Prevention of Sudden Cardiac Death in Patients with Cardiomyopathy (Classification, Evaluation and Management of Cardiomyopathies) Part 2

Primary prevention of SCD in patients with non-ischemic cardiomyopathy

The early trials examining the effects of ICD therapy compared to antiarrhythmic therapy in patients with non-ischemic cardiomyopathy (NICM) were small and not powered enough to show mortality benefit. The first trial is the Amiodarone versus Implantable Cardioverter Defibrillator Trial (AMIOVERT) which compared ICD therapy in 103 patients with NICM (with the diagnosis made > 6 months before enrollment) and non-sustained VT to amiodarone. The primary endpoint was all cause mortality. There was no difference in survival between the ICD group and the amiodarone group. The trial was terminated due to futility. The second trial is the The Cardiomyopathy trial (CAT) which was carried out in Germany and enrolled 104 patients with non-ischemic cardiomyopathy who were diagnosed within 9 months of enrollment. Mean follow up was 5.5 years and the primary end point was all cause mortality. Again there was no difference in survival between the ICD group and the control group. Both AMIOVERT and CAT trials were underpowered to detect a difference between groups, and in both of them the observed mortality was far lower than the predicted mortality used to design these trials.

Table 3. Primary prevention trials of ICD therapy in patients with coronary artery disease. VT is for ventricular tachycardia, VF is for Ventricular Fibrillation, NSVT is for non sustained VT, LVEF is for left Ventricular ejection Fraction. HR is for hazard Ratio, CI is confidence interval.

The Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation Trial (DEFINITE) studied the efficacy of ICD therapy to prevent all cause mortality in patients with LVEF < 35% and non-sustained VT or frequent premature ventricular contractions (PVCs) on ambulatory monitoring (Kadish et al. 2004). A total of 488 patients (229 in the ICD group and 229 in the conventional treatment group) were enrolled and the primary end point was death from any cause and the secondary endpoint was sudden cardiac death. Most of the patients were receiving beta blocker (85%) and ACE-I (86%) There was a 35% relative risk reduction in mortality in the ICD group compared to the medical therapy group (HR, 0.65; 95% CI, 0.40 to 1.06; P=0.08) but it did not reach statistical significance. A significant reduction in SCD was observed (HR 0.20; 95% CI, 0.06 to 0.71; P=0.006). The DEFINITE trial didn’t specify duration of heart failure as an entry criterion and it only required absence of a reversible cause of cardiomyopathy for enrollment. Patients in DEFINITE who had a recent diagnosis of non-ischemic cardiomyopathy (Using a 3 months cut point or a 9 months cut point) had similar benefit from ICD therapy when compared to patients who had a remote diagnosis of non-ischemic cardiomyopathy (p0.25) (Kadish et al. 2006)

The largest trial conducted to examine the effects of ICD therapy on sudden cardiac death prevention in patients with ischemic and non-ischemic cardiomyopathy is the Sudden Cardiac Death-Heart Failure (SCD-HeFT) trial. This trial enrolled 2521 patients with LVEF < 35%, NYHA II-III and it had similar proportion of patients with ischemic cardiomyopathy (52%) and non-ischemic cardiomyopathy (48%). Patients were randomized to receive a single chamber ICD (n=829), Amiodarone (n=845) or placebo (n=847) (Bardy et al. 2005). Patients with recent MI or revascularization (<1 month) were not eligible. Nearly 96% of patients were on ACE-I or angiotensin receptor blockers and 69% were receiving beta-blocker therapy. The primary endpoint was all cause mortality. The ICD group was programmed to shock therapy only. After mean follow up of 45.5 months, the ICD group had lower mortality compared to placebo (HR 0.77, 95% CI of 0.62-0.96, p=0.007) while amiodarone had no effect on mortality compared to placebo (HR 1.06, 95% CI 0.86 to 1.30, p=0.53) (Figure 2). Annual rate of appropriate ICD shocks occurred in 68% of patients with an average annual rate of 5.1%. The absolute reduction in mortality was similar in patients with ischemic (7.2%) and non-ischemic cardiomyopathy (6.5%).

Fig. 2. Kaplan-Meier Estimates of Death from Any Cause. CI denotes confidence interval.

In a pooled analysis of 10 primary prevention trials (AMIOVERT, MADIT-I, MUSTT, MADIT-II, CABG PATCH, CAT, SCD-HeFT, COMPANION, DEFINITE and DINAMIT), ICD therapy was associated with 25% relative risk reduction of all cause mortality (RRR 9% to 37%, p=0.003) compared to the medical treatment group. The absolute risk reduction was 7.9%, which means 13 ICDs need to be implanted to save one life over about 3 years. This was not sensitive to removal of the any of the trials from the analysis. The benefit of ICD therapy in sudden cardiac death prevention is above and beyond the mortality benefit associated with use of beta blocker and ACE-I in patients with systolic heart failure (Nanthakumar et al. 2004). Table 4 summarized the primary prevention trials of ICD therapy in patients with non-ischemic cardiomyopathy.

Table 4. Primary prevention trials of ICD therapy in patients with non-ischemic cardiomyopathy. VT is for ventricular tachycardia, VF is for Ventricular Fibrillation, PVC is for premature ventricular contractions, LVEF is for left Ventricular ejection Fraction. HR is for hazard Ratio, CI is confidence interval.

Cost effectiveness of ICD therapy

ICD therapy adds to the costs of care of patients with cardiomyopathy. Analysis of cost effectiveness in the SCD-HeFT trial showed that ICD therapy is cost effective, with incremental cost of $38,400 (95% CI of $25,217 to $80,160). This was similar in patients with ischemic and non-ischemic cardiomyopathy (Mark et al. 2006). In a pooled analysis of eight primary prevention trials, ICD therapy was not found to be cost effective in CABG PATCH and in DINAMIT, which are the trials that showed no mortality benefit from ICD therapy compared to conventional medical therapy. When examining the primary prevention trials that showed mortality benefit (MADIT-I, MUSTT, MADIT-II, COMPANION, DEFINITE and SCD-HeFT), ICD therapy was found to be cost effective, adding between 1.01 and 2.99 quality-adjusted life years with costs ranging from $34,000 to $70,200. This analysis takes into account that the ICD generator will be replaced every 5 years and assumes that the mortality benefit persists throughout the patient’s life time (Sanders et al. 2005). Careful patient selection with a focus on patients who best fit the trials and are likely to die from arrhythmia and not from other non cardiac causes is important to insure the best utilization of this important and life saving therapy.

Defibrillator therapy in less common types of cardiomyopathy

Some of the inherited cardiomyopathies carry an increased risk of sudden cardiac death. We will review in this section the data behind ICD therapy in patients with two inherited disorders, first is Hypertrophic cardiomyopathy (HCM) and the second is arrhythmogenic right ventricular dysplasia./ cardiomyopathy (ARVD/C).

Hypertrophic cardiomyopathy

Hypertrophic cardiomyopathy is an autosomal dominant disorder diagnosed by two-dimensional echocardiography and is characterized by hypertrophied and non-dilated LV in the absence of other causes of hypertrophy (no history of hypertension or aortic stenosis or any other cardiac or systemic disease causing hypertrophy) (Maron 2002; 2003). Histologically, there is myocardial disarray, abnormal microvascular circulation with mismatch between myocardial mass and blood supply as well as interstitial fibrosis (Maron et al. 1986; Varnava et al. 2001). All of these predispose to ventricular tachycardia and ventricular fibrillation putting the patients at risk of sudden cardiac death (Varnava et al. 2001). The disease can present at any age in life. Earlier registries from tertiary care centers overestimated the risk of sudden cardiac death due to selection bias (with the annual risk of death thought to be 3 to 6%). More recent population studies of unselected patients from community centers suggest a more benign prognosis (annual risk of death of 1%) (Maron et al. 1999; Kofflard et al. 2003). Despite all recent data, there is a subset of patients with HCM who are at high risk of sudden cardiac death. In fact HCM remains the most common etiology for SCD in patients younger than 40 years and it can be the first presentation in patients with HCM (Maron 2003). Patients who survived a cardiac arrest are particularly at high risk of dying suddenly. Data from registries suggest a number of markers that increase the risk of sudden cardiac death. These markers include one or more of the following: LV wall thickness > 30 mm (Maron et al. 1999), syncope (particularly exertional syncope) (Kofflard et al. 2003), family history of SCD, non-sustained VT on ambulatory holter of > 120 bpm and a blunted blood pressure response to exercise. High LV outflow gradient (> 30 mmHg) has also been considered as risk marker for sudden cardiac death (Maron et al. 2003).

There are no randomized trials examining the benefits of ICD therapy in this patient population, so the data supporting ICD therapy is derived from registry data in patients with HCM who received an ICD when found to be high risk by their treating cardiologist / Electrophysiologist (Maron et al. 2000). The last update from the registry included 506 patients with HCM with a mean age of 42 years. Patients received and ICD if they had survived a cardiac arrest due to ventricular tachycardia or ventricular fibrillation (secondary prevention cohort of 123 patients) or if they had one or more risk factors of sudden cardiac death: unexplained syncope, family history of sudden cardiac death in one or more first degree relatives, massive LV hypertrophy or non-sustained VT on holter monitoring (Primary prevention cohort of 383 patients) (Maron et al. 2007). Based on this registry, patients with HCM who survived cardiac arrest have a high appropriate ICD discharges (10.6% per year). This risk is lower in patients with HCM who had an ICD for primary prevention (3.6% per year). A third of these patients were 30 years or younger. Amiodarone was used based on the physicians judgment and it did not prevent arrhythmia occurrence (27% of patients who were on amiodarone had appropriate shocks). A third of the patients who received an ICD for primary prevention and had appropriate ICD discharges had one risk factor only for sudden cardiac death. There was no difference between the risk factors in the prediction of SCD. Since this is a young population of patients, they are at risk of inappropriate shocks, which occurred in 27% of the patients and were mainly due to sinus tachycardia, atrial fibrillation or lead malfunction. ICD implantation was shown to be safe with a rate of infection of 3.8% and the rate of lead fracture or dislodgement of 6.7%. Implantation of ICD has become an acceptable therapy in patients with HCM who survived cardiac arrest or who have two or more of the aforementioned risk factors of sudden cardiac death. For patients with only one risk factor of sudden cardiac death, the decision to implant an ICD is left to the physician’s judgment and a careful discussion with the patient and his / her family in regards to the risks, benefits and alternatives of ICD therapy. In our experience, the presence of only one risk factor for SCD does not guarantee a recommendation for implanting an ICD. The clinical scenario is to be taken into account, as well as the patient’s age and his / her wishes. The decision to implant is more favorable in a young patient with HCM with family history of sudden cardiac death or in a young patient with severe LV wall thickness (>30 mm) but still the discussion should take into account the young age of the patient and his / her wishes. On the other hand the discussion is more careful in an old patient with HCM in his / her 60s with history of non-exertional syncope that seems to be vasovagal in origin, the fact that the patient survived to that age without any major cardiac arrest indicates a more benign prognosis. Table 2 lists the current recommendations for ICD implantation for patients with hypertrophic cardiomyopathy.

Arrhythmogenic Right Ventricular Dysplasia / Cardiomyopathy (ARVD/C)

Arrhythmogenic right ventricular Dysplasia/ Cardiomyopathy (ARVD/ C) is an inherited myopathy characterized by fibrofatty infiltration of the right ventricular (RV) wall, with left ventricular involvement over time in some patients (Gemayel et al. 2001; Sen-Chowdhry et al. 2004). The RV wall becomes thin, and the most areas affected are the RV inflow, apex and RV outflow, which form what is called the triangle of dysplasia. Ventricular tachycardia in general has left bundle branch morphology and is caused by macro-reentry and there is evidence that adrenergic stimulation acts as a trigger for these arrhythmias (Leclercq et al. 1996). ARVC/D accounts for 3 to 10% of death occurring in patients younger than 65 years (Tabib et al. 2003) and is one of the causes of sudden cardiac death in athletes (Maron 2003). The Most common presentation is with palpitations (due to frequent ventricular ectopy and ventricular tachycardia), chest pain, and syncope (mostly exertional). With time patients might develop RV dilatation, LV involvement and heart failure. Most of the data are obtained from registries in the United States and Europe (either single centers or multicenter registries). Diagnosis of ARVD/C is based on the European Task Force criteria (McKenna et al. 1994). Risk factors of sudden cardiac death include prior cardiac arrest, hemodyamically unstable VT and prior syncope. Some studies suggested that LV involvement, development of heart failure and marked RV dilatation are risk factors for sudden cardiac death (Hulot et al. 2004). The role of electrophysiologic testing in risk stratification is less clear, with some studies showing a high positive predictive value and some showing low positive and low negative predictive values in predicting arrhythmias and appropriate ICD shocks (Corrado et al. 2003; Roguin et al. 2004). Beta blockers and sotalol were thought to be the best in suppressing these arrhythmias; however, this is challenged in more recent studies (Marcus et al. 2009). ICD therapy is clearly indicated in patients who survived cardiac arrest or have sustained VT and is a class IIa indication in patients with ARVD/C who have unexplained syncope. Some patients experience repetitive shocks requiring administration of antiarrhythmic drug therapy as well as VT ablation. Since this is a young population, they are also likely to experience inappropriate shocks due to sinus tachycardia or other supraventricular arrhythmias. In general ICD therapy is life saving and is well tolerated and has become accepted standard of care in patients with ARVD/C who experience cardiac arrest, sustained VT, unexplained syncope or marked RV dilatation or LV involvement (Epstein et al. 2008). Table 2 lists the current guidelines for ICD implantation in this patient population.

Cardiac Resynchronization Therapy (CRT) in patients with heart failure and its effects on mortality

Cardiac Resynchronization Therapy (CRT) aims at correcting mostly intraventricular dyssynchrony by stimulating the left ventricle (preferably basal stimulation) or by simultaneously stimulating the left and right ventricles after a sensed or paced atrial beat or during atrial fibrillation. CRT has been shown to improve the cardiac hemodynamics in patients with systolic heart failure, including improvements in the systolic blood pressure and decrease in the pulmonary capillary wedge pressure (by up to 20% in some patients) (Blanc et al. 1997). The early trials had endpoints related to heart failure functional status (including the 6 minute walk test, NYHA functional class), LV systolic function and improvement in the LV dimensions (including LVEF, LV end systolic volume and LV end systolic volume index as well as LV end diastolic dimension). Other studies relied on clinical composite score (which combines death from any cause, recent hospitalization for heart failure, NYHA class as well as the global assessment score) to define response to CRT (Chung et al. 2008). However, there is poor correlation between clinical and echocardiographic measurements of response and there is disagreement about the best way to measure response in patients with heart failure receiving CRT (Fornwalt et al.). So far QRS duration remains an important criterion for patient selection for CRT. Kass and colleagues demonstrated that baseline QRS duration correlated with enhancement in the isovolumetric dP/dtmax (r = 0.6, p = 0.02), while changes in the QRS duration with pacing did not predict hemodynamic response (Nelson et al. 2000). Most of the trials on CRT involved patients with systolic heart failure with LVEF < 35% and NYHA class III or IV as well as a QRS duration of > 120 msec. A trial studying patients with narrow QRS in patients with systolic heart failure failed to show any benefit. Later studies included patients with NYHA class I and class II heart failure with endpoints related to death or hospitalization. This section will focus mainly on the studies that included mortality as an endpoint.