Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/44874
Title: | 心臟超音波指標與心血管事件及總死亡事件之相關探討 – 華人社區民眾世代追蹤研究 Echocardiographic Parameters and the risk of Cardiovascular Events and All-Cause Death - A Longitudinal Cohort Study in An Ethnic Chinese Community |
Authors: | Chao-Lun Lai 賴超倫 |
Advisor: | 簡國龍 |
Keyword: | 心臟超音波,危險因子,心血管疾病,腦中風,死亡, cardiovascular diseases,echocardiography,mortality,risk factors,stroke, |
Publication Year : | 2010 |
Degree: | 博士 |
Abstract: | 背景與目標
由於心血管疾病具有高發生率與高盛行率,又是重大死因,因此如何進行心血管疾病的防治,多年來是已開發國家公共衛生上的重要課題。心臟超音波檢查,具有非侵入性、無痛無害、可反覆檢查等許多優勢,是臨床應用及流行病學研究的重要工具。本研究之主要目標在於:從心臟超音波檢查所獲得的資料中,尋找適合於罹病預測、危險分級的重要指標,以方便未來利用超音波指標,篩選出高危險族群,進行重點防治。 方法 本研究利用台灣金山社區心血管疾病世代研究(Chin-Shan Community Cardiovascular Cohort Study, CCCC Study)的資料,以世代研究(cohort study)的研究設計,分析左心室質量、左心房直徑與主動脈根部直徑,3個心臟超音波指標,與腦中風、心血管事件和總死亡事件之間的關係。針對其中具有獨立的預後預測價值的指標,界定出適用於華人族群的危險閾值。並分析在不同的次群體(如性別、年齡)間,個別指標與臨床事件之間的關係是否有所不同。 結果 在左心室質量的研究中,一共有2604名個案被納入,追蹤時間的中位數達到14.4年。在追蹤了34325人年之後,發生心血管事件的個案數為205位,453名個案死亡。左心室質量一律先以體表面積進行校正後,得到左心室質量指數(left ventricular mass index, LVMI),再將個案依LVMI平分成五等分進行分析。經由多變項回歸調整後,LVMI分層與心血管事件的發生有顯著相關:LVMI最高五等分的個案,相較於LVMI最低五等分的個案,發生心血管事件的相對危險是2.01(95% confidence interval [CI], 1.11 to 3.63)倍(p for trend = 0.001)。此外,LVMI與總死亡事件的相關性,違背了線性關係的假設(p = 0.003 for test of linearity),呈現J型曲線模式。LVMI第二個五等分的個案與LVMI第三個五等分的個案,相較於LVMI最低五等分的個案,擁有比較低的總死亡事件相對危險,其相對危險分別是0.58(95% CI, 0.40 to 0.84)倍及0.68(95% CI, 0.47 to 0.96)倍。男性與女性罹患心血管事件的共同危險閾值之LVMI為105 g/m2。 在左心房直徑的研究中,一共收納了1937名個案,追蹤了21733人年(追蹤時間中位數:11.9年)。發生腦中風的個案數為114位,此外,364名個案死亡。由統計檢定,發現身體質量指數最適於用來對左心房直徑進行體型校正,因此以身體質量指數校正後的左心房直徑指數(left atrial dimension index, LADI)進行分析。 所有個案依LADI平分成三等分。以女性而言,經由多變項回歸調整後,LADI高等分的個案,相較於LADI低等分的個案,發生腦中風的相對危險是2.44(95% CI, 1.11 to 5.36)倍(p for trend = 0.029)。不過,此相關性在加入了LVMI做為調整變項後,有明顯減弱的現象(相對危險:2.11倍, 95% CI, 0.88 to 5.02, p for trend = 0.09)。女性發生腦中風的LADI危險閾值為1.39 mm/(kg/m2)。以男性而言,LADI的分層與腦中風的發生無關。此外,不論男女,LADI的分層與總死亡事件的發生無關。 在主動脈根部直徑的研究中,共收納1851名個案,經過11.9年的追蹤,累積達20800人年,共185名個案發生心血管事件,335名個案死亡。主動脈根部直徑以體表面積進行校正後,得到主動脈根部直徑指數(aortic root dimension index, ARDI),再將個案依ARDI平分成三等分進行分析。雖然在單變項回歸模型中,ARDI與心血管事件及總死亡事件皆有相關,但是,一旦調整了性別與年齡因素後,所有的相關都變成不顯著 (心血管事件:p for trend = 0.12;總死亡事件:p for trend = 0.23)。在次群體分析則發現:在年齡<65歲的個案中,ARDI分層與總死亡事件有顯著相關。在多變項回歸模型中,ARDI高等分的個案,相較於ARDI低等分的個案,發生總死亡事件的相對危險是1.88(95% CI, 1.04 to 3.40)倍(p for trend = 0.037)。發生總死亡事件的ARDI危險閾值為18.0 mm/m2。相反地,在年齡>=65歲的個案中,ARDI分層與總死亡事件無關(p for trend = 0.14)。不論年齡層高低,ARDI的分層與心血管事件的發生無關。 結論 本研究以台灣金山地區成年華人族群為研究對象,發現:左心室質量指數與心血管事件的發生,呈線性相關;而與總死亡事件的發生,呈現J型曲線相關。左心房直徑指數與女性個案罹患腦中風的危險性呈正相關。另外,針對年齡<65歲的個案,主動脈根部直徑指數與總死亡事件呈顯著的正相關。此三個心臟超音波指標,與臨床事件具有顯著的相關性,可以做為進一步發展臨床事件預測模型的基礎。 Background and Objectives: Because of the great burden of cardiovascular deaths, prevention and treatment of cardiovascular diseases are important in developed countries. Echocardiography possesses the advantage of non-invasiveness and rapid acquisition of parameters about cardiac function. Several population studies have investigated the predictive value of echocardiographic measures for cardiovascular morbidity and mortality. However, some issues including the methods of indexation of echocardiographic parameters are not clarified in the literature. This research was proposed to investigate the relationship between different echocardiographic parameters and clinical outcomes in an ethnic Chinese population. Methods: We recruited participants in the Chin-Shan Community Cardiovascular Cohort (CCCC) study who had received echocardiography without previous cardiovascular events as the study population. Three echocardiographic parameters, including left ventricular mass, left atrial dimension and aortic root dimension, were measured through two different sessions of echocardiographic examinations on 1992~1993 and 1994~1995. The end-points were all-cause death and incident cardiovascular events including coronary heart disease and stroke. We proposed to define the best cut-off values for good clinical predictors for risk stratification in our ethnic Chinese population and to test the different prediction power of individual predictor among different subgroups. Results: We evaluated 2604 participants in the sub-study concerning left ventricular mass. By the end of 2005, with a median follow-up of 14.4 years, 34325 person-years were accrued. We identified 205 participants with cardiovascular events and 453 deaths. We used body surface area as the indexation to calculate left ventricular mass index (LVMI). The results of multivariate Cox regression analysis showed a linear relationship between LVMI and cardiovascular events. The adjusted relative risk was 2.01 (95% confidence interval [CI], 1.11 to 3.63) for the highest quintile of LVMI compared with the lowest quintile (p for trend = 0.001). A J-shape relationship between LVMI and all-cause death was observed, with the test for a linear relationship being rejected (p = 0.003). The adjusted relative risks of all-cause death were significantly lower for the second quintile of LVMI (0.58, 95% CI, 0.40 to 0.84) and for the third quintile of LVMI (0.68, 95% CI, 0.47 to 0.96), compared with the lowest quintile. The proposed cut-off value of LVMI was 105 g/m2 for prediction of cardiovascular events in both genders. Concerning the sub-study of left atrial dimension, 1937 subjects were included and 21733 person-years were accrued during a median follow-up of 11.9 years. We identified 114 subjects with stroke and 364 with all-cause death. We found that the best method of indexation was to divide left atrial dimension by body mass index to obtain left atrial dimension index (LADI). In women, the relative risk of stroke was 2.44 (95% CI, 1.11 to 5.36, p for trend = 0.029) among participants in the upper tertile of LADI compared with participants in the lower tertile of LADI after multivariate adjustment. However, further adjustment for left ventricular mass index attenuated the relation of LADI to stroke (relative risk: 2.11, 95% CI, 0.88 to 5.02, p for trend = 0.09). The proposed cut-off value of LADI was 1.39 mm/(kg/m2) for prediction of stroke in women. In men, tertiles of LADI was not associated with stroke. Besides, LADI was not associated with the risk of all-cause death in both genders. In the sub-study of aortic root dimension, we enrolled 1851 subjects with a median follow-up of 11.9 years. In 20800 person-years of follow-up, 185 subjects were diagnosed as having cardiovascular events and 335 subjects died during the follow-up period. We indexed aortic root dimension by body surface area to obtain aortic root dimension index (ARDI). Although tertiles of ARDI was associated with increased risk of cardiovascular events and all-cause death in univariate analysis, the significance diminished after adjusting for both gender and age (p for trend = 0.12 for cardiovascular events; p for trend = 0.23 for all-cause death). In subgroup analysis, we found a significant association between tertiles of ARDI and risk of all-cause death in the final multivariate Cox regression model in adults < 65 years. The adjusted relative risk was 1.88 (95% CI, 1.04 to 3.40) in participants in the upper tertile of ARDI compared with participants in the lower tertile of ARDI (p for trend = 0.037). The proposed cut-off value of ARDI was 18.0 mm/m2 for prediction of all-cause death in adults < 65 years. In adults >= 65 years, tertile of ARDI was not associated with all-cause death (p for trend = 0.14). Tertiles of ARDI was not associated with cardiovascular events throughout this study. Conclusions: In this ethnic Chinese cohort study, we demonstrated a linear relationship between LVMI and cardiovascular events, and a J-shape relationship between LVMI and all-cause death. An association between increased LADI and incident stroke in women rather than in men was established. Besides, we found a significant association between ARDI and all-cause death in adults aged < 65 years. Since LVMI, LADI, and ARDI were significantly related with clinical outcomes, they can be enrolled into the models for prediction of clinical events in the future studies. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/44874 |
Fulltext Rights: | 有償授權 |
Appears in Collections: | 流行病學與預防醫學研究所 |
Files in This Item:
File | Size | Format | |
---|---|---|---|
ntu-99-1.pdf Restricted Access | 1.65 MB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.