Forty publications were included in the review (approximately 14,413 patients). The quality of the included trials was variable; many trials did not clearly describe blinding or randomisation and some trials had high levels of drop-outs. Trial sample size ranged from 16 to 4,628 patients.
Direct analysis: Only results with comparisons of two or more trials are reported here; other results were presented in the review.
There was no statistically significant difference between any of the drugs in terms of mortality. In terms of stroke, there was a statistically significantly lower risk with dronedarone compared with placebo (OR 0.69, 95% CI 0.47 to 0.99; two trials). For atrial fibrillation recurrences, amiodarone had statistically significantly lower rates compared with sotalol (OR 0.47, 95% CI 0.36 to 0.62; five trials). Compared with placebo, all of the drugs studied had statistically significantly greater withdrawals due to adverse events and proarrhthmia events. Amiodarone and flecainide also had a statistically significantly greater risk of serious adverse events compared with placebo. Further results were presented in the review.
Mixed-treatment analysis: Sotalol had a statistically significantly higher rate of mortality compared with placebo (OR 3.44, 95% CI 1.02 to 11.59). In trials with more than 100 patients, sotalol had a statistically significantly increased risk of mortality (OR 4.32, 95% CI 1.59 to 11.70) and amiodarone had a marginally statistically significantly increased risk of mortality (OR 2.73, 95% CI 1.00 to 7.41) . Dronedarone had a statistically significantly lower risk of stroke compared with placebo (OR 0.69, 95% CI 0.57 to 0.84). All of the drugs studied had statistically significantly reduced atrial fibrillation recurrences compared with placebo. Compared with placebo, there were increased rates of proarrhythmia events with sotalol (Or 6.44, 95% CI 1.03 to 40.24), dronedarone (OR 1.45, 95% CI 1.02 to 2.08) and propafenone (OR 4.06, 95% CI 1.13 to 14.52). There was no significant difference in treatment discontinuations or serious adverse events, but dronedarone, amiodarone and sotalol had statistically significantly greater withdrawals due to adverse events.
There was significant heterogeneity detected in several of the mixed treatment analyses.