New-Onset Atrial Fibrillation Among Patients With Infection in the Emergency Department: A Multicenter Cohort Study of 1-Year Stroke Risk

Abstract

BackgroundPatients with new-onset atrial fibrillation in relation to infection are frequent in emergency departments (EDs) and may require antithrombotic therapy because of the increased risk of stroke. Our objective was to describe the 1-year risk of stroke in patients in the ED with infection, new-onset atrial fibrillation, and no antithrombotic therapy. MethodsThis was a population-based cohort study at 4 EDs in Denmark and Sweden. Atrial fibrillation was identified by electrocardiogram (ECG) upon arrival at the ED, and infection was identified by discharge diagnosis. Patient history was followed for 12 months or until initiation of oral anticoagulant therapy, ischemic stroke, or death. Primary outcome was stroke within 12 months compared to patients with infection and no atrial fibrillation. ResultsIn the analysis, 15,505 patients were included; 48.7% were male and the median age was 71 (IQR, 56-83). Among the included patients, 2107 (13.6%) had atrial fibrillation of any kind and 822 (39.0%) of these had new-onset atrial fibrillation with a median CHA2DS2-VASc score of 3 (IQR 2-4).New-onset atrial fibrillation during infection showed an absolute postdischarge 1-year risk of stroke of 2.7% (95% CI 1.6-4.2), corresponding to a crude hazard ratio (HR) of 1.4 (95% CI 0.9-2.3), a sex and age adjusted HR of 1.0 (95% CI 0.6-1.6), and a CHA2DS2-VASc adjusted HR of 1.1 (95% CI, 0.7-1.8) compared to patients with infection but no atrial fibrillation. ConclusionsPatients in the ED with infection and new-onset atrial fibrillation without current oral anticoagulant therapy had a 2.7% absolute 1-year risk of stroke. Stroke events were mainly related to sex and age and risk factors identified by the CHA2DS2-VASc score.