Exploring the association between volatile organic compound exposure and chronic kidney disease: evidence from...

Area of Science:

• Environmental Health
• Metabolomics
• Nephrology

Background:

• Chronic Kidney Disease (CKD) affects nearly 700 million globally, with environmental factors like volatile organic compounds (VOCs) increasingly recognized as potential contributors.
• Traditional linear methods may underestimate the complex interplay between VOCs and CKD, necessitating advanced analytical approaches.

Purpose of the Study:

• To investigate the association between specific urinary VOC metabolites and the risk of developing CKD.
• To apply interpretable machine learning models to identify key VOC predictors of CKD risk.

Main Methods:

• Analysis of 15 urinary VOC metabolites from the National Health and Nutrition Examination Survey (2011-2020).
• Utilized multivariable logistic regression, LASSO, and machine learning models (LR, RF, XGBoost, KNN, MLP).
• Employed SHapley Additive exPlanations (SHAP) for enhanced model interpretability.

Main Results:

• Significant positive associations found for CEMA, DHBMA, HMPMA, and PGA with CKD risk.
• The XGBoost model showed strong predictive power, with DHBMA identified as a primary predictor via SHAP analysis.
• Inverse associations were noted for AAMA and CYMA in higher concentrations.

Conclusions:

• Specific urinary VOC metabolites, notably DHBMA, are significantly associated with CKD risk.
• Environmental VOC exposure plays a role in CKD pathogenesis.
• Findings may guide the development of targeted prevention strategies for CKD.