Abstract
BACKGROUND: Previous studies on the associations of per- and polyfluoroalkyl substances (PFASs) and heavy metals with lipid profiles among adolescents have been scarce. We sought to investigate the associations of PFASs and heavy metals with blood lipid levels in a representative sample of Korean adolescents. METHODS: Data from the Korean National Environmental Health Survey (2018-2020) were used. Concentrations of PFASs [perfluorooctanoic acid (PFOA), perfluorooctane sulfonic acid (PFOS), perfluorohexane sulfonic acid, perfluorononanoic acid (PFNA), and perfluorodecanoic acid (PFDeA)], lead, and mercury were measured in serum, whole blood, and urine samples, respectively. Linear regression, Bayesian kernel machine regression (BKMR), and k-means clustering analyses were employed to evaluate the associations between pollutants and lipid levels. RESULTS: In the linear regression analyses, PFOA levels were associated with higher low-density lipoprotein cholesterol (LDL-C) levels; PFOS with higher total cholesterol (TC) levels; PFNA with higher TC, LDL-C, and non-high-density lipoprotein cholesterol (non-HDL-C) levels; PFDeA with higher TC, LDL-C, non-HDL-C, and high-density lipoprotein cholesterol levels; and mercury with higher TC and non-HDL-C levels. The BKMR analysis revealed that the PFAS and heavy metal mixture was associated with higher LDL-C levels (1.8% increase in LDL-C at the 75th percentile of all PFAS and heavy metal concentrations compared to their median values, 95% credible interval: 0.5, 3.1), primarily driven by the effect of PFDeA. Compared to individuals in the low pollutant exposure cluster (geometric mean levels of PFOA, PFOS, PFHxS, PFNA, PFDeA, lead, and mercury were 2.7 μg/L, 6.2 μg/L, 1.6 μg/L, 0.7 μg/L, 0.4 μg/L, 0.8 μg/dL, and 0.3 μg/L, respectively), those in the high pollutant exposure cluster (5.1 μg/L, 10.7 μg/L, 3.7 μg/L, 1.3 μg/L, 0.6 μg/L, 0.9 μg/dL, and 0.4 μg/L, respectively) demonstrated higher TC levels (2.5% increase in TC, 95% confidence interval: 0.1, 5.0) in the k-means clustering analysis. CONCLUSION: Due to the limitations of this study, such as its cross-sectional design, these results should be interpreted cautiously and confirmed in future studies before drawing implications for public health strategies aimed at promoting health during adolescence and later in life.