Background: Associations between most single metals and children's intelligence quotient (IQ) scores have been evaluated in previous studies. However, associations between multi-metal exposures and children's IQ scores have not been analyzed.
Objectives: To assess the joint effects of lead (Pb), manganese (Mn), antimony (Sb), tin (Sn) and titanium (Ti) co-exposure on children's IQ scores.
Methods: A prospective cohort study was conducted in Shimen and Huayuan, Hunan Province, China. Urine metals levels were measured by inductively coupled plasma mass spectrometry (ICP-MS) at baseline. Children's IQ scores were repeatedly measured at baseline and follow-up following the method of Raven's Standard Progressive Matrices (SPM) and standardized as z scores. We fitted linear regression models and Bayesian kernel machine regression (BKMR) models to investigate the associations of metal levels with children's IQ scores after adjusting for covariates.
Results: A total of 633 participants aged 7-10 years completed the survey. Urinary Pb (β = -0.028, P = 0.022) and urinary Ti (β = -0.0003, P = 0.001) were inversely associated with children's IQ scores. The BKMR analyses revealed significant negative overall effects of the five metals on children's IQ scores when all the metals were above their median levels, while significant positive associations were shown when all the metal concentrations were below their median levels. The model also showed negative trends of Sn and Ti on children's IQ. Furthermore, Ti and Sn had a synergistic relationship, with a decline in IQ score when Sn exposure was relatively high. The urinary Sn concentration was significantly higher but the urinary Ti concentration was significantly lower in participants from the Shimen area than in those from the Huayuan area. Decreasing trends of the overall effects were observed in both the Shimen and Huayuan areas.
Conclusion: Our findings revealed that multi-metal exposures caused a decline in children's IQ scores according to traditional linear regression models and the BKMR model. Our results provide some evidence of the association between multi-metal exposure and children's IQ. Meanwhile, interactions between multi-metal exposures on children's IQ should be given more attention.
Keywords: Bayesian kernel machine regression; Children; Heavy metals; Intelligence quotient.
Copyright © 2021 The Authors. Published by Elsevier Ltd.. All rights reserved.