Maternal particulate matter exposure and gestational diabetes mellitus: a population–based cohort study

BACKGROUND: Gestational diabetes mellitus prevalence is steadily increasing worldwide, posing a significant threat to the short-term and long-term health of both mother and offspring. Because particulate matter air pollution has been reported to affect glucose metabolism, it was suggested that maternal particulate matter exposure may be associated with the development of gestational diabetes mellitus; however, the evidence is limited and inconsistent. OBJECTIVE: This study aimed to determine the association between maternal exposure to particulate matter of diameter ≤2.5 µm and of diameter of ≤10 µm and the risk of gestational diabetes mellitus, to identify critical windows of susceptibility and to evaluate effect modification by ethnicity. STUDY DESIGN: A retrospective cohort study was conducted including pregnancies of women who delivered at a large tertiary medical center in Israel between 2003 and 2015. Residential particulate matter levels were estimated by a hybrid spatiotemporally resolved satellite-based model at 1 km spatial resolution. Multivariable logistic analyses were applied to study the association between maternal particulate matter exposure in different pregnancy periods and gestational diabetes mellitus risk, while adjusting for background, obstetrical, and pregnancy characteristics. Analyses were also stratified by ethnicity (Jewish and Bedouin). RESULTS: The study included 89,150 pregnancies, of which 3245 (3.6%) were diagnosed with gestational diabetes mellitus. First trimester exposure to both particulate matter of diameter ≤2.5 µm (adjusted odds ratio per 5 μg/m³, 1.09; 95% confidence interval, 1.02–1.17) and particulate matter of diameter of ≤10 µm (adjusted odds ratio per 10 μg/m³, 1.11; 95% confidence interval, 1.06–1.17) was significantly associated with increased risk for gestational diabetes mellitus. In the stratified analyses, the association with first trimester particulate matter of diameter of ≤10 µm exposure was consistent among pregnancies of both Jewish and Bedouin women, whereas the association with first trimester particulate matter of diameter ≤2.5 µm exposure was significant among pregnancies of Jewish women only (adjusted odds ratio per 5 μg/m³, 1.09; 95% confidence interval, 1.00–1.19), as well as association with preconception particulate matter of diameter of ≤10 µm exposure (adjusted odds ratio per 10 μg/m³, 1.07; 95% confidence interval, 1.01–1.14). No association was found between second trimester particulate matter exposure and gestational diabetes mellitus risk. CONCLUSION: Maternal exposure to both particulate matter of diameter ≤2.5 µm and diameter of 10 µm or less during the first trimester of pregnancy is associated with gestational diabetes mellitus, suggesting that the first trimester is a particular window of susceptibility to the effect of particulate matter exposure on gestational diabetes mellitus risk. The effects found in this study differed by ethnic group, emphasizing the importance of addressing ethnic disparities when assessing environmental impacts on health.