Long-term follow-up of patients with indication for a implantable defibrillator for primary prevention of death

Hadid, Claudio; Avellana, Patricia; Di Toro, Darío; Gomez, Claudia Fernández; Visser, Miguel; Prieto, Noemí

doi:10.1590/S0066-782X2008000500006

Abstracts

BACKGROUND: Recent MADIT II and SCD-HeFT trials have led to an expansion of indications for use of prophylactic Implantable Cardioverter Defibrillator (ICD) in patients with severe left-ventricular impairment. This therapy has not been fully adopted in our health care system, mainly due to its high cost. OBJECTIVE: To assess total mortality of SCD-HeFT-like patients from our daily practice who are under stable, optimal medical treatment and who have not received an ICD; and to compare it to that of the placebo arm of the SCD-HeFT Trial. METHODS: SCD-HeFT-like patients identified from office medical records were included in our study. Total mortality was assessed by telephone contact. Statistical analysis was performed by Student's t-Test, Mann-Whitney Test or chi2 test, depending on the type of variable. Cumulative mortality rates were calculated according to the Kaplan-Meier method. RESULTS: Our study comprised 102 patients (seventy-four of which were men) with a median age of 64 years, and an overall median ejection fraction of 25%. We found no differences between our patients and SCD-HeFT patients across these 3 variables. Over a 19.6-month follow-up period, 21 patients died (20.6%) vs. 28.8% of the SCD-HeFT patients. This difference was not statistically significant (p = 0.08). CONCLUSION: SCD-HeFT-like patients from our practice had no difference in mortality rate than patients enrolled in the placebo arm of the SCD-HeFT trial. These results suggest that the SCD-HeFT population is representative of our patients.

Heart failure, congestive; defibrillators, implantable; death, sudden

FUNDAMENTO: Os recentes estudos MADIT II e SCD-HeFT levaram a uma expansão das indicações do uso profilático de cardiodesfibrilador implantável (CDI) em pacientes com grave disfunção ventricular esquerda. Essa terapia não foi totalmente adotada em nosso sistema de saúde, sobretudo em virtude de seu alto custo. OBJETIVO: Avaliar a taxa de mortalidade global de pacientes da nossa prática diária que têm o mesmo perfil dos participantes do estudo SCD-HeFT, estão recebendo tratamento clínico otimizado e não foram submetidos a implante de CDI, e comparar essa taxa com a do grupo de placebo do estudo SCD-HeFT. MÉTODOS: Foram incluídos neste estudo pacientes com o mesmo perfil dos participantes do estudo SCD-HeFT identificados a partir de prontuários médicos. A taxa de mortalidade global foi avaliada por contato telefônico. A análise estatística foi realizada com o teste t de Student, teste de Mann-Whitney ou teste de qui-quadrado, dependendo do tipo de variável. As taxas cumulativas de mortalidade foram calculadas de acordo com o método de Kaplan-Meier. RESULTADOS: Este estudo englobou 102 pacientes, 74 dos quais do sexo masculino. A mediana da idade foi 64 anos e a média da fração de ejeção, 25%. Não encontramos diferenças entre os nossos pacientes e os pacientes do estudo SCD-HeFT em relação a essas três variáveis. Durante o período de acompanhamento de 19,6 meses, 21 pacientes morreram (20,6%), contra 28,8% do estudo SCD-HeFT. Essa diferença não foi estatisticamente significante (p = 0,08). CONCLUSÃO: Não houve diferença na taxa de mortalidade dos nossos pacientes e dos pacientes do grupo de placebo do estudo SCD-HeFT. Esses resultados indicam que a população do SCD-HeFT é representativa dos nossos pacientes.

Insuficiência cardíaca congestiva; desfibriladores implantáveis; morte súbita

ORIGINAL ARTICLE

Hospital Donación Francisco Santojanni, Buenos Aires - Argentina

Mailing address

SUMMARY

BACKGROUND: Recent MADIT II and SCD-HeFT trials have led to an expansion of indications for use of prophylactic Implantable Cardioverter Defibrillator (ICD) in patients with severe left-ventricular impairment. This therapy has not been fully adopted in our health care system, mainly due to its high cost.

OBJECTIVE: To assess total mortality of SCD-HeFT-like patients from our daily practice who are under stable, optimal medical treatment and who have not received an ICD; and to ompare it to that of the placebo arm of the SCD-HeFT Trial.

METHODS: SCD-HeFT-like patients identified from office medical records were included in our study. Total mortality was assessed by telephone contact. Statistical analysis was performed by Student's t-Test, Mann-Whitney Test or c² test, depending on the type of variable. Cumulative mortality rates were calculated according to the Kaplan-Meier method.

RESULTS: Our study comprised 102 patients (seventy-four of which were men) with a median age of 64 years, and an overall median ejection fraction of 25%. We found no differences between our patients and SCD-HeFT patients across these 3 variables. Over a 19.6-month follow-up period, 21 patients died (20.6%) vs 28.8% of the SCD-HeFT patients. This difference was not statistically significant (p = 0.08).

CONCLUSION: SCD-HeFT-like patients from our practice had no difference in mortality rate than patients enrolled in the placebo arm of the SCD-HeFT trial. These results suggest that the SCD-HeFT population is representative of our patients.

Key words: Heart failure, congestive/mortality; defibrillators, implantable; death, sudden.

Introduction

Patients with congestive heart failure (CHF) have an increased risk of all-cause mortality and can die suddenly from arrhythmia despite the use of proven useful medical therapies, such as beta-blockers, angiotensin converting enzyme inhibitors (ACEI) and spironolactone^1-11. Use of amiodarone has also been evaluated in these patients with inconclusive findings^12,13. Although this drug reduced incidence of all-cause mortality and sudden cardiac death in the GESICA trial, beta-blockers were not used as widely as they are nowadays in this kind of patient¹⁴. Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) tested the additional effect of amiodarone and an implantable cardioverter-defibrillator (ICD) for primary prevention of sudden death among patients with New York Heart Association (NYHA) class II-III CHF and left ventricular ejection fraction (LVEF) of 35% or less¹⁵. Outcome in the amiodarone group was similar to that of the placebo group, while ICD implantation was associated with a significant (23%) reduction in all-cause mortality after a 5-year follow-up period. In agreement with these findings, MADIT-II Trial showed reduced mortality following ICD implantation in patients with previous myocardial infarction and reduced LVEF¹⁶. These concordant results have led to an expansion of indications for prophylactic ICD in patients with severe LVEF impairment¹⁷.

However, implementation of this therapy would represent a significant economic impact. This is possibly the main reason hindering broader adoption of ICD therapy for primary prevention of sudden death in CHF patients in our health care system.

In order to test whether SCD-HeFT results can be extrapolated to our population, we set out to assess the outcomes of SCD-HeFT-like patients from our daily practice, and to compare them to those of the placebo arm of the SCD-HeFT Trial.

Methods

Patients from the CHF section of the Cardiovascular Service of our hospital, who fulfilled SCD-HeFT inclusion criteria, were included in the present study. Patients were at least 18 years old and had NYHA class II-III chronic, stable CHF due to ischemic or nonischemic causes and a LVEF of no more than 35 percent. Patients who had already received an ICD or had valvular heart disease were excluded.

Baseline characteristics and clinical data from our patients were obtained from office medical records. Total mortality was assessed by telephone contact.

This study was approved by the Research Ethics Committee of our institution, according to the Declaration of Helsinki.

Statistical analysis

Continuous variables with parametric distribution were expressed as mean ± standard deviation, while nonparametric variables were expressed as median and interquartile range (IR). For statistical analysis we used the Student's t-test for continuous variables with parametric distribution, Mann-Whitney test for those with nonparametric distribution and chi-square test for categorical data. Cumulative mortality rates were calculated according to the Kaplan-Meier method and Cox-Mantel test.

Results

Of 116 patients who fulfilled inclusion criteria, 102 (87.9%) were successfully contacted by telephone and comprised our study group. Mean age was 63.5 ± 10.9 years and 72.5% were male. Etiology of dilated cardiomyopathy was ischemic in 65 patients (63.7%) and nonischemic in the remaining 37 (36.3%). Sixty-nine patients (67%) were in NYHA class II and median LVEF was 25% (IR 21 to 30). With regard to standard treatment for CHF, 89.7% of patients were receiving ACEI, 69.2% beta-blockers, 43.2% spironolactone, 35.3% digoxin and 45.9% amiodarone.

During a median follow-up of 19.6 months (range 3.3 to 74.3) 21 patients died (20.6%). Mortality rate was 23.1% among ischemic patients (15 p) and 16.2% in the nonischemic group (6 p); Cox-Mantel test p = 0.52. (Figure 1).

Comparison with SCD-HeFT patients

Baseline characteristics of both groups are summarized in Table 1. Of patients enrolled in the placebo arm of SCD-HeFT trial (n = 847), 655 (77.3%) were male vs 72.5% of patients included in this study; p = 0.27. Median age (60, IR 51 to 68) and LVEF (25%, IR 20 to 30) were similar to that of our patients. Seventy percent of SCD-HeFT patients were in NYHA class II vs 67% in our study. In both groups ischemic heart disease was the most frequent origin of dilated cardiomyopathy (53.5% in SCD-HeFT vs 63.7% in our population; OR 1.52; p = 0.05). Use of ACEI and beta-blockers was similar in both, whereas digoxin was more commonly used in SCD-HeFT patients.

Thumbnail

Mortality rate of patients enrolled in the placebo arm of SCD-HeFT trial was 28.8% after follow-up period (median 45.5 months) vs 20.6% of patients included in the present study. This difference was not statistically significant: OR 0.64 (CI 95% 0.38-1.05); p = 0.08. Kaplan-Meier estimates of actuarial death from any cause showed similar mortality curves for both populations and an annual mortality rate of 8.25% for our patients vs 7.25% for SCD-HeFT patients (p = NS). (Figure 2)

Discussion

Benefit from ICD therapy for secondary prevention of sudden death in patients with severe LVEF impairment has been shown in several trials^18-20. Early studies showing that ICD improves outcome for primary prevention of sudden death, enrolled patients with coronary artery disease and left ventricular dysfunction who had spontaneous unsustained ventricular tachycardia (VT) and inducible sustained VT at electrophysiological study (EPS)^21,22.

However, ventricular arrhythmia non-inducibility does not necessarily identify a subgroup of patients at lower risk of sudden cardiac death. This fact has been reported both in survivors of cardiac arrest^23-25 and in a recent MADIT-II substudy²⁶. Moreover, MUSTT investigators have shown that EP study is of limited value when LVEF is less than 30%²⁷. These findings confirm that negative predictive value of programmed ventricular stimulation is low in these patients.

MADIT-II trial enrolled patients with previous myocardial infarction and LVEF < 30%¹⁶. Despite ventricular arrhythmias or CHF symptoms not being inclusion criteria, 60% of patients were in NYHA class II-III. In this study, ICD therapy significantly reduced mortality by 31% (p = 0.016). MADIT-II was the first trial to demonstrate benefit from ICD therapy for primary prevention of sudden death in non-selected patients with ischemic dilated cardiomyopathy.

The AMIOVIRT trial examined the role of ICD for primary prevention in patients with non-ischemic severe left ventricular dysfunction and nonsustained VT²⁸. Compared to amiodarone, there was no difference in mortality, probably due to the small size of the patient sample.

The DEFINITE study enrolled patients with similar characteristics and showed a 35% reduction in mortality with ICD therapy, but this difference did not reach statistical significance (p = 0.08)²⁹. Upon more detailed analysis, however, a large reduction in arrhythmic death (RRR 80%; p = 0.006) was observed, whereas there were no differences in death from other causes.

The SCD-HeFT trial included patients with class II-III CHF and LVEF < 35% due to ischemic or non-ischemic causes¹⁵. Patients were randomized to ICD therapy, amiodarone or placebo in 1:1:1 fashion. The ICD arm was conservatively designed, implanting shock-only, single-lead devices. Because of the potential for antibradychardia pacing to worsen CHF, it was initiated only if the intrinsic rate decreased to less than 34 beats per minute. ICD therapy significantly reduced mortality by 23% compared to placebo, with 14 devices needed to implant to save one life. Although the benefit was observed both in ischemic and non-ischemic CHF, it was restricted to class II CHF patients. Amiodarone had no effect on survival.

In the present study we compared the outcomes of CHF patients from our practice to those of the placebo arm of the SCD-HeFT trial who did not receive amiodarone. Although 45% of our patients received this antiarrhythmic drug, mortality rate was similar in the amiodarone and placebo groups of the SCD-HeFT trial. Therefore, it appears safe to assume that this difference in treatment does not invalidate the comparison at the center of our study. Even though the median follow-up period was different (45.5 months in SCD-HeFT vs 19.6 months in our patients), the maximum follow-up was similar in both groups (72.6 vs 74.3).

In regard to the remaining medical therapy, there was a larger use of digoxin in SCD-HeFT patients (probably due to a greater proportion of nonischemic CHF), with no different use of other medications in both groups. Although prescription of ACEI and beta-blockers was very similar to that of SCD-HeFT patients, it was clearly greater than that derived from epidemiological data. Indeed, the EuroHeart Failure Survey showed an underuse of recommended medications both in terms of proportion of patients receiving them and the doses employed. This applies particularly to beta-blockers, which were prescribed in only 37% of patients included in that survey³⁰. This difference in treatment reinforces the need for development of specialized areas of health care, such as a Heart Failure Program.

Chagas disease was found to be the etiology of cardiomyopathy in 8/37 patients (21%) of the nonischemic group and it would have been interesting to analyze this population. Owing to the small size of this sample (8% of the whole study group) we considered that those results may not be reliable.

These results show that CHF patients we usually treat have similar characteristics to those in the SCD-HeFT trial. They further argue against the notion that patients included in large clinical trials are not representative of our domestic population, and suggest that our patients might also benefit from ICD therapy. However, it may not be appropriate to extend these results to the general population with CHF, considering that ACEI and beta-blockers remain under-prescribed, as mentioned above. Moreover, MADIT-II authors have suggested that benefits from ICD therapy may be lower in patients who do not receive beta-blockers or who receive suboptimal doses of these drugs³¹. Likewise, it would be interesting to know the outcomes of MADIT-II-like patients who had not been selected according to the presence of CHF or ventricular arrhythmias.

According to study design, clinical characteristics of our patients were collected retrospectively, and the end point (death from any cause) was assessed by telephone contact. Although 14 patients lacking a telephone line could not be followed, their clinical variables did not differ from those of the 102 patients analyzed. Follow-up was completed in 88% of patients, allowing an appropriate data analysis.

Economic impact is probably the main reason precluding full adoption of ICD therapy in class II-III CHF patients in our health care system. Because these patients do not display ventricular arrhythmias and do not require antibradychardia pacing, SCD-HeFT authors proposed to conduct ICD implantation on an outpatient basis, without defibrillation testing. Other authors have identified more than 5% of patients undergoing ICD implant for primary prevention with an inadequate safety margin, indicating that defibrillation threshold testing remains an important part of ICD implantation and should not be omitted³². Additionally, under our current health care system the device represents the single highest cost component of the procedure. Hence, outpatient insertion can be potentially risky with negligible benefits in terms of costs. In a cost-effectiveness substudy, SCD-HeFT authors reported that ICD therapy is economically acceptable for the North American Health Care System and suggested that implantation of ICD in class III CHF patients would not be cost-effective³³. Unfortunately there are no published data concerning cost-effectiveness analysis in our health care system. Perhaps in the future, as the ICD becomes more affordable, indication of this device for primary prevention of sudden death will not elicit the same level of debate that it currently generates in regard to treatment of NYHA class II-III CHF patients.

Conclusion

SCD-HeFT-like patients from our daily practice had no different mortality rate than that of patients enrolled in the placebo arm of the SCD-HeFT trial. These results may indicate that SCD-HeFT population is representative of our patients.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

There were no external funding sources for this study.

Study Association

This study is not associated with any graduation program.

References

1. CIBIS Investigators and Committees. A randomized trial of beta blockade in heart failure: the Cardiac Insufficiency Bisoprolol Study (CIBIS). Circulation. 1994; 90: 1765-73.
2. CIBIS II Investigators and Committees. The Cardiac Insufficiency Bisoprolol Study (CIBIS II): a randomized trial. Lancet. 1999; 353: 9-13.
3. MERIT-HF Investigators. Effect of metoprolol CR/XL in chronic heart failure: metoprolol CR/XL: randomized intervention trial in congestive heart failure (MERIT-HF). Lancet. 1999; 353: 2001-7.
4. Packer M, Coats AJ, Fowler MB, Katus H, Krum H, Mohacsi P, et al. Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med. 2001; 344: 1651-8.
5. Doval H, Ferrante D. Meta-analysis of trials of beta-blockers in heart failure. In: Doval HY, Tajer C. Evidencias en cardiología IV. 4th. ed. GEDIC; 2005. p. 445-70.
6. The CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure: results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med. 1987; 316: 1429-35.
7. The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fraction and congestive heart failure. N Engl J Med. 1991; 325: 293-302.
8. Garg R, Yusuf S, for the Collaborative Group on ACE Inhibitor Trials. Overview of randomized trials of angiotensin-converting enzyme inhibitors on mortality and morbidity in patients with heart failure. JAMA. 1995; 273: 1450-6.
9. The Acute Infarction Ramipril Efficacy (AIRE) Study Investigators. Effect of ramipril on mortality and morbidity of survivors of acute myocardial infarction with clinical evidence of heart failure. Lancet. 1993; 342: 821-8.
10. Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone Evaluation Study (RALES) Investigators. N Engl J Med. 1999; 341: 709-17.
11. Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, et al. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. EPHESUS Trial. N Engl J Med. 2003; 348: 1309-21.
12. Singh SN, Fletcher RD, Fisher SG, Singh BN, Lewis HD, Deedwania PC, et al. Survival trial of Antiarrhythmic Therapy in Congestive Heart Failure (CHF-STAT). Amiodarone in patients with congestive heart failure and asymptomatic ventricular arrhythmia. N Engl J Med. 1995; 333: 77-82.
13. Amiodarone Trials Meta Analysis Investigators. Effect of prophylactic amiodarone on mortality after acute myocardial infarction and in congestive heart failure: meta-analysis of individual data from 6500 patients in randomized trials. Lancet. 1997; 350: 1417-24.
14. Doval H, Nul D, Grancelli H, Perrone S, Bortman G, Curiel R. Randomized trial of low-dose amiodarone in severe congestive heart failure. Grupo de Estudio de la Sobrevida en la Insuficiencia Cardiaca en Argentina (GESICA). Lancet. 1994; 344: 493-8.
15. Bardy G, Lee K, Mark D, Poole J, Packer D, Boineau R, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. The Sudden Cardiac Death in Heart Failure (SCD-HeFT) Trial. N Engl J Med. 2005; 352: 225-37.
16. Moss A, Zareba W, Jackson Hall W, Klein H, Wilber D, Cannom D, et al. for the Multicenter Automatic Defibrillator Implantation Trial (MADIT) II Investigators. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Eng J Med. 2002; 346: 877-83.
17. ACC/AHA/ESC 2006 Guidelines for Management of Patients with Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death. J Am Coll Cardiol. 2006; 48: 1064-108.
18. The Antiarrhythmics versus Implantable Defibrillators (AVID) Investigators. A comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from near fatal ventricular arrhythmias. N Engl J Med. 1997; 337: 1576-83.
19. Connolly S, Gent M, Roberts R, Dorian P, Roy D, Sheldon R, et al. Canadian implantable defibrillator study (CIDS): a randomized trial of the implantable cardioverter defibrillator against amiodarone. Circulation. 2000; 101: 1297-302.
20. Kuck K, Cappato R, Siebels J, Ruppel R, for the CASH Investigators. Randomized comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from cardiac arrest. The Cardiac Arrest Study Hamburg (CASH). Circulation. 2000; 102: 748-54.
21. Moss A, Hall W, Cannom D, Daubert J, Higgins S, Klein H, et al. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmia. Multicenter Automatic Defibrillator Implantation Trial Investigators. N Eng J Med. 1996; 335: 1933-40.
22. Buxton A, Lee K, Fisher J, Josephson M, Prystowsky E, Hafley, et al. A randomized study of the prevention of sudden death in patients with coronary artery disease. Multicenter Unsustained Tachycardia Trial Investigators. N Engl J Med. 1999; 341: 1882-90.
23. Brodsky M, Mitchell L, Halperin B, Raitt M, Hallstrom A, AVID Investigators. Prognostic value of baseline electrophysiology studies in patients with sustained ventricular tachyarrhythmia: the Antiarrhythmics Versus Implantable Defibrillators (AVID) Trial. Am Heart J. 2002; 144: 478-84.
24. Wilber D, Garan H, Finkelstein D, Kelly E, Newell J, McGovern B, et al. Out-of-hospital cardiac arrest: use of electrophysiologic testing in the prediction of long-term outcome. N Engl J Med. 1988; 318: 19-24.
25. Roy D, Waxman H, Kienzle M, Buxton A, Marchlinski M, Josephson M. Clinical characteristics and long-term follow up of 119 survivors of out-of-hospital cardiac arrest: relation to inducibility at electrophysiologic testing. Am J Cardiol. 1983; 52: 969-74.
26. Daubert J, Zareba W, Hall J, Schuger C, Corsello A, Leon A, et al. Predictive value of ventricular arrhythmia inducibility for subsequent ventricular tachycardia or ventricular fibrillation in multicenter automatic defibrillator implantation trial (MADIT) II patients. J Am Coll Cardiol. 2006; 47: 98-107.
27. Buxton A, Lee K, Hafley G, Wyse G, Fisher J, Lehmann M, et al. for the MUSTT Investigators. Relation of ejection fraction and inducible ventricular tachycardia to mode of death in patients with coronary artery disease: an analysis of patients enrolled in the Multicenter Unsustained Tachycardia Trial. Circulation. 2002; 106: 2466-72.
28. Strickberger S, Hummel J, Bartlett T, Frumin H, Schuger C, Beau S, et al. Amiodarone versus implantable cardioverter-defibrillator: randomized trial in patients with nonischemic dilated cardiomyopathy and asymptomatic nonsustained ventricular tachycardiaAMIOVIRT. J Am Coll Cardiol. 2003; 41: 1707-12.
29. Kadish A, Dyer A, Daubert J, Quigg R, Estes N, Anderson K, et al. for the Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation (DEFINITE) Investigators. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med. 2004; 350: 2151-8.
30. Komajda M, Follath F, Swedberg K, Cleland J, Aguilar JC, Cohen-Solal A, et al. for the Study Group on Diagnosis of the Working Group on Heart Failure of the European Society of Cardiology. The EuroHeart Failure Survey programme - a survey on the quality of care among patients with heart failure in Europe. Part 2: Treatment. Eur Heart J. 2003; 24: 464-74.
31. Brodine WN, Tung RT, Lee JK, Hockstad ES, Moss AJ, Zareba W, et al. for the MADIT II Research Group. Effects of beta-blockers on implantable cardioverter defibrillator therapy and survival in the patients ischemic Cardiomyopathy (from the Multicenter Automatic Defibrillator Implantation Trial-II). Am J Cardiol. 2005; 96: 691-5.
32. Toal S, Harris L, Asta J, Hill A, Cameron D. Routine defibrillation threshold testing at implant in patients receiving defibrillators for primary prevention: is this necessary? (Abstract). Heart Rhythm. 2006; 3: S157.
33. Mark DB, Nelson CL, Anstrom KJ, Al-Khatib SM, Tsiatis AA, Cowper PA, et al. for the SCD-HeFT Investigators. Cost-effectiveness of defibrillator therapy or amiodarone in chronic stable heart failure. Circulation. 2006; 114: 135-42.