Stroke risk assessment in different atrial fibrillation subtypes: CHA2DS2-VASc score and beyond

Patients with atrial fibrillation (AF) have been known to have an inhomogeneous risk of stroke, with several well-documented risk factors including hypertension, history of stroke, and diabetes mellitus that have been incorporated into the CHA2DS2–VASc score to stratify risk levels. Accumulating evidence has suggested that AF subtype, namely persistent AF and paroxysmal AF, also have an impact on stroke development.

This study focused on the impact of AF subtype on stroke risk in a large, long-term follow-up cohort as to provide insight of the impact on ischemic stroke risks. Such findings may play a role in future ischemic stroke risk stratification in selected patients.

A total of 7346 AF patients were followed up for 10 years from January 2010 to December 2020 (mean follow-up time 2.74 years). Among these patients, 2460 (33.5%) had persistent or permanent AF (PerAF) and 4886 (66.5%) had paroxysmal AF (PAF). The study evaluated the risk of ischemic stroke between the two groups using Cox models adjusting for all components of the CHA2DS2–VASc score. Analyses were further stratified into the usage of oral anticoagulation (OAC) or no OAC.

Comparing AF patients based on subtype, it was found that PerAF patients were older, had a higher prevalence of previous stroke, congestive heart failure, PAD, DM, and were more often prescribed anticoagulants. Univariate analysis found patients with PerAF at a higher risk for ischemic stroke (HR 1.262, p<0.001). The results remained significant after multivariate analysis with factors of the CHA2DS2–VASc score (HR 1.234, p<0.001). After stratifying patients according to anticoagulation prescription, PerAF and PAF patients were found to have similar rates of developing ischemic stroke in patients under DOAC.

Patients with PerAF have a higher risk of stroke during follow-up independent of components of the CHA2DS2VASc score. In patients with OAC use, there was no difference in stroke risk between AF subtypes.

Stroke Risk in OAC-naive patients