Cryptogenic stroke frequent and recurrent

This does not give me a warm fuzzy feeling at all. I'm sure your doctor has long ago given up on trying to explain your cryptogenic stroke.
http://www.news-medical.net/news/20150804/Cryptogenic-stroke-frequent-and-recurrent.aspx
Cryptogenic ischaemic strokes and transient ischaemic attacks frequently recur, despite having no identifiable cause, show findings from the Oxford Vascular Study (OXVASC).
“[I]n view of the high rate of recurrent cryptogenic stroke, randomised trials of available preventive treatments, such as the ongoing trials of new anticoagulants, are justified despite uncertainty about cause”, write the researchers in The Lancet Neurology.
Patients in the first-phase of OXVASC, from 2002 to 2010, were classified as having cryptogenic stroke only if they had no identifiable cause after undergoing at least brain imaging, electrocardiography and extracranial imaging, and those in the later phase also had to undergo intracranial vascular imaging, R test and echocardiography.

Yet of the total 2555 patients with first ischaemic events, 812 (32%) were classed as having cryptogenic stroke, report Peter Rothwell (John Radcliffe Hospital, Oxford, UK) and co-researchers.
In an accompanying commentary, Jose Ferro (University of Lisbon, Portugal) says the study highlights the importance of a comprehensive work-up, noting that the work-up used in OXVASC is not even routine in all of the UK and Europe, and in developing countries is “a distant mirage”.
He stresses that neurologists should “not be surprised to find no cause in a third of ischaemic strokes” but should “not label a patient as having a cryptogenic stroke before completion of a comprehensive investigation of possible causes”.
Patients were as likely to die of cryptogenic stroke as noncardioembolic stroke, with 1-year mortality rates of 6.0% for a cryptogenic cause, 0.9% for small-vessel disease and 14.7% for large-vessel disease. They also had similar rates of disability at 6 months.
The cryptogenic stroke group had the lowest rate of most cardioembolic and atherosclerotic risk factors, and had fewer risk factors overall than the 668 patients with cardioembolic stroke and the 597 with noncardioembolic stroke.
Nevertheless, they were just as likely to have a recurrent ischaemic stroke, with 5-year rates of 23.3% compared with 20.0% to 25.3% for the other groups.
Of note, 80% of cryptogenic stroke patients in the second phase of OXVASC underwent ambulatory home cardiac monitoring, but only 6% had paroxysmal atrial fibrillation lasting longer than 30 seconds.
Ferro says that this “contrasts with the present enthusiasm for the use of technology to detect paroxysmal atrial fibrillation and for the possibility to prevent recurrent cryptogenic stroke with anticoagulants.”