Estimated incidence of previously undetected atrial fibrillation on a 14-day continuous electrocardiographic monitor and associated risk of stroke

Abstract

Aims: There is uncertainty about whether and how to perform screening for atrial fibrillation (AF). To estimate the incidence of previously undetected AF that would be captured using a continuous 14-day ECG monitor and the associated risk of stroke.

Methods and results: We analysed data from a cohort of patients >65 years old with hypertension and a pacemaker, but without known AF. For each participant, we simulated 1000 ECG monitors by randomly selecting 14-day windows in the 6 months following enrolment and calculated the average AF burden (total time in AF). We used Cox proportional hazards models adjusted for CHA2DS2-VASc score to estimate the risk of subsequent ischaemic stroke or systemic embolism (SSE) associated with burdens of AF > and <6 min. Among 2470 participants, the median CHA2DS2-VASc score was 4.0, and 44 patients experienced SSE after 6 months following enrolment. The proportion of participants with an AF burden >6 min was 3.10% (95% CI 2.53-3.72). This was consistent across strata of age and CHA2DS2-VASc scores. Over a mean follow-up of 2.4 years, the rate of SSE among patients with <6 min of AF was 0.70%/year, compared to 2.18%/year (adjusted HR 3.02; 95% CI 1.39-6.56) in those with >6 min of AF.

Conclusions: Approximately 3% of individuals aged >65 years with hypertension may have more than 6 min of AF detected by a 14-day ECG monitor. This is associated with a stroke risk of over 2% per year. Whether oral anticoagulation will reduce stroke in these patients is unknown.

Keywords: Atrial fibrillation; Holter monitor; Oral anticoagulation; Screening; Simulation; Stroke.