Options
慢性發炎疾病的發炎反應導致中風之追蹤研究： 臺灣健康保險資料庫之追蹤研究
Other Title
Evidence for the contribution of chronic inflammatory processes toward the development of stroke utilizing Taiwan’s National Health Insurance Research Database
Type
thesis
Date Issued
2016-06-05
Author(s)
柯捷
Advisor
白其卉
Subjects
系所名稱：公共衛生學系暨研究所
Description
學位別：博士
語文別：英文
指導教授：白其卉
共同指導教授：
口試委員： 邱弘毅;李建宏;潘信良;連立明
中文關鍵字：中風；慢性發言疾病，健保資料庫
英文關鍵字：stroke;chronic inflammatory processes, national health insurance research database
語文別：英文
指導教授：白其卉
共同指導教授：
口試委員： 邱弘毅;李建宏;潘信良;連立明
中文關鍵字：中風；慢性發言疾病，健保資料庫
英文關鍵字：stroke;chronic inflammatory processes, national health insurance research database
Abstract
Introduction:
Chronic inflammatory diseases have been suggested to contribute to the development of stroke. However, there is significant controversy in the literature regarding these associations. Moreover, inflammatory diseases are heterogeneous in the immuno-inflammatory responses they elicit which may affect subsequent stroke risk. Moreover, while the epidemiology of stroke is well-known to differ with regard to race and ethnicity, the majority of the evidence in the literature has been obtained from studies conducted in western populations. Therefore, this study set out to investigate the association between three chronic inflammatory diseases (Crohn’s disease (CD), Ulcerative colitis (UC), and Ankylosing spondylitis (AS)) mediated by different immune-inflammatory responses (Th1, Th2, and Th2/Th17) and stroke to help settle controversies in the literature, garnish more evidence regarding immune response type and stroke risk, and to provide Asian population-specific risk estimates.
Methods:
This work comprises three separate cohort studies investigating the effect of CD, UC, and AS on the subsequent risk of stroke utilizing data captured from the Taiwan National Health Insurance Database. Cox proportional hazards regression analyses were conducted to appropriately adjust for confounding. Risk estimates are presented in terms of hazard ratios and Kaplan Meir plots elucidate survival.
Results:
Increased risks for stroke were observed among all three investigated conditions. For AS, after adjusting for chronic lower respiratory diseases, type 2 diabetes mellitus, hypertension, hyperlipidemia, renal disease, coronary heart disease, atrial fibrillation, income, and urbanization, compared with matched non-AS comparison patients, the hazard ratio for subsequent stroke among patients with AS was 2.3 (95 % CI 1.9-2.8). We also stratified our results by both gender and pharmaceutical prescription, but did not find a statistically significant difference for the risk of subsequent stroke either between men and women, or between AS patients taking various pharmaceutical regimens and the overall AS population.
For UC, after adjusting for selected medical co-morbidities and recent prescriptions of selected pharmaceuticals, the hazard ratio (HR) for subsequent stroke among patients with UC was 2.045 (95 % confidence interval (CI) = 1.374-3.043) than that among matched non-UC comparison subjects. While we did not detect an association between stroke and UC among patients aged 30-40 or 40-50 years, we did detect increased risks for stroke among UC patients aged over 50 years (HR = 2.045). We also found the association to remain significant for both men (HR = 2.153) and women (HR = 2.750).
For CD, after adjusting for selected medical co-morbidities and recent prescriptions of selected pharmaceuticals, the hazard ratio (HR) for subsequent stroke among patients with CD was found to be 1.911 (95 % confidence interval (CI) = 1.65-2.22) that of matched non-CD comparison subjects.
While we did not detect an association between stroke and CD among patients aged 30-40 years, we did detect increased risks for stroke among CD patients aged 40-50 years (HR = 2.29) and those aged over 50 years (HR = 1.88). We also found women (HR = 2.39) to be at a greater risk than men (HR = 1.50).
Conclusion:
Similar increased risks for stroke were observed in all three conditions, including those classified as producing immune responses deleterious toward stroke, as well as those classified as producing neuroprotective immune responses.
Chronic inflammatory diseases have been suggested to contribute to the development of stroke. However, there is significant controversy in the literature regarding these associations. Moreover, inflammatory diseases are heterogeneous in the immuno-inflammatory responses they elicit which may affect subsequent stroke risk. Moreover, while the epidemiology of stroke is well-known to differ with regard to race and ethnicity, the majority of the evidence in the literature has been obtained from studies conducted in western populations. Therefore, this study set out to investigate the association between three chronic inflammatory diseases (Crohn’s disease (CD), Ulcerative colitis (UC), and Ankylosing spondylitis (AS)) mediated by different immune-inflammatory responses (Th1, Th2, and Th2/Th17) and stroke to help settle controversies in the literature, garnish more evidence regarding immune response type and stroke risk, and to provide Asian population-specific risk estimates.
Methods:
This work comprises three separate cohort studies investigating the effect of CD, UC, and AS on the subsequent risk of stroke utilizing data captured from the Taiwan National Health Insurance Database. Cox proportional hazards regression analyses were conducted to appropriately adjust for confounding. Risk estimates are presented in terms of hazard ratios and Kaplan Meir plots elucidate survival.
Results:
Increased risks for stroke were observed among all three investigated conditions. For AS, after adjusting for chronic lower respiratory diseases, type 2 diabetes mellitus, hypertension, hyperlipidemia, renal disease, coronary heart disease, atrial fibrillation, income, and urbanization, compared with matched non-AS comparison patients, the hazard ratio for subsequent stroke among patients with AS was 2.3 (95 % CI 1.9-2.8). We also stratified our results by both gender and pharmaceutical prescription, but did not find a statistically significant difference for the risk of subsequent stroke either between men and women, or between AS patients taking various pharmaceutical regimens and the overall AS population.
For UC, after adjusting for selected medical co-morbidities and recent prescriptions of selected pharmaceuticals, the hazard ratio (HR) for subsequent stroke among patients with UC was 2.045 (95 % confidence interval (CI) = 1.374-3.043) than that among matched non-UC comparison subjects. While we did not detect an association between stroke and UC among patients aged 30-40 or 40-50 years, we did detect increased risks for stroke among UC patients aged over 50 years (HR = 2.045). We also found the association to remain significant for both men (HR = 2.153) and women (HR = 2.750).
For CD, after adjusting for selected medical co-morbidities and recent prescriptions of selected pharmaceuticals, the hazard ratio (HR) for subsequent stroke among patients with CD was found to be 1.911 (95 % confidence interval (CI) = 1.65-2.22) that of matched non-CD comparison subjects.
While we did not detect an association between stroke and CD among patients aged 30-40 years, we did detect increased risks for stroke among CD patients aged 40-50 years (HR = 2.29) and those aged over 50 years (HR = 1.88). We also found women (HR = 2.39) to be at a greater risk than men (HR = 1.50).
Conclusion:
Similar increased risks for stroke were observed in all three conditions, including those classified as producing immune responses deleterious toward stroke, as well as those classified as producing neuroprotective immune responses.