Associations of exposure to cotinine, polycyclic aromatic hydrocarbon, and metal(loid) mixtures during pregnancy with birth weight: the MIEHR study

BACKGROUND: Prenatal exposure to chemical mixtures may have greater fetal health impacts than single chemicals due to complex, interacting biological mechanisms. Hydroxylated polycyclic aromatic hydrocarbon metabolites (OH-PAHs) and metals/metalloids share exposure sources, yet evidence of their combined effects on perinatal outcomes is limited. OBJECTIVE: To examine associations between individual cotinine, OH-PAHs, and metals/metalloids and birth weight, and to investigate their overall and interactive effects. METHODS: We included 1120 mothers from the Maternal and Infant Environmental Health Riskscape (MIEHR) study in Houston, Texas. A large suite of chemical constituents, including cotinine, six OH-PAHs, and 18 metals/metalloids, was measured in spot urine samples collected at delivery. Birth weight, abstracted from medical records, was converted into z-scores standardized by gestational age and sex. Linear regression was employed to examine associations between each constituent and birth weight, adjusted for covariates. Bayesian kernel machine regression (BKMR) was used to evaluate individual, joint, and interactive effects of multiclass chemical mixture on birth weight. RESULTS: Based on the BMKR analysis, no overall mixture effect on birth weight was observed. The exposure-response functions showed no clear patterns between OH-PAHs and birth weight. After accounting for the other constituents in the mixture, molybdenum, thallium, and strontium were negatively associated with birth weight, whereas barium and tungsten were positively associated. In comparing the 75th versus 25th percentiles of the distribution with other constituents fixed at their median levels, the strongest associations were observed for molybdenum and barium with a change in birth weight z-scores of -0.12 (95% credible interval: -0.23, -0.01) and 0.14 (0.02, 0.27), respectively. No interactions between individual mixture constituents were identified. SIGNIFICANCE: In the MIEHR study, with relatively lower urinary OH-PAH and metals/metalloids concentrations compared with women from other U.S. pregnancy cohorts, molybdenum was associated with decreased birth weight, and barium with increased birth weight. Given that both metals are relatively understudied, additional investigation is warranted. IMPACT: Few studies have examined associations between prenatal exposure to mixtures of PAH metabolites and birth weight, particularly in combination with metals/metalloids, despite shared exposure sources and potential biological mechanisms. We found heterogeneous associations for specific PAH metabolites and metals, most notably molybdenum and barium, revealing complex environmental chemical effects on birth weight within a mixture framework. Given pregnant women's simultaneous exposure to sources from industry and traffic and beneficial trace elements, our findings underscore the need to consider co-exposure effects to better integrate the effects of constituents on infant health and inform public health surveillance.