Anticoagulation in patients with Embolic Stroke of Unknown Source

When warfarin was the mainstay of anticoagulation for the prevention of cardioembolic stroke, the paradigm was essentially "we mustn't anticoagulate anyone unless we prove that the stroke was cardioembolic." Now that direct-acting oral anticoagulants are available, the paradigm should change. The risk of stroke is highest soon after the initial event, particularly in patients with more than one infarction. Direct-acting oral anticoagulants are not significantly more likely than aspirin to cause severe hemorrhage, and it is now clear that patients with paradoxical embolism are better treated with anticoagulant than aspirin. Percutaneous closure of a patent foramen ovale is better than aspirin, but not better than anticoagulant, and some patients with paradoxical embolism may be better treated with anticoagulant than with percutaneous closure, which cannot prevent pulmonary embolism. Patients in whom cardioembolic stroke is strongly suspected should probably be anticoagulated pending the results of investigations such as echocardiography and prolonged cardiac monitoring for atrial fibrillation, and some of them, in whom the suspicion of a cardioembolic source is very strong, should probably be anticoagulated long term, even if such investigations do not confirm a cardiac source.