Current issues of ACP Journal Club are published in Annals of Internal Medicine


An implantable cardioverter-defibrillator was effective for cardiac arrest in postinfarction sudden death survivors

ACP J Club. 1996 Jan-Feb;124:8. doi:10.7326/ACPJC-1996-124-1-008

Source Citation

Wever EF, Hauer RN, van Capelle FJ, et al. Randomized study of implantable defibrillator as first-choice therapy versus conventional strategy in postinfarct sudden death survivors. Circulation. 1995 Apr 15;91:2195-203.



To compare the effectiveness of an implantable cardioverter-defibrillator (ICD) as first-choice therapy with that of the conventional strategy of starting with antiarrhythmic drugs in patients resuscitated from cardiac arrest > 4 weeks after myocardial infarction (MI).


Randomized controlled trial.


2 university hospitals in the Netherlands.


60 patients (mean age 58 y, 90% men) were studied. Inclusion criteria were age < 75 years, ≥ 1 successful resuscitation from cardiac arrest caused by ventricular fibrillation or rapid ventricular tachycardia without antiarrhythmic drug treatment, presenting arrhythmia related to previous MI with ≥ 4 weeks between the 2 events, study entry < 3 months from onset of arrhythmia, inducibility of ventricular tachyarrhythmia by programmed electrical stimulation, and amenability to ≥ 1 oral antiarrhythmic drug regimen. Exclusion criteria were frequent need for defibrillation during baseline, indication for surgery or angioplasty, unipolar pacing device, life expectancy < 1 year, or other serious illness. Median follow-up was 24 months.


Patients were stratified by presence of < 4 and ≥ 4 of 9 normokinetic segments at left ventricular cineangiography and allocated to ICD implantation (n = 29) or to conventional antiarrhythmic drug therapy (n = 31).

Main Outcome Measures

Primary end points were total mortality, prolonged syncope, and extreme pump failure requiring cardiac transplantation.

Main Results

Patients in the ICD and conventional therapy groups did not differ for total deaths (4 vs 11), prolonged syncope (0 vs 1), or cardiac transplantation (0 vs 1). When the primary end points were combined, patients in the ICD group had fewer total primary events than patients in the conventional therapy group (4 [14%] vs. 13 [42%] {95% CI for the 28% absolute risk reduction 6% to 48%, P < 0.02; relative risk reduction 67%, CI 17% to 88%; number needed to treat 4, CI 2 to 18}*; hazard ratio 0.27, CI 0.09 to 0.85, P = 0.02).


In patients resuscitated from cardiac arrest after myocardial infarction, cardioverter-defibrillator implantation as a first-choice therapy was more effective in reducing the number of main outcome events than conventional antiarrhythmic drugs.

Sources of funding: Dutch National Sick-Fund Council and CPI.

For article reprint: Dr. E.F. Wever, Department of Cardiology, Heart-Lung Institute, University Hospital Utrecht, 100 Heidelberglaan, PO Box 85500, 3508 GA Utrecht, the Netherlands. FAX 31-30-542-155.

*Numbers calculated from data in article.


In the evolving evidence of how patients with previous MI who are resuscitated from sudden death should be treated, this is the first randomized trial to attempt to clarify the approaches that were in effect as of 1989: ICD or trials with class I agents. Treatment outcomes were compared as part of a cost-effectiveness analysis. The results showed that compared with drug use guided by electrophysiologic study (EPS), early ICD use was associated with reductions in the combined main outcome of death, recurrent prolonged syncope, and cardiac transplantation.

This study has several limitations. First, the limited grant support allowed only 30 ICDs, necessitating the combined use of multiple end points to achieve statistically significant differences between the groups. The benefits are probably underestimated, however, as 52% of patients in the conventional therapy group were crossed over to ICD because drug therapy did not control arrhythmia.

Second, amiodarone was not used in lower doses in this study, although recent data suggest this drug may be more effective and associated with fewer spontaneous arrhythmias, deaths, and side effects than class I agents. There was also an underuse of angiotensin-converting enzyme inhibitors (used in only 47% of the patients, who had a mean ejection fraction of 30%) and an overuse of class I agents, which are now known to adversely affect patients with previous MI.

What is learned from this study is that early ICD use for patients resuscitated from sudden death > 4 weeks after MI is probably better than the routine use of EPS-guided class I antiarrhythmic agents. In his accompanying editorial, Zipes (1) points out that the need for randomized studies to investigate the correct use of ICD compared with class III (amiodarone) agents is currently being addressed in 3 studies: the Cardiac Arrest Study Hamburg (2), the Canadian Implantable Defibrillator Study (3), and the Antiarrhythmias Versus Implantable Defibrillators Study (4).

While we await the results of these investigations to establish the utility of a new technology, I believe that either amiodarone therapy or the judicious early use of an ICD is appropriate. I tend to favor the use of an ICD because, with the newer transvenous leads and smaller devices, the procedure is now less invasive and drugs can easily be added to the patient's regimen.

Robert J. Kornberg, MD
Wayne State UniversityDetroit, Michigan, USA


1. Zipes DP. Circulation. 1995;91:2115-7.

2. Siebels J, Kuck KH. Am Heart J. 1994;127:1139-44.

3. Connolly SJ, Gent M, Roberts RS, et al. Am J Cardiol. 1993;72:103F-8F.

4. AVID Investigators. Am J Cardiol. [In press].