Prenatal heavy metal exposures and atopic dermatitis with gender difference in 6-month-old infants using multipollutant analysis

Abstract

Background: Prenatal exposure to heavy metals during critical developmental phases has been implicated in allergic phenotypes. However, few studies have been conducted on the gender-specific association of prenatal heavy metal exposure with atopic dermatitis (AD) in infants. Objective: To examine the gender-specific association of prenatal exposure to multiple heavy metals with AD incidence in 6-month-old infants using data from the Mothers and Children's Environmental Health (MOCEH). Methods: We evaluated 738 mother–child pairs from the MOCEH study, an ongoing prospective birth cohort. The concentrations of three heavy metals (lead, mercury and cadmium) in maternal blood samples were measured during early and late pregnancy. Each quartile of heavy metal concentration was used to consider the possible nonlinear association with AD. For assessing the multi-pollutant model, we constructed the multivariate regression model including all three heavy metals at both early and late pregnancy. Further, the group Lasso model was used to perform the variable selection with categorized exposures and assess the effect of multiple pollutants including their pairwise interactions. Results: A total of 200 incident cases of AD were diagnosed in 6-month-old infants. In the multivariate regression model of the boy group, adjusted odds ratios comparing the second, third and fourth quartile of lead exposure in boys with the first quartile were 1.83 (95% CI: 1.00, 3.38), 1.04 (0.91, 3.32) and 2.40 (1.18, 4.90), respectively. However, the only second quartile of lead exposure compared to first quartile was significantly associated with AD in girls. In addition, the results of the group Lasso model were similar with the results of multivariate regression model. Conclusion: The results suggest that lead exposure in late pregnancy increases risk of AD in 6-month-old boys although the strength of association is weak. Further studies are needed to confirm the susceptibility window and gender differences in lead-induced AD. © 2021 Elsevier Inc.