Differential characteristics, stroke recurrence, and predictors of covert atrial fibrillation of embolic strokes of undetermined source.

Background and purpose Identifying embolic strokes of undetermined source (ESUS) patients likely to harbor atrial fibrillation may have diagnostic and therapeutic implications. Our aim was to examine differences between ESUS and cardioembolic strokes, to evaluate stroke recurrence rate among ESUS and to identify baseline characteristics of ESUS patients who were later diagnosed with atrial fibrillation. Materials and methods We assessed all ischemic stroke patients admitted between June 2012 and November 2013. ESUS were compared to cardioembolic strokes at discharge. After at least 12-month follow-up, ESUS patients diagnosed with atrial fibrillation were compared to those who remained as ESUS. Results There were 236 ischemic strokes, 32.6% were ESUS. Compared to cardioembolic strokes, ESUS were younger (p < 0.0001), had milder strokes (p < 0.05), less prevalence of hypertension (p < 0.05), peripheral vascular disease (p < 0.05), and previous ischemic stroke (p < 0.05). After follow-up, 15% of ESUS patients had stroke recurrences and 12% evidenced paroxysmal atrial fibrillation. ESUS patients diagnosed with atrial fibrillation in the follow-up were older (p < 0.0001), had higher erythrocyte sedimentation rate (p < 0.05), and were more likely to have ≥2 infarcts in the same arterial territory in the initial magnetic resonance imaging (p < 0.05). Conclusions Older age, small-scattered infarcts on initial magnetic resonance imaging and high erythrocyte sedimentation rate levels appear to identify ESUS patients more likely to be diagnosed of atrial fibrillation during follow-up.