A negative combined effect of exposure to maternal Mn-Cu-Rb-Fe metal mixtures on gestational anemia, and the mediating role of creatinine in the Guangxi Birth Cohort Study (GBCS): Twelve machine learning algorithms.

Journal: Ecotoxicology and environmental safety
Published Date:

Abstract

The link between individual metals and gestational anemia has been established, but the impact of metal mixtures and the mediating role of renal function on gestational anemia remain inconclusive. The concentrations of 20 blood essential trace and nonessential metals and 7 serum kidney function indicators were measured among 2000 pregnant women from the Guangxi Birth Cohort Study. Maternal hemoglobin < 110 g/L and hematocrit < 0.33 were defined as gestational anemia. We utilized twelve machine learning (ML) algorithms to independently screen for effective metal mixtures, assess their combined impacts and dose-response relationships on gestational anemia, and estimate the mediating role of kidney function. In the total population, manganese (Mn), copper (Cu), rubidium (Rb), and iron (Fe) were identified as significant metals with independent effects on gestational anemia by seven ML algorithms. The results of the Mn-Cu-Rb-Fe metal mixture ML models revealed that Rb (P = 0.008) and Fe (P < 0.001) were linearly and nonlinearly negatively associated with gestational anemia, respectively, whereas Cu (P = 0.069) showed a borderline positive association. The results for the Mn-Cu-Rb-Fe metal mixtures from both the first and second trimesters were consistent with their significance in the total population. Moreover, the protective effect of Rb increased while that of Fe decreased as pregnancy progressed; simultaneously, the risk effect of Cu diminished. In the third trimester, linear positive combined effects of tin (Sn) (P = 0.002) and cadmium (Cd) (P = 0.004) on gestational anemia were observed in the Sn-Rb-vanadium-Cd-Fe metal mixture ML models. Furthermore, we found that creatinine mediated the association between Fe and gestational anemia in the second trimester (mediated proportion = 3.7 %, P = 0.032). Hence, exposure to Mn-Cu-Rb-Fe mixtures inversely correlated with gestational anemia, which is mediated by creatinine. Sustained Rb and Fe supplementation may avert anemia and ameliorate metal toxicity.

Authors

  • Yuen Zhong
    School of Public Health, Guangxi Medical University, Nanning, Guangxi, China; Center for Genomic and Personalized Medicine, Guangxi Key Laboratory for Genomic and Personalized Medicine, Guangxi Collaborative Innovation Center for Genomic and Personalized Medicine, Guangxi Medical University, Nanning, Guangxi, China.
  • Yu Bao
    Laboratory of Mathematical Bioinformatics, Bioinformatics Center, Institute for Chemical Research, Kyoto University, Gokasho, Uji, Kyoto, 611-0011, Japan. houu@kuicr.kyoto-u.ac.jp.
  • Hong Cheng
    Department of Neurology, First Affiliated Hospital of Nanjing Medical University, Nanjing, 210029, China. ch8706@sohu.com.
  • Chaoqun Liu
    School of Public Health, Guangxi Medical University, Nanning, Guangxi, China.
  • Shengzhu Huang
    Center for Genomic and Personalized Medicine, Guangxi Key Laboratory for Genomic and Personalized Medicine, Guangxi Collaborative Innovation Center for Genomic and Personalized Medicine, Guangxi Medical University, Nanning, Guangxi, China.
  • Hualong Qiu
    Guangdong Provincial Key Laboratory of Silviculture, Protection and Utilization, Guangdong Academy of Forestry, Guangzhou, China.
  • Honglin Huang
    School of Public Health, Guangxi Medical University, Nanning, Guangxi, China; Center for Genomic and Personalized Medicine, Guangxi Key Laboratory for Genomic and Personalized Medicine, Guangxi Collaborative Innovation Center for Genomic and Personalized Medicine, Guangxi Medical University, Nanning, Guangxi, China.
  • Jiajun Ren
    School of Instrumentation and Opto-Electronic Engineering, Beihang University, Beijing, China.
  • Hailiu Jin
    School of Public Health, Guangxi Medical University, Nanning, Guangxi, China; Center for Genomic and Personalized Medicine, Guangxi Key Laboratory for Genomic and Personalized Medicine, Guangxi Collaborative Innovation Center for Genomic and Personalized Medicine, Guangxi Medical University, Nanning, Guangxi, China.
  • Caitong He
    Department of science and education, Maternal & Child Health Hospital of Yulin, Yulin, Guangxi, China.
  • Long Tian
    Department of science and education, Maternal & Child Health Hospital of Qinzhou, Qinzhou, Guangxi, China.
  • Yu Zhang
    College of Marine Electrical Engineering, Dalian Maritime University, Dalian, China.
  • Bangzhu Luo
    Department of Medical Services Section, Maternal & Child Health Hospital of Guigang, Guigang, Guangxi, China.
  • Tao Liang
    Department of Pharmaceutical Sciences, University of Maryland School of Pharmacy, Baltimore, MD, 21201, USA.
  • Mujun Li
    Guangxi Reproductive Medical Center, The First Affiliated Hospital of Guangxi Medical University, Nanning, China.
  • Zengnan Mo
    Center for Genomic and Personalized Medicine, Guangxi Medical University, Nanning, Guangxi, China.
  • Longman Li
    School of Public Health, Guangxi Medical University, Nanning, Guangxi, China; Center for Genomic and Personalized Medicine, Guangxi Key Laboratory for Genomic and Personalized Medicine, Guangxi Collaborative Innovation Center for Genomic and Personalized Medicine, Guangxi Medical University, Nanning, Guangxi, China. Electronic address: lilongman@gxmu.edu.cn.
  • Xiaobo Yang
    School of Public Health, Guangxi Medical University, Nanning, Guangxi, China.