Stroke of Undetermined Cause: Workup and Secondary Prevention

Christian Weimar


Curr Opin Neurol. 2016;29(1):4-8. 

In This Article

Abstract and Introduction


Purpose of review The purpose of this review is to update the reader on current concepts of workup and secondary prevention in patients with stroke of undetermined cause.

Recent findings Clinical research in patients with cryptogenic stroke has been hampered by the lack of standardized, widely accepted diagnostic criteria. The new definition of 'Embolic stroke of undetermined etiology' postulates an embolic mechanism of ischemic stroke. It is based on the exclusion of lacunar infarction by brain imaging, arterial stenosis more than 50% or dissection of the respective brain-supplying artery by computed tomography/magnetic resonance-angiography or ultrasound, atrial fibrillation by at least 24 h EKG monitoring, as well as some rare etiologies such as vasculitis, drug abuse, or coagulopathies. However, it still comprises many patients with atherosclerotic etiologies (but <50% stenosis) as well as covert paroxysmal atrial fibrillation which can be detected by repeated Holter EKG or an implantable device. A patent foramen ovale can be found in up to 58% of cryptogenic stroke patients, but causality in an individual patient remains uncertain and can only be statistically inferred.

Summary The new concept of embolic stroke of undetermined etiology enabled three ongoing randomized controlled trials which investigate oral anticoagulation versus aspirin for secondary stroke prevention.


Classification of ischemic stroke serves several purposes, namely describing and determining case mix in a clinical trial, phenotyping for genetic studies, and making treatment decisions in clinical routine. About 25 years ago, the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria were introduced and provided a first widely adopted operational definition of presumed etiologies for ischemic stroke.[1] However, this classification considers only the one cause thought to be directly and causally related to the ischemic stroke, neglecting other underlying diseases not deemed to be direct causes. As a consequence, patients with unknown stroke causes, incomplete workup or more than one possible cause are lumped together in a category called cryptogenic, which is thus only defined by what it is not. As a consequence, an incomplete workup may also qualify for a classification of cryptogenic stroke. Despite many modifications and adaptations, interrater reliability for the TOAST criteria remains low and no randomized trials have defined optimum antithrombotic prevention strategies, specifically in patients with cryptogenic stroke. Subsequent classification systems, such as the Oxford classification, ASCO, or ASCOD phenotyping of ischemic stroke omit the term undetermined or cryptogenic altogether because it is only negatively defined.[2–4] Although these classification systems define specific diseases, no causal relationship is inferred for low causality grades, leaving the physician with the same uncertainty about secondary preventive treatment. Furthermore, these classifications have not become widely used in secondary prevention trials, and therefore, no clinical data are available yet. Thus, current guidelines do not specifically comment on cryptogenic stroke or recommend antiplatelet therapy only.[5–8] The new concept of Embolic Stroke of Undetermined Source (ESUS) in contrast provides an operational definition and comprises nonlacunar strokes of presumed embolic or thromboembolic origin.[9]