P834Long term monitoring for atrial fibrillation in patients with transient ischemic attack

Abstract

Background: Atrial fibrillation (AF) is a major cause of stroke. Use of continuous cardiac monitoring (CCM) with implantable loop recorders (ILR) significantly enhances AF detection in stroke patients. Patients with transient ischemic attack (TIA) are often included in stroke trials even though the causality between TIA and AF is poorly understood and prospective studies are lacking. Purpose: To determine the one‐year incidence of AF in patients with TIA by CCM in addition to standard care with 12‐lead ECG (ECG‐12) and 72‐hour Holter‐monitoring (HM). Secondly, to determine predictors for developing AF. Methods: Prospective cohort study of patients with TIA with normal ECG and HM. Exclusion criteria were: Age <18 or > 81 years; history of stroke or AF; ongoing oral anticoagulation (OAC) therapy or a contraindication for OAC; significant carotid artery stenosis; TIA diagnosis not confirmed by two independent neurologists. Eligible patients were offered inclusion and implanted with a loop recorder and followed for one year. Results: From November 2013 to October 2015 809 patients were diagnosed with TIA. We excluded 574, leaving 235 eligible patients. Of these, nine (3.8%) were diagnosed with AF on ECG‐12 or HM. Another 121 refused ILR implantation. In total, 105 patients (median age 65.4 (range 27.1‐80.8) years) received an ILR. Seven (6.7%) patients were diagnosed with AF, an increase of 78% compared to ECG‐12 and HM alone. The median time from device implantation to first AF episode was 21 days (range 9‐146). Predictors for AF were a history of prior TIA (Odds ratio (OR) 11.5, 95% confidence interval (CI) 2.1‐63.6; p = 0.005) and heart failure (OR 12.7, 95% CI 1.71‐96.83; p = 0.013). Conclusions: Addition of CCM with an ILR for one year increases the detection rate of AF in TIA patients with 78%. However, AF incidence is still relatively low in TIA patients compared with reported AF incidences in stroke patients. Predictors for developing AF were prior TIA and heart failure. Further studies are needed to confirm these findings, especially randomized clinical trials to determine the true yield of CCM with an ILR in a TIA population.