Thirteen studies (6,553 participants) were included: 8 studies (5,101 participants) of patients who had recently had a MI, and 5 studies (1,452 participants) of patients with CHF. Individual patient data were not available for two of these (small) studies; summary data from the original reports was used for these.
Total mortality was reduced by 13% (OR 0.87, 95% CI: 0.78, 0.99, p=0.03) when using the fixed-effect analysis, and by 15% (OR 0.85, 95% CI: 0.71, 1.02, p=0.081) when using the random-effects approach; both were in favour of amiodarone. Amiodarone reduced arrhythmic/sudden death by 29% when using the fixed-effect approach (OR 0.71, 95% CI: 0.59, 0.85, p=0.0003), or by 31% when using the random-effects approach (OR 0.69, 95% CI: 0.55, 0.87, p=0.0016). There was evidence of heterogeneity for this outcome (p=0.058), which the authors attributed to two small studies and one large study that used high-dose intravenous amiodarone in the immediate acute MI period. Following the exclusion of these studies there was no evidence of heterogeneity (p=0.093), but no summary OR was presented. There was no effect of amiodarone treatment on non-arrhythmic/non-sudden deaths (OR 1.02, 95% CI: 0.87, 1.19, p=0.84).
The subgroup analyses showed there was no difference in treatment effect between post-MI and CHF studies. The effectiveness of amiodarone was not influenced by left-ventricular ejection fraction, New York Heart Association class, or the presence or absence of asymptomatic ventricular arrhythmias on the Holter electrocardiogram.
The adverse effects associated with amiodarone use included hypothyroidism (net absolute difference 5.9%; p=0.00005 for OR), hyperthyroidism (net absolute difference 0.9% p=0.0043 for OR). Peripheral neuropathy (p=0.071), lung infiltrates (p=0.0003), bradycardia (p=0.0003) and liver dysfunction (p=0.0072) were all more common in the amiodarone groups. In the double-blind placebo trials, 41% of the amiodarone-assigned participants and 27% of controls permanently discontinued study medication (14% difference). This was primarily related to the adverse experiences associated with amiodarone.