Association of prenatal exposure to PM_2.5 and NO₂ with gestational diabetes in Western New York

Background: Although many studies have examined the association between prenatal air pollution exposure and gestational diabetes (GDM), the relevant exposure windows remain inconclusive. We aim to examine the association between preconception and trimester-specific exposure to PM_2.5 and NO₂ and GDM risk and explore modifying effects of maternal age, pre-pregnancy body mass index (BMI), smoking, exercise during pregnancy, race and ethnicity, and neighborhood disadvantage.

Methods: Analyses included 192,508 birth records of singletons born to women without pre-existing diabetes in Western New York, 2004-2016. Daily PM_2.5 and NO₂ at 1-km² grids were estimated from ensemble-based models. We assigned each birth with exposures averaged in preconception and each trimester based on residential zip-codes. We used logistic regression to examine the associations and distributed lag models (DLMs) to explore the sensitive windows by month. Relative excess risk due to interaction (RERI) and multiplicative interaction terms were calculated.

Results: GDM was associated with PM_2.5 averaged in the first two trimesters (per 2.5 μg/m³: OR = 1.08, 95% CI: 1.01, 1.14) or from preconception to the second trimester (per 2.5 μg/m³: OR = 1.10, 95% CI: 1.03, 1.18). NO₂ exposure during each averaging period was associated with GDM risk (per 10 ppb, preconception: OR = 1.10, 95% CI: 1.06, 1.14; first trimester: OR = 1.12, 95% CI: 1.08, 1.16; second trimester: OR = 1.10, 95% CI: 1.06, 1.14). In DLMs, sensitive windows were identified in the 5^th and 6^th gestational months for PM_2.5 and one month before and three months after conception for NO₂. Evidence of interaction was identified for pre-pregnancy BMI with PM_2.5 (P-for-interaction = 0.023; RERI = 0.21, 95% CI: 0.10, 0.33) and with NO₂ (P-for-interaction = 0.164; RERI = 0.16, 95% CI: 0.04, 0.27).

Conclusion: PM_2.5 and NO₂ exposure may increase GDM risk, and sensitive windows may be the late second trimester for PM_2.5 and periconception for NO₂. Women with higher pre-pregnancy BMI may be more susceptible to exposure effects.