Abstract：
Per- and poly-fluoroalkyl substances (PFAS) have been reported to have hepatotoxic effects. However, it is unclear whether they are linked to non-alcoholic fatty liver disease (NAFLD). This nested case-control study focused on the epidemiological links between PFAS and the prevalence of NAFLD. We selected 476 new cases of NAFLD and 952 age- and sex-matched controls from the Jinchang cohort population between 2014 and 2019. Serum concentrations of PFAS were measured using high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS). Only PFAS with a detection rate of ≥90 % were included for analysis, which included PFPeA, PFOA, PFNA, PFHxS, PFOS, and 9Cl-PF3ONS. The relationship between single and co-exposure to PFAS and the occurrence of NAFLD was evaluated using conditional logistic regression, Quantile g-computation (QgC), and Bayesian kernel machine regression (BKMR) model. Logistic regression indicated that PFPeA, PFOA, and 9Cl-PF3ONS were positive correlation with the incidence of NAFLD after adjusting for confounders, with odds ratios (OR) and 95 % confidence interval (CI) of 3.13 (95 % CI: 2.53, 3.86), 1.39 (95 % CI: 1.12, 1.73), and 1.41 (95 % CI: 1.20, 1.66), respectively. PFNA, PFHxS, and PFOS were nonlinearly and negatively associated with the incidence of NAFLD, with OR (95 % CI) of 0.53 (0.46, 0.62), 0.83 (0.73, 0.95), and 0.52 (0.44, 0.61), respectively. QgC showed a significant joint effect of PFAS mixture on NAFLD onset (OR: 1.52, 95 % CI: 1.24, 1.88). BKMR showed a weak positive trend between PFAS mixtures and NAFLD incidence. Positive correlations were primarily driven by PFPeA and 9Cl-PF3ONS, while negative correlations were mainly influenced by PFNA and PFOS. The BKMR model also suggested that there was an interaction between PFOS and PFNA and other four PFAS compounds. In conclusion, our findings suggest that individual and co-exposure to PFAS is associated with a risk of NAFLD onset.
摘要：
已报道过和多氟烷基物质（PFAS）具有肝毒性作用。然而,目前尚不清楚它们是否与非酒精性脂肪性肝病(NAFLD)相关.这项巢式病例对照研究的重点是PFAS与NAFLD患病率之间的流行病学联系。我们从2014年至2019年的金昌队列人群中选择了476例新的NAFLD病例和952例年龄和性别匹配的对照。使用高效液相色谱-串联质谱法(HPLC-MS/MS)测量PFAS的血清浓度。仅纳入检出率≥90%的PFAS进行分析，其中包括PFPeA，PFOA,PFNA,PFHxS,全氟辛烷磺酸,和9Cl-PF3ONS。使用条件逻辑回归评估单次和共同暴露于PFAS与NAFLD发生之间的关系，分位数g计算(QgC)，和贝叶斯核机回归(BKMR)模型。Logistic回归表明PFPeA，PFOA,和9Cl-PF3ONS在校正混杂因素后与NAFLD的发生率呈正相关，优势比（OR）和95%置信区间（CI）为3.13（95%CI：2.53，3.86），1.39（95%CI：1.12，1.73），和1.41（95%CI：1.20，1.66），分别。PFNA,PFHxS,和全氟辛烷磺酸与NAFLD的发病率呈非线性负相关，OR（95%CI）为0.53（0.46，0.62），0.83（0.73，0.95），和0.52（0.44，0.61），分别。QgC显示PFAS对NAFLD发病有显著的联合作用(OR:1.52,95%CI:1.24,1.88)。BKMR在PFAS混合物和NAFLD发生率之间显示出微弱的积极趋势。正相关主要由PFPeA和9Cl-PF3ONS驱动，而负相关主要受PFNA和PFOS的影响。BKMR模型还表明，PFOS和PFNA以及其他四种PFAS化合物之间存在相互作用。总之,我们的研究结果表明,个体和共同暴露于PFAS与NAFLD发病风险相关.
