Comprehensive Cross-Sectional Study of the Triglyceride Glucose Index, Organophosphate Pesticide Exposure, and Cardiovascular Diseases: A Machine Learning Integrated Approach.

Abstract

Using NHANES data from 2003 to 2008, 2011 to 2012, and 2015 to 2020, we examined the relationship between urinary organophosphate pesticide (OPP) metabolites and the triglyceride glucose (TyG) index. The TyG index evaluates insulin resistance, a crucial factor in metabolic diseases. Linear regression analyzed urinary metabolites in relation to the TyG index and OPPs. An RCS (restricted cubic spline) model explored the nonlinear relationship of a single OPP metabolite to TyG. A weighted quantile regression and quantile-based g-computation assessed the impact of combined OPP exposure on the TyG index. XGBoost, Random Forest, Support Vector Machines, logistic regression, and SHapley Additive exPlanations models investigated the impact of OPPs on the TyG index and cardiovascular disease. Network toxicology identified CVD targets associated with OPPs. This study included 4429 participants based on specific criteria. Linear regression analysis indicated that diethyl thiophosphate was positively correlated with the TyG index. The positive correlation between OPP metabolites and the TyG index at low to moderate concentrations was confirmed by WQS and QGC analyses. The machine learning results aligned with traditional statistical findings. Network toxicology identified PTGS3, PPARG, HSP40AA1, and CXCL8 as targets influenced by OPPs. OPP exposure influences IR and cardiometabolic health, highlighting the importance of public health prevention.