Management of Patients with Embolic Stroke of Unknown Source: Interpreting the Evidence in the Light of Clinical Judgement.

The failure of randomized trials to show benefit of anticoagulation in ESUS is probably due to misclassification of large artery atherosclerosis (LAA) as ESUS, as defined by a stenosis ≥ 50%. There are important differences among DOACs. There are a number of problems with dabigatran, and rivaroxaban and edoxaban are not suitable for once-daily dosing. Recent evidence from real-world practice indicates that apixaban is more effective and safer than rivaroxaban. Plaque burden should be included in the definition of LAA. Patients in whom a cardioembolic source is strongly suspected should be anticoagulated; antiplatelet agents are not significantly safer than DOACs, and are not effective in cardioembolic stroke.