Ventricular ectopy is considered an electrical trigger of sustained ventricular arrhythmias and potentially SCD. Several trials evaluated the role of ventricular arrhythmia suppression and its effect on cardiac mortality in patients with coronary artery disease.

The International Mexiletine and Placebo Antiarrhythmic Coronary Trial (IMPACT)⁸ included 630 patients with recent myocardial infarction who were randomly assigned to treatment with mexiletine or placebo. After an average follow up of nine months, the mortality on mexiletine was 7.6% and on placebo was 48% (P = NS). The Cardiac Arrhythmic Suppression Trial (CAST)⁹ enrolled 1727 patients with recent onset of myocardial infarction and with asymptomatic, or mildly symptomatic, ventricular arrhythmia (≥6 ventricular premature ventricular complexes per hour), potentially suppressible by a class I anti-arrhythmic drug (encainide, flecainide or moricizine), who were randomized to the active antiarrhythmic drug or placebo and followed for arrhythmic death. The trial was halted early because of an increased incidence of arrhythmic cardiac death and non-fatal cardiac arrests in patients treated with encainide and flecainide (4.5% vs 1.2%, risk ratio (RR) 3.6, 95% confidence interval (CI) 1.7–8.5). In a meta-analysis of the results of 138 trials of antiarrhythmic prophylactic therapy in patients after myocardial infarction,¹⁰ there were 660 deaths among 11 712 patients allocated to receive class I agents and 571 deaths among 11 517 corresponding control patients (51 trials: odds ratio (OR) 1.14; 95% CI 1.01 –1.28; P = 0.03). Based on this evidence, class I antiarrhythmic agents have been abandoned for the reduction of the the risk of SCD in patients with ischemic cardiomyopathy and non-sustained ventricular arrhythmia.

Class III antiarrhythmic agents were tested in a series of randomized studies. The two largest trials using amiodarone are the Canadian Amiodarone Myocardial Infarction Arrhythmia Trial (CAMIAT)¹¹ and the European Myocardial Infarction Amiodarone Trial (EMIAT).¹² Both showed a reduction in arrhythmic death but no significant reduction in overall death. Meta-analysis of data from all 13 randomized controlled trials of amiodarone (89% of patients, after myocardial infarction) showed a significant reduction in total mortality (OR 0.87, 95% CI 0.78–0.99) and a significant reduction in arrhythmic death (OR 0.71, 95% CI 0.59–0.85).^13,14

Analysis of the interaction between the treatment and baseline factors suggested an important positive relationship between beta-blocker use and amiodarone effect,¹⁵ such that patients on beta-blockers received a significantly greater benefit from amiodarone than those not on betablockers. More recently, the association of amiodarone and beta-blockers has been demonstrated to markedly reduce the incidence of appropriate ICD therapy. In the Optimal Pharmacological Therapy in Cardioverter Defibrillator Patients (OPTIC) trial,¹⁶ amiodarone plus beta-blocker sig-nificantly reduced the risk of appropriate ICD shocks compared with beta-blocker alone (HR 0.27; 95% CI 0.14–0.52; P < 0.001) or sotalol (HR 0.43; 95% CI 0.22–0.85; P = 0.02).

D-Sotalol, a pure class III agent, was evaluated for prevention of SCD in a placebo-controlled trial of 3121 patients with recent myocardial infarction and LVEF <40%, or symptomatic heart failure with a remote myocardial infarction (Survival with Oral d-sotalol, SWORD trial).¹⁷ Among 1549 patients assigned to d-sotalol there were 78 deaths (5.0%) compared to 48 (3.1%) among the 1572 patients assigned to placebo (RR 1.65, 95% CI 1.15 –2.36). Presumed arrhythmic deaths (RR 1.77, 95% CI 1.15–2.74) accounted for the excess mortality in the d-sotalol group. This proarrhythmic fatal effect of d-sotalol was greater in patients with a left ventricular ejection fraction of 31–40% than in those with lower ejection fractions (RR 4.0 vs 1.2, P = 0.007). In a multicenter double-blind randomized study, Julian et al reported the effect of sotalol 320 mg once daily compared with placebo in patients surviving an AMI.¹⁸ Treatment was started 5–14 days after infarction in 1456 patients (60% being randomized to sotalol and 40% to placebo). Patients were followed for 12 months. The mortality rate was 7.3% (64 patients) in the sotalol group and 8.9% (52 patients) in the placebo group. The mortality was 18% lower in the sotalol than in the placebo group, but this difference was not statistically significant.

Another pure class III compound, dofetilide, was tested in patients with symptomatic heart failure. In the Danish Investigations of Arrhythmia and Mortality on Dofetilide (DIAMOND), 1518 patients were randomized to dofetilide or placebo.^19,20 The study treatment was initiated in hospital and included three days of cardiac monitoring and dose adjustment. During a median follow-up of 18 months, 311 patients in the dofetilide group (41%) and 317 patients in the placebo group (42%) died (OR 0.95; 95% CI 0.81–1.11). Treatment with dofetilide significantly reduced the risk of hospitalization for worsening congestive heart failure (CHF), rate of conversion and risk of recurrence of atrial fibrillation. There were 25 cases of torsades de pointes in the dofetilide group (3.3%), compared to none in the placebo group. Dofetilide has also been tested in a randomized trial of 1510 patients with severe LVEF ≤35% after recent myocardial infarction (DIAMOND MI trial).²¹ The primary endpoint was all-cause mortality. No significant difference was found between the dofetilide and placebo groups in overall mortality (31% vs 32%). The cardiac mortality (26% vs 28%) and arrhythmic mortality (17% vs 18%) were also similar. There were seven cases of torsades de pointes ventricular tachycardia, all in the dofetilide group. In the Azimilide Postinfarct Survival Evaluation (ALIVE) trial the effect of azimilide, another pure class III agent, was evaluated in 3717 patients with recent myocardial infarction, with LVEF < 35% and with low heart rate variability. Azimilide had no effect on mortality (HR 1.0).²²

Other studies have tested antiarrhythmic agents in patients with sustained ventricular arrhythmias. The endpoints were arrhythmic death, recurrence of sustained ventricular arrhythmia or reinducibility of ventricular tachycardia during programmed electrical stimulation. Steinbeck et al²³ conducted a prospective randomized trial in 170 patients to investigate whether electrophysiologic study (EPS)-guided antiarrhythmic therapy improved the long-term outcome of patients with spontaneous and inducible sustained ventricular arrhythmia compared with metoprolol therapy not guided by EPS. In 55 patients ventricular tachycardia was not inducible during the baseline EPS, thereby precluding further serial drug testing. These patients were treated with metoprolol. The two-year incidence of the composite outcome was the same for EPS-guided therapy compared to metoprolol (46% vs 48%).

The Electrophysiological Study Versus Electrocardiographic Monitoring (ESVEM) study^24,25 included patients with inducible ventricular tachycardia and any of the following: (i) history of cardiac arrest, (ii) sustained ventricular tachycardia or (iii) syncope. They were randomly assigned to undergo serial testing to determine efficacy of antiarrhythmic drugs either by performing EPS or by 24-hour ambulatory ECG monitoring. Patients (n = 486) received long-term treatment with the first antiarrhythmic drug that was determined to be effective on the basis of either repeated EPS or 24-hour Holter. The ESVEM study concluded that therapies guided by EPS or Holter monitoring were comparable. The secondary outcome, related to the efficacy of individual antiarrhythmic drugs, concluded that sotalol was more effective than the class I drugs tested. The actuarial probability of a recurrence of arrhythmia after a drug was deemed efficacious by either strategy was significantly lower for patients treated with sotalol than for patients treated with other antiarrhythmic agents (RR 0.43; 95% CI 0.29–0.62). Sotalol significantly reduced the risk of all-cause mortality (RR 0.50; 95% CI 0.30–0.80), cardiac mortality (0.50; P = 0.02) and arrhythmic death (0.50; P = 0.04).

In the Cardiac Arrest in Seattle: Conventional versus Amiodarone Drug Evaluation (CASCADE) study, antiar-rhythmic drug therapy was evaluated in survivors of out-of-hospital ventricular fibrillation (VF).²⁶ Amiodarone without EPS or Holter guidance was compared to class I antiarrhythmic agents (quinidine, procainamide, their combination, or flecainide), selected by serial EPS or Holter monitoring. Most of the 228 randomized patients had coronary artery disease with a prior myocardial infarction, and the mean LVEF was 35 ± 10%. During a mean follow-up of six years, amiodarone improved survival compared to class I agents (53% vs 40%, P = 0.007).

The Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) randomly assigned 2521 patients with NYHA class II or III CHF and a LVEF of 35% or less to conventional therapy for CHF plus placebo (847 patients), conventional therapy plus amiodarone (845 patients), or conventional therapy plus a conservatively programmed, shock-only, single-lead ICD (829 patients).²⁷ Placebo and amiodarone were administered in a double-blind fashion. The primary endpoint was death from any cause. Mortality was 29% in the placebo group and 28% in the amiodarone group. As compared with placebo, amiodarone was associated with a similar risk of death (HR1.06; 97.5% CI 0.86–1.30; P = 0.53).

Beta-blockers are the agent most frequently studied in postmyocardial infarction patients for the prevention of death, with more than 12 large trials reported. A meta-analysis of the beta-blocker trials, reported in 1985, showed a significant reduction in mortality during treatment after myocardial infarction.²⁸ This meta-analysis also indicated a highly significant 30% reduction in sudden cardiac death with beta-blockers. The risk of non-sudden death was also decreased by 12%, but this difference was not significant. Recent beta-blocker trials in CHF patients also show a reduction in both overall deaths and SCD.²⁹–³² Beta-blockers are effective against arrhythmic death (20–30% reduction) and non-arrhythmic deaths, and reduce overall mortality significantly. Beta-blocker therapy is indicated in all patients at high risk for SCD.

In summary, antiarrhythmic drugs have been extensively evaluated in randomized trials as prophylactic agents against death, but little tested against recurrence of arrhythmia. Class I antiarrhythmic drugs are harmful and are proscribed in patients with ischemic heart disease and reduced LVEF. Amiodarone has a moderate effect against SCD and a neutral effect on other deaths, therefore its overall effect on total mortality is modest. Pure class III agents are at best neutral and, in the case of d-sotalol, actually harmful. Pure class III antiarrhythmic agents clearly do not reduce mortality or SCD when used prophylacti-cally in high-risk patients. The different results of d-sotalol, dofetilide and azimilide trials are probably due to differences in the design of the studies and differences in risk of torsades de pointes between the agents.

The effect of angiotensin-converting enzyme (ACE) inhibitors on the risk of SCD following myocardial infarction has been demonstrated in randomized trials.^33–35 A recent meta-analysis incorporated data from 15 trials that included 15 104 patients having 900 SCD.³⁵ ACE inhibitor therapy resulted in a significant reduction in total mortality (OR 0.83, 95% CI 0.71–0.97), cardiovascular death (OR 0.82, 95% CI 0.69–0.97) and SCD (OR 0.80, 95% CI 0.70–0.92). Also, the meta-analysis suggested that a reduction in SCD risk with ACE inhibitors was an important component of overall survival benefit, the magnitude of effect on SCD being the same as on overall mortality.

Interestingly, in the Heart Outcome Prevention Evaluation (HOPE) Study³⁶ involving cardiovascular patients without a significant decrease in left ventricular function (LVEF >40%), the ACE inhibitor ramipril significantly decreased the incidence of cardiac arrest (RR 0.62, 95% CI 0.41–0.94). The mechanism by which ACE inhibitors reduce SCD is poorly understood. In addition to attenuation of remodeling, thereby reducing the substrate for ventricular tachyarrhythmia, they provide significant neurohumoral modulation and protection from future ischemic events.

Angiotensin receptor blockers (ARB) are potentially antiarrhythmic; this effect can be explained by blockade of the angiotensin II receptor type I leading to inhibition of the proarrhythmic effects of angiotensin II.³⁷ Several trials of different angiotensin receptor blockers have demonstrated equivalent effects to those of ACE inhibitors on cardiac and overall mortality. However, with the exception of the CHARM trials, the addition of an ARB to an ACE inhibitor has not yielded any added benefits.³⁸ In summary, ARBs should be used as alternative therapy in patients intolerant to ACE inhibitors.

In the Randomized Aldactone Evaluation Study (RALES),³⁹ aldactone was evaluated in patients having NYHA III–IV. After a mean follow-up of 24 months, the incidence of SCD was significantly decreased (RR 0.71, 95% CI 0.54–0.95). The magnitude of this effect was similar to the effect on total mortality (RR 0.70, 95% CI 0.68–0.72). In the EPHESUS trial patients with an acute myocardial infarction complicated by symptomatic LV dysfunction were randomized to standard therapy versus standard therapy plus eplerenone.⁴⁰ Eplerenone significantly reduced all-cause mortality, cardiovascular death and the risk of cardiovascular death or hospitalization. Interestingly SCD was also significantly reduced (RR 0.79, 95% CI 0.64–0.97). Either spironolactone or eplerenone is recommended as adjunctive therapy in patients with heart failure with LV dysfunction and NYHA class III–IV.

The role of n-3 polyunsaturated fatty acids for the prevention of SCD has been assessed in a number of randomized trials. The Diet And Reinfarction Trial randomized 2033 men with recent myocardial infarction and reported a significant reduction in both cardiac and total mortality (RR 0.71, 95% CI 0.54–0.93).⁴¹ In a larger study the GISSI-Prevenzione Trial,⁴² treatment with n-3 polyunsaturated fatty acids in 11 324 postmyocardial infarction patients significantly decreased the incidence of SCD (RR 0.74, 95% CI 0.58–0.93), also significantly decreasing total cardiac mortality and coronary mortality by a similar extent (RR 0.78 and 0.80, respectively, both significant). A recent systematic review of 48 RCTs (36 913 participants) and 41 cohort studies did not document a clear effect of omega 3 fats on total mortality, combined cardiovascular events or cancer.⁴³ The pooled estimate of this meta-analysis showed no strong evidence of reduced risk of total mortality (relative risk 0.87, 95% CI 0.73–1.03) or combined cardiovascular events (0.95, 0.82–1.12) in participants taking additional omega 3 fats. When data from the subgroup of studies of long-chain omega 3 fats were analyzed separately, total mortality (0.86, 0.70–1.04; 138 events) and cardiovascular events (0.93, 0.79–1.11) were not clearly reduced. Neither RCTs nor cohort studies suggested increased risk of cancer with a higher intake of omega 3 (trials: 1.07, 0.88–1.30; cohort studies: 1.02, 0.87–1.19), but clinically important harm could not be excluded.

In summary, the effects of fish oils on SCD and cardiovascular morbidity and mortality remain debated. The protective effect of fish oils may be limited to patients with a previous myocardial infarction. Current guidelines recommend that patients with documented coronary artery disease consume approximately 1 g of EPA+ DHA per day, preferably from oily fish although supplements can also be considered.⁴⁴

Thus, accumulated evidence supports the wide use of upstream interventions in appropriate patients.

Several studies have assessed the effectiveness of ICDs for the prevention of SCD in high-risk populations. The primary risk stratifier is a LVEF <35%. In some instances the presence of non-sustained VT or the induction of VT during programmed electrical stimulation have been used as inclusion criteria. These trials may be classified according to the underlying anatomic substrate:

• Coronary artery disease and depressed LVEF: MADIT,⁴⁵ CABG-Patch,⁴⁶ MUSTT,⁴⁷ MADIT II,⁴⁸ DINAMIT⁴⁹
  
• Dilated non-ischemic cardiomyopathy (DCM) and depressed LVEF: CAT,⁵⁰ AMIOVIRT,⁵¹ DEFINITE⁵²
  
• LVEF <35% independent of the etiology of the underlying cardiomyopathy: SCD-HeFT.²⁷

Primary prevention of SCD in ischemic cardiomyopathy

The rationale for the CABG-Patch trial was developed at a time when a thoracotomy was required for implantation of an ICD. Thus patients scheduled for CABG and with LVEF ≤35% were further stratified for risk of arrhythmic death by signal-averaged ECG.⁴⁶ There were 900 high-risk patients randomized to receive an ICD or not at the time of CABG. Antiarrhythmic drug use was similar between the two groups. There were 52 patients randomized to ICD who either never received a device or who had it removed. There were 198 deaths (102 in the ICD group and 96 in the control group) for an overall mortality rate of 21.8% during an average follow-up of 32 ± 16 months. The hazard ratio was 1.07 (95% CI 0.81–1.42), indicating no benefit from the ICD in this patient population. Secondary analysis showed that the ICD did reduce arrhythmic death; this potential benefit was offset by an unexplained increase in non-arrhythmic deaths.

The Multicenter Automatic Defibrillator Implantation Trial (MADIT)⁴⁵ included patients with LVEF <35% and recent myocardial infarction who were screened with programmed ventricular stimulation. Patients with inducible VT or VF were enrolled in the study if inducibility of VT could not be suppressed by procainamide. There were 196 patients randomized to receive either an ICD or “conventional” therapy. The choice of conventional therapy was at the discretion of the investigator. Amiodarone and beta-blockers, the only drugs proven effective against VT and VF, were used predominantly but sporadically (in 45% and 5%, respectively, of “conventional” patients at last contact). The trial was terminated prematurely when about 75% of planned enrollment had occurred, because of marked benefit derived from ICD treatment (HR 0.46; 95% CI 0.26–0.82; P = 0.009). When the specific causes of death were examined, the ICD not only reduced arrhythmic death, but also appeared to reduce non-arrhythmic cardiac death, and deaths of unknown cause, which is unexplained and biologically implausible.

The Multicenter Unsustained Tachycardia Trial (MUSTT)⁴⁷ was a randomized trial of electrophysiologiprophylacticcally guided antiarrhythmic therapy in patients with coronary artery disease, LVEF ≤0.40 and asymptomatic, non-sustained ventricular tachycardia. Seven hundred and four patients who satisfied these criteria, and in whom sustained VT was induced by programmed stimulation, were randomly assigned to receive either antiarrhythmic drug tailored by electrophysiologic testing, including drugs and ICDs (if drugs failed to suppress inducibility), or no antiarrhythmic therapy. The primary endpoint of cardiac arrest or death from arrhythmia was reached in 25% of those receiving electrophysiologically guided therapy, and in 32% of those assigned to no antiarrhythmic therapy (relative risk 0.73, 95% CI 0.53–0.99). Five-year total mortality was 42% in patients receiving EPS-guided therapy, versus 48% in controls (RR 0.80, 95% CI 0.64–1.01). In a non-randomized analysis the primary endpoint was less frequent among patients who received ICDs compared to patients discharged without an ICD (relative risk 0.24; 95% CI 0.13–0.45; P < 0.001). In contrast, the primary endpoint was similar among those receiving antiarrhythmic therapy compared to those on no therapy.

The second Multicenter Automatic Defibrillator Implantation Trial (MADIT II)⁴⁸ included patients with LVEF <30% but without evidence of sustained VT/VF and excluded those with recent (<1 month) myocardial infarction, coronary artery bypass graft (CABG) or percutaneous transluminal coronary angioplasty (PTCA) (<2 months). There were 1232 patients randomly assigned to receive an ICD (742 patients) or conventional medical therapy (490 patients). During an average follow-up of 20 months, the mortality rates were 19.8% in the conventional therapy group and 14.2% in the defibrillator group (HR 0.69, 95% CI 0.51–0.93). MADIT II was a landmark trial expanding ICD indications to patients with severe LVEF impairment due to ischemic cardiomyopathy but with no evidence of sustained ventricular arrhythmia.

Identifying those subjects with coronary artery disease (CAD) and severe LV dysfunction who will derive the greatest benefit from prophylactic ICD therapy remains a challenge. Several MADIT II subgroup studies have been reported including survival in elderly patients, QRS duration, and clinical risk factors, and time after MI.⁵³–⁵⁷ An important question is whether ICD therapy maintains its efficacy in elderly subjects, addressed by Huang et al in a MADIT II substudy⁵⁴ Among 1232 patients enrolled with prior infarct and LVEF ≤0.30, 204 were ≥75 years old and of those, 121 underwent ICD implant.

The hazard ratio for mortality in patients >75 years assigned to ICD compared with those in conventional therapy was 0.56 (95% CI 0.29–1.08; P = 0.08) after a mean follow-up of 17.2 months. Comparatively the hazard ratio in patients < 75 years assigned to ICD was 0.63 (0.45–0.88; P = 0.01) after 20.8 months. The ICD was associated with an equivalent reduction of mortality in elderly and younger patients. Dhar et al evaluated the prognostic significance of prolonged QRS relative to arrhythmic outcomes in medically and ICD-treated patients enrolled in MADIT II.⁵⁵ In the medically treated arm, prolonged QRS was a significant independent predictor of SCD (HR 2.12; 95% CI 1.20–3.76; P = 0.01). However, in the ICD-treated arm, prolonged QRS did not predict SCD or rapid VT/VF (HR 0.77; 95% CI 0.47–1.24; P = 0.28). The difference in the prognostic effect of prolonged QRS in these two groups was significant (P < 0.01). These results suggest that in MADIT II patients, prolonged QRS does not predict SCD/VT/VF in ICD-treated patients but does predict SCD in medically treated patients. This underscores the non-equivalence of VT/VF and SCD and the need for caution in inferring risk of SCD when using non-randomized databases that include only patients with ICDs.

Clinical risk factors may be useful to identify subjects who would benefit from a prophylactic ICD implant. Goldenberg et al developed a simple clinical risk score (NYHA class >II, age >70 years, BUN >26mg/dl, QRS duration >0.12s, and atrial fibrillation) using a proportional-hazards regression analysis for the end point of all-cause mortality in patients allocated to the conventional therapy arm of MADIT II.⁵⁶ A prespecified subgroup of very high-risk (VHR) patients (BUN >50mg/dL and/or serum creatinine >2.5 mg/dL) was excluded. The benefit of the ICD was then assessed within risk score categories and separately in VHR patients. Crude mortality rates in the conventional group were 8% and 28% in patients with 0 and >1 risk factors, respectively, and 43% in VHR patients. ICD therapy was associated with a 49% reduction in the risk of death (P < 0.001) among patients with >1 risk factors (n = 786), whereas no ICD benefit was identified in patients with 0 risk factors (n = 345; HR 0.96; P = 0.91) and in VHR patients (n = 60; HR 1.00; P > 0.99). Patients that derive the greatest benefit from a prophylactic ICD are at intermediate risk, in contrast, low-risk and VHR patients have attenuated efficacy with an ICD.