Pharmacologic conversion of atrial fibrillation: a systematic review of available evidence

This report reviews the efficacy of currently available antiarrhythmic agents for conversion of atrial fibrilation (AF) to normal sinus rhythm (NSR). A systematic search of literature in the English language was done on computerized databases, such as MEDLINE, EMBASE, and Current Contents, in reference lists, by manual searching, and in contact with expert informants. Published studies involving humans that described the use of antiarrhythmic therapy for conversion of AF to NSR were considered and only studies that examined the use of agents currently available in the United States were included. Studies exclusively describing antiarrhythmic therapy for conversion of postsurgical AF were excluded. The methodology and results of each trial were assessed and attempts were made to acquire additional information from investigators when needed. Assessment of methodological quality was incorporated into a levels-of-evidence scheme. Eighty-eight trials were included, of which 34 (39%) included a placebo group (level I data). We found in recent-onset AF of less than 7 days, intravenous (i.v.) procainamide, high-dose i.v. or high-dose combination i.v. and oral amiodarone, oral quinidine, oral flecainide, oral propafenone, and high-dose oral amiodarone are more effective than placebo for converting AF to NSR. In recent-onset AF of less than 90 days, i.v. ibutilide is more effective than placebo and i.v. procainamide. In chronic AF, oral dofetilide converts AF to NSR within 72 hours, and oral propafenone and amiodarone are effective after 30 days of therapy. We conclude than for conversion of recent-onset AF of less than 7 days, procainamide may be considered a preferred i.v. agent and propafenone a preferred oral agent. For conversion of recent-onset AF of longer duration (less than 90 days), i.v. ibutilide may be considered a preferred agent. For patients with chronic AF and left ventricular dysfunction, direct current cardioversion is the preferred conversion method. Larger, well-designed randomized controlled trials with clinically important endpoints in specific populations of AF patients are needed.