Effect Modification By Maternal Diabetes during Pregnancy on the Association between Regional Air Pollution Exposure and Autism Spectrum Disorder

Background: Recent studies found that both maternal diabetes during pregnancy and early life exposure to air pollution increase the risk of autism spectrum disorder (ASD). Potential common biological mechanisms of air pollution and maternal immune activation-related pregnancy complications such as diabetes may interact, including inflammation and systemic oxidative stress, which could adversely affect neurodevelopment. However, no study has evaluated the joint effects of maternal diabetes and air pollution on ASD.

Objectives: We evaluated ASD risk associated with prenatal and first year of life exposures to regional air pollution by maternal diabetes during pregnancy.

Methods: This retrospective cohort study included 246,420 singleton children born in Kaiser Permanente Southern California (KPSC) hospitals between 1999 and 2009. Children were followed from birth until age five. Using the local regulatory air monitoring data with inverse distance-weighted monthly averages interpolated to each geocoded residential birth address, we estimated the exposures to ambient ozone and other regional air pollutants, including particulate matter less than 2.5 μm in aerodynamic diameter (PM_2.5), PM less than 10 μm (PM₁₀), and nitrogen dioxide (NO₂) during each trimester and first year of life. Exposure-associated increases in the relative risks of ASD were estimated by hazard ratios (HRs) using Cox regression models to adjust for birth year, KPSC medical center service areas, and relevant maternal medical, pregnancy, sociodemographic, and child characteristics. Interactions of estimated exposures and a 4-level categorical variable for maternal diabetes (none, gestational diabetes mellitus (GDM) diagnosed <24 weeks’ gestation, GDM diagnosed ≥24 weeks’ gestation, and pre-existing type-2 diabetes) were tested. For each observed exposure window with statistically significant global interaction, HRs associated with the index pollutant exposure were estimated separately for each subcohort defined by the category of maternal diabetes during pregnancy.

Results: There were 2471 children diagnosed with ASD. Statistically significant interactions between maternal diabetes and ozone exposure during the first trimester (p=0.047) and first year of life (p=0.007) were observed. Risk of ASD was associated with ozone levels among mothers with GDM <24 weeks’ gestation with adjusted HRs per 15.7 ppb ozone 1.50 (95% CI, 1.08-2.09) during the first trimester and 2.01 (95% CI, 0.67-6.07) during the first year of life. No statistically significant associations of ASD with ozone were observed in offspring of pregnancies without diabetes, GDM diagnosed at ≥24 weeks’ gestation, or pre-gestational diabetes. No statistically significant interactions of diabetes were observed with PM_2.5, PM₁₀, and NO₂.

Conclusions: GDM diagnosed earlier in pregnancy may increase children’s susceptibility to prenatal and early life ozone associated ASD risk.