Long-Term Comparison of ICD vs Amiodarone

Abstract & Commentary

Synopsis: The benefit of ICD therapy, compared to amiodarone therapy in patients with life-threatening arrhythmia, continues to increase over time, and long-term data support the use of an ICD as first line therapy for secondary prevention of sudden cardiac death.

Source: Bokhari F, et al. Long-Term Comparison of the Implantable Cardioverter Defibrillator vs Amiodarone: Eleven-Year Follow-Up of a Subset of Patients in the Canadian Implantable Defibrillator Study (CIDS). Circulation. 2004;110:112-116.

The Canadian Implantable Defibrillator Study (CIDS) was a randomized comparison of amiodarone and implantable cardioverter Defibrillators (ICD) as initial therapy in patients with documented, sustained ventricular arrhythmias or syncope with inducible ventricular tachycardia. Patients in the study had to have either survived a cardiac arrest with ventricular fibrillation (VF), had hemodynamically significant sustained ventricular tachycardia (VT), or syncope with left ventricular dysfunction and inducible VT. CIDS showed a benefit with ICD therapy that did not reach statistical significance. However, a meta-analysis of 3 trials, CIDS, the Antiarrhythmics vs Implantable Defibrillators Trial (AVID), and the Cardiac Arrest Study Hamburg (CASH) showed a significant reduction in death from any cause in the ICD group, with a summary hazard ratio of 0.72. In this paper, Bokhari and colleagues from St. Michael’s Hospital in Toronto, Ontario, Canada, report their single-center experience with longer follow-up of patients in CIDS. At the time the CIDS main trial results were published, Bokhari et al, for this report, with their institutional review board’s approval, elected to continue patients who had enrolled in CIDS at their hospital on their previously assigned therapy. The primary end point for this extended follow-up study was all cause mortality. Secondary end points were cause specific mortality, amiodarone related side effects, amiodarone discontinuation, and ventricular arrhythmia recurrence. In patients with an ICD, each stored arrhythmia episode was reviewed and classified as appropriate or inappropriate.

There were 120 patients entered at St. Michael’s Hospital in the CIDS Trial. The amiodarone group and the ICD group were well matched in terms of age, gender, ejection fraction, and other clinical parameters. Twenty-seven patients in the amiodarone group presented with VF, compared to 18 in the ICD group. In contrast, 35 patients presented with VT in the ICD group vs 23 in the amiodarone group. The mean total daily amiodarone dose was 398 ± 39 mg at 2 months and 306 ± 89 mg at last follow-up.

During follow-up, there were 28 deaths (47%) in the amiodarone group, compared with 16 deaths (27%) in the ICD group (P = 0.02)—a 43% lower risk of all cause mortality in the latter. There were only 2 presumed arrhythmic deaths in the ICD group, compared with 12 in the amiodarone group. Arrhythmic cardiac deaths, vascular deaths, and noncardiac deaths were almost evenly matched between the 2 groups.

By the end of a median follow-up 5.92 years, 40 patients assigned to amiodarone had either symptomatic nonfatal arrhythmia recurrence (n = 12) or had died. Of note, 25 of the 28 deaths were not preceded by a nonfatal symptomatic arrhythmia recurrence.

Side effects thought to be related to amiodarone were reported in 49 of 60 patients (82%). Of these, 30 patients had side effects requiring dose reduction or discontinuation. In 13 of these patients, the adverse events were considered to be serious.

Among the amiodarone patients, there were 19 patients who crossed over to ICD therapy because of either adverse effects which required drug discontinuation or a nonfatal arrhythmia recurrence. In the ICD group, there were 3 pocket infections, 18 lead failures or dislodgements, 1 pneumothorax, 1 deep vein thrombosis, and 1 pocket hematoma. Amiodarone was added to ICD therapy in 26 patients in the ICD group to decrease the frequency of atrial or ventricular arrhythmias. Appropriate ICD therapy was observed during follow-up in 70% of the ICD therapy group. One or more episodes of inappropriate therapy delivery was noted in 30 patients. By using a Cox proportional hazards model, 2 variables were identified as significant independent predictors of survival, ICD therapy and the absence of coronary artery disease. Bokhari et al conclude that the benefit of ICD therapy, compared to amiodarone therapy in patients with life-threatening arrhythmia, continues to increase over time, and that their long-term data support the use of an ICD as first line therapy for secondary prevention of sudden cardiac death.

Comment by John P. DiMarco, MD, PhD

The 3 major secondary prevention trials that compared ICD’s to antiarrhythmic drugs for secondary prevention of sudden cardiac death (AVID, CIDS and CASH) included relatively few patients who were followed for more than 4 years. In this paper, from a single center that enrolled in CIDS, Bokhari et al elected to keep patients in their assigned group after publication of the main trial results. They observed an increased incidence of serious adverse effects using amiodarone, and also documented a relatively high frequency of complications related to transvenous defibrillators. Another publication from CIDS, which previously showed that the benefits of the initial study, was largely seen among the oldest patients and those with the lowest ejection fractions and the poorest functional status. This led to a proposal that the ICD be the first option only in those with depressed ventricular function, with other patients selected on a case-by-case basis. The data presented here suggest that, over time, the ICD will prove superior for most, but the group is not large enough to test if outcomes for those with preserved ventricular function will be satisfactory.

We should learn several lessons from this report. The data from all the secondary prevention trials show us that neither drug therapy nor ICD therapy is the perfect option for all patients. Drug therapy was frequently limited by side effects, some of which may be dose related, and imperfect efficacy. ICD therapy will be poorly tolerated if frequent events occur. Many patients are best treated with a combined approach. The ICD is the primary treatment, and antiarrhythmics are used to modify the frequency of ventricular, and also supraventricular, arrhythmias. This is illustrated by this paper, in which 19 of the 60 patients in the amiodarone group eventually crossed over to ICD therapy and almost half of the patients (26) in the ICD group eventually had amiodarone added. Patients with rare arrhythmia episodes can be managed just with ICDs. Patients with frequent arrhythmias should have drugs added. The drugs should not be used in high doses, which will increase the frequency of side effects. Rather, the minimal dosage needed to make ICD therapy tolerated should be employed. Combining drugs and devices, in a balanced approach, should lead to the best outcomes.

Dr. DiMarco, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville, is on the Editorial Board of Clinical Cardiology Alert.