Environment International (Feb 2023)

Serum per- and polyfluoroalkyl substances and abnormal lipid metabolism: A nationally representative cross-sectional study

  • Bing Wu,
  • Yitao Pan,
  • Zheng Li,
  • Jinghua Wang,
  • Saisai Ji,
  • Feng Zhao,
  • Xiaochen Chang,
  • Yingli Qu,
  • Yuanduo Zhu,
  • Linna Xie,
  • Yawei Li,
  • Zheng Zhang,
  • Haocan Song,
  • Xiaojian Hu,
  • Yidan Qiu,
  • Xulin Zheng,
  • Wenli Zhang,
  • Yanwei Yang,
  • Heng Gu,
  • Fangyu Li,
  • Jiayi Cai,
  • Ying Zhu,
  • Zhaojin Cao,
  • John S.Ji,
  • Yuebin Lv,
  • Jiayin Dai,
  • Xiaoming Shi

Journal volume & issue
Vol. 172
p. 107779

Abstract

Read online

Background: The associations of legacy per- and polyfluoroalkyl substances (PFAS) with lipid metabolism are controversial, and there is little information about the impact of emerging PFAS (6:2 Cl-PFESA) on lipid metabolism in China. Objectives: We aimed to explore the associations of legacy and emerging PFAS with lipid profiles and dyslipidemia in Chinese adults. Methods: We included 10,855 Chinese participants aged 18 years and above in the China National Human Biomonitoring. The associations of 8 PFAS with 5 lipid profiles and 4 dyslipidemia were investigated using weighted multiple linear regression or weighted logistic regression, and the dose–response associations were investigated using restricted cubic spline model. Results: Among the 8 PFAS, the concentration of PFOS was the highest, with a geometric mean of 5.15 ng/mL, followed by PFOA and 6:2 Cl-PFESA, which were 4.26 and 1.63 ng/mL, respectively. Legacy (PFOA, PFOS, PFUnDA) or emerging (6:2 Cl-PFESA) PFAS were associated with lipid profiles (TC, LDL-C, HDL-C, non HDL-C) and dyslipidemia (high LDL-C, high TC, low HDL-C), and their effects on TC were most obvious. TC concentration increased by 0.595 mmol/L in the highest quartile (Q4) of PFOS when compared with the lowest quartile (Q1), (95 % CI:0.396, 0.794). Restricted cubic spline models showed that PFAS are nonlinearly associated with TC, non HDL-C, LDL-C and HDL-C, and that the lipid concentrations tend to be stable when PFOS and PFOA were > 20 ng/mL well as when the 6:2 Cl-PFESA level was > 10 ng/mL. The positive associations between PFAS mixtures and lipid profiles were also significant. Conclusions: Single and mixed exposure to PFAS were positively associated with lipid profiles, and China's unique legacy PFAS substitutes (6:2 Cl-PFESA) contributed less to lipid profiles than legacy PFAS. In the future, cohort studies will be needed to confirm our findings.

Keywords