Evaluation of a novel cardioversion intervention for atrial fibrillation: the Ottawa AF cardioversion protocol.

Electrical cardioversion is commonly performed to restore sinus rhythm in patients with atrial fibrillation (AF), but it is unsuccessful in 10-12% of attempts. We sought to evaluate the effectiveness and safety of a novel cardioversion protocol for this arrhythmia. Consecutive elective cardioversion attempts for AF between October 2012 and July 2017 at a tertiary cardiovascular centre before (Phase I) and after (Phase II) implementing the Ottawa AF cardioversion protocol (OAFCP) as an institutional initiative in July 2015 were evaluated. The primary outcome was cardioversion success, defined as ≥2 consecutive sinus beats or atrial-paced beats in patients with implanted cardiac devices. Secondary outcomes were first shock success, sustained success (sinus or atrial-paced rhythm on 12-lead electrocardiogram prior to discharge from hospital), and procedural complications. Cardioversion was successful in 459/500 (91.8%) in Phase I compared with 386/389 (99.2%) in Phase II (P < 0.001). This improvement persisted after adjusting for age, body mass index, amiodarone use, and transthoracic impedance using modified Poisson regression [adjusted relative risk 1.08, 95% confidence interval (CI) 1.05-1.11; P < 0.001] and when analysed as an interrupted time series (change in level +9.5%, 95% CI 6.8-12.1%; P < 0.001). The OAFCP was also associated with greater first shock success (88.4% vs. 79.2%; P < 0.001) and sustained success (91.6% vs 84.7%; P=0.002). No serious complications occurred. Implementing the OAFCP was associated with a 7.4% absolute increase in cardioversion success and increases in first shock and sustained success without serious procedural complications. Its use could safely improve cardioversion success in patients with AF. www.clinicaltrials.gov ID: NCT02192957.

Informations de copyright

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: [e-mail].