Outdoor Air Pollution, Perivascular Space Morphology, and Cognition in Preadolescence
Morrel, J.; Sukumaran, K.; Torgerson, C.; Rosario, M. A.; Lan, H.; Schwartz, J.; Chen, J.-C.; Choupan, J.; Herting, M. M.
Show abstract
BackgroundAmbient air pollution exposure is associated with structural brain differences and poorer cognition in children; however, mechanisms of toxicity remain unclear. Perivascular spaces (PVS), key for brain waste clearance, may play a role in the neurotoxicity of air pollution. This study explored associations between air pollution exposure, PVS morphology, and cognition in preadolescents. MethodsWe analyzed cross-sectional Adolescent Brain Cognitive DevelopmentSM (ABCD) Study(R) data from 6,949 9-10-year-old participants. Annual average exposures to PM2.5, O3, NO2, and 15 PM2.5 components were estimated using spatiotemporal models mapped to residential addresses. PVS count and volume were derived from T1w and T2w MRI, and cognition was estimated using NIH Toolbox scores. Linear mixed-effects models examined independent associations between air pollution, PVS, and cognition; weighted quantile sum regression assessed co-exposure effects of PM2.5 mixtures. FindingsLinear models revealed that exposures to Zn, NH4 +, and Br were positively associated with PVS count in several regions. Higher PVS count in five key regions was associated with poorer cognitive performance across several NIH Toolbox domains. Higher Ca, Zn, and NH 4 + exposures were associated with poorer cognition (PFDR < 0.01). Higher frontal lobe PVS count mediated the association between Zn exposure and poorer total cognition (P < 0.01). Co-exposure models revealed that PM2.5 mixtures were associated with higher temporal and cingulate PVS count, and poorer working memory and crystallized intelligence (P < 0.01). InterpretationOutdoor air pollution was associated with higher PVS count and reduced cognition, suggesting that brain clearance may be a novel mechanism linking pollution to neurodevelopmental harm in preadolescents. FundingThis work was supported by the National Institutes of Health (NIH) National Institute of Environmental Health Sciences (NIEHS) (Grant Nos. R01ES032295 and R01ES031074 [to MMH]; T32ES013678 [to JM]; P30ES07048 [to JM and MAR]; 3P30ES000002-55S [to MAR]), National Institute of Mental Health (NIMH) (Grant RF1MH123223 [to JC]), National Institute of Neurological Disorders and Stroke (Grant R01NS128486 [to JC]), and EPA grants (Grant Nos. 83587201 and 83544101 [to JS]).
Matching journals
The top 7 journals account for 50% of the predicted probability mass.