节点文献
亚临床甲状腺功能减退与血脂紊乱相互影响的临床研究
The Interaction between Subclinical Hypothyroidism and Dyslipidemia in Middle Aged and Elderly People
【作者】 赵萌;
【作者基本信息】 山东大学 , 内科学(内分泌与代谢病), 2016, 博士
【摘要】 研究背景:亚临床甲状腺功能减退(subclinical hypothyroidism, SCH),简称亚甲减,是一种轻度的甲状腺功能衰退,表现为血清促甲状腺激素(thyroid-stimulating hormone,TSH)水平升高,而游离甲状腺素(free thyroxine, FT4)的水平处于正常参考值范围内,这是一种最常见的甲状腺功能紊乱。亚甲减的患病率近年来在全球范围内呈逐渐上升的趋势(1972-1974年,英格兰为5.0%;1995年,美国为9.0%;2012年,印度高达19.3%),但迄今为止其病因尚未完全阐明。既往研究提示,自身免疫反应、碘摄入过多或不足、炎症等多种因素均可引起甲状腺功能紊乱,但这些原因难以完全解释亚甲减在全球范围内高发的现象。Jankovic等的研究发现,合并高甘油三酯血症的肥胖患者血清TSH水平升高;令人惊讶的是,当这些患者经过减重手术后,血清甘油三酯(triglyceride, TG)水平下降,TSH的水平也随之下降。在动物实验中也观察到了类似的现象:与普通饮食喂养的动物相比,高脂饮食喂养的动物血清甲状腺激素(thyroxine, T4)及三碘甲状腺原氨酸(triiodothyronine, T3)的水平均有下降。这些证据提示我们:高甘油三酯血症可能与亚甲减的发生相关。血中的TG水平升高,可导致其水解产物游离脂肪酸(free fatty acid, FFA)增多,当超出机体对游离脂肪酸的处理能力及脂肪组织对TG的储存能力,大量TG和游离脂肪酸在非脂肪组织异位沉积,继而造成组织细胞的慢性损伤,最终引起靶器官的功能障碍,这种现象被称为脂毒性(Lipotoxicity)。脂毒性因其对人类健康广泛、严重的危害而引起全世界的普遍关注。众所周知,脂毒性是多种代谢性疾病(例如,2型糖尿病,肥胖,非酒精性脂肪肝)共同的致病机制。近年来,不断有研究发现,脂毒性在心衰、慢性肾疾病、内皮细胞功能障碍等疾病的发生、发展过程中也扮演着重要的角色。那么,脂毒性对甲状腺是否也有影响呢?迄今为止,尚无研究关注这一问题。既往已有研究提示亚甲减是导致血脂异常的重要危险因素之一。Colorado甲状腺疾病调查研究显示,随血清TSH水平升高,总胆固醇(total cholesterol, TC)及低密度脂蛋白胆固醇(low-density lipoprotein cholesterol, LDL-C)水平逐渐上升。Whickham研究则发现,与甲状腺功能正常者相比,亚甲减患者的TC水平明显升高。经过左甲状腺素(L-thyroxine, L-T4)治疗后,亚甲减患者血清TSH水平控制在正常参考值范围内,血脂水平随之下降。我们的前期临床研究也在冠心病患者及甲状腺功能正常人群中证实了TSH与血脂的正相关关系(校正甲状腺激素和年龄、性别等一般混杂因素)。这提示我们,TSH可能有独立于甲状腺激素之外的调节血脂的作用。年龄是亚甲减和血脂异常共同的危险因素。根据美国第三次国家健康与营养调查(National Health and Nutrition Examination Survey III, NHANES III)报道,与20-29岁人群相比,60-69岁人群的血清TSH水平升高约40%。国内的大规模现况调查(N=54,240)显示,血脂水平与年龄呈线性正相关。然而,既往的研究在分析TSH与血脂谱的关系时,大多将年龄作为一个潜在的混杂因素进行校正,而鲜少有研究关注年龄和亚甲减对血脂的影响是否可能存在交互作用,即亚甲减对血脂的影响在青年人与老年人中是否相同?在与年龄密切相关的血脂异常中,亚甲减又是否发挥了重要的作用?本研究首先从两个方面来阐述甲状腺功能异常与血脂紊乱的关系:一方面,采用基于人群的病例-对照设计,观察血清TG水平与亚甲减患病风险的相关性,从而揭示高甘油三酯血症是否是亚甲减患病风险升高的危险因素。另一方面,采用横断面设计,在社区人群中验证血清TSH与血脂谱的正相关关系。另外,在上述研究的基础上,进一步分析年龄与TSH对血脂谱的交互作用:将研究人群分别按照年龄和甲状腺功能状态进行分组,在各年龄组中比较TSH与血脂的相关系数是否相同,同时在甲状腺功能状态不同的人群中观察随年龄增长,血脂水平升高的幅度是否一致。本研究将为探索亚甲减与血脂异常的关系提供新的角度和方向,同时为疾病的筛查和预防提供新的策略和思路。目的:1.观察血清TG水平与亚甲减患病风险的相关性,明确高甘油三酯血症是否是亚甲减患病风险升高的危险因素。2.验证血清TSH与血脂谱的相关性。3.比较各年龄组TSH与血脂的相关系数是否相同,同时观察甲状腺功能状态不同的人群中随年龄增长,血脂水平升高的幅度是否一致,明确年龄与TSH对血脂的影响是否具有交互作用。研究方法:1.研究对象:研究人群来自2011-2012年全国代谢性疾病流行病学调查REACTION (Risk Evaluation of cAncers in Chinese diabeTic Individuals:A lONgitudinal Study)项目。我们选择了项目组中碘营养状态稳定且接近的三个社区,共纳入研究对象24,100例。经过严格的纳排标准筛选后,研究的第一部分首先确定了5,033例亚甲减患者作为病例组,然后根据年龄、性别及地域按照1:1的比例匹配了相同例数的非甲状腺功能减退者作为对照组;研究的第二部分则纳入17,046例进行横断面分析。2.资料获取:通过调查问卷记录一般资料,包括年龄、性别、既往史及药物使用情况等。通过体格检查获取身高、体重、腰围、臀围、血压等资料。所有研究对象均采空腹血进行甲状腺功能及血脂的检测;甲状腺功能检测采用电化学发光法(Cobas E601; Roche),血脂检测采用雅培ARCHITECT ci16200系统进行;检测的批间差异和批内差异均控制在5%以下。3.研究指标:甲状腺功能的正常参考值范围:FT3,3.1-6.8 pmol/L (2.01-4.42 pg/mL); FT4, 12-22 pmol/L (0.94-1.72 ng/dL); TSH,0.27-4.2 mIU/L。亚甲减定义为TSH水平升高(>4.2mIU/L),FT4水平处于正常参考值范围内(12-22 pmol/L);其中TSH在4.2-10.0 mIU/L范围内定义为轻度亚甲减(mild SCH), TSH> 10.0 mIU/L定义为重度亚甲减(severe SCH)。根据美国国家胆固醇教育计划成人治疗组第三次报告(National CholesterolEducation Program Adult Treatment Panel III)的诊断标准,高TC、高TG、高LDL-C及低HDL-C的诊断标准分别为:TC> 6.20 mmol/L, TG> 1.69 mmol/L, LDL-C> 4.13 mmol/L, HDL< 1.04 mmol/L(男)/< 1.30 mmol/L(女)。4.统计分析:采用SPSS 18.0软件进行统计分析。亚甲减患病风险采用条件Logistic回归模型(Conditional logistic regression model)进行分析;TSH与血脂的相关性采用线性回归模型(Linear regression model)进行分析;年龄与血脂异常患病率的相关性采用线性趋势卡方检验(Linear-by-linear association chi-square test)进行分析;TSH每改变1 mIU/L所对应的血脂变化采用偏最小二乘法(Partial least squares)进行估计;各组血脂或甲状腺功能的估计边界均值采用一般线性模型(General linear model)进行计算。结果:1-1高甘油三酯血症患者的亚甲减患病风险增加校正年龄等混杂因素后,男性高甘油三酯血症患者的亚甲减患病风险较TG水平正常人群升高了32.7%(OR=1.327,95%CI=1.003-1.756,P=0.047),在女性则升高39.2%(OR=1.392,95%CI=1.212-1.598,P<0.001)。1-2亚甲减患病风险随TG/HDL-C比值的升高而逐渐增加根据TG/HDL-C的四分位数将研究人群分为四组。在男性中,与第一组相比,第二、三、四组的OR值分别为1.865(95%CI=1.376-2.528,P<0.001)、1.978(95%CI=1.409-2.776,P<0.001)、2.376(95%CI=1.534-3.681,P<0.001),在女性则分别为1.052(95%CI=0.909-1.217,P=0.499)、1.270(95%CI=1.075-1.502,P=0.005)和1.476(95%CI=1.177-1.852,P=0.001)。1-3血清TG与TSH水平呈显著正相关根据TG的四分位数将研究人群分为四组。校正混杂因素后,logTSH的估计边界均值从TG第一分位数组到第四分位数组在男性分别为0.466、0.495、0.523和0.554(Linear coefficient=0.065, P=0.008),在女性则分别为0.510、0.520、0.539和0.569(Linear coefficient=0.044, P=0.001). logFT3和logFT4无明显线性趋势。2血清TSH与TC、LDL-C、TG水平呈显著正相关校正甲状腺激素及年龄等一般混杂因素后,TSH与TC(r=0.0102,P<0.0001)、LDL-C (r=0.0070, P<0.0001 )及logTG (r=0.0017, P=0.0001 )呈显著正相关,与HDL-C (r=-0.0001, P=0.8820)不相关。3-1年龄与血脂异常的患病率呈线性相关局TC( Linear trend coefficient=0.070, P<0.001)、高TG( Linear trend coefficient=0.017,P=0.024)、高LDL-C (Linear trend coefficient=0.059, P<0.001)的患病率也随年龄增长呈线性升高,而低HDL-C的患病率则随年龄增长而逐渐降低(Linear trend coefficient=-0.022, P=0.003)。3-270岁以下人群中随年龄增长,TSH与TC及LDL-C的回归系数逐渐增大校正甲状腺激素及年龄等一般混杂因素后,TSH每改变1 mlU/L对应TC的变化量在40-49岁、50-59岁、60-69岁及>70岁各年龄组分别为0.0147 mmol/L (P= 0.006), 0.0395 mmol/L (P=0.004), 0.0551 mmol/L (P=0.048), 0.0521 mmol/L(P=0.104)。TSH与LDL-C的相关性与TC类似:TSH每改变1 mIU/L对应LDL-C的变化量在各年龄组分别为0.0098 mmol/L (P = 0.031 ), 0.0290 mmol/L(P=0.021)、 0.0343mmol/L (P=0.070)、0.0332 mmol/L (P=0.075)。3-3与甲状腺功能正常状态相比,亚甲减状态下,年龄对TC及LDL-C的影响更加明显在甲状腺功能正常状态下,60-69岁人群TC水平较40-49岁人群升高7.39%;而在轻度亚甲减和重度亚甲减状态下,60-69岁人群与40-49岁人群相比,TC水平分别升高7.52%和9.88%。亚甲减对LDL-C的影响是类似的:甲状腺功能正常、轻度亚甲减和重度亚甲减状态下,60-69岁人群与40-49岁人群相比,LDL-C的水平分别升高9.06%、10.76%和14.83%。结论:1.血清TG水平与亚甲减患病风险显著正相关。2.血清TSH与TC、LDL-C及TG呈正相关。3.年龄与TSH对血清TC和LDL-C具有交互作用。
【Abstract】 Background:Subclinical hypothyroidism (SCH), defined by elevated serum thyroid-stimulating hormone (TSH) and normal serum free thyroxine (FT4), is the most common thyroid dysfunction.SCH continues to progressively increase all over the world (the prevalence has increased from less than 5.0% in England in 1972-1974 to 9.0% in the United States in 1995 and 19.3% in India in 2012) and was regarded as an independent cardiovascular risk factor in a large number of studies. Therefore, it is considerably necessary to further investigate the etiology of SCH, which has not been fully elucidated. Several risk factors, i.e., autoimmunity, abnormal iodine intake and inflammation, have been associated with SCH. However, the presence of SCH could not be fully explained by the above-stated reasons in the following interesting phenomenon. Jankovic et al. observed that obese subjects with hypertriglyceridemia exhibited the increased serum TSH concentrations; interestingly, the TSH levels significantly decreased after the serum triglyceride (TG) levels decreased after bariatric surgery. Similarly, the decreased serum T3 and T4 concentrations, representing the impaired thyroid function, were also observed in animals fed a high-fat diet. These lines of evidence suggested that the presence of SCH might be associated with hypertriglyceridemia.High circulating TG levels lead to the increased importation of fatty acids into nonadipose tissues, contributing to ectopic intracellular lipid accumulation, also known as lipotoxicity. Lipotoxicity attracts great attention worldwide for its serious and extensive impact on human health. Apart from its important role in the pathogenesis of metabolic diseases (such as type 2 diabetes mellitus, obesity and nonalcoholic fatty liver disease), the effect of lipotoxicity in other diseases, such as heart failure, chronic kidney disease and endothelial cell dysfunction has also drawn intensive interest. However, so far, no research has explored whether thyroid is another organ affected by lipotoxicity.In the other hand, previous studies have demonstrated that SCH is an important risk factor for dyslipidemia. Among 25,862 participants in a statewide health fair in Colorado, fasting total cholesterol (TC) and low-density lipoprotein cholesterol (LDI-C) were gradually elevated with the increasing TSH levels. Whickham survey found that serum TC levels were significantly higher in SCH subjects than in euthyroid subjects. With L-thyroxine replacement therapy, TSH concentrations were decreased into the reference range and dyslipidemia was improved accordingly. Our previous study also illustrated the positive association between TSH and lipid after adjustment for thyroxine and other confounders in patients with coronary heart diseases or subjects with euthyroidism. It suggested that TSH might play a role in lipid metabolism independent of thyroid hormones.Age is the common risk factor for both SCH and dyslipidemia. According to the National Health and Nutrition Examination Survey III (NHANES III), TSH levels were increased by approximately 40% in adults aged 60-69 years compared with individuals aged 20-29 years. In addition, large-scale epidemiological survey in China showed that age was positively associated with serum lipid profile. However, few studies explored the interaction of age and SCH on lipid profiles.The present study investigated the association between thyroid dysfunction and dyslipidemia. First, we assessed the association between serum TG levels and risk for SCH. Second, we identified the association between serum TSH and lipid profiles. Furthermore, we compared the regression coefficients of TSH and lipid parameters in different age groups, as well as evaluated the serum lipid elevation associated with age in subjects with different thyroid status. Our study might be helpful in providing attractive and feasible approaches to resist SCH and dyslipidemia.Objectives:1. To assess the association between serum TG levels and risk for SCH.2. To identify the association between serum TSH levels and lipid frofile.3. To investigate the potential interaction of age and SCH on lipid parameters.Methods:1. Study ParticipantsThe present study was part of the community-based Risk Evaluation of cAncers in Chinese diabeTic Individuals:A 1ONgitudinal (REACTION) Study aiming to investigate the epidemiology of metabolic diseases that was conducted across China from April 1, 2011, through May 30,2012. We selected three communities from the east, west, and south of China with similar and stable iodine nutrition status. A total of 24,100 participants provided an overnight fasting blood sample and were asked to complete a self-reported questionnaire. In the Part I, after exclusion,5,033 cases were defined as those participants with diagnoses of SCH from the eligible subjects, and the controls were matched one to one with the cases by age, gender and region from subjects without either SCH or overt hypothyroidism. In the Part II,17,046 participants were included for the cross-sectional analysis.2. Data CollectionBlood samples were collected from all participants between 8:00 AM and 10:00 AM after a minimum 10-hour fasting. Chemiluminescent methods (Cobas E601; Roche) were used to quantitate thyroid function based FT3, FT4, and TSH levels. The serum lipid profile was determined using the ARCHITECT ci16200 Integrated System (Abbott). The intraassay and interassay coefficients of variation were always below 5% for all of the above parameters.3. Study OutcomesThe laboratory reference ranges were 3.1-6.8 pmol/L (2.01-4.42 pg/mL) for FT3, 12-22 pmol/L (0.94-1.72 ng/dL) for FT4, and 0.27-4.2 mIU/L for TSH. SCH was defined as a combination of high TSH and normal FT4. Based on the serum TSH level, SCH was further divided into mild SCH (TSH:4.2-10.0 mIU/L) and severe SCH (TSH> 10.0 mIU/L).According to the National Cholesterol Education Pro gram Adult Treatment Panel Ⅲ, high TC, high TG, high LDL-C and low HDL-C were defined as TC> 6.20 mmol/L, TG> 1.69 mmol/L, LDL-C> 4.13 mmol/L, HDL< 1.04 mmol/L (Male)/< 1.30 mmol/L (Female).4. Statistical AnalysesStatistical tests were performed using SPSS version 18.0 for Windows (SPSS Inc). The conditional logistic regression was used to estimate the risk for SCH. The linear regression model was used to estimate the association between TSH and lipid parameters. The linear-by-linear association chi-square test was used to analyze the association between age and the prevalence of dyslipidemia. The relationships between TSH and serum lipid parameters were evaluated using partial correlation analysis. We generated a general linear model to estimate the marginal mean of lipid or thyroid functions.Results:1-1 The risk for SCH was higher in subjects with hypertriglyceridemiaAfter adjustment for confounders, hypertriglyceridemia was associated with a 32.7% (OR= 1.327,95%CI= 1.003-1.756, P= 0.047) increased risk for SCH in men and a 39.2% (OR= 1.392,95%CI= 1.212-1.598,P<0.001) increased risk in women.1-2 The risk for SCH was progressively increased following higher TG/HDL-C ratioThe population was stratified according to the quartiles of TG/HDL-C ratio. In male, compared with Quartile 1, the ORs for Quartile 2, Quartile 3 and Quartile 4 were 1.865, 1.978 and 2.376, respectively. In female, the ORs were 1.052,1.270 and 1.476, respectively.1-3 Positive correlation between serum TG levels and TSH concentrationsThe population was stratified according to the quartiles of TG levels. The log-transformed serum TSH levels showed a significantly positive association with the TG levels, whereas the FT4 and FT3 levels were not correlated significantly.2 The positive relationship between TSH and TC, LDL-C and TGAfter adjustment for thyroid hormones and other confounders, TSH was positively related to TC (r= 0.0102, P< 0.0001), LDL-C (r= 0.0070, P< 0.0001) and logTG (r= 0.0017, P= 0.0001), but not related to HDL-C (r=-0.0001, P= 0.8820).3-1 The linear association between age and the prevalence of dyslipidemiaFollowing the increasing age, the prevalence of high TC (Linear trend coefficient= 0.070,P< 0.001), high TG (Linear trend coefficient= 0.017, P= 0.024) and high LDL-C (Linear trend coefficient= 0.059, P< 0.001) was linearly elevated, whereas the prevalence of low HDL-C was decreased (Linear trend coefficient=-0.022, P= 0.003).3-2 With the increasing age, regression coefficients of TSH and TC or LDL-C were elevatedAfter adjustment for thyroid hormones and other confounders, each 1 mlU/L increase in TSH was estimated to elevate the TC level by 0.0147 mmol/L,0.0395 mmol/L, and 0.0551 mmol/L,0.0521 mmol/L, respectively, in individuals aged 40-49 years,50-59 years,60-69 years and≥70 years. The relationship was similar for LDL-C: each 1 mIU/L increase in TSH was estimated to elevate the LDL-C level by 0.0098 mmol/L,0.0290 mmol/L, and 0.0343 mmol/L,0.0332 mmol/L, respectively, in individuals aged 40-49 years,50-59 years,60-69 years and≥70 years.3-3 Compared with euthyroidism, the effect of age on TC or LDL-C was greater in SCHIn euthyroidism, compared with subjects aged 40-49 years, TC levels in subjects aged 60-69 years was increased by 7.39%. In mild and severe SCH, TC levels in subjects aged 60-69 years was increased by 7.52% and 9.88%, respectively. The effect on LDL-C was similar:in euthyroidism, mild SCH and severe SCH, LDL-C levels in subjects aged 60-69 years was increased by 9.06%,10.76% and 14.83%, respectively compared to subjects aged 40-49 years.Conclusion:1. Hypertriglyceridemia was positively associated with the risk for SCH.2. Serum TSH levels were positively related to TC, LDL-C and TG concentrations.3. The interaction of age and TSH on TC and LDL-C was identified in the population.