P3762The impact of type and burden of atrial fibrillation on stroke occurrence in patients with atrial fibrillation: from a prospective cohort of atrial fibrillation patients (CODE-AF Registry)

Abstract

Background and purpose Although several studies reported that stroke risk in patients with paroxysmal atrial fibrillation (AF) is similar to those with persistent or permanent AF, there is still controversy on the relationship of AF type and stroke occurrence. We investigated the effect of AF type on AF burden and stroke risk in patients with non-valvular AF. Methods Within the CODE-AF prospective, outpatient registry (COmparison study of Drugs for symptom control and complication prEvention of Atrial Fibrillation), we identified 8,883 patients ≥18 years of age with non-valvular AF and eligible follow-up visits. We compared AF burden and stroke risk among patients with 3 types of AF: paroxysmal (n=5,808) or persistent (n=2,806) or permanent (n=269). Results The median age of the overall population was 68.0 (interquartile range, 60.0–75.0); 36.0% were female. Patients with persistent and permanent AF were older and had higher CHA2DS2-VASc scores and anticoagulation rate than those with paroxysmal AF. Compared with permanent AF (5.2±16.4%), the arrhythmic burden of AF on 24hrs Holter monitoring was significantly lower in paroxysmal AF (2.1±7.2%, p<0.001) and persistent AF (2.0±7.5%, p<0.001). During median follow-up period of 1.38 years (interquartile range: 0.96–1.67), total 82 (0.92%) patients experienced ischemic stroke with incidence rates of 0.51, 1.04 and 1.69 events per 100 person-years for paroxysmal, persistent and permanent AF, respectively. Compared with paroxysmal AF, the risk of ischemic stroke was increased in persistent AF with clinical variable adjusted hazard ratio (aHR) of 1.94 (95% confidence intervals [CI], 1.23–3.07; P=0.005) and permanent AF with aHR of 2.64 (95% CI, 1.09–6.41; P=0.03). AF type and HR of stroke occurrence Paroxysmal (n=5,808) Persistent (n=2,806) Permanent (n=269) Stoke events 39 37 6 Person years (PYs) 7673 3544 356 /100 PYs 0.51 1.04 1.69 HR (95% CI), p-value HR (95% CI), p-value HR (95% CI), p-value Unadjusted HR 1 (Reference) 2.05 (1.27–3.31), 0.003 3.32 (1.15–7.90), 0.02 Clinical variables adjusted HR 1 (Reference) 1.94 (1.23–3.07), 0.005 2.64 (1.09–6.41), 0.03 PYs: Person years; HR: Hazard ratio. Conclusion Persistent and permanent AF was associated with the increased risk of stroke than paroxysmal AF, after adjustment of clinical variables including age, sex, comorbidities and anticoagulation rate. These results suggest that AF type and burden might be related with the risk of ischemic stroke and should be considered in the stroke prevention of AF.