SNOWMASS, COLO. — The value of T-wave alternans testing as a risk stratification tool for selecting implantable cardioverter defibrillator candidates has been cast into serious doubt by the disappointing results of two recent large clinical trials.
This test is clearly not ready for prime-time application in clinical practice. Further well-designed studies are needed to resolve the existing confusion surrounding its role. And yet the Centers for Medicare and Medicaid Services has reimbursed for T-wave alternans (TWA) testing for risk stratification in ICD candidates since 2006, in a noteworthy instance of the “federal follies,” Dr. William H. Spencer III observed at a conference sponsored by the Society for Cardiovascular Angiography and Interventions.
“Talk about paradoxical reactions by our government,” commented Dr. Spencer, professor of medicine at the Medical University of South Carolina, Charleston.
The hypothesis underlying TWA testing is that it noninvasively identifies patients having an electrophysiologic substrate for reentrant ventricular tachyarrhythmias. If true, that would allow more selective placement of costly ICDs than is currently possible. Many patients who receive an ICD for primary prevention solely on the basis of the current criterion of a left ventricular ejection fraction below 35% will never use the device during their lifetime.
However, the results of the recent Microvolt T-Wave Alternans Testing for Risk Stratification of Post-MI Patients (MASTER 1) and Alternans Before Cardioverter Defibrillator (ABCD) trials raise a question as to whether the test merely identifies a population that's relatively sick and therefore at increased risk for all-cause mortality rather than specifically for the arrhythmic sudden cardiac deaths (SCDs) that ICDs are designed to prevent.
In other words, TWA testing might not provide incremental value over clinical markers of increased mortality risk, such as comorbid conditions, advanced age, and lower left ventricular ejection fraction (EF), the cardiologist said.
MASTER 1, a Medtronic Inc.-sponsored trial presented at the 2007 American Heart Association meeting, involved 575 patients who received an ICD because they had a prior MI and an EF below 30%. At 3 years' follow-up, the primary end point—the rate of life-threatening ventricular tachyarrhythmias—was not significantly different between patients with a positive or indeterminate TWA and those who were TWA-negative.
“Very disappointing. TWA did not point out the kind of patient we would put an ICD in,” Dr. Spencer said at the meeting, which was cosponsored by the American College of Cardiology.
A positive or indeterminate TWA result did predict, however, significantly increased risk of all-cause mortality. “There are two alternative explanations for this. Maybe the people with indeterminate or positive tests drove their cars faster. But it's more likely that they had more markers for mortality. It tells you that there are clinical variables we're not measuring with TWA,” Dr. Spencer said.
The St. Jude Medical Inc.-sponsored ABCD study was presented at the March 2006 meeting of the ACC yet remains unpublished to date. It involved 566 patients with ischemic heart disease and a low EF who met current criteria for an ICD. All underwent both TWA and invasive electrophysiologic testing. The results showed that the two tests were equivalent at predicting SCD or appropriate ICD shocks at 12 and 24 months. Dr. Spencer's reaction: So what? Saying TWA is equivalent to electrophysiologic testing is faint praise, because most electrophysiologists no longer routinely do invasive testing in an effort to identify ICD candidates because of its lack of incremental value.
The TWA test reimbursed by CMS is conducted on a treadmill using a modified Bruce protocol at a heart rate of 110–120 beats per minute. The test relies on proprietary equipment marketed by Cambridge Heart Inc. The results can be interpreted with minimal training.
Earlier observational studies of TWA yielded conflicting results. MASTER 1 and ABCD were supposed to clear matters up but have had the opposite effect, in Dr. Spencer's view.
Dr. Michael R. Gold predicted that in the long run, genetic testing for predisposition to SCD will be far more useful than TWA or any of the other risk-stratification methods for ICD candidates now being investigated. He pointed to several intriguing studies that suggest family history might be an important risk factor for cardiac arrest.
That's a provocative finding because physicians haven't traditionally been trained to ask about family history of SCD in patients with coronary disease, noted Dr. Gold, professor of medicine and director of adult cardiology at the Medical University of South Carolina.
In a retrospective case-control study, Dr. Kari S. Kaikkonen and colleagues at the University of Oulu (Finland) scrutinized 138 consecutive individuals who experienced SCD, 254 consecutive patients who survived an acute MI, and 470 healthy controls. Individuals with a history of SCD in a first-degree relative were 2.2 times more likely to experience SCD than were controls. A family history in two or more first-degree relatives was associated with an 11.3-fold increased risk (Circulation 2006;114:1462–7).