OBJECTIVES: To describe the methodology of a systematic review and meta-analysis evaluating the efficacy and safety of anti-arrhythmic drugs (AADs) in the treatment of atrial fibrillation (AF). Outcomes of interest included AF recurrence, cardiovascular (CV) morbidity, all-cause mortality, discontinuations, serious adverse events (SAE), persistence/compliance and health related quality of life (HRQoL). METHODS: Electronic databases (Cochrane Library, Medline, EMBASE; accessed March 2009) and manual bibliographic searches were conducted to identify relevant randomised clinical trials (RCTs). Comparators of interest included all AADs, rate and rhythm strategies or ablation in comparison with AADs. In addition to dronedarone, the primary AADs of interest were restricted to Class IC (flecainide and propafenone) and Class III (amiodarone and sotalol). Relevant data were extracted by two independent reviewers. Data were analysed on an intention-to-treat basis and meta-analysis performed using the Peto odds ratio (OR)/fixed-effect model. The Bucher method was employed for indirect comparisons. Direct comparisons and/or indirect comparisons via non-active control were conducted where appropriate, in relation to the outcomes of interest. RESULTS: In total, 145 separate publications met the pre-defined inclusion criteria and were included in the systematic review. Of these, 71 were related to one of the five AADs of primary interest and were analysed. SAEs were reported in 24 publications, AF recurrence in 44, all-cause mortality in 52 and discontinuations in 62. Data relating to other morbidity outcomes such as hospitalisations or to persistence/compliance and HRQoL were very limited; therefore, meta-analyses were not possible. These findings are consistent with results of a previous Cochrane review (2007). CONCLUSIONS: Based on the results of this current systematic review, limited meta-analyses were possible. The majority of studies were designed to assess AF recurrence only; therefore, few publications assessed relevant clinical markers of AF progression or complications such as CV morbidity.