448 - Association of Early Life Air Pollutants with Adolescent Magnetic Resonance Spectroscopy Concentrations

Publication Number: 448.3749

Kim M. Cecil, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Chloe Dunseath, University of Cincinnati College of Medicine, Cincinnati, OH, United States; Yingying Xu, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Kimberly Yolton, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Joseph Braun, Brown University, Providence, RI, United States; Aimin Chen, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States; Bruce Lanphear, Simon Fraser University, Vancouver, BC, Canada; Patrick Ryan, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States

Background: An ever-increasing number of studies document the neurotoxicity and contribution of air pollution to several conditions including autism spectrum disorder, deficits in attention-related behaviors, and increased anxiety and depressive symptoms in childhood. Early life exposure to pollutants may contribute to atypical neurodevelopment resulting in altered intrinsic neurochemistry.

Objective: We estimated the association of key air pollutants at multiple timepoints and proton magnetic resonance spectroscopy (MRS) derived metabolite concentrations from participants in a prospective birth cohort longitudinally followed to adolescence.

Design/Methods: With IRB approval, we conducted this study with participants of the Cincinnati Combined Childhood Cohorts (C4).

We applied validated spatiotemporal models to estimate daily exposures to particulate matter less than 2.5 microns (PM2.5), nitrogen dioxide (NO2) and ozone (O3) at the homes of participants. Daily exposure estimates for each pollutant were aggregated to derive average gestational, first year of life and annual timepoints.

During the 12-year study visit, participants completed anatomical T1-weighted imaging using a 3 Tesla MR scanner. Short echo point resolved spectroscopy (PRESS) was acquired within a single voxel (8 cc) positioned within the frontal lobe cortex. Raw spectra were quantified using LCModel software. Adjustments for fractional tissue volumes, water and metabolite relaxation rates were applied to generate metabolite concentrations.

We developed linear regression models to evaluate associations between exposure concentrations of each pollutant and metabolite concentrations for the participants (n=334, 167 female, mean (SD) age 12.2 (0.7) years). All effect estimates are presented corresponding to each unit increase in exposure concentration. Final models were adjusted for adolescent sex, maternal education, neighborhood deprivation index and greenspace.

Results: Adjusted models found composite glutamate and glutamine concentrations were inversely associated with NO2 (β -0.076 (95% confidence interval (CI): -0.11, -0.042) for gestational, -0.061 (95% CI: -0.1, -0.021) for year 1) and PM2.5 (β -0.115 (95% CI: -0.19, -0.039) for gestational, -0.46 (95% CI: -0.61, -0.311) for year 1) and O3 (β -0.048 (95% CI: -0.085, -0.011 for gestational) concentrations.

Conclusion(s): Higher concentrations of NO2, PM2.5 and O3 during early life are associated with reduced brain glutamate concentrations at age 12. This may reflect reduced numbers of glutamatergic receptors arising from early-life exposure to air pollutants perturbing intrinsic neurochemistry.

Proton Magnetic Resonance Spectrum
Example of a C4 study participant single voxel proton magnetic resonance spectrum acquired within the frontal lobe cortex (shown as yellow box on a single slice from a T1-weighted magnetic resonance image of the brain (top right)). The short echo spectrum demonstrates resonances for N-acetyl-aspartate (NAA), glutamate and glutamine (GLX), creatine and phosphocreatine (Cr), cholines (Cho) and myo-inositol (mI).

Adjusted Difference in MRS Concentrations by Each Unit Increase of Air Pollutants
Adjusted estimates and 95% confidence interval for difference in concentrations of N-acetyl-aspartate (NAA), glutamate (Glu) and glutamine (GLX), creatine and phosphocreatine (Cr), cholines (Cho) and myo-inositol (mI), in association with each unit increase in NO2, PM2.5 and O3 concentrations. Final models were adjusted for adolescent sex, maternal education, neighborhood deprivation index and greenspace.