[Ambient PM2.5 during pregnancy and risk on preterm birth]

To investigate the association between ambient fine particulate matter with aerodynamic diameter less than 2.5 μm (PM2.5) and the risk on preterm birth.

A total of 1 882 pregnant women with local residency of Taiyuan city and underwent delivery at the First Hospital of Shanxi Medical University with the dates of conception between January 1 and December 31, 2013, were enrolled in the study. Information on general demographics, home address and history on pregnancy, lifestyle and related environmental factors were collected through in-person interview. Birth outcomes and maternal complications were abstracted from medical records. Data on the amount of daily average PM2.5 from 8 monitor points in Taiyuan city, between March 1, 2012 and December 31, 2013 were also collected. Individual exposure during pregnancy were calculated using the inverse-distance weighting method, based on home address. Multivariate unconditional logistic regression model was used to examine the associations among PM2.5 exposure, risk of preterm birth and related clinical subtypes.

The overall incidence of preterm birth was 8.21% (151/1 839)in 1 839 pregnant women. Exposure to ambient PM2.5 during the second week prior to delivery was associated with an increased risk of preterm birth (OR=1.087, 95% CI: 1.001-1.182 per 10 μg/m(3) increase) and mild preterm birth (OR=1.099, 95% CI: 1.007-1.200 per 10 μg/m(3)). Compared to data from the China Environmental Air Quality Standard, higher level of exposure (≥75 μg/m(3)) of PM2.5 during the second week before delivery was associated with an increased risk of preterm birth (OR=1.008, 95%CI: 1.000-1.017) but the association was mainly seen for mild preterm birth (OR=1.010, 95%CI: 1.001-1.018).

RESULTS from our study showed that exposure to high level of PM2.5 during late pregnancy would increase the risk of preterm birth. Future large studies are needed to examine the association by preterm clinical subtypes and to elucidate potential underlying mechanisms.