Pharmacologic cardioversion of recent-onset atrial fibrillation: a systematic review and network meta-analysis

Aims: We sought to identify the most effective antidysrhythmic drug for pharmacologic cardioversion of recent-onset atrial fibrillation (AF). Methods and results: We searched MEDLINE, Embase, and Web of Science from inception to March 2019, limited to human subjects and English language. We also searched for unpublished data. We limited studies to randomized controlled trials that enrolled adult patients with AF <= 48h and compared antidysrhythmic agents, placebo, or control. We determined these outcomes prior to data extraction: (i) rate of conversion to sinus rhythm within 24h, (ii) time to cardioversion to sinus rhythm, (iii) rate of significant adverse events, and (iv) rate of thromboembolism within 30days. We extracted data according to PRISMA-NMA and appraised selected trials using the Cochrane review handbook. The systematic review initially identified 640 studies; 30 met inclusion criteria. Twenty-one trials that randomized 2785 patients provided efficacy data for the conversion rate outcome. Bayesian network meta-analysis using a random-effects model demonstrated that ranolazine + amiodarone intravenous (IV) [odds ratio (OR) 39.8, 95% credible interval (CrI) 8.3-203.1], vernakalant (OR 22.9, 95% CrI 3.7-146.3), flecainide (OR 16.9, 95% CrI 4.1-73.3), amiodarone oral (OR 10.2, 95% CrI 3.1-36.0), ibutilide (OR 7.9, 95% CrI 1.2-52.5), amiodarone IV (OR 5.4, 95% CrI 2.1-14.6), and propafenone (OR 4.1, 95% CrI 1.7-10.5) were associated with significantly increased likelihood of conversion within 24h when compared to placebo/control. Overall quality was low, and the network exhibited inconsistency. Probabilistic analysis ranked vernakalant and flecainide high and propafenone and amiodarone IV low. Conclusion: For pharmacologic cardioversion of recent-onset AF within 24h, there is insufficient evidence to determine which treatment is superior. Vernakalant and flecainide may be relatively more efficacious agents. Propafenone and IV amiodarone may be relatively less efficacious. Further high-quality study is necessary.