Implantable Defibrillators

1. CMS has determined that the evidence is adequate to conclude that an implantable cardioverter-defibrillator (ICD) is reasonable and necessary for the following:
   • Patients with ischemic dilated cardiomyopathy (IDCM), documented prior myocardial infarction (MI), New York Heart Association (NYHA) Class II and III heart failure, and measured left ventricular ejection fraction (LVEF) ≤ 35%;
   • Patients with nonischemic dilated cardiomyopathy (NIDCM) > 9 months, NYHA Class II and III heart failure, and measured LVEF ≤ 35%;
   • Patients who meet all current CMS coverage requirements for a cardiac resynchronization therapy (CRT) device and have NYHA Class IV heart failure;

   For each of these groups, the following additional criteria must also be met:

   1. Patients must be able to give informed consent;
   2. Patients must not have:
      • Cardiogenic shock or symptomatic hypotension while in a stable baseline rhythm;
      • Had a coronary artery bypass graft (CABG) or percutaneous transluminal coronary angioplasty (PTCA) within the past 3 months;
      • Had an acute MI within the past 40 days;
      • Clinical symptoms or findings that would make them a candidate for coronary revascularization;
      • Irreversible brain damage from preexisting cerebral disease;
      • Any disease, other than cardiac disease (e.g. cancer, uremia, liver failure), associated with a likelihood of survival less than one year;
   3. Ejection fractions must be measured by angiography, radionuclide scanning, or echocardiography;
   4. Myocardial infarctions must be documented and defined according to the consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction;¹
   5. The beneficiary receiving the ICD implantation for primary prevention is enrolled in either an FDA-approved category B IDE clinical trial (42 CFR §405.201), a trial under the CMS Clinical Trial Policy (NCD Manual §310.1) or a qualifying data collection system including approved clinical trials and registries. Initially, an ICD database will be maintained using a data submission mechanism that is already in use by Medicare participating hospitals to submit data to the Iowa Foundation for Medical Care (IFMC)a Quality Improvement Organization (QIO) contractorfor determination of reasonable and necessary and quality improvement. Initial hypothesis and data elements are specified in this decision (Appendix VI) and are the minimum necessary to ensure that the device is reasonable and necessary. Data collection will be completed using the ICDA (ICD Abstraction Tool) and transmitted via QNet (Quality Network Exchange) to the IFMC who will collect and maintain the database. Additional stakeholder-developed data collection systems to augment or replace the initial QNet system, addressing at a minimum the hypotheses specified in this decision, must meet the following basic criteria:
      1. Written protocol on file;
      2. Institutional Review Board review and approval, if required;
      3. Scientific review and approval by two or more qualified individuals who are not part of the research team;
      4. Certification that investigators have not been disqualified.
      For purposes of this coverage decision, CMS will determine whether specific registries or clinical trials meet these criteria.
   6. Providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient medical record.
   
   
2. CMS has determined that the evidence, though less compelling at this time, is adequate to conclude that an ICD is reasonable and necessary for patients with NIDCM > 3 months, NYHA Class II or III heart failure, and measured LVEF ≤ 35%, only if the following additional criteria are also met:
   1. Patients must be able to give informed consent;
   2. Patients must not have:
      • Cardiogenic shock or symptomatic hypotension while in a stable baseline rhythm;
      • Had a CABG or PTCA within the past 3 months;
      • Had an acute MI within the past 40 days;
      • Clinical symptoms or findings that would make them a candidate for coronary revascularization;
      • Irreversible brain damage from preexisting cerebral disease;
      • Any disease, other than cardiac disease (e.g. cancer, uremia, liver failure), associated with a likelihood of survival less than one year;
   3. Ejection fractions must be measured by angiography, radionuclide scanning, or echocardiography;
   4. Myocardial infarctions must be documented and defined according to the consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction;²
   5. The beneficiary receiving the ICD implantation for this indication is enrolled in either an FDA-approved category B IDE clinical trial (42 CFR §405.201), a trial under the CMS Clinical Trial Policy (NCD Manual §310.1) or a prospective data collection system meeting the following basic criteria:
      1. Written protocol on file;
      2. Institutional Review Board review and approval;
      3. Scientific review and approval by two or more qualified individuals who are not part of the research team;
      4. Certification that investigators have not been disqualified.
      For purposes of this coverage decision, CMS will determine whether specific registries or clinical trials meet these criteria.
   6. Providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient medical record.

All other indications for ICDs not currently covered in accordance with this decision will continue to be covered under Category B IDE trials (42 CFR 405.201) and the CMS routine clinical trials policy (CIM 30-1, NCD 130.1).

To:		Administrative File: CAG 00157R3  
		Implantable Defibrillators  
  
From:	Steve E. Phurrough, MD, MPA  
		Director, Coverage and Analysis Group  
  
		Marcel E. Salive, MD, MPH  
		Director, Division of Medical & Surgical Services  
  
		JoAnna F. Baldwin  
		Lead Analyst, Division of Medical & Surgical Services  
  
		Joseph Chin, MD, MS  
		Medical Officer, Division of Medical & Surgical Services  
  
Subject:		Coverage Decision Memorandum for Implantable Cardioverter Defibrillators  
  
Date:		January 27, 2005

I.    Decision

1. CMS has determined that the evidence is adequate to conclude that an implantable cardioverter-defibrillator (ICD) is reasonable and necessary for the following:
   • Patients with ischemic dilated cardiomyopathy (IDCM), documented prior myocardial infarction (MI), New York Heart Association (NYHA) Class II and III heart failure, and measured left ventricular ejection fraction (LVEF) ≤ 35%;
   • Patients with nonischemic dilated cardiomyopathy (NIDCM) > 9 months, NYHA Class II and III heart failure, and measured LVEF ≤ 35%;
   • Patients who meet all current CMS coverage requirements for a cardiac resynchronization therapy (CRT) device and have NYHA Class IV heart failure;

   For each of these groups, the following additional criteria must also be met:

   1. Patients must be able to give informed consent;
   2. Patients must not have:
      • Cardiogenic shock or symptomatic hypotension while in a stable baseline rhythm;
      • Had a coronary artery bypass graft (CABG) or percutaneous transluminal coronary angioplasty (PTCA) within the past 3 months;
      • Had an acute MI within the past 40 days;
      • Clinical symptoms or findings that would make them a candidate for coronary revascularization;
      • Irreversible brain damage from preexisting cerebral disease;
      • Any disease, other than cardiac disease (e.g. cancer, uremia, liver failure), associated with a likelihood of survival less than one year;
   3. Ejection fractions must be measured by angiography, radionuclide scanning, or echocardiography;
   4. Myocardial infarctions must be documented and defined according to the consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction;¹
   5. The beneficiary receiving the ICD implantation for primary prevention is enrolled in either an FDA-approved category B IDE clinical trial (42 CFR §405.201), a trial under the CMS Clinical Trial Policy (NCD Manual §310.1) or a qualifying data collection system including approved clinical trials and registries. Initially, an ICD database will be maintained using a data submission mechanism that is already in use by Medicare participating hospitals to submit data to the Iowa Foundation for Medical Care (IFMC)a Quality Improvement Organization (QIO) contractorfor determination of reasonable and necessary and quality improvement. Initial hypothesis and data elements are specified in this decision (Appendix VI) and are the minimum necessary to ensure that the device is reasonable and necessary. Data collection will be completed using the ICDA (ICD Abstraction Tool) and transmitted via QNet (Quality Network Exchange) to the IFMC who will collect and maintain the database. Additional stakeholder-developed data collection systems to augment or replace the initial QNet system, addressing at a minimum the hypotheses specified in this decision, must meet the following basic criteria:
      1. Written protocol on file;
      2. Institutional Review Board review and approval, if required;
      3. Scientific review and approval by two or more qualified individuals who are not part of the research team;
      4. Certification that investigators have not been disqualified.
      For purposes of this coverage decision, CMS will determine whether specific registries or clinical trials meet these criteria.
   6. Providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient medical record.
   
   
2. CMS has determined that the evidence, though less compelling at this time, is adequate to conclude that an ICD is reasonable and necessary for patients with NIDCM > 3 months, NYHA Class II or III heart failure, and measured LVEF ≤ 35%, only if the following additional criteria are also met:
   1. Patients must be able to give informed consent;
   2. Patients must not have:
      • Cardiogenic shock or symptomatic hypotension while in a stable baseline rhythm;
      • Had a CABG or PTCA within the past 3 months;
      • Had an acute MI within the past 40 days;
      • Clinical symptoms or findings that would make them a candidate for coronary revascularization;
      • Irreversible brain damage from preexisting cerebral disease;
      • Any disease, other than cardiac disease (e.g. cancer, uremia, liver failure), associated with a likelihood of survival less than one year;
   3. Ejection fractions must be measured by angiography, radionuclide scanning, or echocardiography;
   4. Myocardial infarctions must be documented and defined according to the consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction;²
   5. The beneficiary receiving the ICD implantation for this indication is enrolled in either an FDA-approved category B IDE clinical trial (42 CFR §405.201), a trial under the CMS Clinical Trial Policy (NCD Manual §310.1) or a prospective data collection system meeting the following basic criteria:
      1. Written protocol on file;
      2. Institutional Review Board review and approval;
      3. Scientific review and approval by two or more qualified individuals who are not part of the research team;
      4. Certification that investigators have not been disqualified.
      For purposes of this coverage decision, CMS will determine whether specific registries or clinical trials meet these criteria.
   6. Providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient medical record.

All other indications for ICDs not currently covered in accordance with this decision will continue to be covered under Category B IDE trials (42 CFR 405.201) and the CMS routine clinical trials policy (CIM 30-1, NCD 130.1).

II.   Background

In June 2003, CMS released a National Coverage Determination (NCD) on ICDs that expanded coverage to specific patient populations for both primary and secondary prevention of sudden cardiac death.³ Since our prior decision, new evidence has been presented and/or published on the use of ICDs in primary prevention.

In March 2004, CMS received a request from Medtronic Inc. for reconsideration of the prior decision on ICDs, based largely upon the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) results. SCD-HeFT was a prospective, randomized trial to test the hypothesis that “amiodarone or a conservatively programmed shock-only, single-lead ICD would decrease the risk of death from any cause in a broad population of patients with mild-to-moderate heart failure.”⁴ It included patients with NYHA Class II and III heart failure, LVEF ≤ 35%, and nonischemic dilated cardiomyopathy (NIDCM) in addition to patients with ischemic dilated cardiomyopathy (IDCM).^5,6 Since our prior decision limited coverage to patients with ICDM, LVEF ≤ 30% and prolonged QRS > 120 milliseconds, this request for reconsideration includes assessment of etiology of dilated cardiomyopathy, level of LVEF and QRS duration. As noted in the prior decision, the use of ICDs for the secondary prevention of SCD has been well studied and accepted. Therefore, secondary prevention will not be further evaluated in this memorandum.

Based on the National Health and Nutrition Examination Survey III, it has been estimated that 5 million Americans have congestive heart failure.⁷ Congestive heart failure (CHF) is often a symptom of dilated cardiomyopathy, which is “characterized by ventricular remodeling that produces chamber dilation, with normal or decreased wall thickness, and diminution in systolic function.”⁸ Dilated cardiomyopathy may be further classified by etiology into IDCM (due to prior myocardial infarction and coronary artery disease) and NIDCM (due to non-ischemic conditions such as infections, inflammation, familial or genetic conditions or idiopathic causes); however, these two categories are not necessarily mutually exclusive.

Since there are various types of commercially available ICDs, a closer look at the type of ICD used in the various trials may provide insight into the necessary features of the device suitable for primary prevention. In general, current ICDs can be classified into single-chamber devices and dual-chamber devices, depending on the number of leads and lead placement. Single-chamber ICDs typically have one lead that ends in the right ventricle. Dual-chamber ICDs have at least one additional lead, usually in the right atrium. Once a ventricular tachyarrhythmia is detected, two methods may be used to treat the arrhythmia – antitachycardia pacing and direct current shocks.⁹ The ICD can deliver one or more bursts of pacing to end ventricular tachycardia and restore a normal rate and rhythm. ICDs can also deliver “either synchronized, usually low-energy shocks (less than 5J) or unsynchronized high-energy shocks.”¹⁰ Another feature of some ICDs is antibradycardia pacing; however, the MADIT II¹¹ and DAVID¹² trials suggested that antibradycardia pacing may not be necessary and in some instances, may be detrimental. In SCD-HeFT single-lead defibrillators were used.

A closely related device is the combined cardiac ventricular resynchronization and defibrillator (CRT-D), also referred to as a combined biventricular pacing and defibrillator device. Cardiac ventricular resynchronization refers to pacing techniques that aim to alter the amount of intraventricular asynchrony, most commonly identified by prolonged QRS interval. Although cardiac resynchronization therapy (CRT) devices were not completely reviewed in this decision, CRT-D devices were considered since these devices also have defibrillator functions.

III.    History of Medicare Coverage

The Centers for Medicare & Medicaid Services (CMS), issued a Medicare National Coverage Determination (NCD) in 1986 providing limited coverage of implantable defibrillators. The policy has expanded over the years with revisions in 1991, 1999 and 2003. The most recent expansion of coverage is discussed in the June 6, 2003 decision memorandum that is available on our web site at www.cms.gov/coverage. This decision became effective for services on or after October 1, 2003 and expanded coverage to patients with a previous myocardial infarction, low ejection fraction and a wide QRS interval. The policy was also expanded to include coverage to patients enrolled in an Investigational Device Exemption Category B device trial. A follow up decision memorandum to clarify this specific aspect of the policy was published March 12, 2004 and is also available on our web site.

The benefit category for ICDs has been previously determined to fall within the prosthetic devices category.

On March 30, 2004, CMS accepted a request from Medtronic Inc. to expand coverage for ICDs. Medtronic Inc. made this request based on the results of the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) and specifically requested that Medicare expand coverage to the trial population. On December 16, 2004, CMS announced its concern on the absence of publication of the SCD-HeFT data and the potential for closing the NCD. Publication did not occur prior to the December 28, 2004 deadline so on that day CMS posted a final DM that continued current coverage. CMS opened a reconsideration of that decision on December 29, 2004 in anticipation of the SCD-HeFT publication and announced that a final decision based on SCD-HeFT data would be made shortly after publication.

Since an NCD already exists for ICDs, this review is a reconsideration of the current policy. The current policy is:

1. Covered Indications
   1)  Documented episode of cardiac arrest due to ventricular fibrillation (VF), not due to a transient or reversible cause (effective July 1, 1991);
   2)  Documented sustained ventricular tachyarrhythmia (VT), either spontaneous or induced by an electrophysiology (EP) study, not associated with an acute myocardial infarction (MI) and not due to a transient or reversible cause (effective July 1, 1999);
   3) Documented familial or inherited conditions with a high risk of life-threatening VT, such as long QT syndrome or hypertrophic cardiomyopathy (effective July 1, 1999);
   

   Additional indications effective for services performed on or after October 1, 2003:

   4) Coronary artery disease with a documented prior MI, a measured left ventricular ejection fraction ≤ 0.35, and inducible, sustained VT or VF at EP study. (The MI must have occurred more than 4 weeks prior to defibrillator insertion. The EP test must be performed more than 4 weeks after the qualifying MI.);
   5) Documented prior MI and a measured left ventricular ejection fraction ≤ 0.30 and a QRS duration of > 120 milliseconds.  Patients must not have:

   1. New York Heart Association classification IV;
   2. Cardiogenic shock or symptomatic hypotension while in a stable baseline rhythm; 
   3. Had a coronary artery bypass graft (CABG) or percutaneous transluminal coronary angioplasty (PTCA) within the past 3 months;
   4. Had an enzyme-positive MI within the past month;
   5. Clinical symptoms or findings that would make them a candidate for coronary revascularization;
   6. Any disease, other than cardiac disease (e.g. cancer, uremia, liver failure), associated with a likelihood of survival less than 1 year.
2. All patients considered for implantation of a defibrillator must not have irreversible brain damage, disease or dysfunction that precludes the ability to give informed consent.
3. Myocardial infarctions must be documented by elevated cardiac enzymes or Q-waves on an electrocardiogram.  Left ventricular ejection fractions must be measured by angiography, radionuclide scanning, or echocardiography.
4. All other indications remain noncovered except in Category B IDE clinical trials (42 CFR 405.201) or as a routine cost in clinical trials defined under the NCD Manual §310.

IV.    Timeline of Recent Activities

V.    FDA Status

The FDA approved the first implantable defibrillator in 1985 while the first implantable cardioverter defibrillators were approved in 1988 and 1989.¹³ The FDA approves each device individually and has granted premarket approvals (PMA)¹⁴ for implantable defibrillators for the indications of providing antitachycardia pacing and ventricular defibrillation for automated treatment of life-threatening ventricular arrhythmias.

CMS assesses relevant health outcomes, above and beyond the safety and effectiveness regulatory mandate of the FDA. Although a device must receive FDA approval or clearance for at least one indication to be eligible for Medicare coverage, except for a category B device under an investigational device exemption (IDE) clinical trial (60 FR 48417, September 19, 1995), FDA approval/clearance alone does not entitle that device to coverage. The device must fall under a Medicare benefit category and be determined to be reasonable and necessary for the diagnosis or treatment of an illness or injury or to improve the functioning of a malformed body member to be covered by CMS. CMS has the authority to conduct a separate assessment of a device’s appropriateness for Medicare coverage, including whether it is reasonable and necessary specifically for its intended use for Medicare beneficiaries (see e.g., 60 FR 48417, 48420 September 19, 1995). Under a premarket approval application (PMA) review, the FDA determines whether or not there is reasonable assurance of safety and effectiveness for the device’s intended use that is stated in its proposed labeling. Medicare NCDs consider the medical benefit and clinical utility of an item or service in determining whether the item or service is considered reasonable and necessary under the Medicare program. CMS determines whether or not the intervention improves net health outcomes in the Medicare population at least as well as established treatments. Thus, FDA PMA approval by itself is not sufficient for making a determination concerning Medicare coverage.

As we similarly stated in 66 FR 58788, 58797 (November 23, 2001) with regard to FDA 510(k) clearance, "[t]he criteria the FDA uses in making determinations related to substantial equivalency under section 510(k) of the Food, Drug, and Cosmetic Act is significantly different from the scientific evidence we consider in making "reasonable and necessary" determinations under Medicare. FDA does not necessarily require clinical data or outcomes studies in making a determination of substantial equivalency for the purpose of device approval under section 510(k) of the Food, Drug, and Cosmetic Act. Medicare NCDs consider medical benefit and clinical utility of an item or service in determining whether the item or service is considered reasonable and necessary under the Medicare program. Thus, a substantial equivalency approval under section 510(k) of FDA is not sufficient for making determination concerning Medicare coverage."

VI.   General Methodological Principles

When making NCDs, CMS evaluates relevant clinical evidence to determine whether or not the evidence is of sufficient quality to support a finding that an item or service is reasonable and necessary. The overall objective for the critical appraisal of the evidence is to determine to what degree we are confident that: 1) the specific assessment questions can be answered conclusively; and 2) the intervention will improve net health outcomes for patients.

A detailed account of the methodological principles of study design the agency staff utilizes to assess the relevant literature on a therapeutic or diagnostic item or service for specific conditions can be found in Appendix III. In general, features of diagnostic studies that improve quality and decrease bias include the selection of a clinically relevant inception cohort, the consistent use of a single good reference standard, the inclusion of patients with and without the disorder in question, and the blinding of readers of the index test and of reference test results.¹⁵

VII.    Evidence

A. Introduction

There have been numerous trials on the use of defibrillators to prevent sudden cardiac death. These trials have predominantly used mortality as the primary outcome. We have thus focused this reconsideration on mortality evidence published since our prior decision in June 2003.

B. Discussion of evidence reviewed

1. Questions

The development of an assessment in support of Medicare coverage decisions is based on the same general question for almost all requests: “Is the evidence sufficient to conclude that the application of the technology under study will improve final health outcomes for Medicare patients?”

The formulation of specific questions for the assessment recognizes that the effect of an intervention can depend substantially on how it is delivered, to whom it is applied, the alternatives with which it is being compared and the delivery setting. In this reconsideration, CMS sought to address the following questions:

• Is there evidence to conclude that ICDs decrease mortality for patients with ischemic dilated cardiomyopathy (IDCM) and reduced LVEF?


• Is there evidence to conclude that ICDs decrease mortality for patients with nonischemic dilated cardiomyopathy (NIDCM) and reduced LVEF?

2. External technology assessments

There was no external technology assessment commissioned.

3. Internal technology assessments

Medline was searched iteratively from 2002 using the following keywords: defibrillator with and without implantable. Studies on animal subjects and reports in languages other than English were excluded. Six original randomized clinical trials and several review articles were reviewed and classified into primary prevention and secondary prevention.

Bansch D, Antz M, Boczor S, et al. Primary prevention of sudden cardiac death in idiopathic dilated cardiomyopathy: the Cardiomyopathy Trial (CAT). Circulation 2002;105:1453-1458.

The Cardiomyopathy Trial (CAT) was a randomized trial to compare ICD therapy to standard medical therapy in patients with idiopathic dilated cardiomyopathy and impaired ejection fraction. Inclusion criteria were idiopathic dilated cardiomyopathy ≤ 9 months, LVEF ≤ 30%, NYHA Class II or III. Exclusion criteria included coronary artery disease, prior myocardial infarction, history of VT or VF, and significant valvular disease. The primary endpoint was all-cause mortality at 1 year of follow-up.¹⁶

A total of 104 patients were enrolled in the CAT pilot phase: 50 patients were randomly assigned to ICD treatment and 54 to the control group. Mean follow-up was 5.5 years. Mean age was 52 years. Men comprised 83% of the study population. Mean LVEF was 24%. Approximately 65% of the patients had NYHA Class II heart failure and 35% had NYHA Class III.

There were a total of 30 deaths: 13 in the ICD group compared to 17 in the control group (p-value = 0.554). The investigators concluded “this trial did not provide evidence in favor of prophylactic ICD implantation in patients with DCM of recent onset and impaired left ventricular ejection fraction.”¹⁷

In this study, Kaplan-Meier and Cox proportional regression analyses were used. No patients with NYHA Class IV were enrolled. Eleven patients received adequate ICD therapy. The sample size was relatively small. The authors found that “short- and long-term overall mortality rates in patients with DCM and significantly impaired LV function were surprisingly low.”¹⁸ The trial was stopped “after the inclusion of 104 patients because the all-cause mortality at 1 year did not reach the expected 30% in the control group.”¹⁹ The authors further noted, “even if 1348 patients had been included, as initially planned, the trial would have been underpowered.”²⁰

Strickberger SA, Hummel JD, Bartlett TG, et al. Amiodarone versus implantable cardioverter-defibrillator: randomized trial in patients with nonischemic dilated cardiomyopathy and asymptomatic nonsustained ventricular tachycardia – AMIOVIRT. J Am Coll Cardiol 2003;41:1707-1712.

The Amiodarone Versus Implantable Cardioverter-Defibrillator Trial (AMIOVIRT) was a prospective, randomized study “to compare the total mortality rates of patients with non-ischemic dilated cardiomyopathy (NIDCM)²¹ and asymptomatic nonsustained ventricular tachyarrhythmia (NSVT) who were randomized to therapy with amiodarone or an ICD.”²² Inclusion criteria were nonischemic dilated cardiomyopathy, asymptomatic non-sustained ventricular tachycardia, LVEF ≤ 35% and NYHA Class I-III. Exclusion criteria included syncope, pregnancy, a contraindication to amiodarone or ICD, therapy with class I antiarrhythmic drug, or NIDCM within 6 months. The primary endpoint was total mortality.

A total of 103 patients were enrolled: 51 patients were randomly assigned to ICD treatment and 52 to amiodarone. Mean follow-up was 2.0 years. Mean age was 59 years. Men comprised 70% of the study population. Mean LVEF was 23%. Approximately 63% of the patients had NYHA Class II and 20% had Class III.

There were a total of 13 deaths: 6 in the ICD group compared to 7 in the amiodarone group (p-value = 0.8). The investigators reported: “The study was stopped when the prospective stopping rule futility was reached. The percent of patients surviving at one year (90% vs. 96%) and three years (88% vs. 87%) in the amiodarone and ICD groups, respectively, were not statistically different (p = 0.8).”²³

In this study, there was no standard medical therapy group. The authors further stated: “The relatively small sample size in AMIOVIRT permits a mortality difference observed with a power of only 3%. The study was stopped due to the prospective rule used to identify inability to differentiate between ICD and amiodarone therapy.”²⁴

Kadish A, Dyer A, Daubert JP, et al. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med 2004;350:2152-2158.

The Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation (DEFINITE) trial was a prospective, randomized study to test “the hypothesis that an ICD will reduce the risk of death in patients with nonischemic cardiomyopathy and moderate-to-severe left ventricular dysfunction.”²⁵ Inclusion criteria were a LVEF < 36%, the presence of ambient arrhythmias,²⁶ a history of symptomatic heart failure and the presence of nonischemic dilated cardiomyopathy. Patients were excluded if they had New York Heart Association (NYHA) Class IV congestive heart failure, were not candidates for the implantation of a cardioverter–defibrillator, had undergone electrophysiological testing within the prior three months, or had permanent pacemakers. Patients in whom cardiac transplantation appeared to be imminent, those with familial cardiomyopathy associated with sudden death and patients with acute myocarditis or congenital heart disease were also excluded. The primary endpoint was death from any cause. The secondary endpoint was sudden death from arrhythmia.

A total of 458 patients were enrolled: 229 patients were randomly assigned to receive standard medical therapy and 229 to receive standard medical therapy plus a single-chamber ICD. Mean follow up was 29.0 months. Mean age was 58.3 years. Men comprised 71.2% of the study population. Mean LVEF was 21%. Approximately 57% of the patients had NYHA Class II heart failure.

“There were 68 deaths: 28 in the ICD group, as compared with 40 in the standard-therapy group (hazard ratio, 0.65; 95 percent confidence interval, 0.40 to 1.06; P=0.08). The mortality rate at two years was 14.1 percent in the standard-therapy group (annual mortality rate, 7 percent) and 7.9 percent in the ICD group. There were 17 sudden deaths from arrhythmia: 3 in the ICD group, as compared with 14 in the standard therapy group (hazard ratio, 0.20; 95 percent confidence interval, 0.06 to 0.71; P=0.006).”²⁷

The authors concluded, “in patients with severe, nonischemic dilated cardiomyopathy who were treated with ACE inhibitors and beta-blockers, the implantation of a cardioverter–defibrillator significantly reduced the risk of sudden death from arrhythmia and was associated with a nonsignificant reduction in the risk of death from any cause.”²⁸

In this study, FDA approved, single chamber devices were used. The ICDs were programmed to back up VVI pacing at a rate of 40 beats per minute and to detect ventricular fibrillation at a rate of 180 beats per minute. Forty-one patients (17.9%) received appropriate ICD shocks. Forty-nine patients (21.4%) received inappropriate ICD shocks, primarily for atrial fibrillation or sinus tachycardia.

Bristow MR, Saxon LA, Boehmer J, et al. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140-2150.

The Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) trial was a randomized trial to test the hypothesis that prophylactic cardiac-resynchronization therapy with a pacemaker (CRT) or with a pacemaker-defibrillator (CRT-D) would reduce the risk of death and hospitalization among patients with advanced chronic heart failure and intraventricular conduction delays. Enrollment criteria included New York Heart Association (NYHA) Class III or IV heart failure from either ischemic or nonischemic cardiomyopathy, LVEF ≤ 0.35, an electrocardiographically measured QRS interval ≥ 120 msec and a PR interval > 150 msec, sinus rhythm, no clinical indication for a pacemaker or implantable defibrillator, and a hospitalization for the treatment of heart failure or the equivalent in the preceding 12 months. Exclusion criteria included myocardial infarction within 60 days of randomization, CAD with surgical or percutaneous correction within 60 days of randomization, progressive or unstable angina, uncontrolled blood pressure and surgically uncorrected primary valvular heart disease.²⁹ The primary endpoint was a composite of “all-cause mortality and all-cause hospitalization, in which all-cause mortality is defined as death from all causes and all-cause hospitalization is defined as admission to a hospital for any reason.”³⁰ The secondary endpoints included all-cause mortality and cardiac morbidity.

A total of 1,520 patients were enrolled: 308 were randomly assigned to receive optimal pharmacologic therapy, 617 to optimal pharmacologic therapy plus CRT and 595 to optimal pharmacologic therapy plus CRT-D. Median follow-up for the primary endpoint was 11.9 months in the pharmacologic-therapy group, 16.2 months in the pacemaker group and 15.7 months in the pacemaker–defibrillator group (significantly longer for the groups that received devices compared to medical therapy). Mean age was approximately 67 years. Men comprised approximately 68% of the study population. Mean LVEF was approximately 21%. Approximately 55% of patients had ischemic cardiomyopathy.

For the primary endpoint, the 12-month rate of death from any cause or hospitalization for any cause was 68 percent in the pharmacologic therapy group as compared with 56 percent in the CRT group (hazard ratio = 0.81; 95% CI 0.69 - 0.96; p-value = 0.014) and 56 percent in the CRT-D group (hazard ratio = 0.80; 95% CI 0.68 - 0.95; p-value = 0.010).

For the secondary endpoint, there were 77 death (25%) in the pharmacologic therapy group, 131 deaths (21%) in the CRT group (hazard ratio = 0.76; 95% CI 0.58 - 1.01; p-value = 0.0059), and 105 deaths (18%) in the CRT-D group (hazard ratio = 0.64; 95% CI 0.48 - 0.86; p-value = 0.003).

The authors concluded, “in patients with advanced heart failure and a prolonged QRS interval, cardiac-resynchronization therapy decreases the combined risk of death from any cause or first hospitalization and, when combined with an implantable defibrillator, significantly reduces mortality.”³¹

In this study, there was a 1:2:2 randomization weighted towards device therapy. Kaplan-Meier analysis was used. In the pharmacologic therapy group, 26% of the patients withdrew from the study. Subsequently the study was modified to attempt to address this issue. The authors reported: “For the mortality endpoint analysis, data on patients whose vital status was not known at the end of the study were censored on the date of the last known contact.”³² The definition of all-cause hospitalization was changed after the trial had started, potentially influencing the primary endpoint of the trial.

Hohnloser SH, Kuck KH, Dorian R, et al. Prophylactic use of an implantable cardioverter-defibrillator after acute myocardial infarction. N Engl J Med 2004;351:2481-2488.³³

The Defibrillator in Acute Myocardial Infarction Trial (DINAMIT) was a randomized trial to test the hypothesis that an ICD will reduce the risk of death in patients with recent MI who are at high risk of arrhythmic death due to extensive myocardial scarring (low LVEF) and autonomic imbalance (low heart rate variability or high resting heart rate). Inclusion criteria included recent acute MI (6-40 days), LVEF ≤ 35%, and depressed HRV or elevated heart rate. Exclusion criteria included NYHA class IV heart failure, CABG surgery or 3-vessel PTCA since qualifying MI, and candidates for heart transplant. The primary outcome was mortality from all causes. Secondary outcomes included arrhythmic death, quality of life and cost effectiveness.

A total of 674 patients were enrolled; 332 were randomly assigned to ICD therapy and 342 to conventional medical therapy. Mean follow-up was approximately 30 months. Mean age was 62 years. Men comprised 76% of the study population. Mean LVEF was about 28%.

For the primary endpoint, there were 62 deaths in the ICD group compared to 58 deaths in the conventional therapy group (hazard ratio = 1.08; 95% CI=0.76-1.55; p-value = 0.66). For the secondary endpoint, there were 12 arrhythmic deaths in the ICD group compared to 29 deaths in the conventional therapy group (hazard ratio = 0.42; 95% CI=0.22-0.83; p-value = 0.009). The authors concluded: “Prophylactic ICD therapy does not reduce overall mortality in high-risk patients who have recently had a myocardial infarction. Although ICD therapy was associated with a reduction in the rate of death due to arrhythmia, that was offset by an increase in the rate of death from nonarrhythmic causes.³⁴

Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352:225-237.

SCD-HeFT was a prospective, randomized controlled trial to test the hypothesis that “amiodarone or a conservatively programmed shock-only, single-lead ICD would decrease the risk of death from any cause in a broad population of patients with mild-to-moderate heart failure.”³⁵ The primary endpoint was all cause mortality based on a minimum of 2.5 years of follow-up. The secondary endpoints included arrhythmic cardiac mortality, non-arrhythmic cardiac mortality, morbidity, cost-effectiveness, and quality of life.

Patient inclusion criteria included heart failure duration of at least 3 months, stable symptomatic CHF (NYHA class II and class III) due to ischemic or nonischemic dilated cardiomyopathy, CHF present for at least 3 months treated with a vasodilator, and LVEF ≤ 35%. Exclusion criteria included asymptomatic patients; history of cardiac arrest; restrictive, infiltrative or hypertrophic cardiomyopathy; and unexplained syncope.

A total of 2,521 patients were enrolled; 847 were randomly assigned to placebo plus conventional heart failure therapy, 845 to amiodarone plus conventional heart failure therapy, and 829 to single lead ICD plus conventional heart failure therapy. Randomization was stratified by cause of heart failure and NYHA class. Median follow-up was 45.5 months. Median age was about 60 years. Men comprised 77% of the study population. Median LVEF at baseline was 25%. Approximately 52% of patients had ischemic cardiomyopathy. About 70% of patients were in NYHA Class II and 30% in NYHA Class III.

For the primary endpoint at the end of the follow-up period, there was a significant reduction in mortality in the ICD group compared to the placebo group (hazard ratio compared to control = 0.77; 97.5% CI = 0.62-0.96; p-value = 0.007). ³⁶

For patients with IDCM, there was a reduction in mortality hazard ratio for ICD therapy compared to control but it was not statistically significant (hazard ratio = 0.79; 97.5% CI = 0.60-1.04). For patients with NIDCM, there was a reduction in the mortality hazard ratio for ICD therapy compared to control but it was also not statistically significant (hazard ratio = 0.73; 97.5% CI = 0.50-1.07).³⁷ Kaplan-Meier survival curves by etiology of CHF are shown in Appendix IV. There also were several hazard ratio analyses performed by the investigators, as shown in Appendix V. For device therapy, the authors noted: “Of the 829 patients in the ICD group, 259 (31 percent) were known to have received shocks from their device for any cause, with 177 (68 percent of those shocked, or 21 percent of the ICD group) receiving shocks for rapid ventricular tachycardia or fibrillation.”³⁸ There was a 10% inappropriate shock rate for patients with ICDs.

The authors concluded that “in patients with NYHA class II or III CHF and LVEF of 35 percent or less, amiodarone has no favorable effect on survival, whereas single-lead, shock-only ICD therapy reduces overall mortality by 23 percent.”³⁹

In this trial, there was equal randomization. The significance level was reduced to account for the additional comparisons in a 3-group trial. The follow-up time was extended by 1 year. Single lead defibrillators were used in all patients in the ICD group. No dual chamber or bi-ventricular devices were used.

4. Medicare Coverage Advisory Committee (MCAC)

CMS did not convene an MCAC for this issue.

5. Evidence-based guidelines

At this time, professional society guidelines addressing the SCD-HeFT patient population are not available. We have been informed that the guidelines published in 2002 by American College of Cardiology (ACC), the American Heart Association (AHA) and the North American Society for Pacing and Electrophysiology (NASPE) will be reviewed and updated in the near future.

In 2002, the ACC/AHA recommendations for implantable defibrillator therapy, expressed in the standard format, ⁴⁰ are as follows:

6. Open Comment Period

Initial Comments

CMS held three open comment periods during this national coverage analysis. Each comment period encouraged responses to different issues regarding ICDs. Most comments are available in full text on our web site at http://www.cms.hhs.gov/mcd/viewtrackingsheet.asp?id=108.

The first comment period was at the start of this national coverage analysis. CMS allowed the first 30 days for the public to comment on any issue related to the request to expand Medicare coverage of ICDs to the SCD-HeFT population. In addition to general comments, CMS encouraged commenters to focus on clinical trials and device selection.

In the second comment period, CMS specifically requested comments on the results of the COMPANION and DEFINITE trials.

In the third comment period, CMS asked for comments on three specific questions regarding the evidence of threshold testing, evidence of risks and benefits of anti-tachycardia pacing (ATP) and the justification of an ICD registry to better identify predictors of device firing (as a proxy for sudden cardiac death).

Comments received during these periods are summarized below according to classification of the commenter as a professional society, expert commenter or public commenter.

Professional Society Position Statements

The American College of Cardiology (ACC) comments that the SCD-HeFT data made available prior to the peer-reviewed publication is favorable but that review of peer reviewed data should be used to determine coverage. In addition, the ACC strongly supports that cardiovascular device selection be at the discretion of the treating physician. In comments by the Heart Failure Society of America (HFSA), the society points out that recent clinical trials show patients with heart failure and left ventricular dysfunction (with and without coronary artery disease as the underlying cause) but no history of a life-threatening arrhythmia treated with ICDs benefit from a reduced risk of death. The NASPE-Heart Rhythm Society supports full Medicare coverage of the SCD-HeFT patient population based on the currently available data. The society also strongly supports that cardiovascular device selection be at the discretion of the treating physician, particularly since some patients have multiple indications that may require a dual chamber defibrillator or biventricular pacing.

The Heart Rhythm Society submitted a clinical competency statement containing guidelines for non-electrophysiologists implanting defibrillators. They note that as the demand for these devices increases it is important to ensure patient safety through appropriate training of implanting physicians. The guidelines restrict non-electrophysiologist implanting physicians to those that meet high volume requirements of standard pacemakers in addition to receiving training for ICDs and cardiac resynchronization devices.

In comments from the Heart Failure Society of America, the society discusses recommendations for ICDs in patients eligible for cardiac resynchronization therapy (CRT) and in class IV heart failure, regardless of etiology. Both the ACC and Heart Rhythm Society support this recommendation.

In the third round of public comments the ACC and Heart Rhythm Society submitted joint comments. Both support threshold testing and note that clinical trials have used testing as part of their protocol and do not support the removal of threshold testing until clinical trials are available in support of changing clinical practice. The societies state that testing is necessary to determine lead re-positioning, need for device upgrade and assessment of device sensing. They also provide data in response to concern over ATP and note that ATP reduces the number of inappropriate shocks and should be available in defibrillators intended for the primary prevention of SCD. Although ACC and Heart Rhythm Society agree a registry could be beneficial they note many barriers to implementation.

Expert Opinion

Of the expert respondents in the first comment period, all support expanded coverage of ICDs. Although most experts cite results from SCD-HeFT, some experts note recent results from DEFINITE, COMPANION and MADIT-II as support for expanded Medicare coverage. CMS was cautioned against performing subgroup analyses of SCD-HeFT data. Some experts support defibrillator implantation in patients with LVEF ≤35% while one supports coverage of ICD implantation in patients with LVEF ≤39% regardless of etiology. Another expert supports extending coverage to all patients with a prior MI and LVEF ≤30% irrespective of QRS width in addition to patients with Class II or Class III CHF and LVEF≤35% irrespective of the etiology of heart failure or the QRS width. Others support coverage of both ischemic and non-ischemic patients and Class II and Class III heart failure patients. One expert suggests that at least two different methods of measuring ventricular function be used to confirm LVEF. Real-world population registries are the suggestion of one expert to confirm the trial results. Overall, respondents strongly support that cardiovascular device selection be at the discretion of the treating physician.

Expert opinions received in the second comment period emphasize that MADIT II and SCD-HeFT prove ICDs beneficial to both ischemic and non-ischemic patients. In addition, they maintain that physician judgment, not grouping by QRS criteria, remains the most appropriate determinant of treatment. One expert supported coverage for the entire MADIT II and SCD-Heft populations and another notes a cost-effectiveness analysis that favors expanded coverage.

In response to the third comment period, Medicare received many comments from experts on the issue of threshold testing. Experts generally agree that threshold testing remains necessary to ensure the device will defibrillate the patient and make changes in the device setting or lead positioning if required. Another issue is the concern that physicians have appropriate specialty training to perform defibrillator implants. In regard to ATP, experts state that this function reduces inappropriate shocks and provides initial therapy for VT. However, some caution that inappropriately programmed ATP can increase risk of syncope or injury. For this reason, some recommend that only electrophysiologists be able to perform the procedure. Data is provided from MADIT II that demonstrates the value of ATP through the number of VT/VF episodes that were terminated by ATP. One expert argues that the primary goal of using ICDs as primary prevention is not to manage complex VT, therefore, recommends broad coverage of all class II and III patients with a simpler device. Expert comments varied on the issue of a registry. Some express that a registry at this point in time is unnecessary and that clinical trials have already proven the need for devices in these patients. Others state that additional information on clinical predictors of SCD would be valuable. Experts advise CMS that microvolt t-wave alternans testing offers a negative predictive value to rule out patients who will not benefit from an ICD.

Public Commenters

In the first comment period, the majority of commenters support expanded coverage of ICDs to the SCD-HeFT population while some also recommend coverage of MADIT II patients. Many note support of ICDs for patients with LVEF ≤35% while some support coverage of patients with Class II and Class III heart failure. One respondent does not feel that ICDs should be routinely provided to all patients meeting SCD-HeFT criteria and feels strongly that an ICD should be available to selected patients with mild symptoms of heart failure with low ejection fraction, and otherwise good prognosis for meaningful survival extending past two years. Another respondent does not believe that CMS should cover ICDs for all patients with low LVEF or clinical heart failure, but should cover SCD-HeFT patients only with QRS > 0.12 sec. CMS is cautioned against performing subgroup analyses of the SCD-HeFT data by some commenters. Some commenters note the need for more data. Of the commenters that discussed device selection, most agreed that CMS should not determine device selection through coverage policy or cost information but that the treating physician should continue to make the decision. A few commenters, however, note that there should be cost restrictions on cardiovascular devices. Microvolt T-wave alternations is suggested as a method of identifying patients who will not benefit from ICD therapy.

CMS received few public comments in response to the second comment period. Some of the public comments strongly support the COMPANION trial and suggest its results imply NYHA Class IV patients will be disenfranchised by a narrow CMS coverage determination. In order to allow another route to access for these patients, such commenters suggest Class IV patient coverage be left to local contractor discretion. Others recommend coverage of all or a combination of the MADIT II, SCD-HeFT, DEFINITE and COMPANION populations based on the favorable results of those trials. Commenters generally recommend the elimination of QRS as a condition of coverage and recommend using only LVEF. One commenter maintains that the differences in results between the DEFINTE (35% reduction in overall mortality) and SCD-Heft (25% reduction in overall mortality) trials are within expected limits for repeated trials. The same commenter points out a suggested benefit for Class III patients, though admitting this trend was statistically insignificant. Some commenters point out the evidence does not demonstrate a clear benefit for women treated with ICDs. Finally, most of the public commenters remain opposed to CMS determining device selection, stating that such decisions are best left to physician discretion.

In the third comment period, public commenters note that threshold testing is necessary to make sure it can deliver the required shock to the patient. The concern of some commenters is that removing threshold testing could encourage physicians to implant devices that are not adequately trained to do so. ATP comments received agree that this function reduces inappropriate shocks and contributes to a better quality of life for the patient since otherwise painful shocks are avoided. Commenters also note that clinical trials utilize threshold testing as part of the protocol therefore results of those studies cannot be extrapolated to patients who do not receive threshold testing. Most public commenters express concern for the reliability of information from a registry and do not expect it to provide more information than what is already available. Others determine that clinically useful information can come out of a registry but are unsure of how CMS would implement such a massive undertaking. Public commenters recommended the use of microvolt t-wave alternans testing to exclude certain patients from implants based on a prediction that they will not benefit from the device. They refer to the test’s high negative predictive value. The need for the initial database is discussed below.

Comments on Proposed Decision Memorandum

In response to the reconsideration proposed decision memorandum posted on September 28, 2004, we received comments from 374 individuals and groups during the required statutory period ending October 28, 2004. Commenters included major national professional associations (e.g., electrophysiologists, cardiologists, heart failure specialists), patient advocacy groups, national associations of health plans and of device manufacturers, academic researchers, practicing professionals, and other individuals including patients and caregivers. The majority of commenters supported the expanded coverage proposed in the draft decision memorandum. Many commenters, however, did not agree completely with the proposed coverage and had additional concerns. These were taken into consideration in the final analysis. A summary of the comments is provided below.

General Comments

Of the total, 261 commenters stated that CMS was improving access to ICDs by expanding coverage but the expansion did not include all patients at risk for sudden cardiac death. The commenters encouraged CMS to cover all patients at risk; however, characteristics of the population at risk were not identified.

T-Wave Alternans (TWA) Testing

CMS received 29 comments regarding the use of TWA as a risk-stratifier for sudden cardiac death. Two commenters recommended against TWA as a risk-stratifying mechanism. One commenter specifically stated that the technology had not undergone rigorous scrutiny, the test could not be easily applied and the results were not reproducible. Remaining commenters supported the use of TWA and about half of the supporters wanted TWA to be part of the ICD registry. One commenter made the suggestion to learn more about the predictive value of the test but stated that it is not yet ready for use in practice to determine ICD eligibility. This decision memorandum does not specifically evaluate the use of TWA. As noted in our June 2003 memo, TWA is a promising technology to identify patients who are at low risk for any ventricular tachyarrhythmias. Further consideration of TWA is warranted. CMS encourages the inclusion of TWA in clinical trials, registries and other data collection systems.

End of Life Care

Two commenters were concerned with the use of these devices in nursing home patients and patients nearing the end of life. They specifically addressed the need for physicians to discuss end of life care with patients and their families including options of not having the device implanted or turning the device “off” at a later time. One commenter requested that we collect end of life information in the ICD registry to inform us about the number of patients that choose to have their devices turned off. This is an important concern that was given careful thought in previous ICD decisions which led to noncoverage of implantations in patients with irreversible brain damage, disease, or dysfunction that precludes the ability to give informed consent or any other disease in which the likelihood of survival is less than one year. CMS will apply these requirements to the coverage indications outlined in this decision.

New York Heart Association Class IV CHF

CMS received approximately 20 comments concerning the noncoverage of ICDs for Class IV CHF patients particularly in combination with CRT devices. Commenters recommended that CMS change the coverage decision to include this class of patient for coverage. Most suggested that these patients should be eligible to receive CRT-D therapy with the potential of improving their heart failure class to III. By noncovering Class IV, commenters noted that physicians would be forced to implant a CRT device to improve the patient to Class III CHF and then implant the CRT-D therefore exposing the patients to multiple procedures. Commenters also addressed the benefits of CRT-D in Class IV CHF patients such as less hospitalizations and lower use of medications. One commenter stated that Class IV CHF patients should be covered for CRT-D only when that patient has a reasonable expectation of improving to Class III CHF. CMS carefully reviewed the evidence regarding coverage of ICDs in Class IV patients as noted in the analysis section and will modify our proposed decision to address this population.

New York Heart Association Class I CHF

One commenter recommended that CMS exclude coverage of Class I CHF patients since studies have not demonstrated a benefit in this group. The exclusion of Class I CHF patients is outlined in this document and reflected in the final decision.

NIDCM for 9 months

CMS received approximately 20 comments on the coverage requirement that patients have NIDCM for at least 9 months prior to ICD implantation. Slightly more than half of these commenters favored the SCD-HeFT requirements, that patients have NIDCM for 3 months prior to implantation. They believed that the overall outcome of SCD-HeFT demonstrated that these patients benefit from ICD therapy and that subgroup analysis should not be used to exclude them from coverage. Of the commenters that favored the CMS requirement of 9 months, they agreed that a 9 month interval was appropriate to exclude patients with reversible disease and allow time for evaluating the response to treatment with optimal medical therapy. CMS carefully reviewed the data from multiple trials and outlines the reasoning behind this decision to require NIDCM for 9 and 3 months in the analysis section of this decision memorandum.

Device Selection

Approximately 40 comments were received regarding device selection with one-third of those in support of CMS issuing guidance or policy regarding appropriate use. Some commenters believed that CMS policy would reduce overuse of sophisticated devices when they are not truly indicated. The requirement of documenting the reason for device selection was favored by some. The majority of commenters strongly suggested that CMS remove the term “shock-only” from the coverage determination. “Shock-only” implied to some commenters that ATP and RV pacing were not options on a simple ICD. Commenters wanted the option of programming these functions on simple, single-chamber devices. Many commenters were concerned that inability to program ATP would result in unnecessary shocks. Other commenters were concerned that initial implantation of a simple device would later be followed by implantation of a more sophisticated device as the patient’s condition worsens or the need for additional leads is indicated thus exposing the patient to multiple procedures. It was noted that SCD-HeFT was not designed to study device selection and coverage restrictions should not be developed from the results of the study. Since single lead devices were used in SCD-HeFT, this type of device would be appropriate for the majority of patients. Although CMS does not specifically exclude other types of devices, physicians and providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient medical record. We will modify our decision language to address these concerns.

LVEF

CMS received approximately 40 comments regarding ejection fraction requirements. Commenters stated that CMS should change the coverage determination to include patients with ejection fractions of ≤ 0.35 rather than ≤ 0.30. It was pointed out that the population was included SCD-HeFT and COMPANION and those trials showed overall benefit. Commenters strongly suggested that subgroup analyses should not be used to determine the ejection fraction limits. In addition, one-fourth of comments regarding ejection fraction recommended the use of cardiac magnetic resonance imaging to measure ejection fraction. CMS carefully reviewed the data related to ejection fraction across trials, including MADIT II, which enrolled only patients with LVEF ≤ 0.30. Since SCD-HeFT included patients with LVEF 30-35%, coverage will be expanded.

MI < 1 month, PTCA < 3 months

CMS received approximately 10 comments concerning the restriction of coverage to patients with an MI greater than 1 month prior to ICD implantation and PTCA greater than 3 months prior to implantation. Commenters stated that this requirement should not apply to patients that already met the criteria for implantation prior to their most recent event or procedure. However, most commenters agreed that patients presenting with a new MI should be required to wait 1 month. Based on results of DINAMIT, ICDs have not been shown to improve health outcomes when implanted in patients within 40 days of an AMI and may, in some instances, be harmful in these patients. Therefore, we have decided to not cover ICD implantation within 40 days of an AMI.

Physician Criteria

Approximately 20 comments were received concerning physician criteria for implanting ICDs. There was a range of criteria recommendations with about half supporting that implanters must be board certified electrophysiologists and the other half supporting the recent HRS guidelines for non-electrophysiologist implanters. Others noted that experienced, non-electrophysiologists that are currently implanting should be grandfathered into the requirement and one commenter noted the need to consider concerns of partially disabled physicians that implant with assistance of other physicians. CMS believes only appropriately trained and competent physicians should be implanting devices and that hospitals have responsibility for assuring patient safety and protection. Therefore we will not implement specific physician certification requirements at this time.

Threshold Testing

Regarding threshold testing, CMS received approximately 10 comments with each supporting the need for testing at the time of implant to ensure appropriate device programming and lead placement. CMS does not intend to issue guidance on threshold testing as it was not the focus of this decision.

Data Collection

Approximately 45 comments discussed the CMS requirement that patients receiving an ICD for primary prevention of sudden cardiac death be enrolled in a registry. Many commenters supported a registry with certain conditions while others were opposed. Commenters that did not agree with a registry requirement stated that the evidence from clinical trials already demonstrate ICD benefit for primary prevention and further study through a registry is not necessary and would add no value. Others commented that a registry would not be possible at their institutions because institutional review boards would not approve participation, they would violate privacy and confidentiality requirements of the Health Insurance Portability and Accountability Act or the burden on their resources would be too great. Many commenters raised questions about details of the database that were not outlined in the draft decision memorandum such as hypotheses, data elements, funding, management, analysis and access. CMS will ensure that uses and disclosures of information for any future data collection system be made in compliance with the Standards for Privacy of Individually Identifiable Health Information and that all issues related to patient confidentiality, privacy, and compliance with other Federal laws will be resolved prior to the collection of any data.

Some commenters were concerned that requiring a registry would delay coverage and limit patient access. Their recommendations included allowing a grace period between the effective date of coverage and registry participation, decoupling registry participation from payment and requiring a sampling of ICD patients rather than the entire Medicare primary prevention population.

Registry supporters stated that an ICD registry could provide data on real-world outcomes, assist in answering questions that require a large sample size such as subgroups of women and patients over age 80 and may be hypothesis generating in regard to risk stratification. Commenters made various recommendations on implementing a registry. One commenter suggested that only one registry be used to facilitate tracking and auditing rather than allowing multiple registries to collect data. Other commenters suggested that a valuable registry could be implemented if it was done carefully, with good planning, cooperation from stakeholders and had credible oversight.

CMS will require that Medicare patients receiving an ICD for the primary prevention of sudden cardiac death have certain data submitted to a database. CMS will not delay coverage as a database has been created and made available to all hospitals. HIPAA and Privacy Act requirements have been met for this database which will be maintained using a data submission mechanism that is already in use by Medicare participating hospitals to submit data to the Iowa Foundation for Medical Care (IFMC)a Quality Improvement Organization (QIO) contractorfor determination of reasonable and necessary and quality improvement.

Initial Public Comments on December 29, 2004 Reconsideration

CMS received three public comments in the comment period beginning December 29, 2004 in association with the reconsideration for ICDs while awaiting publication of the SCD-HeFT results. Two commenters recommended the inclusion of T-Wave Alternans testing in the ICD registry. They discuss the value of T-Wave as a mechanism of risk stratification. One commenter suggested that a registry could incorporate the data collection that already takes place by the device industry such as they already collect a significant amount of information regarding the patient and device implantation.

VIII.   CMS Analysis

A. Introduction

National coverage determinations (NCDs) are determinations by the Secretary with respect to whether or not a particular item or service is covered nationally under title XVIII of the Social Security Act § 1869(f)(1)(B). In order to be covered by Medicare, an item or service must fall within one or more benefit categories contained within Part A or Part B, and must not be otherwise excluded from coverage. Moreover, with limited exceptions, the expenses incurred for items or services must be “reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member.” § 1862(a)(1)(A).

For this reconsideration of ICDs, the results of 6 trials (CAT, AMIOVIRT, DEFINITE, SCD-HeFT, COMPANION, DINAMIT) involving defibrillators have been published since our prior decision in June 2003. DINAMIT studied patients with IDCM and will be discussed in the section on IDCM. CAT and AMIOVIRT studied patients with NIDCM and will be discussed in the section on NIDCM. COMPANION and SCD-HeFT studied both patients with IDCM and patients with NIDCM and are discussed below.

The COMPANION trial was a randomized controlled trial involving 1520 patients with nonischemic or ischemic dilated cardiomyopathy, LVEF ≤ 35%, QRS interval ≥ 120 msec. and PR interval > 150 msec. It demonstrated a reduction in mortality in the CRT-D therapy group compared to the optimal pharmacologic therapy group (18% versus 25%, respectively; hazard ratio =0.64; 95% CI=0.48-0.86; p-value=0.003).⁴² However, there were several issues with the design, conduct and analysis of the trial. First there was an unequal, 1:2:2 randomization ratio weighted towards device therapy. An equal 1:1:1 randomization format is generally considered more neutral. Second, the definition of hospitalization was changed during the course of the trial without notifying the FDA. This potentially has a direct impact on the primary outcome since hospitalization was the dominating factor for the composite endpoint. Data should have been collected and reported using both definitions to determine if the change favored one group over another. Third, a high number of patients withdrew from the pharmacologic therapy group, many of whom obtained device therapy. Patients who were subsequently lost to follow-up were censored in the analyses, which may have lead to an inaccurate estimation of the mortality rate. These issues hamper the strength of the findings of the COMPANION trial. Since it was the only trial to evaluate mortality for patients with CRT and CRT-D therapy, additional research is needed to support the findings of this trial.

SCD-HeFT was a randomized controlled trial involving 2,521 patients with both IDCM and NIDCM and LVEF ≤ 35%. It demonstrated a reduction in mortality in the ICD therapy group compared to the placebo group (hazard ratio =0.77; 97.5% CI=0.62-0.96; p-value=0.007) at the end of follow-up.⁴³

Overall, the absolute reduction in mortality was modest for a trial with a median follow-up of 45.5 months. Several potential explanations have been presented for the modest overall effect compared to prior ICD trials. In SCD-HeFT, appropriate medications for heart failure were recommended for all patients. SCD-HeFT also included patients with nonischemic dilated cardiomyopathy. As with other studies on patients with NIDCM, SCD-HeFT showed that overall mortality is lower for patients with NIDCM compared to patients with IDCM.⁴⁴

In SCD-HeFT, the follow-up period was extended by one year but the reason to do so has not been fully explained. In general, the extension of a clinical trial presents potential challenges: “Towards the scheduled end of a study, the investigator may find that he has nearly statistically significant results. He may be tempted to extend or expand the trial in an effort to make the test significant. Such a practice is not recommended. A strategy of extending assumes that the observed relative differences in rates of response will continue. The observed differences which are projected for a larger sample may not hold. In addition, because of the multiple testing issue and the design change, the significance level should be adjusted downward. However, appropriate adjustments in the significance level to account for the design changes may not easily be determined. Since a more extreme significance level should be employed, and since future responses are uncertain, extension may leave the investigator without the expected benefits. Whatever adjustments are made to either sample size or the length of follow-up should be done as early in the trial as possible. Early adjustments would diminish the criticism that the monitoring committee waited until the last minute to see whether the results would achieve some prespecified significance level before changing the study design.”⁴⁵

Overall SCD-HeFT was a well-conducted trial that had a large sample size and a lengthy follow-up period. The available results provide evidence on the benefits of a simple, single lead ICD. Additional information such as ICD firing data, patient characteristics at last follow-up and disease progression should be forthcoming.

B. Questions

For this reconsideration on ICDs, CMS focused on the following questions:

Question 1: Is there evidence to conclude that ICDs decrease mortality for patients with ischemic dilated cardiomyopathy (IDCM) and reduced LVEF?

• What can we conclude from evidence for patients with either QRS < 120 milliseconds or LVEF < 35%?

For patients with IDCM, evidence from DINAMIT, SCD-HeFT and COMPANION is relevant. DINAMIT was a randomized controlled trial involving 674 patients with recent myocardial infarction, LVEF ≤ 35% and depressed heart rate variability. It did not show a reduction in mortality when an ICD was implanted early after a myocardial infarction (6-40 days).⁴⁶

Overall, SCD-HeFT showed a significant reduction in mortality. SCD-HeFT included 884 patients with IDCM and LVEF ≤ 35%. There was a reduction in mortality hazard ratio for ICD therapy compared to control but it was not statistically significant (hazard ratio=0.79; 97.5% CI=0.60-1.04). Overall, the COMPANION trial showed a significant benefit. COMPANION included 842 patients with IDCM, LVEF ≤ 35%, PR >150 ms. and QRS interval ≥ 120 msec. There was also a reduction in mortality hazard ratio for the CRT-D group compared to optimal pharmacologic therapy but it was not statistically significant (hazard ratio=0.73; 95%CI=0.52-1.04).⁴⁷

In our June 2003 decision, we used prolonged QRS interval as a risk stratifier based on our concerns with the conduct of MADIT II [inclusion of patients with a high likelihood of benefit that should have been excluded by trial design (see previous decision memorandum - CAG-00157N: http://www.cms.hhs.gov/mcd/viewdecisionmemo.asp?id=39)] and subgroup analyses of the MADIT II data. The main results of the COMPANION trial are consistent with the use of prolonged QRS since it was one of the inclusion criteria; SCD-HeFT did not base inclusion on QRS interval. In subgroup analyses, SCD-HeFT showed that patients with QRS ≥ 120 msec. had a significant reduction in mortality (hazard ratio=0.67; 97.5% CI=0.49-0.93).⁴⁸ Based on the findings from these trials, QRS interval appears to be a predictor of benefit; that is, in general, patients with longer QRS intervals have a greater benefit. This is also consistent with Nudell’s statement: “When one looks at the combined results seen in MADIT II and SCD-HeFT, we estimate that patients with abnormal QRS widths (>120 milliseconds) have about 4 times the absolute ICD mortality benefit of that seen amongst patients with normal complex widths (<120 milliseconds).”⁴⁹ However, with the additional data from SCD-HeFT, one can conclude that patients with narrow QRS intervals also obtain a small benefit from ICD therapy (hazard ratio for QRS duration < 120 ms. = 0.84; 97.5% CI=0.62-1.14).⁵⁰ Thus, we believe that the evidence is now sufficient to remove the QRS coverage limitation.

In addition to QRS prolongation in our June 2003 decision, we limited coverage to patients with LVEF ≤ 30% based upon the MADIT II inclusion criterion. SCD-HeFT included patients with LVEF ≤ 35%. Data from SCD-HeFT are included in the following table:

The other trials reviewed had LVEF inclusion criteria from 30 – 35% but did not stratify their results by LVEF categories. Other previously reviewed trials had varying LVEF inclusion criteria from 30% to 40%. The vast majority of patients in all trials had actual LVEFs in the mid to low 20% range. However, due to the inclusion of patients with LVEF 30-35% in prior trials and the public comments that were received urging CMS to cover this population, we will expand coverage to these patients.

Question 2: Is there evidence to conclude that ICDs decrease mortality for patients with nonischemic dilated cardiomyopathy and reduced LVEF?

For patients with nonischemic dilated cardiomyopathy, the evidence is not as substantial as for patients with IDCM since few defibrillator trials have been conducted on patients with NIDCM. In addition, prior reports have indicated that patients with NIDCM have better overall survival and outcomes than patients with IDCM. For patients with NIDCM, evidence from AMIOVIRT, CAT, DEFINITE, SCD-HeFT and COMPANION is relevant.

AMIOVIRT was a randomized controlled trial involving 103 patients with nonischemic dilated cardiomyopathy, asymptomatic non-sustained ventricular tachycardia and LVEF ≤ 35%. There was no statistically significant difference in mortality. The investigators reported: “Furthermore, in asymptomatic patients with ischemic cardiomyopathy, prophylactic implantation of an ICD is superior to treatment with amiodarone. This is in contrast to the observation from AMIOVIRT, where patients with NIDCM, left ventricular dysfunction and asymptomatic NSVT, had comparable survival with an ICD or amiodarone. These findings suggest that patients with cardiomyopathy secondary to coronary artery disease may respond differently to treatment than patients with NIDCM, highlighting the notion that ischemic and non-ischemic myocardial arrhythmogenic substrates may be fundamentally different.”⁵¹

CAT was a randomized controlled trial involving 104 patients with recent onset nonischemic dilated cardiomyopathy and LVEF ≤ 30%. The investigators concluded: “ICD therapy did not reveal any survival benefit in the setting of DCM of recent onset and impaired LV function (EF ≤ 30%). This was most likely due to the low overall mortality rate in the control group. However, even in patients with a significantly increased mortality rate caused by a lower EF and nonsustained VTs, ICD therapy did not reveal any survival benefit. Therefore, the results of CAT do not favor prophylactic ICD implantation in patients with DCM of recent onset and impaired LVEF without any further risk stratification.”⁵²

The DEFINITE trial was a randomized controlled trial involving 458 patients with nonischemic dilated cardiomyopathy, LVEF < 36%, and PVC/NSVT. The investigators reported that “fewer patients died in the ICD group than in the standard-therapy group (28 vs. 40), but the difference in survival was not significant (P=0.08 by the log-rank test).”⁵³ The investigators concluded: “On the basis of our results, the routine implantation of a cardioverter–defibrillator cannot be recommended for all patients with nonischemic cardiomyopathy and severe left ventricular dysfunction. However, our findings of a reduction in sudden death from arrhythmia and an apparent benefit of ICDs in subgroup analyses suggest that the use of these devices should be considered on a case-by case basis.”⁵⁴

AMIOVIRT, CAT and the DEFINITE trial did not demonstrate a clear role for the use of a defibrillator for patients with NIDCM. SCD-HeFT enrolled 792 patients with NIDCM, as one prespecified subgroup, and showed a reduction in the mortality hazard ratio for ICD therapy compared to control but it was not statistically significant (hazard ratio=0.73; 95% CI= 0.50-1.07).⁵⁵ In the COMPANION trial, which added prolonged PR and QRS interval as inclusion criteria, there were 678 patients with NIDCM and a significant reduction in mortality in the CRT-D group compared to the optimal pharmacologic therapy group (hazard ratio =0.50; 95% CI=0.29-0.88). ⁵⁶

Considered together, CAT, AMIOVIRT, DEFINITE, SCD-HeFT and COMPANION present somewhat conflicting evidence on the use of defibrillators for patients with NIDCM and reduced LVEF. CAT, AMIOVIRT, DEFINITE indicated the need for further risk stratification of this population. CAT studied patients with recent onset NIDCM (≤ 9 months duration) and did not provide evidence to support the use in these patients. DEFINITE, COMPANION and SCD-HeFT evaluated patients with chronic NIDCM (average duration = 2.8 years; 3.6 years; and 2 years, respectively). COMPANION supported the use of prolonged PR and QRS duration as risk stratifiers; and showed that the mortality risk reduction was the greatest in the patients with prolonged QRS interval and NIDCM.⁵⁷

Of the 5 relevant studies, 2 (COMPANION, SCD-HeFT) showed a significant reduction in mortality overall but not specifically for patients with NIDCM. SCD-HeFT presented evidence for a broad approach yet this evidence is tempered by the negative findings of AMIOVIRT, CAT and DEFINITE and the restricted population of the COMPANION trial. Based on these results, a more thoughtful approach may be more appropriate for patients with NIDCM compared to patients with IDCM. The firing data from SCD-HeFT for patients with NIDCM is forthcoming and should provide some reassurance that the devices are firing as expected in these patients.

Based on the overall results of SCD-HeFT with support from the COMPANION trial, there is evidence that ICDs decrease mortality for patient with nonischemic dilated cardiomyopathy and reduced LVEF. Therefore, coverage will be expanded to include this group of patients.

C. Patient and Device Selection

Although SCD-HeFT demonstrated a statistically significant reduction in mortality, the absolute reduction was modest. In addition, a relatively small proportion (21.4%) of the defibrillator group received an appropriate shock over the course of the trial. This firing rate was slightly higher than the overall firing rate (19%) seen in MADIT II, which had a shorter average follow-up time (20 months).

Since the large proportion of patients who receive an ICD never received any therapy from their device, consideration of additional risk stratification methods would be reasonable. During SCD-HeFT’s lengthy follow-up period, cardiac disease probably progressed and other relevant characteristics likely changed in many patients. Pending data from the last follow-up or measurement may help to predict which patients will likely receive defibrillator therapy and when therapy will likely occur.

1. Risk Reduction

In SCD-HeFT, a conscious attempt was made to ensure that all patients received optimal medical therapy. Appropriate use of beta-blocker therapy, angiotensin converting enzyme therapy, aldosterone blocking diuretics, aspirin and statin therapy was encouraged. The optimal medical therapy likely had an important role in reducing overall mortality rates in all groups. The mortality rate in the placebo medical therapy group was lower than the rates in most prior trials. Thus, optimizing medications for all patients should be emphasized. In addition, reduction of other major risk factors for sudden cardiac death should also be emphasized.

2. Pooled Data Analysis

CMS strongly encourages the sponsors and principal investigators of ICD trials to engage an independent, reputable cardiology research center to pool the databases from their respective trials and conduct analyses to identify patient selection, device related issues and other research questions that need further study. This information would be important for improving decision-making about defibrillator implantation. With the implementation of the initial database, compilation of this data would provide a valuable comparative database.

3. QRS Interval

QRS prolongation has been studied as a potential risk stratifier for patients with congestive heart failure. Zareba and Moss reported that “sudden cardiac death occurs as a result of a complex interplay of changes in myocardial substrate, imbalance of autonomic regulation of the heart, and myocardial vulnerability.”⁵⁸ The authors further explained that “electrical manifestation of changes in myocardial substrate include QRS and QTc prolongation, presence of conduction disturbances, presence of late potentials, abnormalities of repolarization morphology, and presence on nonsinus rhythm, namely atrial fibrillation.”⁵⁹

Given the total body of evidence from ICD and CRT trials, prolonged QRS interval remains a potentially useful risk stratifier, which could be considered for defining the level of baseline risk and likely benefit from an ICD. With further research on CRT and CRT-D devices, more information may be available.

4. LVEF

As previously noted, LVEF is often considered an important prognostic indicator. In 2000, Moss reported that “the findings from MADIT, AVID, MUSTT, and CIDS paint a very clear picture – it is the sickest patients who benefit the most from ICD therapy.”⁶⁰ Based on survival analysis of MADIT I data, Moss also noted “the survival benefit of ICD therapy was significantly greater than conventional therapy only in the subgroup with an ejection fraction < 26%.”⁶¹ In 2002, Skenkman and colleagues also reported “a linear relationship between QRS duration and decrease ejection fraction”⁶² based on a prospective observational study of 3,471 patients with congestive heart failure. The authors also noted “systolic dysfunction was associated with graded increases in mortality across ascending levels of QRS prolongation.”⁶³

5. Acute Myocardial Infarction

As demonstrated in DINAMIT, implantation of an ICD within 40 days of an acute myocardial infarction has not been shown to reduce mortality. Based on the data from DINAMIT,⁶⁴ an ICD should not be implanted within 40 days of an AMI.

6. NYHA Class

Most studies on defibrillator therapy have only enrolled patients with NYHA Class I-III. Although it is likely to be a fairly subjective measure, it has been routinely used as a patient inclusion criterion. MADIT II enrolled patients with NYHA Class I-III with LVEF set at ≤ 30%. SCD-HeFT enrolled patients with NYHA Class II-III with LVEF ≤ 35%. Patients with NYHA Class I and LVEF 30-35% were not included in either trial. Since SCD-HeFT provided additional evidence to expand coverage, eligibility will thus follow the SCD-HeFT inclusion criteria closely. Patients with NYHA Class IV have been excluded in the large primary prevention trials such as SCD-HeFT and MADIT II.

The COMPANION trial, a resynchronization therapy trial, was the only one of the trials reviewed in this decision and the prior decision that included patients with NYHA Class IV. Separate subgroup analyses of COMPANION patients with NYHA Class IV have not been reported, but only 14% of patients (219 of the 1520) were classified in NYHA Class IV. However, due to the inclusion of patients with NYHA Class IV heart failure in COMPANION and the industry and public comments that were received urging CMS to cover this population, we will expand ICD coverage to patients who meet all current CMS coverage requirements for cardiac resynchronization therapy.

7. Microvolt T-Wave Alternans (MTWA)

Microvolt T-wave alternans refers to microvolt variations in the morphology of the electrocardiographic T-wave on an alternate beat basis during exercise. MTWA testing involves measurements of these T-wave variations during exercise stress testing. The testing is non-invasive but requires specific equipment. Several studies and reports have been published on the use of MTWA as a risk stratifier for defibrillator therapy.⁶⁵ Most recently, Bloomfield and colleagues reported that “among MADIT II-like patients, a microvolt T wave alternans test is better than QRS duration at identifying a high-risk group and also better at identifying a low-risk group unlikely to benefit from ICD therapy.”⁶⁶ Peer-reviewed journal publications are expected in the near future, which may provide further information on the utility of this test as a screening tool. We received several comments suggesting that MTWA be part of a national registry.

We do strongly encourage the inclusion of MTWA in subsequent clinical trials, registries and other data collection protocols in order to further evaluate this promising risk-stratification technology and will work with the stakeholders involved in the subsequent data collection systems to include this information. CMS will continue to support these studies that collect this type of information..

8. Age

Although chronological age is generally not an appropriate criterion for coverage, the implantation of any device should not be routinely recommended for patients with limited life expectancy (an exclusion criterion in most trials). Very few patients over the age of 75 years old have been enrolled in defibrillator studies. Of the two largest trials (MADIT II and SCD-HeFT), only 375 (10%) of the combined patients were over the age of 75 years. The actual benefits and harms of defibrillators in patients aged 75 years and older have not been adequately demonstrated. Therefore, the implantation of a defibrillator in the most elderly patients should be carefully considered and not routinely recommended. However, although additional evidence of risks and benefits is desirable, we will not restrict coverage by age in this decision.

9. Gender

Like age, gender is usually not an appropriate criterion for coverage; however, female patients have been under-represented in all ICD trials. There may be different considerations for implantation of devices, type of device, size of device, longevity, shock energy and adverse events for women compared to men. No published study has shown a significant improvement in mortality for women from ICD therapy compared to control therapy. A thorough assessment of whether the benefits and harms of ICDs differ by gender has not been published but is needed.

10. Type of Defibrillator and Functions

In SCD-HeFT, single lead defibrillators with basic programming were used. These devices were not programmed for antitachycardia pacing. The evidence on the benefits of ATP is sparse and inconclusive. ATP has been found to be harmful in several clinical scenarios where pacing was initiated for a benign arrhythmia and resulted in ventricular fibrillation. In addition, the number of adverse events, including lead fractures, increases with the number of leads implanted. Since SCD-HeFT demonstrated a significant reduction in mortality from a single lead device and it enrolled by far the most patients of any trial, a single lead device is clinically appropriate and sufficient for primary prevention of sudden cardiac death. Bardy and colleagues reported: “ICD therapy cannot be considered a single intervention, given the numerous possible permutations of this approach. Consequently, we cannot emphasize too strongly that we evaluated only very conservatively programmed ICDs with a conservative detection algorithm and shock-only therapy. We found strong evidence that this approach works; however, considerable caution should be used in extrapolating our results to other approaches to ICD therapy, such as those involving dual-chamber or biventricular pacing, since, as reported previously, they may not afford the same benefit or, for that matter, any benefit.”⁶⁷

As mentioned in the prior decision, indiscriminate pacing may increase the risk of adverse events such as hospitalization for heart failure. This is consistent with the SCD-HeFT results, which had a lower adverse event rate compared to prior trials such as the DAVID trial and MADIT II.

CMS requested public comments on appropriate device selection (as noted in the public comments section). Few favored any device restrictions. At this time, CMS will not limit coverage based on type of defibrillator. However, providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient medical record. The data collection that we initiate with this NCD as well as subsequent studies will provide better evidence on experience with different types of devices.

11. Physician and Provider Credentialing and Certification

As with any invasive procedure, physicians who insert ICDs must be appropriately trained and fully competent to perform the implantation. For an acceptable risk-benefit consideration, patients should not be harmed by the implantation of the device. CMS strongly encourages credentialing and certification of physicians who insert ICDs by appropriate national organizations, such as the Heart Rhythm Society (HRS) or boards of medical specialties, to ensure the safety of Medicare beneficiaries. CMS also believes that provider credentialing and certification should be tracked and included in any and all registries and data collection systems. This information is valuable for informing patients as part of effective clinical decision-making and will provide useful data on procedural outcomes associated with different levels of provider training and expertise.

12. Cardiac resynchronization plus defibrillator therapy (biventricular pacing plus defibrillator)

Several prior trials on cardiac resynchronization therapy for advanced heart failure have been published.⁶⁸ The primary outcomes have focused predominantly on quality of life and functional status. The COMPANION trial studied the addition of a defibrillator to CRT on the outcomes of all-cause mortality and all-cause hospitalizations. As noted above, there were various issues with the COMPANION trial. The change in the definition of hospitalization during the trial is a potentially critical flaw influencing the primary outcome. The all-cause mortality outcome may have been influenced by the considerable withdrawal rate in the optimal pharmacological therapy group. Since there are no other published or reported trials powered to corroborate the findings of the COMPANION trial on the outcome of mortality, the evidence on the benefit of adding CRT to defibrillator therapy is insufficient. In addition, since CRT alone did not significantly reduce mortality, the observed benefit in the COMPANION trial from CRT-D was probably due to the defibrillator. Further research on CRT and CRT-D is needed.

13. National ICD Database

To maximize the benefits from ICD use, we desire to ensure that defibrillator implantation only occurs in those patients who are most likely to benefit and that the procedures are done only by competent providers in facilities with a history of good outcomes and a quality assessment/improvement program to identify providers with poor outcomes and other areas for improvement. As mentioned above, we are concerned that the available evidence does not provide a high degree of guidance to providers to target these devices to patients who will clearly derive benefit. In order to provide maximum net benefit to our beneficiaries, CMS will require that reimbursement for ICDs for primary prevention of sudden cardiac death occur only if the beneficiary receiving the defibrillator implantation is enrolled in either a FDA approved category B IDE clinical trial, a trial under the CMS Clinical Trial Policy or a qualifying data collection system including approved clinical trials and registries.

The submission of data on patients receiving an ICD for primary prevention to a data collection process is needed to assure patient safety and protection and to determine that the ICD is reasonable and necessary. These patient protections and safeguards require that data be made available in some form to providers and practitioners to inform their decisions, monitor performance quality, benchmark and identify best practices. The reasonable and necessary determination requires that patients meet the ICD implantation criteria set forth in this decision memorandum and are consistent with the trials discussed. Collection of these data elements allows that determination to be made. Covered entities under HIPAA are required to comply with the Standards for Privacy of Individually Identifiable Health Information (Privacy Rule) when using or disclosing protected health information for purposes of such data collection systems; and all issues related to patient confidentiality, privacy, and compliance with other Federal laws will be resolved prior to the collection of any data. There will be an initial ICD database so that data collection can begin with the posting of this decision. A data submission mechanism will be used that is already in use by Medicare participating hospitals to submit quality data. Initial hypotheses to be addressed by the database will include the following:

1. The clinical characteristics of the patients receiving ICDs are similar to those of patients involved in the primary prevention randomized clinical trials.
2. The indications for ICD implantation in patients are similar to those in the primary prevention randomized clinical trials.
3. The in-hospital procedure related complications for patients are similar to those in the primary prevention randomized clinical trials.
4. Certified providers competent in ICD implantation are implanting ICD devices in patients.
5. Patients who receive an ICD represent patients for which current clinical guidelines and the evidence base recommend implantation.
6. The clinical characteristics and indications for ICD implantation do not differ significantly among facilities.
7. The clinical characteristics and indications for ICD implantation do not differ significantly among providers.
8. The in-hospital procedure related complications for ICD implantation do not differ significantly among facilities.
9. The in-hospital procedure related complications for ICD implantation do not differ significantly among providers.
10. The in-hospital procedure related complications for ICD implantation do not differ significantly among device manufacturer, types, and/or programming.

Data elements necessary to address these hypotheses are attached to this decision (Appendix VI) and are the minimum necessary to determine that the ICD is reasonable and necessary. These hypotheses and elements may change over time as other evidence becomes available. Initially, an ICD database will be maintained using a data submission mechanism that is already in use by Medicare participating hospitals to submit data to the Iowa Foundation for Medical Care (IFMC)a Quality Improvement Organization (QIO) contractorfor determination of reasonable and necessary and quality improvement. Data collection will be completed using the ICDA (ICD Abstraction Tool) and transmitted via QNet (Quality Network Exchange) to the IFMC who will collect and maintain the database. We believe this system of data collection has a low burden for facilities. We will be monitoring the costs of submitting data and encourage hospitals to use the virtually no-cost electronic submission method. CMS will post additional information on data submission on its coverage website, through the MedLearn system, and through the QNET education program.

This database is only an initial data collection process. A follow-on database that will replace the QNET database and address additional hypotheses is currently being explored with stakeholders including specialty societies, industry, health plans and hospital associations. The follow-on registry would include clinical outcome information that can be obtained from Medicare databases, and would consider inclusion of device firing information. In addition, industry has committed to developing a system to more closely evaluate the benefit in patients with LVEF 30-35%, NYHA Class IV in CRT-D, or NIDCM of 3-9 months duration. Specialty societies have indicated interest in more clearly defining appropriate facility and provider standards. CMS will continue to encourage the public discussion of the appropriate replacement database. We will also ensure that uses and disclosures of information for any future data collection system be made in compliance with the Standards for Privacy of Individually Identifiable Health Information and that all issues related to patient confidentiality, privacy, and compliance with other Federal laws will be resolved prior to the collection of any data.

Finally, technology exists to easily capture the type of data collected in our initial database and to prevent repeated entry of identical data into the several trials or registries in which hospitals participate. CMS is interested in public input into how the Agency might assist the healthcare community in creating a single data entry system.

Furthermore, to provide greater clarity about how CMS will choose data collection mechanisms to cover, the CMS Council on Technology and Innovation will now begin to develop a draft guidance document on this policy approach in order to make the process more systematic, predictable and transparent. We will shortly announce an open door forum and separately convene an expert panel. Comments on this policy can be submitted through the CTI website at http://www.cms.hhs.gov/providers/cti. An initial draft guidance will be issued in March 2005, at which time additional public feedback will be solicited.

IX.   Conclusions

ICD therapy has been shown in several trials to improve survival and prevent sudden cardiac death in a significant number of patients. Therefore, we will expand coverage as outlined below. However, there still is a considerable mortality rate for patients who have received defibrillators. Patients treated with an ICD in SCD-HeFT had a 22% mortality rate overall. Patients treated with a CRT-defibrillator in COMPANION had a 17.6% mortality rate at 1 year. Since ICDs only treat ventricular tachyarrhythmias, do not prevent death from other cardiac or noncardiac diseases, and may cause adverse events, such as inappropriate shocks and worsening heart failure, they should not be perceived as or projected to be an ideal technology that eliminates significant health risks for patients with heart failure. Responsible use of ICDs should be encouraged. Risk factor reduction and optimal medical therapy, as encouraged in SCD-HeFT, remain crucial in reducing overall mortality from sudden cardiac death.

The Centers for Medicare and Medicaid Services has made the following determination regarding the use of ICDs:

1. CMS has determined that the evidence is adequate to conclude that an implantable cardioverter-defibrillator (ICD) is reasonable and necessary for the following:
   • Patients with ischemic dilated cardiomyopathy (IDCM), documented prior myocardial infarction (MI), New York Heart Association (NYHA) Class II and III heart failure, and measured left ventricular ejection fraction (LVEF) ≤ 35%;
   • Patients with nonischemic dilated cardiomyopathy (NIDCM) > 9 months, NYHA Class II and III heart failure, and measured LVEF ≤ 35%;
   • Patients who meet all current CMS coverage requirements for a cardiac resynchronization therapy (CRT) device and have NYHA Class IV heart failure;
   
   
   For each of these groups, the following additional criteria must also be met:
   1. Patients must be able to give informed consent;
   2. Patients must not have:
      • Cardiogenic shock or symptomatic hypotension while in a stable baseline rhythm;
      • Had a coronary artery bypass graft (CABG) or percutaneous transluminal coronary angioplasty (PTCA) within the past 3 months;
      • Had an acute MI within the past 40 days;
      • Clinical symptoms or findings that would make them a candidate for coronary revascularization;
      • Irreversible brain damage from preexisting cerebral disease;
      • Any disease, other than cardiac disease (e.g. cancer, uremia, liver failure), associated with a likelihood of survival less than one year;
   3. Ejection fractions must be measured by angiography, radionuclide scanning, or echocardiography;
   4. Myocardial infarctions must be documented and defined according to the consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction;⁷⁰
   5. The beneficiary receiving the ICD implantation for primary prevention is enrolled in either an FDA-approved category B IDE clinical trial (42 CFR §405.201), a trial under the CMS Clinical Trial Policy (NCD Manual §310.1) or a qualifying data collection system including approved clinical trials and registries. Initially, an ICD database will be maintained using a data submission mechanism that is already in use by Medicare participating hospitals to submit data to the Iowa Foundation for Medical Care (IFMC)a Quality Improvement Organization (QIO) contractorfor determination of reasonable and necessary and quality improvement. Initial hypothesis and data elements are specified in this decision (Appendix VI) and are the minimum necessary to ensure that the device is reasonable and necessary. Data collection will be completed using the ICDA (ICD Abstraction Tool) and transmitted via QNet (Quality Network Exchange) to the IFMC who will collect and maintain the database. Additional stakeholder-developed data collection systems to augment or replace the initial QNet system, addressing at a minimum the hypotheses specified in this decision, must meet the following basic criteria:
      1. Written protocol on file;
      2. Institutional Review Board review and approval, if required;
      3. Scientific review and approval by two or more qualified individuals who are not part of the research team;
      4. Certification that investigators have not been disqualified.
      For purposes of this coverage decision, CMS will determine whether specific registries or clinical trials meet these criteria.
   6. Providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient medical record.
   
   
2. CMS has determined that the evidence, though less compelling at this time, is adequate to conclude that an ICD is reasonable and necessary for patients with NIDCM > 3 months, NYHA Class II or III heart failure, and measured LVEF ≤ 35%, only if the following additional criteria are also met:
   1. Patients must be able to give informed consent;
   2. Patients must not have:
      • Cardiogenic shock or symptomatic hypotension while in a stable baseline rhythm;
      • Had a CABG or PTCA within the past 3 months;
      • Had an acute MI within the past 40 days;
      • Clinical symptoms or findings that would make them a candidate for coronary revascularization;
      • Irreversible brain damage from preexisting cerebral disease;
      • Any disease, other than cardiac disease (e.g. cancer, uremia, liver failure), associated with a likelihood of survival less than one year;
   3. Ejection fractions must be measured by angiography, radionuclide scanning, or echocardiography;
   4. Myocardial infarctions must be documented and defined according to the consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction;²
   5. The beneficiary receiving the ICD implantation for this indication is enrolled in either an FDA-approved category B IDE clinical trial (42 CFR §405.201), a trial under the CMS Clinical Trial Policy (NCD Manual §310.1) or a prospective data collection system meeting the following basic criteria:
      1. Written protocol on file;
      2. Institutional Review Board review and approval;
      3. Scientific review and approval by two or more qualified individuals who are not part of the research team;
      4. Certification that investigators have not been disqualified.
      For purposes of this coverage decision, CMS will determine whether specific registries or clinical trials meet these criteria.
   6. Providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient medical record.

All other indications for ICDs not currently covered in accordance with this decision will continue to be covered under Category B IDE trials and the CMS routine clinical trials policy (CIM 30-1, NCD 130.1).

¹ Alpert and Thygesen et al., 2000.

² Ibid.

³ CMS NCD 2003 http://www.cms.hhs.gov/mcd/viewtrackingsheet.asp?id=39.

⁴ Bardy et al., 2005.

⁵ Bardy et al., 2005.

⁶ New York Heart Association (published by AHA), 1994.

⁷ AHA, 2004.

⁸ Topol EJ (editor), 2002.

⁹ DiMarco, 2003.

¹⁰ Ibid.

¹¹ Moss et al., 2002.

¹² DAVID trial investigators, 2002.

¹³ Farley, Dixie. Implanted Defibrillators and Pacemakers: A Gentler Jolt and Tickle for Trembling Hearts. 29 Jul. 2002 <http://www.fda.gov/bbs/topics/CONSUMER/CON0279b.html>

¹⁴ Medical Device Approvals. 29 Jul 2002 <http://www.fda.gov/cdrh/mda/index.html>

¹⁵ Deek J., 2001.

¹⁶ Bansch et al., 2002.

¹⁷ Ibid.

¹⁸ Ibid.

¹⁹ Ibid.

²⁰ Ibid.

²¹ NIDCM was defined as left ventricular dysfunction in the absence of coronary artery disease or disproportionate to the severity of coronary artery disease.

²² Strickberger et al., 2003.

²³ Ibid.

²⁴ Ibid.

²⁵ Kadish et al., 2004.

²⁶ Ambient arrhythmias were defined by an episode of nonsustained ventricular tachycardia on Holter or telemetric monitoring (3 to 15 beats at a rate of more than 120 beats per minute) or an average of at least 10 premature ventricular complexes per hour on 24-hour Holter monitoring.

²⁷ Kadish et al., 2004.

²⁸ Ibid.

²⁹ Bristow et al., 2000.

³⁰ Ibid.

³¹ Ibid.

³² Bristow et al., 2004.

³³ Hohnloser et al., 2004.

³⁴ Ibid.

³⁵ Bardy et al., 2005.

³⁶ Ibid.

³⁷ Ibid.

³⁸ Ibid.

³⁹ Ibid.

⁴⁰ Gregoratos et al. 2002.

⁴¹ Ibid.

⁴² Bristow et al., 2004.

⁴³ Bardy et al., 2005.

⁴⁴ AMIOVIRT, DEFINITE, SCD-HeFT.

⁴⁵ Friedman, 1996.

⁴⁶ St. Jude presentation.

⁴⁷ Bristow et al., 2004.

⁴⁸ Bardy et al., 2005.

⁴⁹ Nudell, Bernstein Research Call 05/24/04.

⁵⁰ Bardy et al, 2005.

⁵¹ Strickberger et al., 2003.

⁵² Ibid.

⁵³ Kadish et al., 2004.

⁵⁴ Ibid.

⁵⁵ Bardy et al., 2005

⁵⁶ Bristow et al., 2004.

⁵⁷ Ibid.

⁵⁸ Zareba and Moss, 2003.

⁵⁹ Ibid.

⁶⁰ Moss, 2000.

⁶¹ Ibid.

⁶² Shenkman et al., 2002.

⁶³ Ibid.

⁶⁴ Hohnloser et al., 2004.

⁶⁵ MTWA studies.

⁶⁶ Bloomfield et al., 2004.

⁶⁷ Bardy et al., 2005.

⁶⁸ CRT references – MUSTIC, MIRACLE, CONTAK.

⁶⁹ Alpert and Thygesen et al., 2000.

⁷⁰ Ibid.

Alpert and Thygesen, et al., Myocardial Infarction Redefined—A Consensus Document of The Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction. JACC 2000; 36:959-969.

AVID Investigators. A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. N Engl J Med 1997; 337:1576-1583.

Bardy GH. SCD-HeFT: The Sudden Cardiac Death in Heart Failure Trial. ACC meeting, 2004. http://www.acc04online.org/ondemand/trials1/sessions.asp?link=R

Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352:225-237.

Bansch D, Antz M, Boczor S, et al. Primary prevention of sudden cardiac death in idiopathic dilated cardiomyopathy: the Cardiomyopathy Trial (CAT). Circulation 2002;105:1453-1458.

Bigger JT for the (for the Coronary Artery Bypass Graft (CABG) Patch Trial Investigators). Prophylactic use of implanted cardiac defibrillators in patients at high risk for ventricular arrhythmias after coronary artery bypass graft surgery. N Engl J Med 1997; 337:1569-1575.

Bristow MR, Feldman AM, Saxon LA, COMPANION steering committee and investigators. Heart failure management using implantable devices for intraventricular resynchronization: Comparison of Medical Therapy, Pacing, and Defibrillation in Chronic Heart Failure (COMPANION) Trial. J Card Fail 2000;6:276-285.

Bristow MR, Saxon LA, Boehmer J, et al. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140-2150.

Buxton AE, Lee KL, Fisher JD, et al. for the Multicenter Unsustained Tachycardia Trial Investigators. A randomized study of the prevention of sudden death in patients with coronary artery disease. N Engl J Med 1999; 341:1882-1890.

The Criteria Committee of the New York Heart Association. Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels. 9th ed. Boston, Mass: Little, Brown & Co; 1994:253-256. (available at http://www.americanheart.org/presenter.jhtml?identifier=1712)

Connolly SJ. DINAMIT. ACC meeting, 2004. http://www.acc04online.org/ondemand/trials1/sessions.asp?link=R

Connolly SJ, Gent M, Roberts RS, et al. for the CIDS Investigators. Canadian Implantable Defibrillator Study (CIDS). A randomized trial of the implantable cardioverter-defibrillator against amiodarone. Circulation 2000; 101:1297-1302.

DAVID Trial Investigators. Dual-chamber pacing or ventricular backup pacing in patients with an implantable defibrillator. The Dual Chamber and VVI Implantable Defibrillator (DAVID) trial. JAMA 2002; 288:3115-3123.

Deek J. Systematic reviews of evaluations of diagnostic and screening tests. In Egger M et al, editors. Systematic Reviews in Healthcare. BMJ. 2001.

DiMarco JP. Implantable cardioverter-defibrillators. N Engl J Med 2003; 349:1836-1847.

Gregoratos G, Cheitlin M, Conill A, et al. ACC/AHA Guidelines for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices. J Am Coll Cardiol 1998; 31:1175-1209.

Gregoratos G, Abrams J, Epstein AE, et al. ACC/AHA/NASPE 2002 Guideline Update Implantation of Cardiac Pacemakers and Antiarrhythmia Devices. American College of Cardiology Foundation and the American Heart Association, Inc., September 2002.

Hohnloser SH, Kuck KH, Dorian R, et al. Prophylactic use of an implantable cardioverter-defibrillator after acute myocardial infarction. N Engl J Med 2004;351:2481-2488.

Kadish A, Dyer A, Dauber JP, et al. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med 2004; 350:2151-2158.

Kuck KH, 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; 102:748-754.

Moss AJ. Implantable cardioverter defibrillator therapy. The sickest patients benefit the most. Circulation 2000; 101:1638-1640.

Moss AJ, Greenberg H, Case RB, et al. Long-term clinical course of patients after termination of ventricular tachyarrhythmia by an implanted defibrillator. Circulation 2004;110:3760-3765.

Moss AJ, Hall WJ, Cannom DS et al. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmia. N Engl Med 1996; 335:1933-1940.

Moss AJ, Zareba W, Hall JW, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877-883.

Nudell BM. Medical devices: QRS risk stratification & bang for the buck. Sanford C. Bernstein & Co., LLC Research Call, (05/24/2004).

Nudell BM. Cost per life year saved for low risk ICD populations; consequences for European adoption. Sanford C. Berstein & Co., LLC Research Call, (09/14/2004).

Shenkman HJ, Pampati V, Kandelwal AK, et al. Congestive heart failure and QRS duration. Establishing prognosis study. Chest 2002;122:528-534.

Strickberger SA, Hummel JD, Bartlett TG, et al. Amiodarone versus implantable cardioverter-defibrillator: randomized trial in patients with nonischemic dilated cardiomyopathy and asymptomatic nonsustained ventricular tachycardia – AMIOVIRT. J Am Coll Cardiol 2003;41:1707-1712.

Topol EJ (editor). Textbook of Cardiovascular Medicine. Lippincott Williams & Wilkins 2002, ISBN 0-7817-3225-5.

Zareba W and Moss AJ. Noninvasive risk stratification in postinfarction patients with severe left ventricular dysfunction and methodology of the MADIT II noninvasive electrocardiology substudy. J Electrocardiol 2003; 36 Suppl:101-108.