【作者】 孙金辉；

【导师】 向浩； 鲁元安；

【作者基本信息】 武汉大学 ， 流行病与卫生统计学， 2023， 硕士

【副题名】基于队列研究和定组研究

【摘要】 目的:大气细颗粒物(PM_2.5)暴露是影响我国居民血脂异常的重要危险因素,但当前关于PM_2.5对血脂水平急性效应的研究较少且研究结果不一致,同时尚缺乏PM_2.5及其组分长期暴露对血脂异常影响的纵向研究证据。本研究旨在探究PM_2.5长期和短期暴露对我国居民血脂异常的影响,并探索胰岛素抵抗在PM_2.5与血脂关联中的调节作用。。方法:（1）长期暴露于PM_2.5及其组分与血脂异常的关联:基于中国健康与养老追踪调查（China health and retirement longitudinal study,CHARLS）的2011-2018年四次随访数据,共纳入14400名研究对象。从中国大气成分近实时追踪数据集（Tracking Air Pollution in China,TAP）获取各城市10 km空间分辨率的PM_2.5及5种化学组分(硫酸盐SO₄^2-、硝酸盐NO₃^-、铵盐NH₄⁺、黑炭BC、有机颗粒OM)月均浓度,然后通过每个研究对象的家庭地址信息为其匹配污染物的两年暴露情况。基于研究对象每次随访时的自我报告来判断其是否新发生血脂异常。采用时间依存协变量的Cox比例风险回归模型评估PM_2.5及组分长期暴露与血脂异常发病风险的关联,并通过分层分析探索人口学特征、行为生活方式等因素对PM_2.5与血脂异常关联的影响。（2）PM_2.5短期暴露对血脂水平的影响:在武汉大学招募无急慢性疾病史的健康志愿者开展定组研究,通过基线调查收集研究对象基本信息。从2019年9月至2020年1月共进行8次随访,每两次随访时间间隔控制在1至2周。每次随访时,通过便携式个体PM_2.5检测仪获取个体连续3日实时暴露浓度,在第4日清晨进行医学体检,定量测量研究对象空腹血糖（Fasting blood glucose,FBG）、血清胰岛素（Fasting blood insulin,FINS）、C-反应蛋白（C-reactive protein,CRP）、以及甘油三酯（Triglycerides,TG）、总胆固醇（Total cholesterol,TC）、低密度脂蛋白胆固醇（Low-density lipoprotein cholesterol,LDL-C）和高密度脂蛋白胆固醇（High-density lipoprotein cholesterol,HDL-C）等生物指标水平。采用线性混合效应模型定量评估短期PM_2.5暴露与TG、TC、LDL-C和HDL-C水平之间的关联,并通过交互作用分析探讨稳态模型的胰岛素抵抗指数（Homeostasis model assessment of insulin resistance,HOMA-IR）对PM_2.5与血脂之间关联的调节作用。结果:（1）长期暴露于PM_2.5及其组分与血脂异常的关联:研究期间PM_2.5平均暴露浓度为49.45μg/m³,其中SO₄^2-、NO₃^-、NH₄⁺、BC和OM平均水平分别为9.42μg/m³、10.77μg/m³、7.52μg/m³、2.43μg/m³和12.03μg/m³。在调整多种潜在混杂因素后,PM_2.5及5种化学组分浓度均与血脂异常发生风险呈正相关,PM_2.5两年平均暴露每增加1μg/m³,发生血脂异常的风险比（Hazard ratio,HR）为1.010（95%CI:1.008,1.013）。PM_2.5中的SO₄^2-、NO₃^-、NH₄⁺、BC和OM每增加1μg/m³,发生血脂异常的HR分别为1.050,1.026,1.048,1.294和1.042。分层分析结果显示,年龄、性别、受教育水平、居住地、婚姻状况、吸烟、饮酒状况等因素的组间差异均不显著。（2）PM_2.5短期暴露对血脂水平的影响:研究纳入70名健康成年人,研究期间个体PM_2.5暴露的平均水平为38.34μg/m³。调整混杂因素后,个体PM_2.5暴露与LDL-C水平呈负相关,与HDL-C呈正相关,与TG、TC的关联不显著。例如,滞后2日PM_2.5暴露每增加10μg/m³,LDL-C水平降低1.01%（95%CI:-1.50%,-0.51%）;滞后3日PM_2.5与HDL-C水平增加0.77%（95%CI:0.30%,1.24%）有关。交互作用分析结果显示,TG、TC、LDL-C、HDL-C均与PM_2.5暴露呈负相关,并且这种关联在低HOMA-IR组（HOMA-IR<1.82）更强。结论:长期暴露于PM_2.5以及其化学成分SO₄^2-、NO₃^-、NH₄⁺、BC和OM均会显著增加我国中老年人发生血脂异常的风险。大气PM_2.5短期暴露对血脂水平的影响可能受到胰岛素抵抗的调节作用。本研究的结果进一步补充了大气PM_2.5暴露与我国居民血脂异常关联的流行病学证据,亟待开展相关研究揭示PM_2.5暴露导致血脂异常的生物学机制。更多还原

【Abstract】 Objective The epidemical evidence on the association between fine particulate matter(PM_2.5)and blood lipids was inconclusive.Besides,limited research assessed the effect of PM_2.5 and its chemical constituents on blood lipids,and the underlying mechanism has not been well characterized.Our study aimed to investigate the effects of long-term and short-term exposure to PM_2.5 on dyslipidemia and to explore the association between PM_2.5chemical constituents and the risk of dyslipidemia.Methods（i）A total of 14400 participants of China Health and Retirement Longitudinal Study（CHARLS）from 2011 to 2018 were included in this cohort study.The monthly average concentrations of PM_2.5 and its chemical constituents(SO₄^2-,NO₃^-,NH₄⁺,BC,OM)for each city at 10 km spatial resolution were obtained through the Tracking Air Pollution in China（TAP）.Two years of exposure levels to air pollutants were then matched to each subject by their address.The diagnosis of new dyslipidemia was based on self-reporting by the participants at each visit.A time-varying cox propotional hazard regression model was used to assess the effects of PM_2.5 and its chemical constituents on the incidence of dyslipidemia.Besides,the stratified analyses were applied to explore the susceptibility factors for the aforementioned association.（ii）A longitudinal panel study with 8 visits was conducted from September 2019 to January 2020,with 1 to 2 weeks between each visit.The basic characteristics of participants were collected at the baseline.During each visit,the real-time PM_2.5 concentrations of each individual for 3 consecutive days were measured with a portable PM_2.5monitor,and fasting blood was collected to measure the concentrations of FBG,FIN,CRP,and blood lipids after PM_2.5 exposure monitoring.The linear mixed-effect models were used to exam the association between PM_2.5 blood lipids,and the interaction analyses were performed to estimate the modification effects of HOMA-IR on PM_2.5-lipids association.Results（i）The average concentration of PM_2.5 was 49.45μg/m³,with mean levels of SO₄^2-,NO₃^-,NH₄⁺,BC,and OM of 9.42,10.77,7.52,2.43 and 12.03μg/m³,respectively.After adjusting for several confounders,PM_2.5 and its chemical constituents were positively associated with the risk of dyslipidemia.The hazard ratio（HR）for dyslipidemia was 1.010（95%CI:1.008,1.013）for every 1μg/m³ increase in PM_2.5.For SO₄^2-,NO₃^-,NH₄⁺,BC,and OM（per 1μg/m³）,the HR of dyslipidemia were 1.050,1.026,1.048,1.294,and 1.042,respectively.Stratified analysis showed no significant differences between groups.（ii）A total of 70 healthy adults were included.The PM_2.5 concentrations across the research was38.34μg/m³.Individual PM_2.5 exposure was negatively associated with LDL-C levels,positively with HDL-C,and insignificantly associated with TG and TC within specific exposure time window.For instance,for every 10μg/m³ increment in PM_2.5 lag 2 days,LDL-C decreased by 1.01%（95%CI:-1.50%,-0.51%）.And a 10μg/m³ rise in PM_2.5 lag3 days associated with an increment of 0.77%（95%CI:0.30%,1.24%）in HDL-C.Interaction analysis showed that TG,TC,LDL-C,and HDL-C were negatively associated with PM_2.5,and the effects were stronger in the lower HOMA-IR group（HOMA-IR<1.82）.Conclusions Long-term exposure to PM_2.5 and its chemical constituents were associated with an increased risk of dyslipidemia in middle-aged and elderly.Besides,the effects of short-term exposure to ambient PM_2.5 on blood lipids may be modulated by insulin resistance.The results of this study further add to the epidemiological evidence on the association between PM_2.5 and dyslipidemia.And relevant studies are urgently needed to reveal the biological mechanisms of dyslipidemia due to ambient PM_2.5.更多还原