The antiarrhythmics vs. implantable defibrillators (AVID) study was designed to test the hypothesis that there was no difference in overall mortality among survivors of cardiac arrest or sustained ventricular tachycardia using strategies involving either an implantable cardioverter defibrillator (ICD) or antiarrhythmic drug therapy. Fifty-six clinical centers participated in the trial. Patients who presented with ventricular tachycardia (VT) or ventricular fibrillation (VF) were screened at each of the centers. Patients who had been resuscitated from the following three types of arrhythmias were eligible for enrollment: VF, sustained VT with syncope or sustained VT with a left ventricular ejection fraction of 40% or less, and hypotension.
One thousand sixteen patients were enrolled in the study and were randomly assigned to treatment with an ICD or antiarrhythmic drugs. The clinical characteristics of the group were similar to those for other published series of cardiac arrest or VT survivors. The mean age was 65 years; the group was predominantly male (79%), and 86% were caucasian. VF was the presenting rhythm in 455 patients, and 561 patients presented with sustained VT with syncope or hemodynamic compromise. The mean left ventricular ejection fraction was 32 ± 13%. There was a slight imbalance in the ejection fraction among patients who presented with VF0.36 ± 0.15 in the ICD group vs. 0.33 ± 0.15 in the antiarrhythmic drug group.
Five hundred seven patients were assigned to ICD therapy. A nonthoracotomy lead system was used in 93%, 5% received an epicardial system, and 2% did not receive an implant. Among the 509 patients assigned to antiarrhythmic drug therapy, 356 were started immediately on amiodarone.
The initial antiarrhythmic drug was a random selection between amiodarone and sotalol in 153 patients. However, sotalol therapy had to be guided by either electrophysiologic testing or ambulatory monitoring, and only 13 of the 74 patients assigned to sotalol actually had suppression on the drug. The daily maintenance dose of amiodarone was 389 + 112 mg at three months. Compliance with therapy was good during follow-up. More than 85% of the patients begun on amiodarone continued on drug therapy.
A number of complications were reported on the two study groups. Only one patient developed nonfatal polymorphic VT, which was thought to be due to amiodarone. Pulmonary toxicity due to amiodarone was suspected in 3% of the patients at one year and in 5% of the patients at two years. There was one death due to pulmonary toxicity. Thyroid replacement medication was required in 10% of the amiodarone-treated patients at one year and in 16% of the amiodarone-treated patients at two years.
Significant complications were also noted in the ICD group. Bleeding or hematoma after implant was noted in 19 of the 507 patients. Other complications included infection (10 patients), pneumothorax (8 patients), and cardiac perforation (1 patient). There was no difference in mortality in the first 30 days after therapy assignment. Twelve patients in the defibrillator group (2.4%) died within 30 days of initiation of therapy, as compared to 18 patients (3.5%) in the drug-treated group.
Mortality was lower in the patients treated with the ICD. Over a mean follow-up of 18.2 ± 12.2 months, unadjusted survival rates were 89.3% vs. 82.3%, 81.6% vs. 74.7%, and 75.4% vs. 64.1% at one, two, and three years between the ICD and drug therapy groups, respectively. This represents approximately a 30% decrease in death rate through the course of the study. Automatic pacing or shocks from the ICD were common in patients in the ICD group.
Among the patients with VT, 36% received some form of therapy at three months, 68% at one year, 81% at two years, and 85% at three years. Among the patients with VF, 15% received therapy by three months, 39% by one year, 53% by two years, and 69% by three years. Cross-over between therapies was relatively infrequent. By two years, approximately 25% of the patients who received a defibrillator had had antiarrhythmic drug therapy added to their regimen, whereas 18.9% of the patients assigned to drug therapy had had an ICD implanted. Both groups had frequent need for re-hospitalization. By one year, 60% of patients in the ICD group and 56% of the drug therapy group had been re-hospitalized. A number of pre-specified subgroup analyses were performed. The hazard ratio in favor of the ICD was constant with age, left ventricular ejection fraction, type of heart disease, and presenting arrhythmia.
The investigators conclude that ICD therapy is superior to antiarrhythmic drug therapy for increasing overall survival among survivors of VF or hemodynamically significant sustained VT. (AVID Investigators. N Engl J Med 1997;337:1576-1583.)
COMMENT BY JOHN P. DiMARCO, MD, PhD
Since the introduction of the implantable defibrillator into clinical practice in 1980, ICD therapy has gradually assumed a larger and larger role in the management of patients with life-threatening arrhythmias. From the time of its introduction, there has been virtually no question that an ICD could effectively terminate individual episodes of sustained VT or VF. However, questions about the effect of an ICD on total mortality and long-term patient tolerance still remained. In the last five years, there have been significant changes in ICD hardware that have made them easier and safer to implant and enhanced their diagnostic and therapeutic functions. Therefore, the AVID trial was planned as a study to compare ICD therapy to what is generally considered to be the best available antiarrhythmic drug strategiesempiric amiodarone or guided sotalol therapy.
The data presented in this paper clearly show that an ICD should be the first consideration in survivors of serious sustained ventricular arrhythmias. ICD therapy was associated with approximately a 30% decrease in mortality, and this mortality benefit was maintained throughout the relatively short duration of the study.
Although ICD therapy was clearly superior to antiarrhythmic drug therapy, the paper also suggests that drug therapy will continue to have a significant role in managing arrhythmias. Even among the ICD recipients, 25% of the patients eventually received additional drug therapy. The study also notes that many of the patients who were treated with an ICD received therapy. Although not all arrhythmias treated with an ICD will be sustained or produce symptoms, one might say that much of the difference between the frequency of defibrillator therapy in the ICD group and events in the drug group is actually a measure of drug efficacy. In this respect, amiodarone looks to be a reasonably effective agent.
In some sense, the AVID trial was an artificial construct. Antiarrhythmic drugs and ICDs should not be considered as mutually exclusive forms of therapy. Based on the data presented in AVID, the first consideration should be whether the patient is a candidate for and is willing to accept an ICD implant. If so, the ICD should be inserted. There will remain many patients who would benefit from antiarrhythmic drug therapy because of either symptomatic (but nonsustained) ventricular ectopy, atrial arrhythmias, or frequent ICD therapies. In patients who are uncomfortable with the prospect of an ICD, antiarrhythmic drug therapy still provides substantial benefit, and many patients will do well.
Survivors of arrhythmic sudden death exhibit increased survival with:
a. amiodarone.
b. implantable defibrillator.
c. sotalol.
d. anti-tachycardia pacing.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content