Risk Stratification for ICDs: The Debate Rages On

The clinical trials MADIT II and SCD-HeFT definitively proved the benefit of ICDs for preventing the risk of sudden cardiac death and paved the way for reimbursement. But now that the excitement has died down, the confusion has set in. Clinicians aren't entirely sure in which real-world patients to implant devices. The debate about the proper methods of risk stratifying patients for ICD implantation rages on.

Two watershed clinical trials have definitively demonstrated the effectiveness of implantable cardioverter defibrillators (ICDs), particularly those that deliver cardiac resynchronization therapy (CRT), at preventing sudden cardiac death (SCD). The studies provided the basis for the US Centers for Medicare and Medicaid Services' (CMS) decision to increase Medicare reimbursement to cover a larger patient population for those products as well, although not as large a pool as the industry and many clinicians believe this clinical data shows warrant such coverage. Having broadly established the clinical efficacy of ICD/CRT devices for sudden cardiac death, the next battleground is risk stratification: determining which patients meet the criteria qualifying them for this therapy, and that was a hot topic at this year's Heart Rhythm Society (HRS) meeting.

The positive data generated by these trials, along with CMS's decision to increase coverage, provided a tremendous boost for the cardiac rhythm management (CRM) segment, making it one of the fastest growing sectors of the device industry. Indeed, the rapid growth and future prospects of this business was the primary driver behind the race that Boston Scientific Corp. [See Deal] won ahead [See Deal] of Johnson & Johnson's Cordis Corp./B-/ [See Deal] [See Deal] to acquire Guidant Corp. , while also boosting the CRM businesses of Medtronic Inc. and St. Jude Medical Inc. (Guidant, Medtronic, and St. Jude are the three current players in the ICD/CRT space.) Now, with the excitement surrounding the initial market expansion having died down, the confusion has set in. Clinicians are increasingly wrestling with the issue of exactly which patients should receive the devices.

Sudden cardiac death is the number one killer in the US. More than 450,000 die suddenly each year due to sustained ventricular tachycardia or ventricular fibrillation, both characterized by a very rapid heartbeat. The condition is unpredictable, and, based largely on the recent clinical results, the clinical community agrees that ICDs have the power to help a large number of patients. Guidant's MADIT II (Multicenter Automatic Defibrillator Implantation Trial) published in 2001, and the SCD-HeFT (Sudden Cardiac Death in Heart Failure Trial), conducted by Medtronic and the National Institutes of Health and published in January 2005, were randomized, controlled, prospective clinical trials studying more than 3,700 patients. MADIT-II formed the basis for implanting and reimbursing ICDs as secondary prevention therapies, and SCD-HeFT went even further, establishing the device therapy as a prophylactic in certain patients that had not yet suffered a myocardial infarction.

Following SCD-HeFT and after much industry clamor, CMS issued a decision expanding reimbursement coverage to include a larger group of patients, although not as large as industry and many clinicians believed was warranted by the recent clinical data. The new coverage criteria includes patients with ischemic dilated cardiomyopathy, prior myocardial infarction, New York Heart Association Class II and III heart failure, left ventricular ejection fraction less than or equal to 35%, and patients who meet CMS coverage eligibility for cardiac resynchronization therapy that also have a New York Heart Association classification of IV.

But there is still a great deal of controversy over the field. It is only now that analyses on the efficacy and efficiency of the therapies are coming to light. It seems that new studies are now being published in the cardiology journals at a weekly rate. Efficacy is measured in terms of how much devices reduce mortality; and efficiency plays out in the number of lives saved versus the number of devices implanted--an issue which rises to prominence with ICDs more than with other devices because of the high price of ICDs and the potentially large affected population. ICDs cost about $25,000 each, and spread over an enormous heart failure population—SCD-HeFT made eligible 1.3 million patients for ICD implantation—according to some estimates, ICD implantation will cost the health system $5 billion. So the issue of exactly which patients qualify for these new devices, and who will be paying for them (and how much) becomes critical.

At the annual meeting of the Heart Rhythm Society in Boston in May 2006, controversies over ICD implantation dominated several plenary sessions, notably, two debates on the need for further risk stratification in ICD implantation. The gray areas revolved around the high cost of devices, the risk of implanted devices, the discomfort of living with them, and the fact that a large number of patients in clinical trials—21%--got shocks from the devices but didn't need them.

In an HRS session entitled, "Controversies in Devices: All Patients with an Ejection Fraction of 35% or less should Receive an ICD," F. Roosevelt Gilliam III, MD, chief of electrophysiology at Duke University Medical Center , argued for such coverage, taking the position that two good trials indicate that patients with an ejection fraction of less than 35% do better with defibrillators than without, and that ICD implantation is an insurance policy against the unpredictable eventuality of sudden cardiac death.

Antagonist Mark E. Josephson, MD, chief of cardiology at Harvard University 's Beth Israel Deaconess Medical Center rebutted, "If all devices were cheap and pleasant to be implanted, with no side effects, I wouldn't disagree." Josephson points out, however, that researchers at Duke recently looked at the Medicare database and found that in the first month post ICD implantation, there was a mortality rate of .9% which is only slightly less than the 1% sudden death rate for a year in the SCD-HeFT trial. "We are finding a 10.8% complication rate in the first month," he says. Josephson also brings up the issue of inappropriate shocks. In several trials, he says, inappropriate shocks are in the 15-20% range. "If you have two inappropriate shocks, your life is ruined. You are depressed and waiting for the next shock." Other studies document high rates of depression and anxiety in ICD patients.

Unwanted shocks go beyond anxiety and depression. Josephson emphasizes that the DINAMIT (Defibrillator In Acute Myocardial Infarction Trial), a prospective randomized trial (sponsored by St. Jude Medical) designed to assess ICD therapy for the prevention of death in high-risk patients early after acute myocardial infarction, revealed that devices significantly increased mortality due to arrhythmias. "We should not treat all patients with an ejection fraction less than 35% unless we want to inflict on the vast majority of patients who do not need them a potentially proarrhythmic device with variable longevity, which may become infected or develop complications in as many patients as they help." Especially, he adds, because they are "likely to deliver inappropriate shocks in up to 25% of patients, leading to depression and a reduced quality of life at an enormous cost."

In a separate debate, arguing the "pro" side for using low ejection fraction as the basis for implantation, Greg Flaker, MD, Professor of Medicine at the University of Missouri, Columbia, says that the cost effectiveness for the devices falls below the magic number of $50,000 per quality life year saved and improves overall survival in patients with left ventricular dysfunction. (Analyses about the cost-effectiveness of devices vary. Daniel B. Mark, MD, of Duke was the lead author on a paper published in the July 11, 2006 issue of Circulation. Mark and his associates found that the quality-adjusted life year saved for patients with ICDs in SCD-HeFT was about $41,000.) Flaker presented data indicating that any one of several risk stratification methods—left ventricular ejection fraction, Holter monitoring, heart rate variability, Microvolt T-Wave Alternans (MTWA) testing, signal-averaged electrocardiology and electrophysiology studies—did just about as good a job of predicting the 2-year probability of an event as left ventricular ejection fraction, none excelled to a great degree.

The protagonists and antagonists seem to be saying the same thing however: today, current methods of risk stratifying patients lead to inappropriate device implantation, and there is indeed a need for better ways of determining which patients are at risk of sudden cardiac death. This is an opportunity that several companies are trying to address with new diagnostic modalities: Cambridge Heart Inc. , General Electric Co.'s GE Healthcare , and Arrhythmia Research Technologies Inc. Cambridge Heart's technology was the approach most often discussed at the Heart Rhythm Society meeting. The company's noninvasive MTWA test recently received national coverage from CMS. (See "A Wave of Success for Cambridge Heart," IN VIVO, April 2006 (Also see "A Wave of Success for Cambridge Heart " - In Vivo, 1 Apr, 2006.).) The MTWA test measures beat-to-beat fluctuations in the T-wave segment of the heartbeat and it has the potential to not only identify at-risk patients who should receive ICDs, but also to rule out patients who won't benefit from the expensive devices. A study just published in the July 4 issue of the Journal of the American College of Cardiology by lead author Paul S. Chan, MD, of the Veterans Affairs Ann Arbor Healthcare System (Michigan), indicates that the incremental cost-effectiveness ratio for ICD implantation based on MTWA testing was $48,800, much lower than the $55,800 price tag for simply using, as a basis for implantation, risk factors without the test.

GE Healthcare has also developed two new testing algorithms for ICD risk stratification; its own T-Wave Alternans algorithm and a Heart Rate Turbulence algorithm–which, when inputted with ambulatory electrocardiography (ECG) data have been shown to help predict if a person is at risk for SCD, according to the company.

Arrhythmia Research Technology sees an opportunity to predict sudden cardiac death with signal averaged, electrocardiography software that enables late potentials to be amplified and enhanced while eliminating undesired electrical noise.

It's not only the manufacturers of new testing methods that will benefit from new risk stratification strategies, the major CRM device manufacturers, which are backing many of the risk stratification clinical trials, stand to gain as well. Today, the market for ICDs is only 20% penetrated. Cardiac rhythm players believe that an objective test that helps physicians determine the appropriate use of ICDs could further increase their products' reach.

Mary Stuart