Association of urinary metal elements with sarcopenia and glucose metabolism abnormalities: Insights from NHANES data using machine learning approaches

Background: Sarcopenia, a condition marked by the decline of skeletal muscle mass and function, is prevalent in the elderly and closely linked to abnormal glucose metabolism, particularly type 2 diabetes. Hyperglycemia can increase the formation of advanced glycation end-products (AGEs) in muscle pr...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinmin Jin, Lei Li, Xiaoyan Hu, Pengfei Bi, Song Zhang, Qian Wang, Zhongwei Xiao, Hua Yang, Tongtong Liu, Lifang Feng, Jinhuan Wang
Format: Article
Language:English
Published: Elsevier 2025-07-01
Series:Ecotoxicology and Environmental Safety
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0147651325008097
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Background: Sarcopenia, a condition marked by the decline of skeletal muscle mass and function, is prevalent in the elderly and closely linked to abnormal glucose metabolism, particularly type 2 diabetes. Hyperglycemia can increase the formation of advanced glycation end-products (AGEs) in muscle proteins, impairing muscle function. Additionally, deficiencies in trace minerals are associated with the development of sarcopenia. Objectives: This study aimed to explore the association between urinary metal element levels and sarcopenia across different glucose metabolic states using multi-omics clustering algorithms and machine learning models, and to identify diagnostic biomarkers. Methods: Data from the 2011–2014 National Health and Nutrition Examination Survey (NHANES) were used, involving 2390 participants with complete data on urinary metal elements, diabetes, and sarcopenia. Sarcopenia was diagnosed based on established criteria, and diabetes and prediabetes were classified using glycemic thresholds. Participants were stratified into subgroups via multi-omics clustering algorithms (iClusterBayes, moCluster, etc.) within the MOVICS package. Machine learning models (Lasso, RandomForest, etc.) were applied to identify diagnostic metal biomarkers. Associations between key features and sarcopenia were assessed using weighted logistic regression, subgroup analyses, and restricted cubic spline (RCS) analysis. Results: Participants were divided into two subgroups based on urinary metal concentrations. Subgroup 2 showed a significantly higher prevalence of sarcopenia (P < 0.05) in both the overall cohort and diabetes-specific populations. Machine learning models identified four diagnostic biomarkers—urinary lead, dimethylarsinic acid, total arsenic, and molybdenum—that were significantly associated with sarcopenia in prediabetes (IFG/IGT) and diabetes groups (AUC = 0.998–1.0). RCS analysis revealed a nonlinear relationship between urinary lead and sarcopenia incidence, particularly in diabetic individuals (P < 0.05). No significant associations were found in normoglycemic individuals. Conclusion: The study identified urinary lead, dimethylarsinic acid, total arsenic, and molybdenum as key biomarkers for sarcopenia in diabetic and prediabetic populations, with machine learning models demonstrating high diagnostic accuracy. The findings highlight the metabolic state-specific relationship between metal exposure and sarcopenia, suggesting the need for targeted screening strategies in high-risk groups.
ISSN:0147-6513