Treatment of atrial fibrillation with antiarrhythmic drugs or radiofrequency ablation: two systematic literature reviews and meta-analyses.

BACKGROUND: Although radiofrequency catheter ablation (RFA) has evolved from an experimental procedure to an important treatment option for atrial fibrillation, the relative safety and efficacy of catheter ablation relative to that of antiarrhythmic drug (AAD) therapy has not been established.

METHODS AND RESULTS: Two separate systematic reviews were conducted: one on RFA and the other on AAD to provide accurate and broadly representative estimates of the clinical efficacy and safety of both therapies in the treatment of atrial fibrillation. Electronic searches were conducted in EMBASE and MEDLINE from 1990 to 2007. For the RFA review, all study designs were accepted. For the AAD review, articles were limited to prospective studies on the following drugs of interest: amiodarone, dofetilide, sotalol, flecainide, and propafenone. Data were extracted by 1 reviewer, with a second reviewer performing independent confirmation of extracted data. Sixty-three RFA and 34 AAD studies were included in the reviews. Patients enrolled in RFA studies tended to be younger (mean age, 55 versus 62 years), had longer duration of atrial fibrillation (6.0 versus 3.1 years), and had failed a greater number of prior drug trials (2.6 versus 1.7). The single-procedure success rate of ablation off AAD therapy was 57% (95% CI, 50% to 64%), the multiple procedure success rate off AAD was 71% (95% CI, 65% to 77%), and the multiple procedure success rate on AAD or with unknown AAD usage was 77% (95% CI, 73% to 81%). In comparison, the success rate for AAD therapy was 52% (95% CI, 47% to 57%). A major complication of catheter ablation occurred in 4.9% of patients. Adverse events for AAD studies, although more common (30% versus 5%), were less severe.

CONCLUSIONS: Studies of RFA for treatment of atrial fibrillation report higher efficacy rates than do studies of AAD therapy and a lower rate of complications.