Associations between prenatal metal and metalloid mixtures in teeth and reductions in childhood lung function

Abstract

Background: Metal(oid)s have been cross-sectionally associated with lung function outcomes in childhood but there is limited data on their combined effects starting in utero. Child sex may further modify these effects.

Objective: Examine associations between in utero and early life exposure to metals assessed via novel dentine biomarkers and childhood lung function and explore effect modification by child sex.

Methods: Analyses included 291 children enrolled in the Programming Research in Obesity, Growth, Environment and Social Stressors (PROGRESS) study, a longitudinal birth cohort study in Mexico City. Weekly dentine levels of arsenic (As), cadmium (Cd), cobalt (Co), copper (Cu), manganese (Mn), nickel (Ni), and lead (Pb) were measured from 15 weeks pre-birth to 15 weeks post birth in deciduous children's teeth. Lung function was tested at ages 8-14 years and then modeled as age, height and sex adjusted z-scores. Associations were modeled using lagged weighted quantile sum (LWQS) regression to evaluate the potential for a time-varying mixture effect adjusting for maternal age and education at enrollment and exposure to environmental tobacco smoke in pregnancy. Models were also stratified by sex.

Results: We identified a window of susceptibility at 12-15 weeks pre-birth in which the metal mixture was associated with lower FVC z-scores in children aged 8-14 years. Cd and Mn were the largest contributors to the mixture effect (70 %). There was also some evidence of effect modification by sex, in which the mean weights and weighted correlations over the identified window was more evident in males when compared to females. In the male stratum, Cd, Mn and additionally Pb also dominated the mixture association.

Conclusions: Prenatal metal(oid) exposure was associated with lower lung function in childhood. These findings underscore the need to consider both mixtures and windows of susceptibility to fully elucidate effects of prenatal metal(oid) exposure on childhood lung function.

Keywords: Lagged weighted quantile sum regression; Metals; Mixture; Pediatric lung function; Prenatal.

Fig. 1.

Associations of time-varying perinatal metal(loid) mixture exposures and lung function outcomes at age 8–14 years. L-WQS constrained in the negative direction and associated weights for (a) FVC, (b) FEV₁, (c) FEV₁/FVC and (d) FEF_25–75% z-scores. The solid black line represents the smoothed, time-varying association between the metal(loid) mixture and each lung function outcome; gray shaded region indicates 95 % confidence limits. For FVC the relevant time window is highlighted in gray in the weight plot.

Fig. 2.

Time-varying weighted correlations for perinatal metal(loid) mixture exposures and FVC z-score at 8–14 years. The relevant time window is highlighted in gray.

Fig. 3.

Difference in time-varying associations for all lung function parameters between males and females.

Fig. 4.

Time-varying correlation between FVC z-score and metal mixture accounting for sex specific weights. The solid black line represents the smoothed, time-varying association between the metal(loid) mixture and FVC; gray shaded region indicates 95 % confidence limits.

Fig. 5.

Associations of time-varying perinatal metal(loid) mixture exposures and FVC z-score at age 8–14 years stratified by sex. (a) Time varying weights in females (b) time varying weights in males (c) weighted correlations for females and (d) weighted correlations for males. The relevant time window is highlighted in gray in the weight plots.