Exposure to air pollution during preconceptional and prenatal periods and risk of hypertensive disorders of pregnancy: a retrospective cohort study in Seoul, Korea

Background: Previous studies suggested associations between prenatal exposure to air pollution and hypertensive disorders of pregnancy. We explored the associations between ambient concentrations of five major air pollutants during preconceptional and prenatal periods and three hypertensive disorders of pregnancy in Seoul, Korea, using a population-representative cohort. Methods: We obtained heath and demographic data of pregnant women residing in Seoul for 2002-2013 from the Korean National Health Insurance Service-National Sample Cohort. For mother's individual exposures to air pollution, we computed concentrations of particulate matter <= 10 mu m in diameter (PM10), nitrogen dioxide (NO2), carbon monoxide (CO), sulfur dioxide (SO2), and ozone (O-3) during 1, 3, 6, and 12 months to birth using regulatory monitoring data in Seoul. The associations between air pollution and hypertensive disorders were explored by using logistic regression models after adjusting for individual confounders. Results: Among 18,835 pregnant women in Seoul, 0.6, 0.5, and 0.4% of women developed gestational hypertension, preeclampsia, and preeclampsia requiring magnesium sulfate (Mg-preeclampsia), respectively. Although most odds ratios (ORs) were not statistically significant, we found increasing risk gradients with disease severity depending on the pollutant. There was the association between PM10 during 6 months to birth and gestational hypertension (OR for an interquartile range increase = 1.68 [95% confidence interval = 1.09-2.58]). NO2 and ozone during 12 and 1 month, respectively, before birth were associated with Mg-preeclampsia (1.43 [1.01-2.03], 1.53 [1.03-2.27]). Conclusions: We observed positive associations of exposure to some air pollutants before and during pregnancy with hypertensive disorders of pregnancy among the Korean general population. Future studies with refined exposure metrics should confirm our findings.