Use of Microvolt T-wave Alternans Testing
in Clinical Practice to Reduce Sudden
Cardiac Arrest and Death

Richard J. Cohen, MD, PhD
Richard J. Cohen, MD, PhD
Microvolt T-Wave Alternans (MTWA) is a new non-invasive method for identifying patients at increased risk of cardiac arrest and sudden cardiac death from ventricular arrhythmias. MTWA can be measured during a routine exercise stress test during pharmacologic stress or during cardiac pacing. MTWA has been successfully applied to patients both with and without coronary artery disease. Its clinical performance compares favorably with that of other non-invasive risk stratifiers and invasive electrophysiologic study (EPS). The event rate during follow-up among patients who test MTWA-positive is comparable to that of patients with a positive EPS, whereas the event rate among MTWA-negative patients tended to be lower than among EPS-negative patients. In particular, the low event rate during follow-up among patients with a negative MTWA test renders MTWA a suitable non-invasive means for the initial evaluation of patients potentially at risk for ventricular arrhythmias. Patients suitable for MTWA testing include patients presenting with a history suggestive of ventricular arrhythmias, patients with left ventricular dysfunction, and patients at least 4-6 weeks after myocardial infarction. Patients who test MTWA-positive can proceed to invasive study and/or therapy, whereas patients who test negative in most cases can be managed conservatively. In addition, MTWA can be measured during cardiac pacing in the electrophysiology laboratory to provide an additional endpoint in addition to the outcome of programmed ventricular stimulation, or as a means to non-provocatively assess ventricular stability in patients being evaluated or treated for supraventricular arrhythmias. MTWA testing can increase referrals of appropriate patients for further electrophysiologic evaluation and/or therapy. Sudden cardiac death (SCD) from ventricular tachyarrhythmia represents a major public health problem accounting for approximately 300,000-400,000 deaths per year in the United States.1 In recent years, there have been dramatic advances in therapy for the prevention of SCD due to ventricular tachyarrhythmias. Specifically, the development of the implantable cardioverter/defibrillator (ICD) has provided an effective and specific preventative treatment for patients known to be at high risk for sudden cardiac death.2-5 The ICD, however, represents an expensive therapy that is applied only in patients known to be at high risk. Pharmacologic therapy also has advanced. Beta-blockers have been shown to reduce total mortality and arrhythmic death in patients with coronary artery disease.6 Beta-blockers,6,7 angiotensin-converting enzyme (ACE) inhibitors,8 and aldosterone antagonists9 have been shown to reduce total mortality and arrhythmic death in patients with left ventricular dysfunction. Amiodarone has been shown to reduce ventricular tachyarrhythmic events in patients with heart failure.10-12 Sotalol has been shown to reduce the frequency of appropriate ICD discharges for ventricular tachycardia and ventricular fibrillation.13 Until recently, advances in therapeutic modalities have not been paralleled by advances in non-invasive diagnostic technologies to identify high-risk patients. This may partially explain why sudden cardiac arrest and death remains at epidemic levels. Ideally, effective non-invasive diagnostic methods would identify those patients at increased risk of SCD, and then be used to guide prophylactic treatment. The measurement of Microvolt T-Wave Alternans (MTWA) has recently been demonstrated to be a powerful non-invasive predictor of the risk of ventricular tachyarrhythmias and sudden cardiac death in a number of different patient populations. In direct comparison with prior non-invasive diagnostic risk stratification methods, MTWA has been a superior predictor of arrhythmic risk.14-18 In direct comparison with invasive electrophysiologic study, MTWA has been found to be either an equivalent or better predictor of ventricular tachyarrhythmic events and SCD.14,19,20 Moreover, MTWA can now be conveniently measured during exercise or pharmacologic stress testing with commercially available equipment. MTWA testing has been cleared by the United States Food and Drug Administration on the basis of clinical data as a predictor of risk of ventricular tachyarrhythmias and sudden cardiac death. The question now arises: how should MTWA testing be used currently in clinical practice? Background. T-wave alternans is a type of electrical alternans in which there is a beat-to-beat variation in the morphology of the T-wave in an ABABAB… type of pattern. Electrical alternans was described at the very dawn of electrocardiography.22 T-wave alternans, as described below, refers to actual alternation in intrinsic cardiac repolarization processes and should be distinguished from apparent electrical alternans, which results from alternating rotation of the cardiac electrical axis during mechanical alternans as occurs in the setting of pericardial effusion. T-wave alternans now is believed to be due to localized alternation in action potential morphology in particular, alternation in action potential duration.23,24 Alternation in action potential duration, in turn, occurs when the slope of the restitution curve (action potential duration as a function of preceding diastolic interval) exceeds unity at the current value of the diastolic interval25 (Figure 1). Localized alternation in action potential duration in turn is reflected in the surface ECG as T-wave alternans. Localized alternation in action potential duration also results in localized regions of delayed recovery. The resulting spatial dispersion of recovery leads to depolarization wavefront fractionation leading to re-entry. Clinical Interpretation Localized action potential alternation sufficient to increase the risk of spontaneous tachyarrhythmia may result in T-wave alternans in which the variation in T-wave morphology is only a few microvolts in amplitude T-wave alternans of this magnitude would not be detectable by visual inspection of the surface electrocardiogram. Thus, sophisticated signal processing techniques were developed to reliably detect MTWA. The spectral method for detecting T-wave alternans21 allows one to measure MTWA in terms of the alternans voltage, Valt, and also determines the statistical significance of Valt measured in terms of the alternans ratio, which is the number of standard deviations by which Valt exceeds the noise level. Only values of Valt which are greater than 1.9 microvolts and are associated with an alternans ratio greater than three are considered statistically significant. MTWA is highly heart rate dependent, presumably because the restitution curve tends to get steeper at shorter diastolic intervals. Clinical testing for MTWA involves elevating the patient s heart rate by means of exercise stress, pharmacologic stress or cardiac pacing. MTWA, which is present at rest or is consistently present above a patient-specific onset heart rate, which in turn is less than or equal to 110 bpm, is considered clinically significant (Figure 2). Sustained MTWA with an onset heart rate above 110 bpm has not been deemed clinically significant, presumably because heart rates above 110 bpm are reached only infrequently during ambient activity. Patients with sustained MTWA at rest or with onset heart rate of 110 bpm or less are classified as having a positive test, patients who do not test positive and definitively do not have T-wave alternans at elevated heart rates are classified as negative. Remaining patients are classified as indeterminate. Causes of indeterminacy include inadequate heart rate elevation and excessive levels of ectopy or noise. In studies to date, indeterminacy rates have generally exceeded 20%. A number of technical improvements are being implemented that promise to significantly reduce the indeterminacy rate. Clinical Studies in Patients Presenting for Evaluation of Known or Suspected Cardiac Arrhythmias A prospective study19 of the predictive accuracy of MTWA measured during atrial pacing in 83 consecutive patients undergoing EPS revealed that 81% of patients testing T-wave alternans positive sustained a ventricular tachyarrhythmic event (sudden cardiac death, cardiac arrest, or electrocardiographically documented sustained ventricular tachycardia, or appropriate ICD discharge as documented by review of the stored electrocardiogram) within 20 months of follow-up, whereas only 6% of patients who tested T-wave alternans negative sustained an event (Figure 3). Invasive EPS yielded very similar predictive accuracy. The first study14 to prospectively evaluate MTWA measured during exercise stress involved 95 patients receiving ICDs for clinical indications; the endpoint of this study was appropriate ICD discharge as documented by review of the stored electrograms. In addition to MTWA, patients also were risk stratified by means of EPS, left ventricular ejection fraction, baroreceptor sensitivity, signal-averaged ECG, QT dispersion heart rate variability, mean RR interval over 24 hours and the presence of non-sustained ventricular tachycardia during 24-hour Holter monitoring. Of all the risk stratifiers, including EPS, only MTWA was a statistically significant predictor of appropriate ICD discharge over 18 months of follow-up (Figure 4). In a multi-center trial of 313 patients undergoing EPS, Gold et al.20 found that MTWA was a highly significant predictor of ventricular tachyarrhythmic events. Forty-one percent of the patients were being evaluated for syncope or presyncope. During 14-month follow-up, the presence of MTWA was associated with a relative risk of 10.9 compared to 7.1 for EPS for the endpoint of ventricular tachyarrhythmic events (Figure 5). The relative risks for the combined endpoint of ventricular tachyarrhythmic events plus all cause mortality were 13.9 for MTWA and 4.7 for EPS. Clinical Studies in Patients with Left Ventricular Dysfunction Klingeheben17 et al. studied 107 consecutive patients with New York Heart Association congestive heart failure and no prior history of sustained ventricular arrhythmias. Two-thirds of the patients had coronary artery disease and one-third had non-ischemic dilated cardiomyopathy. MTWA was evaluated along with left ventricular ejection fraction, baroreceptor sensitivity, heart rate variability, signal-averaged ECG, baroreceptor sensitivity, mean RR interval over 24 hours, presence of non-sustained ventricular tachycardia on 24-hour Holter monitoring. Of these seven measures, only MTWA was a statistically significant predictor of ventricular tachyarrhythmic events. Twenty-one percent of the patients who tested positive for MTWA had a ventricular tachyarrhythmic event during 18-month follow-up, whereas none of the patients who tested negative had an event (relative risk = ∞) (Figure 6). Several studies26-28 have evaluated patients with non-ischemic dilated cardiomyopathy. These studies have demonstrated a high statistically significant association between the presence of MTWA and the occurrence of spontaneous ventricular tachycardia (Figure 7). Clinical Studies in Patients with Recent Myocardial Infarction Ikeda et al.18 studied 102 consecutive patients with acute myocardial infarction. Patients were studied a median of 20 days after myocardial infarction and were followed for twelve months for the occurrence of ventricular tachyarrhythmic events. T-wave alternans was a high significant predictor (p