Non-vitamin K antagonist oral anticoagulant (NOAC) is the first line therapy to prevent thromboembolism in atrial fibrillation (AF) patients. Previous investigation showed that preceding NOAC therapy was associated with lower severity of ischemic stroke, but with higher in-hospital mortality in intracerebral hemorrhage (ICH), as compared with antithrombotic agent non-users. Measurement of NOAC level upon acute stroke aids the critical decision of acute management. Real-world data regarding the relationship between the NOAC adherence, the appropriateness of NOAC regimen, or NOAC level in acute stroke and the stroke severity or short-term outcome is lacking. Further, optimal selection for long-term stroke prevention among patients with acute stroke during NOAC therapy remains unclear.

Specific purpose: To analyze the association between NOAC adherence or NOAC level upon acute stroke and stroke severity or stroke outcomes, and analyze the impact of starting or withholding antithrombotic therapy after acute stroke on long-term stroke outcomes.

Specific Aim (Year 1):

To investigate the relationship between NOAC adherence or appropriateness of NOAC dose and acute stroke severity or in-hospital mortality based on National Health Insurance Research Database (NHIRD). Another important goal is to prospectively establish a cohort of AF users who developed acute stroke during NOAC therapy in National Taiwan University Hospital (NTUH) (target: around 100 patients annually), measure the NOAC level upon hospital arrival, record stroke severity, 90-days functional outcomes, post-stroke antithrombotic agents and repeat stable NOAC level in patients who restart NOAC treatment.

Specific Aim (Year 2):

To investigate the relationship between post-stroke antithrombotic therapy, especially changing or retaining preceding NOAC and long-term stroke outcomes based on NHIRD. We will also keep enrolling the prospective cohort and follow the 1-year stroke outcome.

Specific Aim (Year 3):

To complete the process of study enrollment (total: 300 patients) and conduct statistical analysis. The main goal is to finish the Aim 1 and 2 based on NHIRD. In addition, to provide data of emergent NOAC level and stroke severity or short-term outcome, and post-stroke antithrombotic therapy and long-term outcomes based on the prospectively enrolled cohort.

Inclusion Criteria:

• Atrial fibrillation
• Under dabigatran, rivaroxaban, apixaban or edoxaban therapy.
• Developing ischemic stroke, transient ischemic attack or intracranial hemorrhage during NOAC therapy.

Exclusion Criteria:

• Refuse to provide blood sample for non-vitamin K antagonist oral anticoagulant (NOAC) concentration measurement.
• Refuse to provide informed consent.

Atrial fibrillation with NOAC therapy and subsequently develops acute stroke.