Conversion of recent-onset atrial fibrillation with intravenous amiodarone: a meta-analysis of randomized controlled trials.

Abstract

To evaluate efficacy and safety of intravenous amiodarone for conversion of recent-onset atrial fibrillation. MEDLINE search of published, randomized, controlled trials assessing the efficacy and safety of intravenous amiodarone in recent-onset (< 7 days) atrial fibrillation, supplemented with searches of reference lists of identified articles and bibliographies of secondary and tertiary review articles. The identified trials were eligible for meta-analysis if they met the following criteria: patients had recent-onset atrial fibrillation; patients were randomized to intravenous amiodarone, placebo, or another antiarrhythmic agent; no other antiarrhythmic agent except digoxin was administered simultaneously with intravenous amiodarone or other active treatments; the number and percentage of conversions to sinus rhythm after treatment began were reported; and the number and type of adverse drug reactions occurring after treatment began were reported. Eighteen randomized controlled trials, including active control cohorts and placebo cohorts, met inclusion criteria. Atrial fibrillation was spontaneous in 13 trials, postoperative in 4, and combined spontaneous and postoperative in 1. A total of 550 patients received intravenous amiodarone, 451 received other antiarrhythmic therapy, and 202 received placebo. Unadjusted averages for cardioversion were as follows: intravenous amiodarone, 417 (76%) of 550 patients; other antiarrhythmics, 324 (72%) of 451 patients; and placebo, 121 (60%) of 202 patients. Pooled estimates of cardioversion for active cohort studies were 72.1% for intravenous amiodarone and 71.9% for other antiarrhythmics (p=0.84). Pooled estimates of cardioversion for placebo cohort studies were 82.4% for intravenous amiodarone and 59.7% for placebo (p=0.03). Unadjusted averages for adverse event rates were intravenous amiodarone, 94 (17%) of 550 patients; other antiarrhythmics, 63 (14%) of 451 patients; and placebo, 23 (11%) of 202 patients. Pooled estimates of adverse event rates for active cohort studies were 12.2% for intravenous amiodarone and 14.0% for other antiarrhythmics (p=0.64). Pooled estimates of adverse event rates for placebo cohort studies were 26.8% for intravenous amiodarone and 10.8% for placebo (p=0.02). The most common adverse drug reactions reported with intravenous amiodarone were infusion phlebitis, bradycardia, and hypotension. The efficacy and safety profile of intravenous amiodarone is similar to that of other antiarrhythmics for cardioversion of recent-onset atrial fibrillation. Intravenous amiodarone is significantly more effective than placebo but is associated with significantly higher frequency of adverse events, although most were not considered to be dose limiting.