The mediation role of blood lipids on the path from air pollution exposure to MAFLD: A longitudinal cohort study

Xinyu Han, Bing Guo, Lele Wang, Kejun Chen, Hanwen Zhou, Shourui Huang, Huan Xu, Xianmou Pan, Jinyao Chen, Xufang Gao, Zhenghong Wang, La Yang, Ciren Laba, Qiong Meng, Yuming Guo, Gongbo Chen, Feng Hong, Xing Zhao, on behalf of the China Multi-Ethnic Cohort (CMEC) collaborative group

Research output: Contribution to journalArticleResearchpeer-review

3 Citations (Scopus)


Background & aims: Recent cross-sectional studies found that exposure to ambient air pollution (AP) was associated with an increased risk of metabolic dysfunction-associated fatty liver disease (MAFLD). The alternation of blood lipids may explain the association, but epidemiological evidence is lacking. We aimed to examine whether and to what extent the association between long-term exposure to AP and incident MAFLD is mediated by blood lipids and dyslipidemia in a prospective cohort. Methods: We included 6350 participants from the China Multi-Ethnic Cohort (CMEC, baseline 2018–2019, follow-up 2020–2021). Three-year average (2016–2018) of AP (PM1, PM2.5, PM10, NO2), blood lipids (TC, LDL-C, HDL-C, TG with their combinations) and incident MAFLD for each individual were assessed chronologically. Linear and logistic regression was used to assess the associations among AP, blood lipids, and MAFLD, and the potential mediation effects of blood lipids were evaluated using causal mediation analysis. Results: A total of 744 participants were newly diagnosed with MAFLD at follow-up. The odds ratios of MAFLD associated with a 10 μm increase in PM1, PM2.5, and PM10 were 1.35 (95 % CI: 1.14, 1.58), 1.34 (1.10, 1.65) and 1.28 (1.14, 1.44), respectively. Blood lipids are important mediators between AP and incident MAFLD. LDL-C (Proportion Mediated: 6.9 %), non-HDL (13.4 %), HDL-C (20.7 %), LDL/HDL (30.1 %), and dyslipidemia (6.5 %) significantly mediated the association between PM2.5 and MAFLD. For PM1, the indirect effects were similar to those for PM2.5, with a larger value for the direct effect, and the mediation proportion by blood lipids was less for NO2. Conclusion: Blood lipids are important mediators between AP and MAFLD, and can explain 5 %–30 % of the association between AP and incident MAFLD, particularly cholesterol-related variables, indicating that AP could lead to MAFLD through the alternation of blood lipids. These findings provided mechanical evidence of AP leading to MAFLD in epidemiological studies.

Original languageEnglish
Article number166347
Number of pages10
JournalScience of the Total Environment
Publication statusPublished - 15 Dec 2023


  • Air pollution
  • Blood lipids
  • Causal mediation effect

Cite this