健康人生物学年龄积分及生物学衰老结构方程模型的统计建模研究
摘要
目的
衰老可以定义为生物系统的退行性过程,机体退行性改变的不可逆性积累和对疾病的脆弱性增加,最终导致死亡。Strehler提出了衰老的四个特点,即普遍性、渐进性、内在性和有害性。衰老是普遍的,所有种类的动物都会经历衰老过程；衰老是渐进性的,在生命的各个时期衰老都是在不断进行的；衰老是内在的,生物的衰老过程和速度有物种固有的特点；衰老是有害的,分子和细胞衰老导致组织的功能储备下降和对应激的适应能力降低。衰老不是疾病,但是它可以降低衰老相关疾病的阈值[3],增加患病机会。
目前,人口老龄化已成为世界各国所面临的严峻挑战,是影响一个国家社会、经济长远发展的战略性问题。从上世纪90年代初开始,欧美、日本等发达国家相继实施了一系列衰老纵向研究。其中代表性的为美国的Baltimore和Health ABC衰老纵向研究、加拿大的VLS研究,日本的7年纵向研究,意大利ILSA研究等。从这些研究的趋势可以看出：(1)重视衰老的个体化评价；(2)确定可靠、易测的衰老生物学标志物是衰老研究的一个重要课题；(3)研究热点向健康老龄化转移；(4)利用系统生物学方法建立整体衰老网络是未来研究的方向；(5)遗传背景、生活方式及饮食习惯是影响个体衰老的重要因素。
因为组织衰老的速度不同,随着年龄的增加,人类个体间的差异性也在增大,时序年龄(Chronological age, CA)不能提供衰老过程的准确的指示。在医学和预防角度来说,研究者关心的是个体的衰老变化,目的是筛查衰老高危个体并及时给予干预。因此为了解决以上的问题,研究者提出了“生物学年龄(Biological age,BA)”,即在个体相对于时序年龄同龄人的功能状态的基础上,能够评价个体的功能状态的生物学指标或参数。个体衰老速度不同导致时序年龄和生物学年龄的差异,因此在任何给定的时序年龄,个体间的生物学年龄的值都会相差很大,并最终可以预期与个体在长寿以及衰老过程的结果的时间和／或幅度的差异性相一致。生物学年龄的研究有助于辨别年龄相关性功能障碍的危险的个体,是定量分析个体化生物学衰老的前提和基础。
人体是一个复杂的生物体,衰老的个体化评价要用到多层次,多系统的生理指标。这种选择和评价越全面,它们的组合预测终点事件的能力就越好。在人体,系统是从整体上提炼出的抽象概念(潜变量),反应了人体的一种综合的功能状态,其本身无法被直接测量,用单一指标进行评价是不科学的,需要用其他的多个指标来表示。结构方程模型(structural equational modeling, SEM)为潜变量的表达和测量准确性的检验提供了方法。其属于多维立体网状结构,结构中所涉及的要素间不但呈上下、左右空间关系,而且呈前后的时间关系,故这一方法适用于纵向观察研究和在逻辑上具有纵向资料的性质的横向资料,如衰老研究。
本研究通过多中心大样本人群横断面研究,在全面检测包括血常规、尿常规、血液生化、心血管超声及炎症和内分泌因子等108项指标的基础上,筛选出中国健康人群衰老的生物学标志物,构建了衰老的个体化生物学评价指标-生物学年龄积分(Biological aging score, BAS),定量分析了中国健康人群生物学衰老规律。并在此基础上,扩大变量入选范围,建立生物学衰老的结构方程模型,为我国衰老的个体化评价研究提供了重要理论依据。
方法
一、生物学年龄积分的建立
1、研究对象
从沈阳、大连、北京三个城市的2876名自我评价健康的人中,筛选出健康居民852人,依据年龄分为4组,≤4岁组、45～59岁组、60～74岁组、≥75岁组。
2、检测指标
检测血压、血常规、尿常规、血液生化、心脏超声、颈动脉超声、内分泌、炎症、营养等共计108项指标。
3、数据处理
在排除二元变量后,观察各项指标随增龄变化情况,进行各指标与年龄的相关分析,与年龄显著相关且相关系数>0.25的变量进入下一步冗余分析和主成分分析,筛选衰老生物学标志物,并在探索性因子分析中利用因子得分赋予各标志物权重以构建衰老生物学年龄积分公式,计算个体生物学年龄积分(Biological aging score, BAS),分析健康人生物学衰老规律。
二、衰老个体化评价结构方程模型的建立
1、研究对象
从沈阳、大连、北京三个城市的2876名自我评价健康的人中,筛选出健康居民852人,依据年龄分为4组,≤44岁组、45～59岁组、60～74岁组、≥75岁组。
2、检测指标
检测血压、血常规、尿常规、血液生化、心脏超声、颈动脉超声、内分泌、炎症、营养等共计108项指标。
3、数据处理
首先选择与时序年龄显著相关且相关系数大于等于0.15的指标进入下一步分析,然后进行冗余分析(排除具有高度共线性的变量),在相关系数大于0.65的变量中,选择与时序年龄相关性最高的指标进入下一步分析。拟定各系统和器官标志物,确定潜变量与观测变量,画出路径图,构想理论模型,进行模型拟合与参数估计,进行模型的修正与再验证,利用各种拟合指数对模型做整体评价,同时检验参数的显著性,评价参数的意义和合理性,计算决定系数,评价方程对数据的解释能力,进行模型修正。最后确定生物学衰老结构方程模型,对模型做出解释,分析各器官和系统生物学衰老的规律。
结果
一、建立生物学年龄积分
108个指标中有58个指标与年龄相关,其中相关系数>0.25的有21个,经冗余分析后有12个指标进入下一步分析,其中雌二醇因为在男性和女性随年龄变化的方向不同,在女性随年龄减少,而在老年男性则随年龄增加,因此不适合整体公式的评价。经过主成分分析,颈动脉内膜中层厚度(IMT)、颈动脉舒张期最大前向血流速度(EDV)、脉压(PP),二尖瓣E峰／二尖瓣A峰比值(E／A)、二尖瓣环E峰侧壁(MVEL)、纤维蛋白原水平(FIB)、血清Cystatin C (CYSC)等7个指标被筛选为衰老的生物学标志物,因子分析显示只有一个特征值大于1的因子,说明这7个生物学标志物共同反映了一个衰老的基本过程。根据因子得分,建立了生物学年龄积分计算公式：BAS=0.248(时序年龄)+0.195(IMT)-0.196(EDV)-0.167(E／A)-0.166(MVEL)+0.188(PP)+0.182(FIB)+0.193(CYSC)。个体生物学年龄积分与时序年龄呈显著正相关(r=0.893,p<0.001)。由生物学年龄积分评价的生物学衰老速度随年龄的增加而加速,在60-74岁组达到峰值,75岁以后进入平台期。
二、建立衰老结构方程模型
108个指标中有58个指标与年龄相关,其中相关系数>0.15的有37个,经冗余分析后有27个指标进入下一步分析(雌二醇和黄体酮因为在男性和女性随年龄变化的方向不同被排除)。建立生物学衰老的结构方程模型,衰老相关性改变表现在血管结构和功能(IMT,EDV,PP)、心脏结构(D, LAAPD, LALRD, AD, MVEDT),心脏功能(E/A, MVEL, MVAS, MVAA)、炎症因子(CRP,IL6,FIB)、肾功能(BUN, CYSC)、肝功能(GPT,PRE)、营养(ALB,HDL,TC,GLU)和血液及内皮功能(MCV,TM,TRF)有关,其中血管因子与衰老的关系最为密切(r=0.90),其次为心脏功能(r=0.69)。
结论
一、建立生物学年龄积分
建立了生物学年龄积分计算公式：BAS=0.248(时序年龄)+0.195(IMT)-0.196(EDV)-0.167(E／A)-0.166(MVEL)+0.188(PP)+0.182(FIB)+0.193(CYSC)。健康人的衰老速度随年龄的增加而增加,至>75岁组达到峰值。
二、建立衰老的个体化评价结构方程模型
衰老相关性改变表现在血管结构和功能、心脏结构、心脏功能、炎症因子、肾功能、肝功能、营养和血液及内皮功能有关,其中血管因子与衰老的关系最为密切(r=0.90),其次为心脏功能(r=0.69)。
Objective
Aging can be defined as a degenerative process of biological system, the accumulation of the body irreversible degenerative changes increased vulnerability to disease and eventually leaded to death. Strehler proposed four characteristics of aging, namely, universal, progressive, intrinsic and harmfu. Aging is universal, none of animals could escape the aging process; aging is gradual in nature, aging are ongoing in various periods of life; aging is inherent, biological aging process and its speed have inherent characteristics of species; Aging is harmful, molecular and cellular senescence declined the functional reserve and reduced ability to adapt to the stress. Aging is not a disease, but it can reduce the threshold of aging-related diseases and thus increased chance of illness.
At present, most of the countries in the world must face the problems of the population aging which were serious challenges to a country's social and economic long-term development. From the early 90s of last century, many developed countries such as Europe, the United States and Japan have implemented a series of longitudinal studies of aging. Representative studies were American Baltimore study, the Health ABC Longitudinal Study,Canada's VLS study Japan's seven-year longitudinal study,the Italian ILSA study and so on. The trends seen from these studies were: (1) emphasis on the individual assessment of aging; (2) the reliable, easily measured biomarkers of aging is an important topic in aging research; (3) research focus to healthy aging transfer; (4) the use of systems biology approach to establish the overall aging network is the direction of future research; (5) genetic background, lifestyle and dietary habits are important factors affecting the aging individual.
Because the organization aged at different speeds, the differences among human individuals increased with increasing age. Chronological age (CA) can not provide the exact instructions for the aging process In medicine and prevention perspective, researchers concerned with aging changes in the individuals. The purposes of them were screening high risk individuals and providing timely intervention. Therefore, to solve the above problems, the researchers proposed the "biological age (BA)" which defined as the fuctional status of individuals relative to their chronological peers on the basement of the biological indicators or parameters Different rates of individual aging leads to the differences in chronological age and biological age so at any given age, biological age of individuals would be showing great differences and ultimately can be expected to be consistent with the time&magnitude of the differences of longevity and results of the aging process. Biological age study was the premise and foundation of quantitative analysis of individual aging and will help identifying the high risk individuals
The human body is a complex organism, aging of the individual evaluation of use to the multi-level, multi-system physiological indicators. This selection and evaluation were more comprehensive, their combined ability to predict the better end of the event. In the body, the system is extracted from the whole of the abstractions (latent variables), the reaction of the human body, a comprehensive functional status, by itself, can not be directly measured, using a single indicator is unscientific to evaluate the need to use other various indicators to be expressed. Structural equation modeling (SEM) for the latent variable expression and measurement accuracy provides a method of testing. It belongs to multi-dimensional three-dimensional network structure, the structure of the elements involved in not only vertically between the left and right spatial relations, but also was the time before and after, so this method is suitable for longitudinal observational studies and in the logical nature of the information with a horizontal longitudinal information, such as aging research.
This study was population-based multi-center cross-sectional study. On the basis of 108 indicators including blood routine, urine routine, blood biochemistry, cardiovascular ultrasound and endocrine factors of healthy people in China, the biomarkers of aging were screened, individual biological evaluation indicators biological age points were constructed, the law of biological aging process in healthy people in China was quantitative analyzed. And on this basis, to expand the scope of the variable selected to establish the structural equation model of biological aging. This study provided the important theoretical basis for our aging individual evaluation studies.
Methods
1. Establishment of biological aging score
(1) Subjects
2876 independently living and apparently healthy community-dwelling participants between 30 and 98 years of age were recruited from Beijing, Shenyang, and Dalian cities in China in 2003. Consequently, a total of 852 participants (392 men and 460 women) were enrolled. They were classified as young (< 45 yr), middle-aged (45-59 yr), old (60-74 yr), and very-old (≥75 yr). Informed consent was obtained from all participants.
(2) Detection indicators
a total of 108 indicators were examined including blood pressure, blood routine, urine routine, blood biochemistry, echocardiography, carotid artery ultrasound, endocrine, inflammation and nutrition.
(3) Data processing
we excluded the binary variablesand observed the changes of variableswith aging. We performed correlation analysis and choose those variables that were significantly correlated with age and the correlation coefficient> 0.25 of the variables for the next redundancy analysis and principal component analysis in order to screen the biomarkers of aging. Then we performed exploratory factor analysis and use factor scores weighting the biomarkers of aging for building the formula to calculate an individual biological age scores and analyzed the laws of biological aging in healthy population.
2. Individual aging evaluation by structural equation modeling
(1) Subjects
2876 independently living and apparently healthy community-dwelling participants between 30 and 98 years of age were recruited from Beijing, Shenyang, and Dalian cities in China in 2003. Consequently, a total of 852 participants (392 men and 460 women) were enrolled. They were classified as young (< 45 yr), middle-aged (45-59 yr), old (60-74 yr), and very-old (≥75 yr). Informed consent was obtained from all participants.
(2) Detection indicators
A total of 108 indicators were examined including blood pressure, blood routine, urine routine, blood biochemistry, echocardiography, carotid artery ultrasound, endocrine, inflammation and nutrition.
(3) Data processing
First we choose those variables that were significantly correlated with age and correlation coefficient was greater than or equal to 0.15 for redundancy analysis (excluding the highly collinear variables). For those variables with the correlation coefficient greater than 0.65, we selected the one with the highest correlation coefficient with age. Then we developed various systems and organs markers to determine the latent variables and observed variables, draw the road map, construct theoretical models, carry out model and estimate the fitting index, make amendment and re-validation by a variety of fit indices of the model to do the whole evaluation, while the significance of test parameters to evaluate the significance and reasonableness of the parameters to calculate the coefficient of determination, the evaluation equation interpretation of the data capacity, model updating. Finalize the biology of aging structural equation model to explain the model, analysis of various organs and systems biology of aging pattern.
Results
1. The establishment of biological aging score
A total of 108 variables including physical characteristics, cardiovascular function, blood and urine biochemical properties, and the state of the inflammation, hormonal and nutritional data, were measured and derived. After having excluded all binary variables and the variables not correlated with CA well (i.e., P>0.05 or r<0.25) or the redundant variables for men and women,12 age-related variables were selected for the analysis of the correlation with CA. At last,8 variables:CA, IMT, EDV, E/A, MVEL, PP, FIB and CYSC, were selected as biomarkers to evaluate BA. The contribution of each variable to the variance of BAS were as follows:CA (20.59%), followed by EDV (12.9%), IMT (12.74%), CYSC (12.47%), PP (11.78%), FIB (11.05%), E/A ratio (9.3%). MVEL variable had the lowest contribution (9.17%). Using the associated factor score coefficients, a weighted composite BAS using 8 age-related variables including CA was created for each individual as follows:
BAS= 0.248(CA)+0.195(IMT)-0.196(EDV)-0.167(E/A)-0.166(MVEL)+ 0.188(PP)+0.182(FIB)+0.193(CYSC).
We calculate individual BAS for 852 healthy individuals plotted against CA with the predicted line and its 95% confidence interval (r=0.893, P<0.001). Biological aging rate predicted by BAS was accelerated with increases in CA and peaked when healthy men and women reached>75 years old.
2. The establishment of structural equation modeling of aging
A total of 108 variables including physical characteristics, cardiovascular function, blood and urine biochemical properties, and the state of the inflammation, hormonal and nutritional data, were measured and derived. After having excluded all binary variables and the variables not correlated with CA well (i.e., P>0.05 or r≤0.15) or the redundant variables for men and women,27 age-related variables were selected for the next analysis, however, estradiol and progesterone in male and female age because of changes in in a different direction are excluded. The establishment of the structural equation model of biological aging, aging-related changes of expression in the vascular structure and function (IMT, EDV, PP), cardiac structure (D, LAAPD, LALRD, AD, MVEDT), cardiac function (E/A, MVEL, MVAS, MVAA), inflammatory factors (CRP, IL6, FIB), renal function (BUN, CYSC), liver function (GPT, PRE), nutrition (ALB, HDL, TC, GLU), and blood and endothelial function (MCV, TM, TRF), of which the relationship between vascular factors and aging was greatest (r= 0.90), followed by cardiac function (r= 0.69).
Conclusion
1. The establishment of biological aging score
Established the integral formula for calculating biological aging score:BAS= 0.248 (timing of age)+0.195 (IMT)-0.196 (EDV)-0.167 (E/A)-0.166 (MVEL)+ 0.188 (PP)+0.182 (FIB)+0.193 (CYSC). Healthy aging rate increases with age to> 75 year-old group then reached its peak.
2. The establishment of the individual aging structural equation modeling
Aging associated changes in expression in the vascular structure and function, cardiac structure, cardiac function, inflammatory factors, renal function, liver function, nutrition, and blood and endothelial function related to vascular factors which most closely with aging (r= 0.90), followed by for the cardiac function (r= 0.69)。
衰老可以定义为生物系统的退行性过程,机体退行性改变的不可逆性积累和对疾病的脆弱性增加,最终导致死亡。Strehler提出了衰老的四个特点,即普遍性、渐进性、内在性和有害性。衰老是普遍的,所有种类的动物都会经历衰老过程；衰老是渐进性的,在生命的各个时期衰老都是在不断进行的；衰老是内在的,生物的衰老过程和速度有物种固有的特点；衰老是有害的,分子和细胞衰老导致组织的功能储备下降和对应激的适应能力降低。衰老不是疾病,但是它可以降低衰老相关疾病的阈值[3],增加患病机会。
目前,人口老龄化已成为世界各国所面临的严峻挑战,是影响一个国家社会、经济长远发展的战略性问题。从上世纪90年代初开始,欧美、日本等发达国家相继实施了一系列衰老纵向研究。其中代表性的为美国的Baltimore和Health ABC衰老纵向研究、加拿大的VLS研究,日本的7年纵向研究,意大利ILSA研究等。从这些研究的趋势可以看出：(1)重视衰老的个体化评价；(2)确定可靠、易测的衰老生物学标志物是衰老研究的一个重要课题；(3)研究热点向健康老龄化转移；(4)利用系统生物学方法建立整体衰老网络是未来研究的方向；(5)遗传背景、生活方式及饮食习惯是影响个体衰老的重要因素。
因为组织衰老的速度不同,随着年龄的增加,人类个体间的差异性也在增大,时序年龄(Chronological age, CA)不能提供衰老过程的准确的指示。在医学和预防角度来说,研究者关心的是个体的衰老变化,目的是筛查衰老高危个体并及时给予干预。因此为了解决以上的问题,研究者提出了“生物学年龄(Biological age,BA)”,即在个体相对于时序年龄同龄人的功能状态的基础上,能够评价个体的功能状态的生物学指标或参数。个体衰老速度不同导致时序年龄和生物学年龄的差异,因此在任何给定的时序年龄,个体间的生物学年龄的值都会相差很大,并最终可以预期与个体在长寿以及衰老过程的结果的时间和／或幅度的差异性相一致。生物学年龄的研究有助于辨别年龄相关性功能障碍的危险的个体,是定量分析个体化生物学衰老的前提和基础。
人体是一个复杂的生物体,衰老的个体化评价要用到多层次,多系统的生理指标。这种选择和评价越全面,它们的组合预测终点事件的能力就越好。在人体,系统是从整体上提炼出的抽象概念(潜变量),反应了人体的一种综合的功能状态,其本身无法被直接测量,用单一指标进行评价是不科学的,需要用其他的多个指标来表示。结构方程模型(structural equational modeling, SEM)为潜变量的表达和测量准确性的检验提供了方法。其属于多维立体网状结构,结构中所涉及的要素间不但呈上下、左右空间关系,而且呈前后的时间关系,故这一方法适用于纵向观察研究和在逻辑上具有纵向资料的性质的横向资料,如衰老研究。
本研究通过多中心大样本人群横断面研究,在全面检测包括血常规、尿常规、血液生化、心血管超声及炎症和内分泌因子等108项指标的基础上,筛选出中国健康人群衰老的生物学标志物,构建了衰老的个体化生物学评价指标-生物学年龄积分(Biological aging score, BAS),定量分析了中国健康人群生物学衰老规律。并在此基础上,扩大变量入选范围,建立生物学衰老的结构方程模型,为我国衰老的个体化评价研究提供了重要理论依据。
方法
一、生物学年龄积分的建立
1、研究对象
从沈阳、大连、北京三个城市的2876名自我评价健康的人中,筛选出健康居民852人,依据年龄分为4组,≤4岁组、45～59岁组、60～74岁组、≥75岁组。
2、检测指标
检测血压、血常规、尿常规、血液生化、心脏超声、颈动脉超声、内分泌、炎症、营养等共计108项指标。
3、数据处理
在排除二元变量后,观察各项指标随增龄变化情况,进行各指标与年龄的相关分析,与年龄显著相关且相关系数>0.25的变量进入下一步冗余分析和主成分分析,筛选衰老生物学标志物,并在探索性因子分析中利用因子得分赋予各标志物权重以构建衰老生物学年龄积分公式,计算个体生物学年龄积分(Biological aging score, BAS),分析健康人生物学衰老规律。
二、衰老个体化评价结构方程模型的建立
1、研究对象
从沈阳、大连、北京三个城市的2876名自我评价健康的人中,筛选出健康居民852人,依据年龄分为4组,≤44岁组、45～59岁组、60～74岁组、≥75岁组。
2、检测指标
检测血压、血常规、尿常规、血液生化、心脏超声、颈动脉超声、内分泌、炎症、营养等共计108项指标。
3、数据处理
首先选择与时序年龄显著相关且相关系数大于等于0.15的指标进入下一步分析,然后进行冗余分析(排除具有高度共线性的变量),在相关系数大于0.65的变量中,选择与时序年龄相关性最高的指标进入下一步分析。拟定各系统和器官标志物,确定潜变量与观测变量,画出路径图,构想理论模型,进行模型拟合与参数估计,进行模型的修正与再验证,利用各种拟合指数对模型做整体评价,同时检验参数的显著性,评价参数的意义和合理性,计算决定系数,评价方程对数据的解释能力,进行模型修正。最后确定生物学衰老结构方程模型,对模型做出解释,分析各器官和系统生物学衰老的规律。
结果
一、建立生物学年龄积分
108个指标中有58个指标与年龄相关,其中相关系数>0.25的有21个,经冗余分析后有12个指标进入下一步分析,其中雌二醇因为在男性和女性随年龄变化的方向不同,在女性随年龄减少,而在老年男性则随年龄增加,因此不适合整体公式的评价。经过主成分分析,颈动脉内膜中层厚度(IMT)、颈动脉舒张期最大前向血流速度(EDV)、脉压(PP),二尖瓣E峰／二尖瓣A峰比值(E／A)、二尖瓣环E峰侧壁(MVEL)、纤维蛋白原水平(FIB)、血清Cystatin C (CYSC)等7个指标被筛选为衰老的生物学标志物,因子分析显示只有一个特征值大于1的因子,说明这7个生物学标志物共同反映了一个衰老的基本过程。根据因子得分,建立了生物学年龄积分计算公式：BAS=0.248(时序年龄)+0.195(IMT)-0.196(EDV)-0.167(E／A)-0.166(MVEL)+0.188(PP)+0.182(FIB)+0.193(CYSC)。个体生物学年龄积分与时序年龄呈显著正相关(r=0.893,p<0.001)。由生物学年龄积分评价的生物学衰老速度随年龄的增加而加速,在60-74岁组达到峰值,75岁以后进入平台期。
二、建立衰老结构方程模型
108个指标中有58个指标与年龄相关,其中相关系数>0.15的有37个,经冗余分析后有27个指标进入下一步分析(雌二醇和黄体酮因为在男性和女性随年龄变化的方向不同被排除)。建立生物学衰老的结构方程模型,衰老相关性改变表现在血管结构和功能(IMT,EDV,PP)、心脏结构(D, LAAPD, LALRD, AD, MVEDT),心脏功能(E/A, MVEL, MVAS, MVAA)、炎症因子(CRP,IL6,FIB)、肾功能(BUN, CYSC)、肝功能(GPT,PRE)、营养(ALB,HDL,TC,GLU)和血液及内皮功能(MCV,TM,TRF)有关,其中血管因子与衰老的关系最为密切(r=0.90),其次为心脏功能(r=0.69)。
结论
一、建立生物学年龄积分
建立了生物学年龄积分计算公式：BAS=0.248(时序年龄)+0.195(IMT)-0.196(EDV)-0.167(E／A)-0.166(MVEL)+0.188(PP)+0.182(FIB)+0.193(CYSC)。健康人的衰老速度随年龄的增加而增加,至>75岁组达到峰值。
二、建立衰老的个体化评价结构方程模型
衰老相关性改变表现在血管结构和功能、心脏结构、心脏功能、炎症因子、肾功能、肝功能、营养和血液及内皮功能有关,其中血管因子与衰老的关系最为密切(r=0.90),其次为心脏功能(r=0.69)。
Objective
Aging can be defined as a degenerative process of biological system, the accumulation of the body irreversible degenerative changes increased vulnerability to disease and eventually leaded to death. Strehler proposed four characteristics of aging, namely, universal, progressive, intrinsic and harmfu. Aging is universal, none of animals could escape the aging process; aging is gradual in nature, aging are ongoing in various periods of life; aging is inherent, biological aging process and its speed have inherent characteristics of species; Aging is harmful, molecular and cellular senescence declined the functional reserve and reduced ability to adapt to the stress. Aging is not a disease, but it can reduce the threshold of aging-related diseases and thus increased chance of illness.
At present, most of the countries in the world must face the problems of the population aging which were serious challenges to a country's social and economic long-term development. From the early 90s of last century, many developed countries such as Europe, the United States and Japan have implemented a series of longitudinal studies of aging. Representative studies were American Baltimore study, the Health ABC Longitudinal Study,Canada's VLS study Japan's seven-year longitudinal study,the Italian ILSA study and so on. The trends seen from these studies were: (1) emphasis on the individual assessment of aging; (2) the reliable, easily measured biomarkers of aging is an important topic in aging research; (3) research focus to healthy aging transfer; (4) the use of systems biology approach to establish the overall aging network is the direction of future research; (5) genetic background, lifestyle and dietary habits are important factors affecting the aging individual.
Because the organization aged at different speeds, the differences among human individuals increased with increasing age. Chronological age (CA) can not provide the exact instructions for the aging process In medicine and prevention perspective, researchers concerned with aging changes in the individuals. The purposes of them were screening high risk individuals and providing timely intervention. Therefore, to solve the above problems, the researchers proposed the "biological age (BA)" which defined as the fuctional status of individuals relative to their chronological peers on the basement of the biological indicators or parameters Different rates of individual aging leads to the differences in chronological age and biological age so at any given age, biological age of individuals would be showing great differences and ultimately can be expected to be consistent with the time&magnitude of the differences of longevity and results of the aging process. Biological age study was the premise and foundation of quantitative analysis of individual aging and will help identifying the high risk individuals
The human body is a complex organism, aging of the individual evaluation of use to the multi-level, multi-system physiological indicators. This selection and evaluation were more comprehensive, their combined ability to predict the better end of the event. In the body, the system is extracted from the whole of the abstractions (latent variables), the reaction of the human body, a comprehensive functional status, by itself, can not be directly measured, using a single indicator is unscientific to evaluate the need to use other various indicators to be expressed. Structural equation modeling (SEM) for the latent variable expression and measurement accuracy provides a method of testing. It belongs to multi-dimensional three-dimensional network structure, the structure of the elements involved in not only vertically between the left and right spatial relations, but also was the time before and after, so this method is suitable for longitudinal observational studies and in the logical nature of the information with a horizontal longitudinal information, such as aging research.
This study was population-based multi-center cross-sectional study. On the basis of 108 indicators including blood routine, urine routine, blood biochemistry, cardiovascular ultrasound and endocrine factors of healthy people in China, the biomarkers of aging were screened, individual biological evaluation indicators biological age points were constructed, the law of biological aging process in healthy people in China was quantitative analyzed. And on this basis, to expand the scope of the variable selected to establish the structural equation model of biological aging. This study provided the important theoretical basis for our aging individual evaluation studies.
Methods
1. Establishment of biological aging score
(1) Subjects
2876 independently living and apparently healthy community-dwelling participants between 30 and 98 years of age were recruited from Beijing, Shenyang, and Dalian cities in China in 2003. Consequently, a total of 852 participants (392 men and 460 women) were enrolled. They were classified as young (< 45 yr), middle-aged (45-59 yr), old (60-74 yr), and very-old (≥75 yr). Informed consent was obtained from all participants.
(2) Detection indicators
a total of 108 indicators were examined including blood pressure, blood routine, urine routine, blood biochemistry, echocardiography, carotid artery ultrasound, endocrine, inflammation and nutrition.
(3) Data processing
we excluded the binary variablesand observed the changes of variableswith aging. We performed correlation analysis and choose those variables that were significantly correlated with age and the correlation coefficient> 0.25 of the variables for the next redundancy analysis and principal component analysis in order to screen the biomarkers of aging. Then we performed exploratory factor analysis and use factor scores weighting the biomarkers of aging for building the formula to calculate an individual biological age scores and analyzed the laws of biological aging in healthy population.
2. Individual aging evaluation by structural equation modeling
(1) Subjects
2876 independently living and apparently healthy community-dwelling participants between 30 and 98 years of age were recruited from Beijing, Shenyang, and Dalian cities in China in 2003. Consequently, a total of 852 participants (392 men and 460 women) were enrolled. They were classified as young (< 45 yr), middle-aged (45-59 yr), old (60-74 yr), and very-old (≥75 yr). Informed consent was obtained from all participants.
(2) Detection indicators
A total of 108 indicators were examined including blood pressure, blood routine, urine routine, blood biochemistry, echocardiography, carotid artery ultrasound, endocrine, inflammation and nutrition.
(3) Data processing
First we choose those variables that were significantly correlated with age and correlation coefficient was greater than or equal to 0.15 for redundancy analysis (excluding the highly collinear variables). For those variables with the correlation coefficient greater than 0.65, we selected the one with the highest correlation coefficient with age. Then we developed various systems and organs markers to determine the latent variables and observed variables, draw the road map, construct theoretical models, carry out model and estimate the fitting index, make amendment and re-validation by a variety of fit indices of the model to do the whole evaluation, while the significance of test parameters to evaluate the significance and reasonableness of the parameters to calculate the coefficient of determination, the evaluation equation interpretation of the data capacity, model updating. Finalize the biology of aging structural equation model to explain the model, analysis of various organs and systems biology of aging pattern.
Results
1. The establishment of biological aging score
A total of 108 variables including physical characteristics, cardiovascular function, blood and urine biochemical properties, and the state of the inflammation, hormonal and nutritional data, were measured and derived. After having excluded all binary variables and the variables not correlated with CA well (i.e., P>0.05 or r<0.25) or the redundant variables for men and women,12 age-related variables were selected for the analysis of the correlation with CA. At last,8 variables:CA, IMT, EDV, E/A, MVEL, PP, FIB and CYSC, were selected as biomarkers to evaluate BA. The contribution of each variable to the variance of BAS were as follows:CA (20.59%), followed by EDV (12.9%), IMT (12.74%), CYSC (12.47%), PP (11.78%), FIB (11.05%), E/A ratio (9.3%). MVEL variable had the lowest contribution (9.17%). Using the associated factor score coefficients, a weighted composite BAS using 8 age-related variables including CA was created for each individual as follows:
BAS= 0.248(CA)+0.195(IMT)-0.196(EDV)-0.167(E/A)-0.166(MVEL)+ 0.188(PP)+0.182(FIB)+0.193(CYSC).
We calculate individual BAS for 852 healthy individuals plotted against CA with the predicted line and its 95% confidence interval (r=0.893, P<0.001). Biological aging rate predicted by BAS was accelerated with increases in CA and peaked when healthy men and women reached>75 years old.
2. The establishment of structural equation modeling of aging
A total of 108 variables including physical characteristics, cardiovascular function, blood and urine biochemical properties, and the state of the inflammation, hormonal and nutritional data, were measured and derived. After having excluded all binary variables and the variables not correlated with CA well (i.e., P>0.05 or r≤0.15) or the redundant variables for men and women,27 age-related variables were selected for the next analysis, however, estradiol and progesterone in male and female age because of changes in in a different direction are excluded. The establishment of the structural equation model of biological aging, aging-related changes of expression in the vascular structure and function (IMT, EDV, PP), cardiac structure (D, LAAPD, LALRD, AD, MVEDT), cardiac function (E/A, MVEL, MVAS, MVAA), inflammatory factors (CRP, IL6, FIB), renal function (BUN, CYSC), liver function (GPT, PRE), nutrition (ALB, HDL, TC, GLU), and blood and endothelial function (MCV, TM, TRF), of which the relationship between vascular factors and aging was greatest (r= 0.90), followed by cardiac function (r= 0.69).
Conclusion
1. The establishment of biological aging score
Established the integral formula for calculating biological aging score:BAS= 0.248 (timing of age)+0.195 (IMT)-0.196 (EDV)-0.167 (E/A)-0.166 (MVEL)+ 0.188 (PP)+0.182 (FIB)+0.193 (CYSC). Healthy aging rate increases with age to> 75 year-old group then reached its peak.
2. The establishment of the individual aging structural equation modeling
Aging associated changes in expression in the vascular structure and function, cardiac structure, cardiac function, inflammatory factors, renal function, liver function, nutrition, and blood and endothelial function related to vascular factors which most closely with aging (r= 0.90), followed by for the cardiac function (r= 0.69)。
引文
1 De la Fuente M. Role of neuroimmunomodulation in aging. Neuroimmunomodulation,2008, 15:213-223.
2 Strehler BL. Time, Cells and Aging. New York:Academic Press,1977.5-30.
3 李虹,白小涓,陈香美.血管衰老机理及检测指标的研究进展.中华医学杂志,2005,85：212-215.
4 Ferrucci L. The Baltimore Longitudinal Study of Aging (BLSA):A 50-Year-Long Journey and Plans for the Future. J Gerontol A Biol Sci Med Sci,2008,63:1416-1419.
5 Njajou OT, Hsueh WC, Blackburn EH, et al. Association between telomere length, specific causes of death, and years of healthy life in health, aging, and body composition, a population-based cohort study. J Gerontol A Biol Sci Med Sci,2009,64:860-864.
6 MacDonald SW, Dixon RA, Cohen AL, et al. Biological age and 12-year cognitive change in older adults:findings from the Victoria Longitudinal Study. Gerontology,2004,50:64-81.
7 Nakamura E, Miyao K. Sex differences in human biological aging. J Gerontol A Biol Sci Med Sci,2008,63:936-944.
8 Inzitari M, Carlo A, Baldereschi M, et al. Risk and predictors of motor-performance decline in a normally functioning population-based sample of elderly subjects:the Italian Longitudinal Study on Aging. J Am Geriatr Soc,2006,54:318-324.
9 Spirduso W, Francis K, MacRae P. Physical Dimensions of Aging.2th ed. Champaign:Human Kinetics,2005.32-35.
10 Anstey KJ, Lord SR, Smith GA. Measuring human functional age:a review of empirical findings. Exp Aging Res,1996,22:245-266.
11 Borkan GA, Norris AH. Assessment of biological age using a profile of physical parameters. J Gerontol,1980,35:177-184.
12 Miller RA. Biomarkers of aging. Sci Aging Knowledge Environ,2001,2001:2.
13 Johnson TE. Resent result:Biomarkers pf aging. Exp Gerontol,2006,41:1243-1246.
14 http://www.infoaging.org/b-biomarker-home.html; 2005.
15 Klemera P, Doubal S. A new approach to the concept and computation of biological age. Mech Ageing Dev.2006; 127(3):240-248.
16 Franklin SS:Ageing and hypertension:the assessment of blood pressure indices in predicting coronary heart disease. J Hypertens Suppl.1999; 17(5):S29-36.
17 Watanabe S, Suzuki N, Kudo A, et al. Influence of aging on cardiac function examined by echocardiography. Tohoku J Exp Med.2005;207(1):9-13
18 Jurasic MJ, Lovrencic-Huzjan A, Bedekovic MR, et al. How to monitor vascular aging with an ultrasound. J Neurol Sci.2007; 257(1-2):139-142.
19 19Tracy RP. Hemostatic and inflammatory markers as risk factors for coronary disease in the elderly. Am J Geriatr Cardiol.2002; 11(2):93-100
20 Fliser D, Ritz E:Serum cystatin C concentration as a marker of renal dysfunction in the elderly. Am J Kidney Dis.2001; 37(1):79-83.
21 Sarnak MJ, Katz R, Fried LF, et al. Cardiovascular Health Study:Cystatin C and aging success. Arch Intern Med.2008; 168(2):147-53.
1 http://www.stats.gov.cn/tjgb/ndtjgb/qgndtjgb/t20090226_402540710.htm.2008年国民经济和社会发展统计公报.国家统计局,2009,2.16
2 De la Fuente M. Role of neuroimmunomodulation in aging. Neuroimmun-omodulation.2008; 15(4-6):213-223.
3 李虹,白小涓,陈香美.血管衰老机理及检测指标的研究进展[J].中华医学杂志,2005；85(3)：212-215.
4 Bai X, Han L, Liu Q, et al. Evaluation of Biological Aging Process-A Population-Based Study of Healthy People in China. Gerontology.2010:56(2): 129-140
5 Han L, Bai X, Lin H, et al. Lack of independent relationship between ageing-related kidney function decline and carotid intima-media thickness in a healthy Chinese population. Nephrol Dial Transplant.2009; DOI:10. 1093/ndt/gfp718.
6 Kriete A, Sokhansanj BA, Coppock DL, et al. Systems approaches to the networks of aging. Ageing Res Rev.2006; 5(4):434-448.
7 West GB, Bergman A. Toward a systems biology framework for understanding aging and health span. J Gerontol A Biol Sci Med Sci.2009; 64(2):205-208.
8 Thomas B. L. Kirkwood. A systematic look at an old problem. Nature.2008; 451: 644-647.
9 Khrapko K. The complexity of aging:are some aging processes more equal than others? Mech Ageing Dev.2007; 128:463-465.
10 Growley SL. Structural equation modeling:basic concepts and applications in personality assessment research. Journal of Personality Assessment.1997; 68(3): 508-531.
11 Hoyle RH. Introduction to the special section:Structural equation modeling in clinical research. Journal of Consulting and Clinical Psychology.1994; 62 (3):427.
12 Carlsson M, Hamrin E. Evaluation of the life satisfaction questionnaire using structural equation modeling. Qual Life Res.2002; 11 (5):415.
13 Sousa KH, Ryu E, Kwok OM, et al. Development of a model to measure symptom status in persons living with rheumatoid arthritis. Nurs Res.2007; 56(6):434-440.
14 Selim M, Jones R, Novak P, et al. The effects of body mass index on cerebral blood flow velocity. Clin Auton Res.2008; 18(6):331-338.
15 Sullivan MD, Kempen GI, Van Sonderen E. Models of health-related quality of life in a population of community-dwelling Dutch elderly. Qual Life Res.2000; 9(7):801-810.
16 Nakamura E, Miyao K. Further evaluation of the basic nature of the human biological aging process based on a factor analysis of age-related physiological variables. J Gerontol A Biol Sci Med Sci.2003; 58(3):196-204.
17 Kriete A. Biomarkers of aging:combinatorial or systems model? Sci Aging Knowledge Environ.2006; 2006(1):1-4
18 Watanabe S, Suzuki N, Kudo A, et al. Influence of aging on cardiac function examined by echocardiography. Tohoku J Exp Med.2005; 207(1):9-13
19 Tracy RP:Hemostatic and inflammatory markers as risk factors for coronary disease in the elderly. Am J Geriatr Cardiol.2002; 11(2):93-100
20 Sarnak MJ, Katz R, Fried LF, et al. Cardiovascular Health Study:Cystatin C and aging success. Arch Intern Med.2008; 168(2):147-153.
21 Komatsumoto S, Onoda A, Nara M. Changes in the level of thrombomodulin with aging. Nippon Ronen Igakkai Zasshi.1996:33(7):512-517.
1 De la Fuente M. Role of neuroimmunomodulation in aging. Neuroimmunomodulation,2008, 15:213-223.
2 Strehler BL. Time, Cells and Aging. New York:Academic Press,1977.5-30
3 李虹,白小涓,陈香美.血管衰老机理及检测指标的研究进展.中华医学杂志,2005,85：212-215
4 Ferrucci L. The Baltimore Longitudinal Study of Aging (BLSA):A 50-Year-Long Journey and Plans for the Future. J Gerontol A Biol Sci Med Sci,2008,63:1416-1419.
5 Njajou OT, Hsueh WC, Blackburn EH, et al. Association between telomere length, specific causes of death, and years of healthy life in health, aging, and body composition, a population-based cohort study. J Gerontol A Biol Sci Med Sci,2009,64:860-864.
6 MacDonald SW, Dixon RA, Cohen AL, et al. Biological age and 12-year cognitive change in older adults:findings from the Victoria Longitudinal Study. Gerontology,2004,50:64-81.
7 Nakamura E, Miyao K. Sex differences in human biological aging. J Gerontol A Biol Sci Med Sci,2008,63:936-944.
8 Inzitari M, Carlo A, Baldereschi M, et al. Risk and predictors of motor-performance decline in a normally functioning population-based sample of elderly subjects:the Italian Longitudinal Study on Aging. J Am Geriatr Soc,2006,54:318-324.
9 Spirduso W, Francis K, MacRae P. Physical Dimensions of Aging.2th ed. Champaign:Human Kinetics,2005.32-35
10 Anstey KJ, Lord SR, Smith GA. Measuring human functional age:a review of empirical findings. Exp Aging Res,1996,22:245-266.
11 Borkan GA, Norris AH. Assessment of biological age using a profile of physical parameters. J Gerontol,1980,35:177-184.
12 Miller RA. Biomarkers of aging. Sci Aging Knowledge Environ,2001,2001:2.
13 Miller RA. Reference standards in evaluating system performance. J Am Med Inform Assoc, 2002,9:87-88.
14 Mitnitski AB, Graham JE, Mogilner AJ, et al. Frailty, fitness and late-life mortality in relation to chronological and biological age. BMC Geriatr,2002,27:1.
15 Uttley M, Crawford MH. Efficacy of a composite biological age score to predict ten-year survival among Kansas and Nebraska Mennonites. Hum Biol,1994,66:121-144.
16 Johnson TE. Resent result:Biomarkers pf aging. Exp Gerontol,2006,41:1243-1246
17 http://www.infoaging.org/b-biomarker-home. html; 2005.
18 Kriete A. Biomarkers of aging:combinatorial or systems model? Sci Aging Knowledge Environ,2006,2006:1.
19 Brant LJ, Verbeke GN. Describing the natural heterogeneity of aging using multilevel regression models. Int J Sports Med,1997,18:S225-S231.
20 Nakamura E, Miyao K. Further evaluation of the basic nature of the human biological aging process based on a factor analysis of age-related physiological variables. J Gerontol A Biol Sci Med Sci,2003,58:196-204.
21 McClearn GE, Vogler GP, Hofer SM. Enviroment-gene and gene-gene interactions.5th ed. San Francisco:Academic Press,2001.423-444.
22 Webster IW, Logie AR. A relationship between functional age and health status in female subjects. J Gerontol,1976,31:546-550.
23 Dubina TL, Dyundikova VA, Zhuk EV. Biological age and its estimation Ⅱ:Assessment of biological age of albino rats by multiple regression analysis. Exp Gerontol,1983,18:5-18.
24白小涓,刘强,韩璐璐,等.中国北方健康人群心血管衰老生物学定量评价与规律分析.中华医学杂志,2007,87：2385-2389
25 Seeman TE, McEwen BS, Rowe JW, et al. Allostatic load as a marker of cumulative biological risk:MacArthur studies of successful aging. Proc Natl Acad Sci U S A,2001,98:4770-4775.
26 Karlamangla AS, Singer BH, McEwen BS, et al. Allostatic load as a predictor of functional decline. MacArthur studies of successful aging. J Clin Epidemiol,2002,55:696-710.
27 Seeman TE, Singer BH, Rowe JW. Price of adaptation-allostatic load and its health consequences. MacArthur studies of successful aging. Arch Intern Med,1997,157:2259-2268.
28 Seeman E, Hopper JL. Genetic and environmental components of the population variance in bone density. Osteoporos Int,1997,7:S10-S16.
29 Anderson JC,Gerbin DW. Structural equation modeling in practice:A review and recommended two-step approach. Psychol Bull,1998,103:411-423.
30 Growley SL, Fan X. Structural equation modeling:basic concepts and applications in personality assessment research. J Pers Assess,1997,68:508-531.
31 侯杰泰.结构方程模型及其应用.北京：教育科学出版社,2004.15-20.
32 Sousa KH, Ryu E, Kwok OM, et al. Development of a model to measure symptom status in persons living with rheumatoid arthritis. Nurs Res,2007,56:434-440.
33 Selim M, Jones R, Novak P, et al. The effects of body mass index on cerebral blood flow velocity. Clin Auton Res,2008,18:331-338.
34 Kriete A, Sokhansanj BA, Coppock DL, et al. Systems approaches to the networks of aging. Ageing Res Rev,2006,5:434-448.
35 Kirkwood TB. A systematic look at an old problem. Nature,2008,451:644-647
36 Sullivan MD, Kempen GI, Van Sonderen E. Models of health-related quality of life in a population of community-dwelling Dutch elderly. Qual Life Res,2000,9:801-810.
37 Guinot C, Malvy DJ, Ambroisine L. Relative contribution of intrinsic vs extrinsic factors to skin aging as determined by a validated skin age score. Arch Dermatol,2002,138:1454-1460.
38 Deshpande N, Metter EJ, Ling S, et al. Physiological correlates of age-related decline in vibrotactile sensitivity. Neurobiol Aging,2008,29:765-773.
2 Strehler BL. Time, Cells and Aging. New York:Academic Press,1977.5-30.
3 李虹,白小涓,陈香美.血管衰老机理及检测指标的研究进展.中华医学杂志,2005,85：212-215.
4 Ferrucci L. The Baltimore Longitudinal Study of Aging (BLSA):A 50-Year-Long Journey and Plans for the Future. J Gerontol A Biol Sci Med Sci,2008,63:1416-1419.
5 Njajou OT, Hsueh WC, Blackburn EH, et al. Association between telomere length, specific causes of death, and years of healthy life in health, aging, and body composition, a population-based cohort study. J Gerontol A Biol Sci Med Sci,2009,64:860-864.
6 MacDonald SW, Dixon RA, Cohen AL, et al. Biological age and 12-year cognitive change in older adults:findings from the Victoria Longitudinal Study. Gerontology,2004,50:64-81.
7 Nakamura E, Miyao K. Sex differences in human biological aging. J Gerontol A Biol Sci Med Sci,2008,63:936-944.
8 Inzitari M, Carlo A, Baldereschi M, et al. Risk and predictors of motor-performance decline in a normally functioning population-based sample of elderly subjects:the Italian Longitudinal Study on Aging. J Am Geriatr Soc,2006,54:318-324.
9 Spirduso W, Francis K, MacRae P. Physical Dimensions of Aging.2th ed. Champaign:Human Kinetics,2005.32-35.
10 Anstey KJ, Lord SR, Smith GA. Measuring human functional age:a review of empirical findings. Exp Aging Res,1996,22:245-266.
11 Borkan GA, Norris AH. Assessment of biological age using a profile of physical parameters. J Gerontol,1980,35:177-184.
12 Miller RA. Biomarkers of aging. Sci Aging Knowledge Environ,2001,2001:2.
13 Johnson TE. Resent result:Biomarkers pf aging. Exp Gerontol,2006,41:1243-1246.
14 http://www.infoaging.org/b-biomarker-home.html; 2005.
15 Klemera P, Doubal S. A new approach to the concept and computation of biological age. Mech Ageing Dev.2006; 127(3):240-248.
16 Franklin SS:Ageing and hypertension:the assessment of blood pressure indices in predicting coronary heart disease. J Hypertens Suppl.1999; 17(5):S29-36.
17 Watanabe S, Suzuki N, Kudo A, et al. Influence of aging on cardiac function examined by echocardiography. Tohoku J Exp Med.2005;207(1):9-13
18 Jurasic MJ, Lovrencic-Huzjan A, Bedekovic MR, et al. How to monitor vascular aging with an ultrasound. J Neurol Sci.2007; 257(1-2):139-142.
19 19Tracy RP. Hemostatic and inflammatory markers as risk factors for coronary disease in the elderly. Am J Geriatr Cardiol.2002; 11(2):93-100
20 Fliser D, Ritz E:Serum cystatin C concentration as a marker of renal dysfunction in the elderly. Am J Kidney Dis.2001; 37(1):79-83.
21 Sarnak MJ, Katz R, Fried LF, et al. Cardiovascular Health Study:Cystatin C and aging success. Arch Intern Med.2008; 168(2):147-53.
1 http://www.stats.gov.cn/tjgb/ndtjgb/qgndtjgb/t20090226_402540710.htm.2008年国民经济和社会发展统计公报.国家统计局,2009,2.16
2 De la Fuente M. Role of neuroimmunomodulation in aging. Neuroimmun-omodulation.2008; 15(4-6):213-223.
3 李虹,白小涓,陈香美.血管衰老机理及检测指标的研究进展[J].中华医学杂志,2005；85(3)：212-215.
4 Bai X, Han L, Liu Q, et al. Evaluation of Biological Aging Process-A Population-Based Study of Healthy People in China. Gerontology.2010:56(2): 129-140
5 Han L, Bai X, Lin H, et al. Lack of independent relationship between ageing-related kidney function decline and carotid intima-media thickness in a healthy Chinese population. Nephrol Dial Transplant.2009; DOI:10. 1093/ndt/gfp718.
6 Kriete A, Sokhansanj BA, Coppock DL, et al. Systems approaches to the networks of aging. Ageing Res Rev.2006; 5(4):434-448.
7 West GB, Bergman A. Toward a systems biology framework for understanding aging and health span. J Gerontol A Biol Sci Med Sci.2009; 64(2):205-208.
8 Thomas B. L. Kirkwood. A systematic look at an old problem. Nature.2008; 451: 644-647.
9 Khrapko K. The complexity of aging:are some aging processes more equal than others? Mech Ageing Dev.2007; 128:463-465.
10 Growley SL. Structural equation modeling:basic concepts and applications in personality assessment research. Journal of Personality Assessment.1997; 68(3): 508-531.
11 Hoyle RH. Introduction to the special section:Structural equation modeling in clinical research. Journal of Consulting and Clinical Psychology.1994; 62 (3):427.
12 Carlsson M, Hamrin E. Evaluation of the life satisfaction questionnaire using structural equation modeling. Qual Life Res.2002; 11 (5):415.
13 Sousa KH, Ryu E, Kwok OM, et al. Development of a model to measure symptom status in persons living with rheumatoid arthritis. Nurs Res.2007; 56(6):434-440.
14 Selim M, Jones R, Novak P, et al. The effects of body mass index on cerebral blood flow velocity. Clin Auton Res.2008; 18(6):331-338.
15 Sullivan MD, Kempen GI, Van Sonderen E. Models of health-related quality of life in a population of community-dwelling Dutch elderly. Qual Life Res.2000; 9(7):801-810.
16 Nakamura E, Miyao K. Further evaluation of the basic nature of the human biological aging process based on a factor analysis of age-related physiological variables. J Gerontol A Biol Sci Med Sci.2003; 58(3):196-204.
17 Kriete A. Biomarkers of aging:combinatorial or systems model? Sci Aging Knowledge Environ.2006; 2006(1):1-4
18 Watanabe S, Suzuki N, Kudo A, et al. Influence of aging on cardiac function examined by echocardiography. Tohoku J Exp Med.2005; 207(1):9-13
19 Tracy RP:Hemostatic and inflammatory markers as risk factors for coronary disease in the elderly. Am J Geriatr Cardiol.2002; 11(2):93-100
20 Sarnak MJ, Katz R, Fried LF, et al. Cardiovascular Health Study:Cystatin C and aging success. Arch Intern Med.2008; 168(2):147-153.
21 Komatsumoto S, Onoda A, Nara M. Changes in the level of thrombomodulin with aging. Nippon Ronen Igakkai Zasshi.1996:33(7):512-517.
1 De la Fuente M. Role of neuroimmunomodulation in aging. Neuroimmunomodulation,2008, 15:213-223.
2 Strehler BL. Time, Cells and Aging. New York:Academic Press,1977.5-30
3 李虹,白小涓,陈香美.血管衰老机理及检测指标的研究进展.中华医学杂志,2005,85：212-215
4 Ferrucci L. The Baltimore Longitudinal Study of Aging (BLSA):A 50-Year-Long Journey and Plans for the Future. J Gerontol A Biol Sci Med Sci,2008,63:1416-1419.
5 Njajou OT, Hsueh WC, Blackburn EH, et al. Association between telomere length, specific causes of death, and years of healthy life in health, aging, and body composition, a population-based cohort study. J Gerontol A Biol Sci Med Sci,2009,64:860-864.
6 MacDonald SW, Dixon RA, Cohen AL, et al. Biological age and 12-year cognitive change in older adults:findings from the Victoria Longitudinal Study. Gerontology,2004,50:64-81.
7 Nakamura E, Miyao K. Sex differences in human biological aging. J Gerontol A Biol Sci Med Sci,2008,63:936-944.
8 Inzitari M, Carlo A, Baldereschi M, et al. Risk and predictors of motor-performance decline in a normally functioning population-based sample of elderly subjects:the Italian Longitudinal Study on Aging. J Am Geriatr Soc,2006,54:318-324.
9 Spirduso W, Francis K, MacRae P. Physical Dimensions of Aging.2th ed. Champaign:Human Kinetics,2005.32-35
10 Anstey KJ, Lord SR, Smith GA. Measuring human functional age:a review of empirical findings. Exp Aging Res,1996,22:245-266.
11 Borkan GA, Norris AH. Assessment of biological age using a profile of physical parameters. J Gerontol,1980,35:177-184.
12 Miller RA. Biomarkers of aging. Sci Aging Knowledge Environ,2001,2001:2.
13 Miller RA. Reference standards in evaluating system performance. J Am Med Inform Assoc, 2002,9:87-88.
14 Mitnitski AB, Graham JE, Mogilner AJ, et al. Frailty, fitness and late-life mortality in relation to chronological and biological age. BMC Geriatr,2002,27:1.
15 Uttley M, Crawford MH. Efficacy of a composite biological age score to predict ten-year survival among Kansas and Nebraska Mennonites. Hum Biol,1994,66:121-144.
16 Johnson TE. Resent result:Biomarkers pf aging. Exp Gerontol,2006,41:1243-1246
17 http://www.infoaging.org/b-biomarker-home. html; 2005.
18 Kriete A. Biomarkers of aging:combinatorial or systems model? Sci Aging Knowledge Environ,2006,2006:1.
19 Brant LJ, Verbeke GN. Describing the natural heterogeneity of aging using multilevel regression models. Int J Sports Med,1997,18:S225-S231.
20 Nakamura E, Miyao K. Further evaluation of the basic nature of the human biological aging process based on a factor analysis of age-related physiological variables. J Gerontol A Biol Sci Med Sci,2003,58:196-204.
21 McClearn GE, Vogler GP, Hofer SM. Enviroment-gene and gene-gene interactions.5th ed. San Francisco:Academic Press,2001.423-444.
22 Webster IW, Logie AR. A relationship between functional age and health status in female subjects. J Gerontol,1976,31:546-550.
23 Dubina TL, Dyundikova VA, Zhuk EV. Biological age and its estimation Ⅱ:Assessment of biological age of albino rats by multiple regression analysis. Exp Gerontol,1983,18:5-18.
24白小涓,刘强,韩璐璐,等.中国北方健康人群心血管衰老生物学定量评价与规律分析.中华医学杂志,2007,87：2385-2389
25 Seeman TE, McEwen BS, Rowe JW, et al. Allostatic load as a marker of cumulative biological risk:MacArthur studies of successful aging. Proc Natl Acad Sci U S A,2001,98:4770-4775.
26 Karlamangla AS, Singer BH, McEwen BS, et al. Allostatic load as a predictor of functional decline. MacArthur studies of successful aging. J Clin Epidemiol,2002,55:696-710.
27 Seeman TE, Singer BH, Rowe JW. Price of adaptation-allostatic load and its health consequences. MacArthur studies of successful aging. Arch Intern Med,1997,157:2259-2268.
28 Seeman E, Hopper JL. Genetic and environmental components of the population variance in bone density. Osteoporos Int,1997,7:S10-S16.
29 Anderson JC,Gerbin DW. Structural equation modeling in practice:A review and recommended two-step approach. Psychol Bull,1998,103:411-423.
30 Growley SL, Fan X. Structural equation modeling:basic concepts and applications in personality assessment research. J Pers Assess,1997,68:508-531.
31 侯杰泰.结构方程模型及其应用.北京：教育科学出版社,2004.15-20.
32 Sousa KH, Ryu E, Kwok OM, et al. Development of a model to measure symptom status in persons living with rheumatoid arthritis. Nurs Res,2007,56:434-440.
33 Selim M, Jones R, Novak P, et al. The effects of body mass index on cerebral blood flow velocity. Clin Auton Res,2008,18:331-338.
34 Kriete A, Sokhansanj BA, Coppock DL, et al. Systems approaches to the networks of aging. Ageing Res Rev,2006,5:434-448.
35 Kirkwood TB. A systematic look at an old problem. Nature,2008,451:644-647
36 Sullivan MD, Kempen GI, Van Sonderen E. Models of health-related quality of life in a population of community-dwelling Dutch elderly. Qual Life Res,2000,9:801-810.
37 Guinot C, Malvy DJ, Ambroisine L. Relative contribution of intrinsic vs extrinsic factors to skin aging as determined by a validated skin age score. Arch Dermatol,2002,138:1454-1460.
38 Deshpande N, Metter EJ, Ling S, et al. Physiological correlates of age-related decline in vibrotactile sensitivity. Neurobiol Aging,2008,29:765-773.