higher exposure to air pollution, mainly during pregnancy and early childhood, was associated with a higher brain functional connectivity among several brain regions in pre-adolescents. Overall, we identified several air pollutants associated with brain structural morphology, structural connectivity, and functional connectivity, such as NOx, NO2, PM of various size fractions (i.e., PM10, PMCOARSE, and PM2.5), PM2.5absorbance, PAHs, OC, three elemental components of PM2.5 (i.e., Cu, Si, Zn), and the oxidative potential of PM2.5.

CONCLUSIONS: The results of this project suggest that exposure to air pollution during pregnancy and childhood play an adverse role in brain development. We observed this relationship even at levels of exposure that were below the European Union legislations. We acknowledge that identifying the independent effects of specific pollutants was particularly challenging. Most of our conclusions generally refer to traffic-related air pollutants. However, we did identify pollutants specifically originating from brake linings, tire wear, and tailpipe emissions from diesel combustion. The current direction toward innovative solutions for cleaner energy vehicles is a step in the right direction. However, our findings indicate that these measures might not be completely adequate to mitigate health problems attributable to traffic-related air pollution, as we also observed associations with markers of brake linings and tire wear.

Original languageEnglish
Number of pages61
Publication statusPublished - Feb 2022

Publication series

SeriesResearch report (Health Effects Institute)

Bibliographical note

© 2022 Health Effects Institute. All rights reserved.


Dive into the research topics of 'Associations of Air Pollution on the Brain in Children: A Brain Imaging Study'. Together they form a unique fingerprint.

Cite this