Stroke event rates in anticoagulated patients with paroxysmal atrial fibrillation

Aims. To test the hypothesis that stroke and systemic embolic events (SEE) in the stroke prevention using an oral thrombin inhibitor in atrial fibrillation (SPORTIF) III and V trials are different between paroxysmal and persistent atrial fibrillation (AF). Methods. Data analysis from two cohorts of patients enroled in the prospective SPORTIF III and V clinical trials (n = 7329); 836 subjects (11.4%) with paroxysmal AF [mean age 70.1 years (SD = 9.5)] were compared with 6493 subjects with persistent AF for this ancillary study. Results. The annual event rates for stroke/SEE are 1.73% for persistent AF and 0.93% for paroxysmal AF. In a multivariate analysis, after adjusting for stroke risk factors, gender and aspirin usage, the differences remained statistically significant with a higher hazard ratio (HR) for stroke/SEE in persistent AF [vs. paroxysmal AF, HR 1.87, 95% confidence interval (CI) 1.04-3.36; P = 0.037]. In 'high risk' patients (with >= 2 stroke risk factors) annual event rates for stroke/SEE were 2.08% for persistent AF and 1.27% for paroxysmal AF (adjusted HR = 1.68, 95% CI 0.91-3.1, P = 0.098). Elderly patients had annual event rates for stroke/SEE of 2.38% for persistent AF and 1.13% for paroxysmal AF (adjusted HR = 2.27, 95% CI 0.92-5.59, P = 0.075). Vitamin K antagonist (VKA)-naive paroxysmal AF patients had a 1.89%/year stroke/SEE rate, compared with 0.61% for previous VKA takers (HR = 0.33, 95% CI 0.11-1.01, P = 0.052). Conclusion. In this large clinical trial cohort of anticoagulated AF patients, those with paroxysmal AF had stroke rates which were lower than for patients with persistent AF, although both groups had broadly similar stroke risk factors. Subjects with paroxysmal AF at 'high risk' had stroke/SEE rates that were not significantly different to persistent AF subjects.