Maternal Exposure to NO2 During Pregnancy and Fetal Growth: A Comparison of Two Exposure Models

Abstract

ISEE-0752 Background and Objective: We have previously reported that nitrogen dioxide (NO2) exposures, as estimated by a geostatistical model and the closest monitoring station approach, showed little to moderate agreement. Here, our aim was to compare these two exposure models in terms of association with fetal growth. Methods: We focused on 428 women from the Eden cohort recruited in two French maternity hospitals, living less than 2 km away from a permanent monitoring station. NO2 levels during each trimester of pregnancy were estimated using a geostatistical model built by combining measures from measurement campaigns with a fine resolution and, to seasonalize these estimates, from permanent monitoring stations. Additionally, we report associations with the more classical approach relying only on the permanent monitoring station closest from the home address. Ultrasound measurements of biparietal diameter (BPD) were performed during each trimester of pregnancy. Associations between exposure and BPD or birth weight (BW) were assessed by linear regression adjusted for potential confounders and gestational age. Results: Compared to the lowest exposure tertile, NO2 levels in the highest tertile were associated with a mean BW lowered by 150 g (95% confidence interval, CI, −266; −34 g) as estimated by the geostatistical model and 110 g (95% CI, −203; −17 g) as estimated by the nearest permanent monitoring station. No clear association was found between NO2 levels estimated by both exposure models and BPD throughout the different trimesters of pregnancy. Conclusion: In spite of their moderate agreement in terms of exposure estimates, both exposure models were consistent in terms of association between estimated NO2 levels and fetal growth. In particular, first trimester exposure levels were more strongly associated with BW than second and third trimester exposures for both models. There was no evidence for an effect of exposure on early BPD measurements.