Amiodarone prophylaxis for atrial fibrillation after cardiac surgery: meta-analysis of dose response and timing of initiation.

STUDY OBJECTIVE To investigate a possible dose-response relationship between amiodarone and reduction in incidence of postoperative atrial fibrillation, and to determine whether pre- or postoperative initiation of amiodarone is superior. DESIGN Meta-analysis of randomized controlled trials. DATA SOURCE MEDLINE and EMBASE databases and the Cochrane Central Register of Controlled Trials for English-language reports published between 1966 and December 2005. MEASUREMENTS AND MAIN RESULTS Of 23 identified randomized controlled trials of amiodarone prophylaxis of postoperative atrial fibrillation, 14 were included in the final analysis. These studies enrolled a total of 2864 patients. For each study, the total administered amiodarone dose--categorized as low (< 3000 mg), medium (3000-5000 mg), or high (> 5000 mg)--and preoperative versus postoperative initiation were aggregated by using meta-analytic techniques. The incidence of postoperative atrial fibrillation was significantly reduced by amiodarone compared with placebo (p<0.001). Although the odds of developing atrial fibrillation appeared to be somewhat higher in the low-dose group, no significant differences were noted in the odds ratios (ORs) of developing atrial fibrillation among the low-, medium-, and high-dose groups: OR 0.58, 95% confidence interval (CI) 0.44-0.77; OR 0.45, 95% CI 0.30-0.69; and OR 0.44, 95% CI 0.33-0.58; respectively (p=0.238). In addition, the ORs for atrial fibrillation development associated with preoperative and postoperative initiation of amiodarone were not significantly different (OR 0.50, 95% CI 0.39-0.63; and OR 0.48, 95% CI 0.37-0.63; respectively, p=0.862). CONCLUSION Total amiodarone doses of 3000 mg or higher may be more effective than lower doses in reducing the rate of postoperative atrial fibrillation after cardiac surgery. Preoperative initiation of amiodarone appears to be unnecessary. These findings require confirmation in prospective, randomized trials.