結核病與心血管疾病風險之關聯：回溯性配對世代研究

Abstract

前言：結核病仍然是一項重要的全球健康挑戰。過去研究指出，結核病患者的心血管疾病風險有所增加。然而，大多數研究由於資料缺乏，未能考慮重要的危險因子，例如身體質量指數、生活型態或社會經濟地位。本研究之目的為利用一個長期追蹤健康檢查資料庫中的族群，探討結核病患者與對照組相比，罹患心血管疾病的風險是否有所增加。 方法：本研究為一個使用臺灣美兆健康資料庫的回溯性配對世代研究，此資料庫包含於美兆健康管理機構接受健檢服務的參加者，並經財團法人演譯基金會授權使用。研究對象包含2006年至2019年間首次在結核病資料庫中通報結核病診斷的患者，對照組則從美兆健康資料庫中選擇，使用風險集抽樣與粗化精確配對法，根據指標月份、年齡、性別、身體質量指數、地理區域、腎絲球過濾率及共病症進行配對，以選出在抽樣時未罹患結核病的對照組。追蹤時間從結核病確診日開始至首次發生心血管疾病、死亡或2020年底為止。使用Cox比例風險模型與逆機率加權存活曲線來比較暴露組與對照組的心血管疾病風險。 結果：本研究以1,923名結核病患者與598,422名未罹患結核病者之資料進行配對，最終分析樣本包含1,894名結核病患者與7,544名對照。追蹤時間之中位數為6.06年(四分位距：3.12–10.48)，共觀察到1,427例新發心血管疾病，結核病患者的心血管疾病發生率較對照組高出22% (分別為每千人年26.12、21.36人)。多變項調整風險比值為1.27 (95%信賴區間：1.11–1.45)，結果顯示結核病患者罹患心血管疾病的風險顯著增加。在三項靈敏度分析中，結果均呈現一致的方向性，本研究進一步分析發現，非動脈粥狀硬化性心血管疾病的風險在結核病患者中為1.65倍。 結論：研究結果顯示，結核病與心血管疾病風險增加有所關聯，建議考慮將心血管疾病風險評估納入結核病治療後的長期照護。未來研究應進一步釐清其潛在致病機轉，並驗證此關聯是否存在於具有不同人口學特徵、地理區域分布或種族背景的族群之中。

Introduction: Tuberculosis (TB) continues to be a major global health issue. Previous studies have indicated an increased risk of cardiovascular disease (CVD) among people diagnosed with TB. However, due to the lack of data, most of these studies did not account for important risk factors, such as body mass index, lifestyle factors, or socioeconomic status. This study aimed to quantify the risk of CVD among individuals with TB compared with matched controls in a large private health examination cohort with longitudinal follow-up. Methods: This retrospective, longitudinal matched cohort study was conducted using data from the MJ Health Database (MJHD) in Taiwan, which was authorized by the MJ Health Research Foundation. MJ Institution is a private service provider that offers annual health examination services to its members. The study population included individuals from the MJHD who had their first documented TB diagnosis in the National TB Registry from 2006 to 2019. Matched controls were selected from the same database. We used a risk set sampling approach with coarsened exact matching on the calendar month, age, sex, body mass index, geographic region, estimated glomerular filtration rate (eGFR), and comorbidities to identify controls free of TB at the time of sampling. The follow-up started on the date of TB and continued until the first event of CVD, death, or the end of 2020, whichever came first. Cox proportional hazards models and inverse probability weighted survival curves were used to compare the risk in the exposed (TB) group with the control group. Results: There were 1,923 eligible exposed and 598,422 unexposed individuals initially identified. After coarsened exact matching, a total of 1,894 individuals with TB and 7,544 matched controls were included in the analysis. During a median follow-up of 6.06 years (IQR: 3.12–10.48), 1,427 cases of incident CVD were observed. The incidence of CVD was 22% higher in the TB-exposed group compared with the controls (26.12 vs. 21.36 per 1,000 person-years). The estimated multivariable-adjusted HR (1.27, 95% CI: 1.11–1.45) indicated a higher risk of CVD among individuals diagnosed with TB. The results were robust across three sensitivity analyses. We found that the risk of non-atherosclerotic CVD was 1.65 times greater in the TB-exposed group compared with controls after multivariable adjustment. Conclusion: Our findings indicated that TB was associated with cardiovascular risk, suggesting the potential need to integrate cardiovascular risk assessment into the long-term care of TB survivors. Further research is required to clarify the biological mechanisms and verify this association across different demographic, geographic, and ethnic populations.