Evidence Based Review Article

The Journal of Informed Pharmacotherapy 2001;5:220-222.

Oral Amiodarone for Prevention of Atrial Fibrillation after Open Heart Surgery: Does a Fist Beat Scissors?

Download a PDF version of this article

Reviewer: Richard S. Slavik, B.Sc.(Pharm.), Pharm.D.
Reviewer's email address:
Reviewer's profession/specialty:
 Pharmacotherapeutic Specialist/Critical Care

Original Citation

Giri S, White CM, Dunn AB, Felton K, Freeman-Bosco L, Reddy P, Tsikouris JP, Wilcox HA, Kluger J. Oral amiodarone for prevention of atrial fibrillation after open heart surgery, the Atrial Fibrillation Suppression Trial (AFIST): a randomised placebo-controlled trial.  Lancet 2001;357:830-6. PubMed Cit

Overall Study Question

The AFIST trial was a prospective, randomized, placebo-controlled study undertaken in a 600-bed tertiary care center to assess the safety and efficacy of a brief oral amiodarone regimen for prevention of post cardiac surgery atrial fibrillation (AF) in patients 60 years and older already receiving beta

-blockers. Patients were included if they were at least 60 years old, in normal sinus rhythm (NSR), hemodynamically stable (SBP >90 mm Hg), had a normal baseline QTc < 440 msec, and were available at least 1 day prior to scheduled cardiopulmonary bypass surgery. Patients were excluded if they had chronic AF, myocardial infarction within the past 3 weeks, HR <45 bpm, advanced heart block, automatic implantable defibrillator, a history of amiodarone toxicity, untreated thyroid disease, hepatic transaminases greater than four times normal, and interacting drugs. Patients were randomized to receive either a slow or fast oral loading regimen of amiodarone or matching placebo prior to surgery and continued for four days post-op. The primary outcome was the incidence of any post-operative AF lasting longer than 5 minutes during the 30-day follow-up.

Are the Results of the Study Valid?

1. Was assignment of patients randomized?

Yes. During a 14-month period, 1


430 patients undergoing angiography for ischemic heart or valvular disease were screened using consecutive sampling. A computer-generated randomization table was used to assign patients to either a slow load or fast load group with each group stratified by presence or absence of valvular surgery. The slow load group (>=

5 days before surgery) received oral amiodarone 200 mg daily for 5 days before surgery, 400 mg twice daily on the day of surgery, and 400 mg twice daily on post-op days 1-4 or matching placebo. The fast load group (<5 but >1 day before surgery) received 400 mg four times daily for 1 day, 600 mg twice daily on the day of surgery, and 400 mg twice daily on post-op days 1-4 or matching placebo.

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Yes. One thousand four hundred and thirty patients were screened and 379 patients met the study criteria. Reasons for ineligibility were provided. Of the 379 eligible patients, 220 were randomized – 120 in the amiodarone group and 100 in the placebo group. The primary outcome was analyzed using the intention to treat principle and no patients were lost to follow-up. Drug withdrawal in the amiodarone and placebo groups was 6.7% and 5%, and overall in-hospital mortality and 30-day mortality rates were 2.3% and 3.6%.

3. Were patients, their clinicians, and study personnel 'blind' to treatment?

Yes. The randomization schedule was sealed in opaque envelopes and kept in the central pharmacy. Only the investigational pharmacist and the unmarked study administrator had knowledge of treatment group allocation, and neither had patient contact nor extracted data. All treatment regimens were double-blinded and an identical matching placebo was used.

4. Were the groups similar at the start of the trial?

Yes. Overall, patients in the experimental and control groups had similar baseline demographic and clinical characteristics, however more patients in the amiodarone group had had a previous MI.

5. Aside from the experimental intervention, were the groups treated equally?

Yes. All surgical and medical procedures were standardized using a predefined critical pathway. The surgical intervention for each group was similar except that patients in the amiodarone group had a slower heart rate during surgery and required fewer intra-operative defibrillations to restore NSR after cross-clamp release. The only co-intervention explicitly stated that was left to the discretion of physicians was the antiarrhythmic treatment of post-op AF.

6. Overall, are the results of the study valid?


What were the Results?

1. How large was the treatment effect? 

The overall risk any AF in the 30-d post op period was 23% in the amiodarone group and 38% in the placebo group (p=0.01), ARR = 15.5% (3.4%-27.6%), NNT = 7 (4-30). Secondary analyses revealed the rate of symptomatic AF was 4% in the amiodarone group and 18% in the placebo group (p=0.001), ARR = 13.8% (5.5%-22.2%), NNT = 8 (5-19). Recurrent AF was 9% in the amiodarone group and 20% in the placebo group (p=0.02), ARR = 10.9% (1.5%-20.2%), NNT = 10 (5-69). The rate of ventricular tachycardia (>30 seconds) and cerebrovascular accidents (CVA) was 1.7% in the amiodarone group and 7.0% in the placebo group (p=0.04), ARR = 5.3% (0.2%-10.8%), NNT = 19 (9-556). 

There were no differences in withdrawal rates due to adverse effects in either group, and there were no differences in bradycardia, heart block, need for pacing, hypotension, or tolerability of upward titration of post-op BB therapy. There was a trend towards more nausea in the amiodarone group (27% vs. 16%, p=0.056). There were no differences in percentage of patients in NSR at discharge, duration of ICU stay, hospital stay, in-hospital mortality, 30-day mortality, or treatment costs between the groups.

2. How precise was the estimate of the treatment effect?

For the primary outcome of any atrial fibrillation and secondary outcomes of symptomatic AF and recurrent AF, the ARR did not cross zero, and the confidence intervals around the point estimate were narrow and represent a clinically significant result. For the analyses of ventricular tachycardia and CVA, the confidence intervals also did not cross zero, however the intervals were wide meaning the clinical significance can be questioned.

Will the Results Help Me in Caring for My Patients?

1. Can the results be applied to my patient care?

Yes. Based on the AFIST trial, pre-operative oral amiodarone begun at least 1 day pre-op and continued for 4 days post-op can reduce any AF, symptomatic AF, and recurrent AF in patients >60 years old (mean 73 years

) receiving pre-operative (mean = 89%)


continued post-operatively. 

This is the first trial comparing the effects of administering an antiarrhythmic drug pre-operatively to eligible patients already receiving the gold standard of pre-op/post-op BB prophylaxis. 

These data should not be extrapolated to younger patients not receiving pre/post-op beta-blockers

as the underlying event rate of atrial fibrillation in these groups would be different.

2. Were all clinically important outcomes considered?

Yes. Although this trial was underpowered to examine the effect of adding amiodarone to the subgroup of patients who all received beta-blockers

, it did determine the effect of adding amiodarone to a group of elderly post cardiac surgery patients in whom pre-op/post-op beta blocker prophylaxis

was attempted. Along with the surrogate endpoint of development of post-op AF, other valuable morbidity, mortality, and cost endpoints were evaluated in secondary analyses. Although adverse drug reactions were prospectively evaluated and surveyed, the trial was underpowered to detect rare, but clinically important adverse effects such as acute lung injury and ARDS.

3. Are the likely treatment benefits worth the potential harms and costs?

Yes. The benefits of reduction of any AF, symptomatic AF, and recurrent AF are clinically important in this population, however a larger trial would be required to clarify the ultimate impact of AF prevention on important benefits such as incidence of CVA, length of stay, and economic outcomes. Careful long-term follow up of post cardiac surgery patients

receiving amiodarone will still be warranted to monitor for rare pulmonary toxicities.


AF is a common complication of cardiac surgery and is associated with a significant impact on symptoms, morbidity, and health care costs. 

Pre-operative and/or post-operative beta blocker

therapy is the gold standard for prevention of post cardiac surgery AF, however, antiarrhythmic agents such as amiodarone and sotalol have also been studied for prevention. 

Conflicting results from studies on the efficacy of amiodarone prophylaxis can be explained by differences in study populations, prophylactic regimens, underutilization of proven beta blocker

prophylaxis, and a falsely elevated AF rates due to post-op

beta blocker withdrawal in some patients. 

The AFIST trial is the first trial specifically designed to evaluate the effectiveness of prophylactic amiodarone added to gold standard BB therapy in an elderly population, and reports significant reductions in post cardiac surgery AF. These benefits are both statistically significant and clinically important. Larger trials will be needed to clarify the impact of this surrogate outcome on more important morbidity and economic outcomes such as incidence of CVA and cost effectiveness of amiodarone prophylaxis. Judicious monitoring of patients receiving any amiodarone prophylaxis regimen should continue due to the potential for rare pulmonary toxicities.

Copyright © 2001 by the Journal of Informed Pharmacotherapy. All rights reserved.