The trial was designed to evaluate the effect of prophylactic ICD therapy on survival in patients with prior MI and LV dysfunction. The primary endpoint was death from any cause.
The trial began on July 11, 1997, and enrolled patients from 76 hospital centers (71 in the United States and 5 in Europe and Israel). The study's principal investigator was Arthur J. Moss, MD, professor of medicine at the University of Rochester Medical Center in Rochester, New York. Eligible study participants included both male and female patients over 21 years of age that had experienced an MI at least 30 days prior to study enrollment and demonstrated an ejection fraction (EF) <30%. Some of the exclusion criteria were prior cardiac arrest, syncope unrelated to an acute MI, NYHA Class IV, CABG or non-CABG revascularization less than 3 months prior to enrollment, and use of antiarrhythmic agents, except for atrial arrhythmias comorbidity associated with decreased survival for trial duration (Table 1).
Table 1. Inclusion / Exclusion Criteria
Inclusion Criteria |
Exclusion Criteria |
||
|
|
||
A total of 1232 patients were randomized into two treatment arms in a 3:2 ratio. Patients in the first arm were treated with conventional drug treatment and also implanted with an ICD. The second group was treated exclusively with conventional drug treatment, as prescribed by the patient's attending physician (control arm). Further, the 3:2 ratio resulted in 742 patients receiving an ICD and 490 patients receiving drug therapy only (Figure 1).
Figure 1. Randomization Cascade1
Patients were seen in the follow-up clinic at 1 month after randomization, and then at 3-month intervals until the termination date of the study or death.
A triangular sequential design that was modified for two-sided alternatives was used with preset boundaries. Termination of the trial was determined by whether there was efficacy, inefficacy, or no difference in mortality between the two treatment groups. The efficacy boundary of the sequential design was reached when 197 deaths were reported. The trial was officially stopped 1 week later, on November 20, 2001.
The log-rank statistic is a measure of the cumulative differences in survival between the two treatments (Figure 2). A positive value for the log-rank statistic indicates that the treatment with the defibrillator was superior to conventional medical therapy. The variance of the log-rank statistic is closely related to the number of deaths. The values of the log-rank statistic continue to increase after the termination of the trial (P=0.016) as a result of a lag in reporting additional survival data and five additional deaths that occurred before the stopping date but were uncovered during the closeout procedure on January 16, 2002.
Figure 2. Sequential Monitoring in the Triangular Design2
(Copyright 2002 Massachusetts Medical Society. All rights reserved.)
Table 2 summarizes some general characteristics of the patients enrolled in the study. The characteristics were balanced across the therapy groups, manifested by no statistical differences and indicating demographically and clinically similar groups.
Table 2. Characteristics of the Study Population2
ICD Patients |
Conventional Therapy | ||
Age (yr) |
64 ± 10 |
65 ± 10 |
|
Gender: male |
84% |
85% |
|
History of CABG |
58% |
56% |
|
Ejection fraction |
23% ± 5% |
23% ± 6% |
|
NYHA CHF Class > II |
65% |
61% |
|
As shown in Table 3, there were no differences in the use of ACE inhibitors, beta blockers, or digitalis therapy between the ICD therapy patients and the conventional therapy patients as reported at last follow-up visit.
Table 3. Medications at Last Contact (%)2
ICD Patients |
Conventional Therapy | ||
ACE Inhibitors |
68 |
72 |
|
Amiodarone |
13 |
10 |
|
Antiarrhythmics (Class I) |
3 |
2 |
|
Beta-blockers |
70 |
70 |
|
Digitalis |
57 |
57 |
|
During an average follow-up of 20 months, the mortality rates in the conventional and defibrillator groups were 19.8% and 14.2%, respectively. The hazard ratio for overall mortality (defibrillator: conventional treatment) was 0.69 (95% confidence interval, 51% to 93%, P=0.016), indicating a 31% reduction in the risk of death at any interval among patients in the defibrillator group as compared with patients in the conventional-therapy group (Table 4).
Table 4. Mortality2
Conventional Group |
ICD Group |
Hazard Ratio |
||
97 (19.8%) |
105 (14.2%) |
0.69 (0.51-0.93) |
0.016 |
|
Figure 3 shows the Kaplan-Meier life-table cumulative survival curves for the two treatment groups. The two survival curves begin to separate at approximately 9 months and separate progressively thereafter. These survival curves represent a decrease in death rates in the defibrillator group (95% confidence limits; P-value) of 12% at 1 year (27 to 40%), 28% at 2 years (4 to 46%), and 29% at 3 years (5 to 46%).
Figure 3. Kaplan-Meier Estimates of the Probability of Survival in
the Group Assigned to Receive an ICD and the Group Assigned
to Receive Conventional Medical Therapy2
(Copyright 2002 Massachusetts Medical Society. All rights reserved.)