Prophylactic ICDs with Coronary Artery Bypass Graft Surgery
ABSTRACT & COMMENTARY
Synopsis: There is no evidence that ICD therapy improves survival after coronary artery bypass surgery in patients with depressed left ventricular ejection fractions and abnormal signal averaged electrocardiograms.
Source: Bigger JT, et al. N Engl J Med 1997;337: 1569-1575.
The coronary artery bypass graft (cabg) patch Trial tested the hypothesis that prophylactic implantation of an implantable cardioverter defibrillator (ICD) in a high-risk group of patients improves long-term survival. The chosen study group consisted of patients with left ventricular (LV) ejection fractions < 36% who were referred for elective coronary artery bypass surgery. An abnormal signal averaged electrocardiogram was used as an arrhythmia qualifier. Patients with preexisting bundle branch block or who were paced were considered to be eligible. No patient had a history of a prior episode of sustained ventricular tachycardia or fibrillation. Thirty-seven clinical centers in the United States and Germany screened all patients undergoing bypass surgery and identified those who were eligible. After enrollment, patients underwent their clinically indicated CABG surgery.
While the patient was still on bypass after completion of the coronary anastomoses, the surgeon decided whether the patient was stable enough to undergo placement of an epicardial ICD.
Nine hundred out of 1055 enrolled patients were thought stable enough for implantation and were randomized at that time in the operating room to ICD implant or no implant. The ICDs used were epicardial systems, and defibrillation efficacy was confirmed in the operating room. The majority of the devices employed were committed therapy devices that did not store electrograms. Long-term follow-up was then continued using an intention-to-treat analysis.
The baseline characteristics of the patients were similar. The mean age was 64 years; the majority of patients were male (79%), and the mean LV ejection fraction was 27%.
There were 44 deaths in the first 30 days after randomization24 in the ICD group and 20 in the control group (P = NS). Cumulative mortality curves showed no long-term benefit of defibrillator therapy. During 32 ± 16 months of follow-up, there were 71 cardiac and 101 total deaths in the defibrillator group and 72 cardiac and 95 total deaths in the control group. There was no group identified by 10 pre-specified clinical characteristics that benefited from ICD therapy.
During follow-up, few patients in either group received class I or class III antiarrhythmic drugs. Beta blocker therapy was used in a similar percentage of patients in both groups. The only significant adverse effect associated with ICD therapy was a slightly increased infection rate, while the control group had slightly more myocardial infarctions during follow-up. Bigger and associates conclude that there is no evidence that ICD therapy improves survival after coronary artery bypass surgery in patients with depressed LV ejection fractions and abnormal signal-averaged electrocardiograms.
COMMENT BY JOHN P. DiMARCO, MD, PhD
The CABG Patch Trial is an interesting contrast to the previously discussed AVID Trial. Most patients with cardiac arrest do not survive their initial episode. In many areas, fewer than 5% of cardiac arrest victims survived to hospital discharge. Therefore, it is critical that a strategy for primary prevention of sudden cardiac death be developed if we are to decrease sudden death mortality in the general population. The CABG Patch Trial indicates that this may not be a simple task.
How could ICD implantation fail to favorably influence survival? In AVID and in the Multicenter Automatic Defibrillator Implantation Trial (MADIT. N Engl J Med 1996;335:1933-1940), ICD therapy was associated with 30% and 52% (respectively) reductions in mortality, compared to drug-treated controls.
The CABG Patch Trial population was a high-risk group with a three-year mortality of 20%, but several factors make this population different from those studied in the other two trials. The most obvious difference relates to the fact that all CABG Patch Trial patients underwent surgical revascularization prior to randomization.
Although many patients in AVID and MADIT had undergone CABG or percutaneous transluminal coronary angioplasty at some point, this may have been remote from their participation in the study. Since revascularization can favorably affect sudden death incidence, it is likely that the proportion of sudden deaths to non-sudden deaths may have been skewed toward the latter in these patients. In fact, more than 70% of the deaths in the CABG Patch Trial occurred in an in-hospital setting. All MADIT and AVID patients had inducible or spontaneous sustained ventricular arrhythmias, while CABG Patch Trial patients only had a positive signal-averaged ECG.
This requirement in AVID and MADIT possibly skewed the sudden to nonsudden death ratio toward the former. CABG Patch also did not have a linear mortality rate. Almost 30% of the cardiac mortality occurred within the first 30 days, of which, much time was spent in-hospital under conditions where an ICD would not be expected to have much benefit. In fact, it is likely that the stress of CABG surgery identified the highest risk patients in the control group, and appropriate drug or ICD treatment of this small but critical group of patients might have decreased the possibility of demonstrating any benefit from the ICD.
Based on the data from these trials, one must conclude that only demonstration of an induced or spontaneous arrhythmia now justifies ICD implant. We eagerly await data from other studies still in progress to see if this narrow indication can be effectively expanded. v
The results of the CABG Patch trial suggest:
a. an ICD is appropriate with documented symptomatic ventricular tachyarrhythmias.
b. the signal averaged ECG is a powerful predictor of mortality.
c. revascularization has little effect on ventricular tachyarrhythmias.
d. pre-operative low EF predicts sudden death mortality after CABG.
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