Abstract
Prenatal air pollution exposure has been associated with adverse childhood cardiometabolic outcomes. It is unknown whether evidence of metabolic disruption associated with air pollution is identifiable at birth. We examined exposure to prenatal ambient air pollution and cord blood cardiometabolic biomarkers among 812 mother-infant pairs in the Healthy Start study. METHODS: Using inverse-distance-weighted interpolation of ambient concentrations obtained from stationary monitors, we estimated daily particulate matter ≤2.5 micrometers (PM(2.5)) and ozone (O(3)) concentrations at participant residences. Daily estimates were averaged by trimester, full-pregnancy, and the 7 and 30 days prior to delivery. Associations of air pollution with the following cord blood biomarkers were estimated via multivariable linear regression: glucose, insulin, glucose/insulin ratio (GIR), leptin, high-density lipoprotein (HDL) cholesterol, non-HDL cholesterol, free fatty acids, and triglycerides. RESULTS: In this Denver-based cohort, PM(2.5) concentrations were lower than in many US urban areas, but O(3) concentrations regularly exceeded federal air quality standards. Higher O(3) concentrations during pregnancy were consistently associated with higher insulin and lower GIR in cord blood. For example, an interquartile range increase in full pregnancy O(3) (6.3 parts per billion [ppb]) was associated with 0.13 log-µIU/ml (95% confidence interval [CI] = 0.04, 0.22) higher cord blood insulin, after adjusting for PM(2.5) and other confounders. We found positive, but generally nonsignificant, associations between PM(2.5) and leptin and isolated associations between pollutants during certain exposure periods and lipids. CONCLUSIONS: In this cohort with moderately high O(3) exposure, prenatal concentrations of O(3) were positively associated with cord blood insulin. Future studies should examine the implications for offspring long-term health.