利用世代研究探討心血管疾病患者在不同治療方針與各項風險因子之預後

Abstract

背景：心血管疾病是全球主要的死亡原因，造成心血管疾病患者死亡的主要因素為冠狀動脈疾病、心臟衰竭、和中風。探討風險因子與心血管疾病風險之間的關聯性很重要，透過風險評估可以改善治療策略，而世代研究可以衡量特定暴露在心血管結局的發生率和危險因子。 方法：我們使用世代研究來探討心血管疾病的危險因素和臨床結果。本博士論文包含三個研究：(1) 第一個研究利用金山社區心血管世代研究調查，探討一般民眾心因性猝死的危險因子 (針對年紀大於35歲、沒有冠心病、和左心室收縮功能< 35%者)。研究重點放在12導程心電圖、標準心臟超音波、和頸動脈超音波等篩檢工具的異常，利用佛萊明風險分數方法來發展預測十年內心因性猝死的評分系統，並使用自助抽樣法驗證。(2) 第二個研究針對心房顫動患者，我們使用臺灣全民健康保險研究資料庫的醫療資料，來發展預測一年內中風評分系統，並進行內部驗證。(3) 第三個研究基於全國大型世代資料，探討臺灣地區罕見疾病類澱粉性沉積症患者的發生率，以及評估長期心室頻脈和心因性死亡的風險。 結果：(1) 嶄新心因性猝死預測分數系統 (CCCC-SCD-Score) 具有良好的十年內心因性猝死預測能力 (高風險切點: > 5; 一致性指數 [C指數]: 0.881, 95 % 信賴區間: 0.805-0.958; Hosmer-Lemeshow 適合度檢定: P值 = 0.82)，分數依年齡組別 (最高4分)、左心室肥大 (1分)、高血壓 (1分)、左心室射出分量 < 40% (1分)、主動脈瓣流速 > 190 cm/s (1分)、以及頸動脈斑塊分數 ≥ 5 (1分) 計算。(2) 透過評估電燒狀態新發展的心房顫動預測中風評分系統 (AF-CA-Stroke: 高風險切點: > 5; 一致性指數 [C指標]: 0.658, 95 % 信賴區間: 0.644-0.675; Hosmer-Lemeshow 適合度檢定: P值 = 0.81)，在預測一年內中風風險方面比傳統評分系統 (CHADS2，一致性指數 [C指數]: 0.577, 95 % 信賴區間: 0.570-0.584) 具有更好的辨別能力 (DeLong測試: P值 < 0.001)，分數依年齡組別 (最高5分)、未接受心房顫動電燒手術 (1分)、過去中風史 (1分)、慢性腎臟病 (1分)、其他過去心臟 (冠心症) 或周邊血管疾病 (1分) 計算。(3) 類澱粉性沉積症患者在臺灣的發生率為每十萬人年6.54人，心臟型類澱粉性沉積症患者在臺灣的發生率為每十萬人年0.61人。罹患有類澱粉性沉積症的患者未來發生心室頻脈的風險 (調整後風險函數比: 7.90, 95% 信賴區間: 4.49-13.9) 和心血管死亡風險 (調整後風險函數比: 5.09, 95 % 信賴區間: 4.23-6.12) 都較沒有罹患類澱粉性沉積症的患者高。 結論：瞭解各種心血管疾病患者的危險因子很重要，可以針對具有心血管事件高風險的患者，進行長期追蹤，以助於提供初級預防處置和治療的策略。

Background: Cardiovascular diseases (CVDs) are among the leading causes of death worldwide. The main causes of death in CVD patients are coronary artery disease (CAD), heart failure, and stroke. Cohort studies are used to measure incidence rates for cardiovascular (CV) outcomes based on a specific exposure, as well as to examine the risk factors and clinical outcomes associated with CVDs. By assessing risk, the treatment strategies can be improved for patients with CVDs. Methods: We investigated several factors and outcomes of CVDs in the cohort studies. The doctoral dissertation included three projects: (1) The objective of the first project was to investigate risk factors for sudden cardiac death (SCD) in a general population aged ≤ 35 years without a prior history of CAD or left ventricular ejection fraction < 35 based on the Chin Shan Community Cardiovascular Cohort, focusing on the screening tools of 12-lead electrogram, standard echocardiography, and carotid artery duplex sonography. By using the Framingham risk score methods, we developed a novel CCCC-SCD-Score to predict incident 10-year SCD. The CCCC-SCD-Score was internally validated using bootstrapping method. (2) The objective of the second project was to develop a novel model-based point scoring system for 1-year stroke prediction in patients with atrial fibrillation (AF) using Taiwan's National Health Insurance Research Database. An internal validation study was performed. (3) The objective of the third project was to investigate the risks of new-onset ventricular tachycardia and CV outcomes in patients with amyloidosis after a long-term follow-up based on a representative national cohort. Results: (1) A CCCC-SCD-Score score is calculated using age groups (maximum points = 4), left ventricular hypertrophy, hypertension, left ventricular ejection fraction < 40%, aortic flow rate > 190 cm/s, and carotid plaque scores ≥ 5 (point = 1 for each risk factor). In predicting 10-year SCD risk, the CCCC-SCD-Score had good prediction performance (cut-off point: > 5; C-index: 0.881, 95% confidence interval [CI]: 0.805-0.958; Hosmer-Lemeshow test: P-value = 0.82). (2) The AF-CA-Stroke scoring system includes important clinical risk factors: age (maximum points = 5), the status of not having undergone AF ablation (point = 1), prior stroke history (point = 1), chronic kidney disease (point = 1), and prior CAD or vascular disease (point = 1). The novel AF-CA-Stroke scoring system using the status of AF ablation (cut-off point: > 5; C-index: 0.658, 95% CI: 0.644-675; Hosmer-Lemeshow test: P-value = 0.81) predicted incident 1-year stroke risk more accurately than conventional CHADS2 scoring system (C-index: 0.577, 95% CI: 0.570-584) (P-value = 0.001, using the DeLong test). (3) The incidence rates of amyloidosis and cardiac amyloidosis were 6.54 and 0.61 per 100000 person-years, respectively. Amyloidosis was associated with higher rates of ventricular tachycardia (adjusted HR: 7.90, 95% CI: 4.49-13.9) and CV deaths (adjusted HR: 5.09, 95% CI: 4.23-6.12). Conclusions: Knowledge of risk factors in patients with various CVDs is essential, and long-term follow-up of patients at high risk for cardiovascular events can aid in primary prevention and guide treatment strategies.