Relationship between device-detected burden and duration of atrial fibrillation and risk of ischemic stroke

Abstract

Wider availability of continuous rhythm monitoring has made feasible the incorporation of metrics of atrial fibrillation (AF) burden and duration into the decision to initiate anticoagulation. However, the relationship between thresholds of burden and duration and underlying risk factors at which anticoagulation should be considered remains unclear. The purpose of this study was to evaluate the relationships of these metrics with each other and the outcome of stroke/transient ischemic attack (TIA). We identified patients with cardiovascular implantable electronic devices (CIEDs) with atrial leads who had at least 1 interrogation in 2016 demonstrating nonpermanent AF and were not receiving oral anticoagulation (OAC). We evaluated the relationship between burden (ie, percentage of time spent in AF), the longest single episode of AF, and risk factors (ie, CHA2DS2-VASc score) in predicting risk of stroke/TIA. The study included 384 patients with mean follow-up of 3.2 ± 0.8 years and incidence of stroke/TIA of 14.8% during follow-up (∼4.6% per year). The burden of AF and the duration of longest episode demonstrated a significant positive correlation to each other but not CHA2DS2-VASc score. Importantly, although the CHA2DS2-VASc score was predictive of stroke/TIA, neither burden nor duration was associated with stroke/TIA. Among patients with CIED-detected AF not receiving OAC, the amount of AF (measured by either burden or duration) does not seem to significantly impact stroke risk, whereas CHA2DS2-VASc score does. These data suggest that among patients with CIED-detected AF, once AF occurs, stroke risk seems to be predominantly driven by underlying risk factors.