Risk of ischaemic stroke according to pattern of atrial fibrillation: analysis of 6563 aspirin-treated patients in ACTIVE-A and AVERROES

Aims The pattern of atrial fibrillation (AF) occurrence-paroxysmal, persistent, or permanent-is associated with progressive stages of atrial dysfunction and structural changes and may therefore be associated with progressively higher stroke risk. However, previous studies have not consistently shown AF pattern to predict stroke but have been hampered by methodological shortcomings of low power, variable event ascertainment, and variable anticoagulant use. Methods and results We analysed the rates of stroke and systemic embolism in 6563 aspirin-treated patients with AF from the ACTIVE-A/AVERROES databases. There was thorough searching for events and adjudication. Multivariable analyses were performed with the adjustment for known risk factors for stroke. Mean age of patients with paroxysmal, persistent, and permanent AF was 69.0 +/- 9.9, 68.6 +/- 10.2, and 71.9 +/- 9.8 years (P < 0.001). The CHA(2)DS(2)-VASc score was similar in patients with paroxysmal and persistent AF (3.1 +/- 1.4), but was higher in patients with permanent AF (3.6 +/- 1.5, P < 0.001). Yearly ischaemic stroke rates were 2.1, 3.0, and 4.2% for paroxysmal, persistent, and permanent AF, respectively, with adjusted hazard ratio of 1.83 (P < 0.001) for permanent vs. paroxysmal and 1.44 (P = 0.02) for persistent vs. paroxysmal. Multivariable analysis identified age >= 75 year, sex, history of stroke or TIA, and AF pattern as independent predictors of stroke, with AF pattern being the second strongest predictor after prior stroke or TIA. Conclusion In a large population of non-anticoagulated AF patients, pattern of AF was a strong independent predictor of stroke risk and may be helpful to assess the risk/benefit for anticoagulant therapy, especially in lower risk patients.