Cryptogenic stroke describes a subset of ischemic stroke for which no cause can be found despite a structured investigation. There are a number of putative mechanisms of cryptogenic ischemic stroke including a covert structural cardiac lesion, paroxysmal atrial fibrillation, hypercoagulable state or undiagnosed malignancy. Because many of these proposed mechanisms are embolic - and based on studies of thrombus history showing commonalities between thrombus composition between cardioembolic and cryptogenic strokes - the concept of embolic stroke of undetermined source (ESUS) (Hart et al. Lancet Neurol. 13(4):429-38, 2014; Stroke. 48(4):867-72, 2017) has been proposed to describe cryptogenic strokes that may warrant systemic anticoagulation. In this review, we discuss the phenomena of cryptogenic stroke, ESUS and a proposed management pathway.1. The concept of ESUS was proposed in 2014 as a potentially useful therapeutic entity. Two recent trials - NAVIGATE-ESUS (Hart et al. N Engl J Med. 378(23):2191-201, 2018) and RESPECT-ESUS (Diener 2018) were proposed based on this concept. They were negative for their primary endpoint and for the secondary endpoint of ischemic stroke recurrence. Post-hoc analysis of the WARSS trial (Longstreth et al. Stroke. 44(3):714-9, 2013) suggested that people with elevated pro-BNP benefited from systemic anticoagulation whereas those with a normal pro-BNP did not. This led to the hypothesis that a subgroup of patients at higher risk for embolism from the left atrium would benefit from anticoagulation, even if the WARSS trial was negative for the primary endpoint. Thus, the ARCADIA trial (Kamel et al. Int J Stroke. 14(2):207-14, 2019) was proposed - a randomized, active-control, multi-center trial comparing apixaban with aspirin for secondary stroke prevention in patients with ESUS and biomarkers of left atrial cardiopathy. This trial is actively recruiting. 2. Carotid web - an intimal form of fibromuscular dysplasia - has come to increased prominence in the literature as a cause of embolic stroke. It is a non-stenosis, non-atherosclerotic lesion in the posterior wall of the internal carotid artery that leads to pooling with stasis of blood distal to the lesion and, as a consequence, embolic stroke. It is not usually detected by a standard stroke workup as it masquerades as non-calcified atherosclerosis and does not cause hemodynamically significant stenosis. There have been two major recent papers - a meta-analysis in Stroke (Zhang et al. Stroke. 49(12):2872-6, 2018) and narrative review in JAMA Neurology (Kim et al. JAMA Neurol. 2018) - that addressed this topic. Cryptogenic stroke describes a stroke for which no cause has been found. ESUS is a more precisely-defined entity that mandates a specific workup and implicates remote embolism as a cause of stroke. In ESUS, the options for further investigation include long-term cardiac monitoring, transesophageal echocardiography, investigation for occult malignancy or arterial hypercoagulability. Options for management include anti-platelet therapy (the current standard of care), empiric anticoagulation or enrollment in to a clinical trial examining the use of NOACs compared with aspirin for secondary prevention (such as ARCADIA or ATTICUS). In a person less than 60 years old with ESUS and a patent foramen ovale the risk of a recurrent stroke is low but recent trials have suggested that percutaneous device closure reduces this risk further with an acceptable complication rate.