[Embolic stroke of undetermined source (ESUS) : How much monitoring is necessary?].

The clinical construct of embolic stroke of unknown source (ESUS) was first described in 2014. It is defined as cryptogenic ischemic stroke after the exclusion of a lacunar infarct, a significant (≥50%) stenosis of extracranial or intracranial arteries and a cardiac source of embolism. Initially, there was hope that these patients would benefit from anticoagulation. This was based on the suspicion that imaging criteria of stroke mimic features of embolism from cardiac sources or the great arteries. In two large randomized trials with 12,600 patients neither rivaroxaban nor dabigatran could reduce the risk of recurrent stroke. Based on these results, current research is focused on paroxysmal atrial fibrillation as a potential cause of stroke in these patients. Several randomized trials could show that by prolongation of monitoring to 30 days atrial fibrillation can be detected in approximately 10% of the patients. Using continuous monitoring (e. g. by implantable loop recorders) atrial fibrillation can even be detected in one quarter of the patients. Not all stroke patients can receive such an intensive monitoring. Therefore, this article summarizes the evidence and presents the resulting recommendations for patient selection and staged rhythm diagnostics and discusses a recently presented algorithm of an expert group for use in daily clinical practice.