Fifteen RCTs (8,522 participants) were included: 10 (7,151 participants) placebo controlled (eight double blind and two single blind); and five (1,371 participants) open label inactive control studies. Study size varied between 34 and 2,521 participants. Follow-up ranged from 12 to 45 months. Twelve studies were considered good quality (scored 6 or more). Studies were published between 1987 and 2006.
There was no evidence of publication bias.
There was no evidence of heterogeneity for any analysis.
Compared to control, amiodarone significantly reduced the incidence of sudden cardiac death (7.2% versus 9.7%, OR 0.72, 95% CI 0.61 to 0.84, p<0.001) and cardiovascular death (14.0% versus 16.3%, OR 0.82, 95% CI 0.71 to 0.94, p=0.004). There was some reduction in all-cause mortality, but this did not reach statistical significance (18.1% versus 19.6%, p=0.093). There was no difference in deaths due to heart failure.
Sensitivity analysis results were similar when the largest trial was removed (see paper for details).
Subgroup analyses revealed no significant differences in results for sudden cardiac death, cardiovascular death or all-cause mortality (see paper for details).
End-organ toxicities were more common with amiodarone compared to control: pulmonary 2.9% versus 1.5% p=0.002; thyroid 3.6% versus 0.4%, p<0.001; hepatic 1.85% versus 0.70%, p=0.015. Bradyarrhythmias (2.8% versus 1.5%, p=0.008) and cancer deaths were also more common (0.7% versus 0.02%, p=0.0501; four trials).
The percentage of those assigned to amiodarone who discontinued treatment was 28.7. In trials that reported discontinuation in placebo groups, the rate was higher for amiodarone than placebo (31.6% versus 21.1%, p<0.0001; nine trials).