Ischaemic stroke of undetermined cause

www.thelancet.com/neurology Vol 14 September 2015 871 The defi nition of cryptogenic stroke or stroke of undetermined origin is aff ected by the current knowledge of the cause and pathogenesis of stroke and by the availability, comprehensiveness, quality, and timeliness of the ancillary investigations undertaken to discover the cause of the stroke. In The Lancet Neurology, Linxin Li and colleagues report fi ndings from a population-based study of the burden, outcome, risk factors, and long-term prognosis of cryptogenic stroke in patients with a fi rst transient ischaemic attack (TIA) or ischaemic stroke. They found that of 2555 fi rst ischaemic events, 812 (32%) were cryptogenic. The study shows that the prognosis of cryptogenic stroke and the risk of recurrence are similar to those of large artery disease and small vessel disease. For example, death or dependency at 6 months was similar after cryptogenic stroke compared with large artery and small vessel subtypes combined (23% vs 27%; p=0·26), as was the 10-year risk of recurrence (32% vs 27%; p=0·91). Secondary stroke prevention must be similarly aggressive and appropriate in stroke with and without identifi ed cause. The fi ndings from this Article question the role of paroxysmal atrial fi brillation as the major cause of cryptogenic stroke, which contrasts with the present enthusiasm for the use of technology to detect paroxysmal atrial fi brillation and for the possibility to prevent recurrent cryptogenic stroke with anticoagulants. The results of this study are especially important for health-care policy makers, because they show the burden of cryptogenic stroke, the rationale for a comprehensive investigation into the cause of every ischaemic stroke, and the need to provide stroke centres with access to the necessary ancillary diagnostic procedures. The diagnostic work up used in this study is unfortunately still not routinely used for every patient with ischaemic stroke, neither in the UK nor in mainland Europe. In developing countries, such diagnostic instruments are a distant mirage. This study has some limitations that hamper the translation of its results into routine practice: both TIAs and strokes were included, and the diagnosis of TIA can be erroneous even when done by a specialist; the causative investigation was incomplete, especially in phase 1; the search for paroxysmal atrial fi brillation was not pursued intensively; the investigation of the cause of stroke in recurrent strokes was incomplete; and the newest devices for prolonged heart rhythm monitoring were not used. All these limitations might have resulted in misclassifi cation of some strokes as cryptogenic and Published Online July 28, 2015 http://dx.doi.org/10.1016/ S1474-4422(15)00149-0