Deciphering the influence of heavy metals on adverse cardiovascular events using machine learning.

This study investigates the association between chronic heavy metal exposure (lead [Pb], arsenic [As], cadmium [Cd]) and major adverse cardiovascular events (MACE) using data from 15,342 participants in the 2003-2018 NHANES survey. Utilizing eight machine learning algorithms (including logistic regression, Random Forest, XGBoost, and LightGBM) combined with SHapley Additive exPlanations (SHAP) analysis, we evaluated the impact of these metals on angina, myocardial infarction (MI), and heart failure (HF), while adjusting for demographic and socioeconomic confounders. Results identified cadmium as the predominant risk factor across all cardiovascular outcomes: RandomForest achieved optimal angina prediction, XGBoost highlighted Cd's strongest association with MI, and LightGBM confirmed Cd's dominance in HF risk stratification. Multi-class modeling integrating logistic regression and RandomForest revealed Cd exposure significantly elevated overall MACE risk, surpassing Pb and As in feature importance. Mechanistic insights suggest Cd exacerbates cardiovascular pathology by disrupting myocardial calcium signaling, amplifying oxidative stress, and promoting inflammation, cardiomyocyte apoptosis, and fibrosis. These findings underscore cadmium's critical role in cardiovascular disease pathogenesis and advocate for prioritized public health interventions targeting environmental cadmium exposure mitigation. The integration of interpretable machine learning models with large-scale epidemiological data provides a robust framework for identifying environmental determinants of chronic diseases. This study identifies cadmium as the predominant heavy metal associated with increased risk of major adverse cardiovascular events, providing critical insights that inform public health interventions aimed at reducing environmental cadmium exposure.