Detection of atrial fibrillation in asymptomatic at-risk individuals

Background: Undiagnosed atrial fibrillation (AF) exposes unsuspecting patients to elevated stroke risks. The op-timal algorithm for identifying patients who should be screened for AF remains undetermined. The objective of this study is to determine the AF burden in an asymptomatic, at-risk population. We also sought to investigate potential predictors of undiagnosed AF. Methods: This registry is a prospective observational study assessing continuous ECG monitoring in screening for AF using a wearable single lead 7-day continuous monitoring device. Patients included were asymptomatic indi-viduals, at risk for AF as determined by either 1) >-65 years of age with >-1 high risk factor or; 2) >-75 years of age and >-2 moderate risk factors. A multivariable logistic regression was used to explore the predictive value of cer-tain patient characteristics in identifying patients susceptible to have undiagnosed AF. Results: Among the 942 patients included, 25 patients (2.7%) had evidence of AF detected. Only 8 patients had AF duration >-24 h. History of perioperative AF (OR: 3.25, 95%CI: 1.08-9.79, p = 0.036), age over 85 (OR: 4.71, 95%CI: 1.31-16.92, p = 0.017) and absence of cardiovascular disease (CVD) (OR: 0.27, 95%CI: 0.10-0.76, p = 0.013) were found to be predictive of undiagnosed AF. Conclusion: This study demonstrates the feasibility of office-based AF screening in at-risk population. The low rate of AF detection suggests that the optimal algorithm for identifying asymptomatic patients who would benefit from continuous screening remains unclear. Advanced age, history of perioperative AF and absence of CVD are variables that could be explored further. (c) 2021 Elsevier B.V. All rights reserved.