Lazarevic Nina, Smurthwaite Kayla S, D'Este Catherine, Lucas Robyn M, Armstrong Bruce, Clements Archie Ca, Trevenar Susan M, Gad Imogen, Hosking Rose, Law Hsei Di, Mueller Jochen, Bräunig Jennifer, Nilsson Sandra, Lane Jo, Lal Aparna, Lidbury Brett A, Korda Rosemary J, Kirk Martyn D
National Centre for Epidemiology and Population Health, College of Health and Medicine, The Australian National University, Canberra, ACT 2600, Australia.
National Centre for Epidemiology and Population Health, College of Health and Medicine, The Australian National University, Canberra, ACT 2600, Australia.
Environ Res. 2023 Jun 1;226:115621. doi: 10.1016/j.envres.2023.115621. Epub 2023 Mar 9.
Per- and polyfluoroalkyl substances (PFAS) have been associated with higher cholesterol and liver function markers in some studies, but the evidence for specific cardiometabolic conditions has been inconclusive.
We quantified the associations of single and combined PFAS with cardiometabolic markers and conditions in a cross-sectional study of three Australian communities with PFAS-contaminated water from the historical use of aqueous film-forming foam in firefighting activities, and three comparison communities.
Participants gave blood samples for measurement of nine PFAS, four lipids, six liver function markers, and completed a survey on sociodemographic characteristics and eight cardiometabolic conditions. We estimated differences in mean biomarker concentrations per doubling in single PFAS concentrations (linear regression) and per interquartile range increase in the PFAS mixture (Bayesian kernel machine regression). We estimated prevalence ratios of biomarker concentrations outside reference limits and self-reported cardiometabolic conditions (Poisson regression).
We recruited 881 adults in exposed communities and 801 in comparison communities. We observed higher mean total cholesterol with higher single and mixture PFAS concentrations in blood serum (e.g., 0.18 mmol/L, 95% credible interval -0.06 to 0.42, higher total cholesterol concentrations with an interquartile range increase in all PFAS concentrations in Williamtown, New South Wales), with varying certainty across communities and PFAS. There was less consistency in direction of associations for liver function markers. Serum perfluorooctanoic acid (PFOA) concentrations were positively associated with the prevalence of self-reported hypercholesterolemia in one of three communities, but PFAS concentrations were not associated with self-reported type II diabetes, liver disease, or cardiovascular disease.
Our study is one of few that has simultaneously quantified the associations of blood PFAS concentrations with multiple biomarkers and cardiometabolic conditions in multiple communities. Our findings for total cholesterol were consistent with previous studies; however, substantial uncertainty in our estimates and the cross-sectional design limit causal inference.
在一些研究中，全氟和多氟烷基物质（PFAS）与较高的胆固醇及肝功能指标有关，但关于特定心脏代谢疾病的证据尚无定论。
在一项横断面研究中，我们对三个澳大利亚社区（这些社区因历史上在消防活动中使用水成膜泡沫而受到PFAS污染）和三个对照社区的单一及联合PFAS与心脏代谢标志物及疾病之间的关联进行了量化。
参与者提供血样以检测9种PFAS、4种血脂、6种肝功能指标，并完成一项关于社会人口学特征和8种心脏代谢疾病的调查。我们估计了单一PFAS浓度每增加一倍（线性回归）以及PFAS混合物每增加一个四分位数间距（贝叶斯核机器回归）时生物标志物平均浓度的差异。我们估计了生物标志物浓度超出参考限值和自我报告的心脏代谢疾病的患病率比（泊松回归）。
我们在暴露社区招募了881名成年人，在对照社区招募了801名。我们观察到血清中单一和混合PFAS浓度越高，总胆固醇的平均水平越高（例如，新南威尔士州威廉敦所有PFAS浓度每增加一个四分位数间距，总胆固醇浓度升高0.18 mmol/L，95%可信区间为-0.06至0.42），不同社区和PFAS的确定性有所不同。肝功能指标的关联方向一致性较差。血清全氟辛酸（PFOA）浓度在三个社区之一中与自我报告的高胆固醇血症患病率呈正相关，但PFAS浓度与自我报告的II型糖尿病、肝病或心血管疾病无关。
我们的研究是少数同时量化多个社区血液PFAS浓度与多种生物标志物及心脏代谢疾病之间关联的研究之一。我们关于总胆固醇的研究结果与先前的研究一致；然而，我们估计中的大量不确定性以及横断面设计限制了因果推断。
