Abstract 22: Incidence Rate of Atrial Fibrillation in Cryptogenic Stroke: Results From the Reveal LINQ Real-World Registry

• Published in: Stroke (1970) 2020-02, Vol.51 (Suppl_1)
• Main Authors: Kasner, Scott E, Verma, Atul, Mittal, Suneet, Natale, Andrea, Beinart, Sean C, Amin, Alpesh, Pouliot, Erika, Franco, Noreli C, Diener, Hans-Christoph
• Format: Article
• Language: English
• ISSN: 0039-2499; 1524-4628
• DOI: 10.1161/str.51.suppl_1.22

Abstract only Objectives: To determine the incidence rate of AF in real-world patients with prior cryptogenic stroke (CS) or transient ischemic attack (TIA) and to assess the impact of using an insertable cardiac monitor (ICM) on treatments. Methods: The Reveal LINQ registry is an ongoing, observational, multi-center study evaluating the role of the ICM in the patient care pathway. Patients with prior CS or TIA were included. Patients without device data or without any contact after ICM insertion, with a history of AF, prior known atrial ablation, or hemorrhagic stroke were excluded. Device data were adjudicated by at least 2 independent reviewers. Patients were considered to have AF if they had a ≥2 min episode, and to not have AF if no AF was found after 1 year of monitoring. Patients without AF and without 1 year of follow-up were excluded from the general analysis due to limited follow-up time but were included in the AF incidence rates calculation for completeness. Results: In total, 192 patients (61±14 years, 61% male, mean CHA2DS2-VASc score 3.8±1.4) from 29 centers in 8 countries were included. Reasons for monitoring were CS (96%) and TIA (4%). During an average follow-up of 19±6 months, AF was detected in 36 (19%) patients and the incidence of first episode of AF ≥2 min, ≥1 h, ≥6 h, and ≥24 h was 16.7%, 7.9%, 3.7% and 1.7% at 24 months, respectively (figure). Patients with AF were older (68±14 years vs. 60±14, p