Cryptogenic stroke - How to make sense of a non-diagnostic entity

Abstract

Secondary preventive strategies in ischaemic stroke depend on the underlying aetiology. However, approximately one-third of ischaemic strokes remain unexplained, or 'cryptogenic'. There is a wide range of possible underlying causes in cryptogenic stroke, and the best approach to secondary prevention of these may differ. To date, though, the widely accepted and uniform secondary preventive strategy in this group consists of modification of vascular risk factors, and of treatment with a combination of antiplatelet therapy and antihypertensive and lipid-lowering medication. Among the potential causes for cryptogenic stroke are occult atrial fibrillation, patent foramen ovale, atrial cardiopathy, aortic arch atheroma and hypercoagulable states. While it is possible to diagnose these conditions, in individual patients there is often uncertainty over whether they have a directly causative role, are markers of disease, or are innocent bystanders. Similarly, even if the cause is found, the best secondary preventive strategies remain uncertain, which questions the benefit of extensive investigations in a clinical setting. More recently, the concept of "embolic stroke of unknown source (ESUS)" has been introduced, in the hope that anticoagulation may offer better secondary prevention than antiplatelet therapy, but trials so far have been negative. At present, there is little justification for introducing extensive new investigative strategies into the management of patients with cryptogenic stroke. Investigations should be targeted at identifying those high-risk conditions which lead to a change in management. Further investigations need to be tailored individually, according to clinical circumstances. This should include identifying patients for participation in clinical trials, as the significance and best management of many of the potential causes for cryptogenic stroke require further research.

Keywords: Aetiology; Ischaemic stroke; Risk factors.