The prevalence of AF, which is tachyarrhythmia, is approximately 2% of the entire population and 5% of the population at the age of 60 or older. AF is the cause of approximately 20% of all events of ischemic stroke, and patients with AF are known to be at 6 to 10% risk of ischemic stroke per year. Patients with valvular AF are known to have a higher incidence of stroke than patients with nonvalvular AF. However, the relevant data are insufficient as large randomized studies comparing NOAC treatment with warfarin, a conventional treatment, did not include many patients with moderate and severe valvular AF.

Ischemic stroke is divided into symptomatic stroke with brain lesions on brain magnetic resonance imaging (MRI) and silent cerebral infarct with lesions on brain MRI but without stroke symptoms. According to a brain MRI follow-up study, the incidence of silent cerebral infarct was 17.7% (254 subjects) over a period of 5 years, with 11.4% of 254 subjects reporting to have experienced symptoms. This means that the incidence of silent cerebral infarct is approximately 9 times that of symptomatic stroke. In addition, patients with a history of silent cerebral infarct are known to be approximately twice more likely to experience stroke in the future than those without a history of silent. Brain microbleed is easily detected by brain MRI and is a well-known independent predictor of intraparenchymal hemorrhage and silent cerebral infarct. The prevention of stroke by the study drug can be indirectly assessed based on the incidence of silent cerebral infarct and brain microbleed on brain MRI. Investigators tried to compare effect of dabigatran with conventional treatment in terms of prevention of stroke by comparing incidences of silent cerebral infarct and brain microbleed and symptomatic stroke using brain MRI.

Amongst different types of cardiac arrhythmia, tachyarrhythmia shows the highest prevalence. The prevalence of AF, which is tachyarrhythmia, is approximately 2% of the entire population and 5% of the population at the age of 60 or older. AF is the cause of approximately 20% of all events of ischemic stroke, and patients with AF are known to be at 6 to 10% risk of ischemic stroke per year. In addition, AF results in heart failure and doubles mortalities. With an increase in the elderly population, there is a worldwide trend of increased hospitalization rate and medical cost caused by cardiac arrhythmia. In addition, whereas AF has been known to occur mainly in the elderly, the incidence of AF is also increasing in younger population, which is thought to be associated with the increased prevalence of heart diseases, dietary changes, and increased environmental pollution.

A. Previous studies related to atrial fibrillation and stroke i. While AF is characterized by a single electrophysiological abnormality, depending on cases, the risk of stroke varies from below 1%/year to 20%/year or above.

ii. New oral anticoagulant (NOAC), left atrial appendage occlusion, etc. have been studied as treatments to replace warfarin and aspirin and their effects have been also demonstrated.

B. Relevant study trends and limitations i. Patients with valvular AF are known to have a higher incidence of stroke than patients with nonvalvular AF. However, the relevant data are insufficient as large randomized studies comparing NOAC treatment with warfarin, a conventional treatment, did not include many patients with moderate and severe valvular AF.

ii. The exclusion of valvular AF from the study is based on the assumption that the pathogenesis of thromboembolism would vary according to the type of AF. However, the difference in pathogenesis of thromboembolism between valvular AF and nonvalvular AF remains unknown.

C. Study features i. Ischemic stroke is divided into symptomatic stroke with brain lesions on brain magnetic resonance imaging (MRI) and silent cerebral infarct with lesions on brain MRI but without stroke symptoms.

ii. According to a brain MRI follow-up study, the incidence of silent cerebral infarct was 17.7% (254 subjects) over a period of 5 years, with 11.4% of 254 subjects reporting to have experienced symptoms. This means that the incidence of silent cerebral infarct is approximately 9 times that of symptomatic stroke.

iii. In addition, patients with a history of silent cerebral infarct are known to be approximately twice more likely to experience stroke in the future than those without a history of silent.

iv. Brain microbleed is easily detected by brain MRI and is a well-known independent predictor of intraparenchymal hemorrhage and silent cerebral infarct.

v. The prevention of stroke by the study drug can be indirectly assessed based on the incidence of silent cerebral infarct and brain microbleed on brain MRI at the start of study and at 1 year follow-up.

D. Significance of the study i. The incidence of stroke is higher in valvular AF than in nonvalvular AF. ii. There is lack of data on dabigatran treatment, an NOAC, in valvular AF. iii. To compare dabigatran with conventional treatment in terms of prevention of stroke by comparing incidences of silent cerebral infarct and brain microbleed and symptomatic stroke at the start of study and at 1 year follow-up using brain MRI.

This is a single center, prospective, randomized study to evaluate the effectiveness and safety of dabigatran versus conventional treatment in patients with valvular AF by comparing the incidence of silent cerebral infarct and brain microbleed using brain MRI. Aspirin or warfarin treatment will be continued in patients randomly allocated to conventional treatment group, and changed to dabigatran treatment in those randomly allocated to dabigatran group.

Inclusion Criteria:

• Voluntary informed consent

• Diagnosis of AF

• Aortic valvular stenosis, aortic valvular regurgitation, mitral valvular stenosis, or mitral valvular regurgitation of moderate or above severity

• Patients must have one of the following:

  • Hypertension requiring medical treatment
  • Symptomatic heart failure, New York Heart Association class 2 or higher in the last 6 mon or ejection fraction <40% documented by echocardiogram
  • Age ≥ 65 years
  • Diabetes mellitus on treatment
  • History of previous stroke, transient ischemic attack, or systemic embolism
  • Previous myocardial infarct, peripheral artery disease, or aortic plaque
  • Enlarged left atrial size ≥ 40mm documented by echocardiogram

Exclusion Criteria:

• An individual involved in planning or conducting this study

• Unable to understand study conduct or study compliance due to dementia, etc.

• Lack of ability to communicate

• Pregnant woman

• Past cardiac valve replacement

• Stroke resulting in severe disabilities within the past 6 months

• Stroke including silent cerebral infarct within the past 2 weeks

• Chronic renal failure with creatinine clearance < 30ml/min

• Active hepatitis (≥2 fold increase in liver enzyme levels, known active hepatitis C, active hepatitis B, active hepatitis A)

• High bleeding risk

  • Major surgery within the past 1 month
  • Planned surgery or procedure within 3 months
  • Intracranial, intraocular, intraspinal, or retroperitoneal bleeding history or atraumatic joint bleeding history
  • Digestive tract bleeding within the past 1 year
  • Symptomatic or endoscopically confirmed gastric or duodenal ulcer within the past 30 days
  • Hematologic abnormalities
  • Requiring anticoagulant treatment for any reasons other than AF
  • Requiring antithrombotics within 48 hours of start of the study
  • Uncontrolled hypertension (systolic blood pressure ≥180mmHg or diastolic blood pressure ≥100mmHg)
  • Recent malignant tumor or 6 months or longer treatment with radiation therapy
  • Mean life expectancy ≤3 years

• Contraindication of anticoagulant treatment

• Screening test results as follows

  • Platelet count <80,000 cells/mm3
  • Hemoglobin level <10g/dL

• Previous experience of major bleeding complications

• Participation in another drug study within 3 months prior to enrollment to this study

• Administration of the study drug within 1 month prior to enrollment to this study