Sex-specific associations of a ferroalloy metal mixture with motor function in Italian adolescents

Abstract

Background: Motor function is critical for children's health, yet remains an understudied neurodevelopmental domain. Exposure to metals has been linked with motor function, but no study has examined the joint effects of metal mixtures.

Methods: We evaluated cross-sectional associations between a metal mixture and motor function among 569 adolescents (10-14 years old) living near the ferroalloy industry. Concentrations of blood lead, hair manganese, hair copper, and hair chromium were quantified using inductively coupled plasma mass spectrometry. Neuropsychologists administered multiple fine motor function assessments: pursuit aiming, finger tapping, visual reaction time (VRT), and subtests from the Luria Nebraska battery. We estimated associations between motor function and the metal mixture using quantile-based g-computation and multivariable linear regression, adjusting for child age, sex, and socioeconomic status. We explored sex-specific associations in stratified models.

Results: Associations between the metal mixture and motor function were mostly null but were modified by sex. We observed a beneficial association among females: a quartile increase in all metals in the mixture was associated with a 2.6% faster average response time on the VRT (95% confidence interval [CI] = -4.7%, -0.5%), driven by Cu and Cr. In contrast, this association was adverse among males (ß = 1.5% slower response time [95% CI = -0.7%, 3.9%]), driven by Cu and Mn.

Conclusions: Results suggest that males may be more susceptible to the adverse effects of metal exposure on motor function during adolescence than females. Future studies, particularly prospective study designs, are warranted to further understand the associations of metal mixtures with motor function.

Keywords: Adolescents; Metals; Mixtures; Motor function; Neurodevelopment.

Figure 1.

Adjusted associations between the metal mixture with adolescent motor function based on quantile g-computation models. Estimates reflect a change in motor function score per quartile increase in the mixture. Asterisk (*) indicates models where outcomes were natural log (ln)-transformed; effect estimates thus reflect percent change in motor function score per quartile increase in the metal mixture (calculated using the equation: ([exp(ß) − 1] × 100). Models including all participants are shown in black and were adjusted for child sex, child age, and family SES. Sex-stratified models were adjusted for child age and family SES.

Figure 2.

Sex-stratified associations between metals mixture and adolescent motor function based on multivariable linear models. Estimates reflect a one-unit change in motor function score per 1-SD increase in metal concentration. Asterisk (*) identifies models where outcomes were natural log (ln)-transformed; these estimates reflect percent change in motor function score per 1-SD increase in metal concentration (calculated using the equation: [(exp(ß) − 1) × 100]). Models with all participants were adjusted for child sex, child age, family SES, and other metals. Sex-stratified models were adjusted for child age, family SES, and other metals.