以不同臨床篩檢方式在預期屬於中度至高度心血管風險族群的效果評估及花費比較

Comparison of the Diagnostic Accuracy and Cost-Effectiveness of Various Screening Tools on Projected Intermediate-to-High Cardiovascular Risk

洪崇烈(Chung-Lieh Hung)

指導教授 ： 蘇喜

國立臺灣大學/公共衛生學院/健康政策與管理研究所/碩士(2011年)

https://doi.org/10.6342/NTU.2011.02374

摘要

背景
由於每年花在心血管疾病相關的費用相當昂貴，對於早期疾病的預防及有效、正確且符合經濟效益的篩檢方式顯得格外重要。
目的
本研究的目標在於探討不同心血管疾病的篩檢工具在無症狀的個體，於預期十年後中度至高度的心血管風險 (利用弗明罕風險指數)的正確診斷及經濟效益。
材料與方法
本研究自2005至2009年總共研究了一千兩百名參與年度體檢且無明顯症狀的個體。藉由病患的基本資料及抽血、生化檢驗及體表心電圖資訊等結果，計算出弗明罕風險指數。另外，也做了人體測量學的定量，並根據血中生化值算出代謝異常指數。我們的研究並額外加入關於高敏感度C蛋白的血清值及超音波頸動脈內膜厚度的分析及粥狀動脈的有無等資訊並且用來和弗明罕風險指數做比較。藉由診斷的正確性及經濟效益的分析，我們試圖找到一個比較有效的篩檢方式。
結果
總共有一千一百零一位參與者加入本研究(平均年齡五十點六歲；百分之三十八點六為女性)。高的弗明罕風險和高的代謝異常指數相關，並且有明顯較高的代謝症候群比率、較高的Hs-CRP及頸動脈內膜厚度值，以及較高比例的頸動脈粥的異常(線性相關及卡方檢定結果皆小於0.001)。男性比起女性有較高的弗明罕風險指數及較高的代謝異常指數、較高的頸動脈內膜厚度值，但是接近的異常頸動脈粥的盛行率。Hs-CRP並無明顯的性別上的差異(p=0.15)。對於篩檢預期十年後中度至高度的心血管風險，使用代謝異常指數大於等於1做為切點，會有最佳的敏感度(94.43%, 95% CI: 92.05 - 96.27)；反之，藉由選定頸動脈內膜厚度大於或等於1毫米，則會有最佳的特異度(98.27, 95% CI: 97.24 - 98.99)。另外，使用代謝異常指數大於等於2或藉由選定頸動脈內膜厚度大於或等於0.65毫米，則會有最大的敏感及特異度的加總。比起最昂貴的超音波頸動脈檢查(62222.2 NTD)及代謝異常指數檢測，在每篩檢一位預期十年後中度至高度的心血管風險陽性的個體，Hs-CRP大於或等於0.1mg/dL卻有最低的花費(1519.3 NTD)。

結論
雖然篩檢一位中度至高度的心血管風險陽性的個體的正確性在不同的篩檢工具藉由選定不同的切點值會有不同的診斷率，選定一個較低的代謝異常指數會有最高的敏感性而異常的頸動脈內膜厚度將會有最佳的特異度。另外，本研究也顯示了在無明顯心血管相關症狀的群體，選定一個低的Hs-CRP值似乎具有最佳的經濟效益。

關鍵字

心血管疾病 ； 弗明罕風險指數 ； 代謝異常指數 ； 血清生物標記 ； 頸動脈檢查 ； 診斷正確性 ； 經濟效益

並列摘要

Abstract
Background
Owing to the high costs spent annually in cardiovascular diseases, there is an urgent need in identifying subjects at an early stage based on view point of preventive medicine. The development of a cost-effective screening tool with adequate diagnostic accuracy is thus crucial.
Goals
To investigate the diagnostic accuracy and cost-effectiveness of various cardiovascular screening tools in the estimation of intermediate-to-high risk Framingham risk score (FRS) subjects in asymptomatic population.
Materials and Methods
We consecutively studied 1200 asymptomatic subjects who underwent health evaluation from 2005-2009. FRS was calculated in all participants based on age, gender, blood pressure, body surface electrocardiography, medical histories, life styles and lipid profiles. We also assessed metabolic scores by additional anthropometric information. Data regarding high-sensitivity C reactive protein (Hs-CRP) serum level and carotid artery Doppler in assessing intima-media-thickness (IMT) and plaque existence were also obtained and correlated with FRS. Diagnostic accuracy and cost-effectiveness analysis were then conducted among these different tools aiming at a more efficient screen of intermediate-to-high Framingham risk population.
Results
Of all, totally 1101 participants (mean age: 50.6 ± 10.4, 38.6% women) were finally entered in our study. Higher Framingham risk score was associated with higher metabolic risk scores, higher prevalence of metabolic syndrome, elevated level of Hs-CRP, higher IMT thickness and higher prevalence of carotid artery plaque existence (all p<0.001 by linear regression or chi square test). In general, male had higher FRS than female gender, as well as higher metabolic risks scores, larger IMT (all p<0.001) though similar prevalence of carotid artery plaques (p=0.134). There was no gender-related difference in Hs-CRP level (p=0.15). By using metabolic score equal or larger than 1 as a cut-off, there seemed to be a very high sensitivity (94.43%, 95% CI: 92.05 - 96.27) in identifying a subject with intermediate-to-high Framingham risk score while the best specificity (98.27, 95% CI: 97.24 - 98.99) was achieved by utilizing carotid IMT equal or larger than 1mm as a cut-off. In addition, a metabolic score cut-off of 2, Hs-CRP of 0.101mg/dL and IMT of 0.65mm seemed to have the highest sum of both sensitivity and specificity. Compared to carotid artery study and metabolic score calculation, Hs-CRP with a cut-off value 0.1 seemed to have lowest cost (1519.3 NTD) in identifying an intermediate-to-high Framingham risk subject with highest screening cost occurred (62222.2 NTD) per positive case by simply using carotid echo-defined abnormal IMT (>=1mm).
Conclusion
Though diagnostic accuracy may differ to some degree by using different cut-off values in various studies, a low metabolic score has the best sensitivity with abnormal IMT had highest specificity in screening a subject at risk for future cardiovascular diseases. In addition, setting a low Hs-CRP serum level seemed to have the best cost-effectiveness in asymptomatic population.

並列關鍵字

Cardiovascular disease ； Framingham risk score ； metabolic risk score ； biomarker ； Carotid artery study ； diagnostic accuracy ； cost-effectiveness

參考文獻

3 American Heart Association. Heart Disease and Stroke Statistics 2008 Update. December 2007.

4 Heidenreich PA, Trogdon JG, Khavjou OA, Butler J, Dracup K, Ezekowitz MD, Finkelstein EA, Hong Y, Johnston SC, Khera A, Lloyd-Jones D, Nelson SA, Nichol G, Orenstein D, Wilson PWF, Woo YJ. Forecasting the future of cardiovascular disease in the United States: a policy statement from the American Heart Association. Circulation. 2011;123:933–944.

5 Fuster V. Elucidation of the role of plaque instability and rupture in acute coronary events. Am J Cardiol. 1995;76(suppl C):24C–33C.

6 McGovern PG, Pankow JS, Shahar E, Doliszny KM, Folsom AR, Blackburn H, Luepker RV, the Minnesota Heart Survey Investigators. Recent trends in acute coronary heart disease: mortality, morbidity, medical care, and risk factors. N Engl J Med. 1996;334:884–890.

7 Walker SH, Duncan DB: Estimation of the probability of an event as a function of several independent variables. Biometrika 1967;54:167-179