Metal mixture exposure, metabolic networks, and hemoglobin reduction: A population-based metabolomics study

Epidemiological evidence has associated metal exposure with altered hemoglobin levels and anemia, but the underlying mechanisms remain unclear. This study evaluated associations of individual metals and metal mixtures with hemoglobin levels and anemia risk, and investigated the potential contributions of metabolites using untargeted metabolomics.

In 2939 participants from Zhejiang, China (2021-2023), we measured 14 urinary metals. Generalized linear and logistic regression models assessed individual metal associations with hemoglobin and anemia, respectively, while weighted quantile sum (WQS) regression evaluated mixture effects. In a subset of 712 participants, weighted gene co-expression network analysis (WGCNA) identified metabolic modules associated with metal exposure, hemoglobin levels, and anemia. Mediation analysis was conducted to explore whether these modules may contribute to the observed associations.

After false discovery rate (FDR) correction, 12 metals were significantly associated with decreased hemoglobin levels, with cobalt and cadmium additionally associated with increased anemia risk. WQS regression indicated that each quartile increase in the metal mixture index was associated with a 3.29 g/L (95% CI: -4.04, -2.54) reduction in hemoglobin and an odds ratio of 1.93 (95% CI: 1.42, 2.63) for anemia, with cobalt contributing the largest mixture weight (45.8% for hemoglobin and 60.8% for anemia). WGCNA identified 14 metabolic modules, among which modules related to steroid metabolism and branched-chain amino acid (BCAA) metabolism were significantly associated with both metal exposure and hematological outcomes. Mediation analysis indicated that these modules may contribute to the associations between metal exposure and lower hemoglobin levels and higher anemia risk.

Metal exposure, particularly to cobalt, is associated with reduced hemoglobin levels and increased anemia risk in the general population. These associations may be partly attributable to metabolic disturbances in steroid and BCAA pathways, offering novel metabolomic insights into metal-related anemia.