Associations of ozone exposure with gestational diabetes mellitus and glucose homeostasis: Evidence from a birth cohort in Shanghai, China

Associations between individual exposure to ozone (O3) and gestational diabetes mellitus (GDM) have rarely been investigated, and critical windows of O3 exposure for GDM have not been identified. We aimed to explore the associations of gestational O3 exposure with GDM and glucose homeostasis as well as to identify the potential critical windows. A total of 7834 pregnant women were included. Individual O3 exposure concentrations were evaluated using a high temporal-spatial resolution model. Each participant underwent an oral glucose tolerance test (OGTT) to screen for GDM between 24 and 28 gestational weeks. Multiple logistic and multiple linear regression models were used to estimate the associations of O3 with GDM risks and with blood glucose levels of OGTT, respectively. Distributed lag nonlinear models (DLNMs) were used to estimate the critical windows of O3 exposure for GDM. Nearly 13.29 % of participants developed GDM. After controlling for covariates, we observed increased GDM risks per IQR increment of O3 exposure in the first trimester (OR = 1.738, 95 % CI: 1.002–3.016) and the first two trimesters (OR = 1.576, 95 % CI: 1.005–2.473). Gestational O3 exposure was positively associated with increased fasting blood glucose (the first trimester: β = 2.964, 95 % CI: 1.529–4.398; the first two trimesters: β = 1.620, 95 % CI: 0.436–2.804) and 2 h blood glucose (the first trimester: β = 6.569, 95 % CI: 1.775–11.363; the first two trimesters: β = 6.839, 95 % CI: 2.896–10.782). We also observed a concentration–response relationship of gestational O3 exposure with GDM risk, as well as fasting and 2 h blood glucose levels. Additionally, 5–10 gestational weeks was identified as a critical window of O3 exposure for GDM development. In summary, we found that gestational O3 exposure disrupts glucose homeostasis and increases the risk of GDM in pregnant women. Furthermore, 5–10 gestational weeks could be a critical window for the effects of O3 exposure on GDM. [Display omitted] •O3 exposure during first trimester and the first two trimester was associated with increased GDM risk.•O3 exposure was positively associated with increased fasting and 2-h blood glucose levels.•Gestational 5–10 weeks could be the critical windows of GDM for O3 exposure.