Frequency of Atrial Fibrillation–Associated Stroke

Study Questions:

What is the prevalence of atrial fibrillation (AF)–associated stroke in different centers around the world?


This study was part of the Embolic Stroke of Undetermined Source Global Registry. At 19 centers in 19 countries, data from consecutive inpatient stroke cases between 2013 and 2014 were reported to the registry. AF–associated stroke was defined as an ischemic stroke and history of AF or detection of AF on monitoring after stroke. Information regarding antithrombotic therapy was not collected. Data were analyzed according to four global regions: Europe (nine sites), North America (two sites), Latin America (three sites), and East Asia/Pacific (nine sites). Sites were also stratified into income groups.


There were 2,144 patients included in the analysis. The average age was 67 (standard deviation, 15.39). The frequency of AF–associated stroke was 28% (95% confidence interval [CI], 25.6-29.5). AF was least prevalent in the Philippines (11%) and most prevalent in Ireland (45%). AF was more prevalent at North American and European sites, and in high income countries. Patients with AF–associated stroke were older, on average, than patients without AF (mean age, 75 vs. 64; p < 0.001). Across sites, older age was correlated with AF–associated stroke (correlation, 0.76; p < 0.001). In the patients with AF–associated stroke, 56% had a history of AF prior to the stroke. Thirty-day mortality was higher in patients with AF–associated stroke (10%; 95% CI, 7.6-12.6) than in stroke patients without AF (4%; 95% CI, 3.3-5.4).


The authors concluded that AF–associated stroke is common and AF is a likely stroke mechanism in older patients.


AF is an important cause of stroke because it is common, associated with more disability and mortality than other stroke types, and stroke due to AF can be effectively prevented with anticoagulation. In this registry, AF was identified in close to 30% of stroke patients. While there were international differences in the prevalence of AF–associated stroke, much of this variation was due to patient age. Older patients are much more likely to have AF–associated ischemic stroke. The results of this study may not be generalizable at the population level due to selection bias. Additionally, the workup to identify AF was not standardized, and this variability tempers the authors’ findings. Despite these limitations, this study is important because it reinforces that AF is a common cause of stroke, particularly in the elderly.

< Back to Listings