Prenatal Exposure to Polycyclic Aromatic Hydrocarbons and Gestational Age at Birth

Thesis (Master's)--University of Washington, 2021Background. Polycyclic aromatic hydrocarbons (PAHs) are chemical compounds that are ubiquitous, probable human carcinogens. Maternal PAH exposure has potential effects on fetal development. We examined associations of prenatal PAH exposure with spontaneous preterm birth (PTB) and gestational age at birth among term births. Given the sex-specific endocrine disrupting properties of PAHs, we additionally examined the extent to which infant sex modifies the association of PAH exposure with spontaneous PTB and gestational age at birth. Methods. Study participants included 1,677 non-smoking women with singleton pregnancies from three cohorts (CANDLE, TIDES, and GAPPS) in the ECHO PATHWAYS Consortium. Seven PAH metabolites (1-hydroxynaphthalene [1-nap], 2-hydroxynaphthalene [2-nap], 2-hydroxyphenanthrene [2-phen], 3-hydroxyphenanthrene [3-phen], combined 1- and 9-hydroxyphenanthrene [1/9-phen], combined 2-, 3-, and 9-hydroxyfluorene [2/3/9-fluo], and 1-hydroxypyrene [1-pyr]) were measured in second trimester maternal urine. PAH values below the limit of detection were imputed using censored likelihood multiple imputation. Logistic and linear regression models were fit for spontaneous PTB and gestational age (days, among terms births), respectively, with individual log10-transformed PAH metabolites as the exposure, adjusted for specific gravity, maternal age, site, race, urinary cotinine, education, income, season of birth, infant sex, alcohol use during pregnancy, pre-pregnancy BMI, parity, and prior PTB. Effect modification by infant sex was assessed using interaction terms and comparison of marginal sex-specific estimates. Results. Most participants were between the ages of 18-29 (N=827, 49.3%) or 30-39 (N=762, 45.4%) and were high school (N=540, 32.2%) or college graduates (N=570, 34.0%). PAH levels varied by cohort and metabolite, with CANDLE having the highest mean levels of all metabolites overall. Overall percent detection was highest for 2-nap (99.8%) and the lowest for 1-pyr (65.2%). The overall prevalence of spontaneous PTB was 5.5% (overall PTB: 9.9%) and cohort-specific prevalence of spontaneous PTB were 6.0%, 3.5% and 8.9% for CANDLE, TIDES, and GAPPS, respectively (overall PTB: 9.7%, 10.4%, 9.4%). Among term births, the mean gestational age was 39.5 weeks (SD= 1.09), with similar means observed across cohorts. We did not find statistically significant associations between PAH exposure and spontaneous PTB, although there was suggestive evidence of a an inverse association with 2-nap. A ten-fold higher exposure to 2-nap was associated with a 1.13-day (95% CI: -2.19, -0.13) lower gestational age at birth. Associations of 1-pyr with PTB and 2-nap and 2/3/9-fluo with gestational age were significant among females (1-pyr: OR=1.96 [95% CI: 1.11, 3.46]; 2-nap: -2.13 days [95% CI: -3.41, -0.85], 2/3/9-fluo: -1.55 days [95% CI: -2.96, -0.14]) but not males (1-pyr: OR=0.99 [95% CI: 0.60, 1.62]; 2-nap: -0.28 [95% CI: -1.50, 0.93]; 2/3/9-fluo: -1.55 [95% CI: -2.96, -0.14]) (interaction term p-values <0.05). Other interactions (interaction term p-values