Serum Polychlorinated Biphenyl Burden and Prevalent Cardiovascular Disease in U.S. Adults: A Cross-Sectional Analysis Using Evidence from NHANES 1999–2002

Abstract

Background: Polychlorinated biphenyls (PCBs) persist in the environment and have been associated with cardiovascular disease (CVD). We examined whether routine cardiometabolic biomarkers statistically accounted for this association. Methods: This cross-sectional analysis used NHANES 1999–2002 data for 2815 adults aged 20 years or older. Lipid-adjusted total PCB burden was the sum of 21 congeners. Survey-weighted logistic regression estimated odds ratios for prevalent CVD, restricted cubic splines examined the exposure–response shape, and mediator-specific structural equation models evaluated indirect and residual pathways through CRP, HbA1c, waist circumference, central obesity, the total cholesterol/HDL ratio, and a binary high-ratio measure. Results: Weighted mean total PCB burden was 234.6 ng/g lipid, and weighted CVD prevalence was 8.9%. A 1-unit increase in the natural log of total PCB burden was associated with higher odds of prevalent CVD after full adjustment (aOR = 1.36, 95% CI: 1.01–1.84). Coronary heart disease and angina showed the clearest subtype associations. The spline analysis did not detect nonlinearity. Indirect estimates through CRP, waist circumference, and the TC/HDL ratio were negative, which is consistent with statistical suppression; estimates through the other biomarkers were small and imprecise. Conclusions: Higher serum PCB burden was associated with prevalent CVD after adjustment for major demographic and cardiometabolic factors. In cross-sectional pathway models, routine biomarkers did not account for most of this association. Longitudinal studies are needed before these statistical pathways can be interpreted as causal mechanisms.