Combined effects of multiple metals on hearing loss: A Bayesian kernel machine regression approach

Ecotoxicol Environ Saf. 2022 Dec 1:247:114279. doi: 10.1016/j.ecoenv.2022.114279. Epub 2022 Nov 8.

Abstract

Background: Exposure to heavy elements is considered one of the risks of hearing loss. The combined effect of multiple metals on different hearing losses has not been extensively studied.

Methods: This study deals with the association between different metals (Ba, Cd, Co, Cs, Mo, Pb, Mn, Sn, Sb, Tl, W) and hearing loss in the 2013-2018 National Health and Nutrition Examination Survey (NHANES). Associations were estimated by a generalized linear regression model (GLM) adjusting for age, gender, race/ethnicity, educational level, marital status, drinking status, hypertension, diabetes, smoking status, noise exposure, body mass index, and income-poverty ratio. The joint effects of mixed exposure were assessed by weighted quantile sum (WQS) model and Bayesian kernel machine regression (BKMR). The effect of multiple metals on speech-frequency hearing loss (SFHL) and high-frequency hearing loss (HFHL) was further calculated. The odds ratio (OR) indicates the risk of every metal exposure.

Results: A total of 2205 adult participants were enrolled from NHANES. Cd, Pb, Sb and Sn had significant effects on total hearing loss. The WQS model found that mixed exposure to heavy elements was significantly positively associated with total hearing loss (OR: 1.136; 95% CI: 1.031, 1.253) after adjusting for various covariates. The ORs of mixed exposure in the SFHL and HFHL groups were 1.066 (95% CI: 0.994, 1.143) and 1.102 (95% CI: 1.013, 1.199), respectively. BKMR found a significant positive association between multiple metals and hearing loss. The results showed that there may be potential interactions between Cd, Pb and other metals.

Conclusions: Multiple metals have joint effects on hearing loss in the United States. The findings provide practitioners with important scientific evidence for possible interventions.

Keywords: Bayesian kernel machine regression; Cadmium; Generalized linear model; Hearing loss; Heavy element; Weighted quantile sum model.

MeSH terms

  • Adult
  • Bayes Theorem
  • Cadmium* / toxicity
  • Hearing Loss* / chemically induced
  • Hearing Loss* / epidemiology
  • Humans
  • Lead
  • Nutrition Surveys

Substances

  • Cadmium
  • Lead