Mediation of the Relationship between Phthalate Exposure and Preterm Birth by Oxidative Stress

Abstract

Mediation of the Relationship between Phthalate Exposure and Preterm Birth by Oxidative StressAbstract Number:2475 Kelly Ferguson*, Thomas McElrath, Yin-Hsiu Chen, Bhramar Mukherjee, John Meeker Kelly Ferguson* University of Michigan, United States, E-mail Address: [email protected] Search for more papers by this author , Thomas McElrath Brigham and Women's Hospital, United States, E-mail Address: [email protected] Search for more papers by this author , Yin-Hsiu Chen University of Michigan, United States, E-mail Address: [email protected] Search for more papers by this author , Bhramar Mukherjee University of Michigan, United States, E-mail Address: [email protected] Search for more papers by this author , and John Meeker University of Michigan, United States, E-mail Address: [email protected] Search for more papers by this author AbstractPhthalate exposure has been linked to preterm birth in a number of studies, but the underlying mechanism remains unclear. We examined the role of oxidative stress as a mediator in the relationship between phthalate exposure across pregnancy and preterm birth in a nested case-control study of women selected from a prospective birth cohort at Brigham and Women’s Hospital in Boston, MA (N=130 cases, 352 controls). Phthalate metabolites were measured in urine samples collected from multiple study visits (median 3.5 samples per subject). Oxidative stress was assessed by 8-isoprostane, a marker of lipid peroxidation, also measured in repeated urine samples. Subject-specific averages of phthalate metabolites and of 8-isoprostane were created for more stable measures of phthalate exposure and oxidative stress during pregnancy. Preterm birth was defined as delivery before 37 weeks, and we also examined spontaneous preterm births alone, as phthalate exposure was more strongly associated with this subset of cases. We established that: 1) phthalate metabolite levels were associated with increased odds of preterm birth; 2) phthalate metabolite levels were associated with increased 8- isoprostane during pregnancy; and 3) 8-isoprostane concentrations were associated with increased odds of preterm birth after adjustment for phthalate metabolite in the same model. We performed a mediation analysis to estimate the proportion of the phthalate- preterm relationship that was mediated by 8-isoprostane. Associations between di-2- ethylhexyl phthalate (DEHP) metabolites and overall preterm birth were 21-29% mediated by 8-isoprostane. Mediation was stronger for phthalate associations with spontaneous preterm birth (ranging between 24-46% for DEHP metabolites). This study is the first to identify through mediation analysis a mechanism between an environmental chemical exposure and preterm birth, and future use of this analysis may be particularly useful as animal models of prematurity are poor.