It is now recognized that in absence of a reversible cause of sudden cardiac death (e.g., acute ischemia causing ventricular tachycardia [VT] or ventricular fibrillation [VF]), the risk of SCD recurrence is high. Three important randomized, controlled clinical trials have demonstrated that when compared
with antiarrhythmic therapy, ICD therapy is associated with a decrease in the incidence of recurrent SCD (Table 21-2) (4,5,6).

In the Antiarrhythmic Versus Implantable Defibrillator (AVID) trial (4), mortality after 2 years was 18.4% in the ICD group, and 25.3% in the antiarrhythmic group, yielding an approximately 30% relative decrease in all-cause mortality, which persisted through 3 years of follow-up. Most of the patients in the AVID antiarrhythmic group received amiodarone. These results were confirmed in the Canadian Implantable Defibrillator (CIDS) trial (5), in which patients were randomly assigned to receive amiodarone or an ICD. A 20% relative risk reduction in all-cause mortality over 3 years of follow-up was observed, although it did not reach statistical significance. Critique of AVID and CIDS included the more frequent use ofβ-blockers in the ICD arms of both trials, which may have biased the findings toward device therapy. An 11-year follow-up from a single-center subset of CIDS demonstrated that the benefit of ICD therapy over amiodarone increases over time (7). Additionally, the majority of patients treated with amiodarone (82%) developed significant adverse effects from the medication, resulting in discontinuation in 50% of the patients. The third trial in the series was the Cardiac Arrest Study Hamburg (CASH) (6), in which patients were randomly assigned to receive a β-blocker metoprolol, amiodarone, or ICD. In this trial, ICD was again superior to antiarrhythmic therapy (both metoprolol and amiodarone), with an approximately 30% reduction in SCD mortality in the device group.

Meta-analysis of AVID, CIDS, and CASH (8) was published in 2005 and demonstrated that among survivors of SCD, defibrillator therapy was associated with a 28% relative risk reduction of all-cause mortality, which was almost entirely to a 50% reduction in the risk of recurrent SCD. Over the follow-up period of 6 years, ICD therapy prolonged patient survival by 4.4 months compared to antiarrhythmic therapy.

Based on the findings of these randomized trials, ICD therapy is a Class I recommendation (beneficial, useful, and effective treatment) for patients who suffered cardiac arrest from VT or VF, or had unstable/high risk VT (Table 21-3).

Survival from an out-of-hospital cardiac arrest remains low despite a significant investment in such effective public health measures as automated external defibrillators (AEDs) (9). Therefore, identification of patients at high risk for a first episode of SCD has been and remains an important scientific and public health priority. In an attempt to define such a population, randomized clinical trials enrolled progressively broader groups of patients and tested ICD therapy as means of primary prevention of SCD (Table 21-4).

Coronary disease in combination with poor ventricular function is thought to portend an especially high risk of malignant ventricular arrhythmias. The first trial to demonstrate benefit of ICD over antiarrhythmic therapy was the small Multicenter Automatic Defibrillator Implantation Trial (MADIT) (10), which enrolled patients with a documented history of MI, ejection fraction (EF) ≤35%, and nonsustained ventricular tachycardia (NSVT). Patients underwent electrophysiological (EP) study, and if VT could be induced but not suppressed, they were randomized to antiarrhythmic therapy or ICD. Over the course of follow-up, patients in the ICD group benefitted from a 54% reduction in mortality compared to placebo. Based on the results of MADIT (and despite the small sample size of the study, as well as the more frequent use of β-blockers in the ICD group), ICD therapy was approved for this indication. A much larger Multicenter Unsustained Tachycardia Trial (MUSTT) (11) was a study comparing a medical approach with an electrophysiology-guided approach for primary prevention of SCD. Patients were enrolled if they had a history of MI, EF ≤40%, and NSVT. Though MUSTT was not designed as a “device trial,” its protocol specified that a subset of patients in the EP-guided group have an ICD implanted. Comparison of this subpopulation with matched antiarrhythmic controls demonstrated a marked benefit to ICD therapy (a 76% decrease in the incidence of SCD), indirectly supporting the results of MADIT. In comparison with MADIT-I and MUSTT, the second MADIT trial (MADIT-II) markedly widened its enrollment criteria (12). It enrolled patients with a history of MI and EF ≤30%. No documentation of NSVT, or proof of VT inducibility in the EP lab was required. Over 1,200 patients were randomized to ICD or conventional therapy, and 31% mortality reduction with ICD therapy compared to control was observed beginning at 9 months after implantation (14.2% vs. 19.8%, P = 0.016). The study was stopped prematurely because of the marked benefit in the ICD group.

How early after the first determination of poor LV function is made should the ICD be implanted? Indeed, do patients with myocardial stunning and low LV EF early after MI represent appropriate candidates for ICD therapy? The DINAMIT trial (13) answered this question in a population of 674 patients with a recent (<40 days) MI and EF ≤35%. Desire to identify an even higher-risk group among this population led to randomization

only of those patients who demonstrated abnormal heart rate variability—a marker of autonomic dysfunction. ICD therapy in this group, when compared to conventional treatment, demonstrated no benefit in reducing mortality after 2.5 years of follow-up.