Associations of Prenatal Exposures to PM2.5 Chemical Constituents and Sources with Stillbirth

Abstract

Background/Aim: The stillbirth rate in the United States is relatively high among developed countries, and it has remained almost unchanged since 2006. Evidence is limited and inconsistent concerning associations between stillbirth and fine particulate matter (PM2.5), its chemical constituents and sources. We estimated the risk of stillbirth associated with prenatal exposures to PM2.5 total mass, chemical constituents and sources. Methods:PM2.5 total mass and twenty chemical constituents were collected from ambient monitors, and five PM2.5 sources were quantified using Positive Matrix Factorization. A case-control study (N=32,262), matched by fetal sex and maternal characteristics (race/ethnicity, age, educational attainment), was conducted using records of livebirths and stillbirths collected from eight locations throughout California from 2002 to 2009. Gestational age-adjusted exposure levels were calculated for each pollutant. Conditional logistic regression was applied to estimate the risk of stillbirth associated with pollutants. As sensitivity analysis, we explored models stratified by gestational length of stillbirth: 20 to 27 weeks (intermediate fetal death) and after 28 weeks (late fetal death). Results: An interquartile range increase in gestational exposure to PM2.5 total mass was associated with an odds ratio (OR) of 1.06 (95% confidence interval: 1.01-1.11) times higher risk of stillbirth. Similarly, associations were found with secondary ammonium sulfate [OR: 1.13 (1.06-1.21)], resuspended soil [OR: 1.11 (1.06-1.15)], and vehicular emissions [OR: 1.06 (1.01-1.12)]. Associations were also found with PM2.5 aluminum, elemental carbon, iron, ammonium ion, nitrate ion, silicon, sulfate ion, titanium, vanadium, and zinc. The results were similar when stillbirths were categorized as intermediate fetal death or late fetal death. Conclusions: Stillbirth was associated with some PM2.5 sources and/or chemical constituents. The findings suggest that certain PM2.5 sources and/or constituents are more toxic than others. Our findings will aid further understanding of chemical constituents and regulating specific emissions sources to prevent stillbirths.