Microvolt T-wave alternans physiological basis, methods of measurement, and clinical utility--consensus guideline by International Society for Holter and Noninvasive Electrocardiology

Abstract

This consensus guideline was prepared on behalf of the International Society for Holter and Noninvasive Electrocardiology and is cosponsored by the Japanese Circulation Society, the Computers in Cardiology Working Group on e-Cardiology of the European Society of Cardiology, and the European Cardiac Arrhythmia Society. It discusses the electrocardiographic phenomenon of T-wave alternans (TWA) (i.e., a beat-to-beat alternation in the morphology and amplitude of the ST-segment or T-wave). This statement focuses on its physiological basis and measurement technologies and its clinical utility in stratifying risk for life-threatening ventricular arrhythmias. Signal processing techniques including the frequency-domain Spectral Method and the time-domain Modified Moving Average method have demonstrated the utility of TWA in arrhythmia risk stratification in prospective studies in >12,000 patients. The majority of exercise-based studies using both methods have reported high relative risks for cardiovascular mortality and for sudden cardiac death in patients with preserved as well as depressed left ventricular ejection fraction. Studies with ambulatory electrocardiogram-based TWA analysis with Modified Moving Average method have yielded significant predictive capacity. However, negative studies with the Spectral Method have also appeared, including 2 interventional studies in patients with implantable defibrillators. Meta-analyses have been performed to gain insights into this issue. Frontiers of TWA research include use in arrhythmia risk stratification of individuals with preserved ejection fraction, improvements in predictivity with quantitative analysis, and utility in guiding medical as well as device-based therapy. Overall, although TWA appears to be a useful marker of risk for arrhythmic and cardiovascular death, there is as yet no definitive evidence that it can guide therapy.

Figure 1. TWA and Nonalternating Fluctuations

(A) Precordial (V₄) electrocardiogram rhythm strip (left) and high-resolution template of QRS-aligned complexes (right) during routine exercise testing from a patient with coronary artery disease. The template illustrates T-wave alternans (TWA) as a separation between ST-T segments in A and B beats. TWA magnitude = 106 μV. Sec = second. (B) Nonalternating fluctuations in T-wave amplitude after ajmaline administration in a Brugada syndrome patient.

Figure 2. Discordant Alternans Leading to VF

Action potential propagation between 2 ventricular sites (A to B) is shown with transition from concordant to discordant alternans and development of ventricular fibrillation (VF). Shaded areas indicate dispersion of repolarization between sites. *Premature beats. L = long action potential duration; S = short action potential duration. Adapted, with permission, from Oshodi et al. (2).

Figure 3. Interlead Discordant Alternans

Mid-myocardial (Mid)/epicardial (Epi) recordings after an abrupt decrease in pacing cycle length from 1,000 to 500 ms. Ventricular tachycardia was initiated by a premature focal discharge from a left ventricular endocardial site, which induced areas of functional conduction block and re-entrant excitation. Numbers above the tracing are calculated activation–recovery intervals (ARI). Numbers below the tracing are dispersion of ARI between mid-myocardial and epicardial sites. *Marked discordant alternans is associated with the onset of ventricular tachyarrhythmia after a premature beat. Reprinted, with permission, from Chinushi et al. (3).

Figure 4. Spectral TWA Method

Schematic representation of T-wave alternans (TWA) assessment of the electrocardiogram (ECG) with the Spectral Method (77). See text for details. FFT = Fast Fourier Transform.

Figure 5. MMA TWA Method

Flow chart of the major components of the Modified Moving Average (MMA) method of T-wave alternans (TWA) analysis. See text for details. Adapted, with permission, from Nearing et al. (19).

Figure 6. Survival Curves for Prospective TWA Studies

(A) The ALPHA (T-Wave Alternans in Patients with Heart Failure) T-wave alternans (TWA) study: Kaplan-Meier cumulative event-free (cardiac death/life-threatening arrhythmias) survival according to TWA Spectral Method testing (37). Patients at risk are shown at selected time points. (B) The REFINE (Risk Estimation Following Infarction, Noninvasive Evaluation) TWA study (54). Risk of cardiac death or resuscitated cardiac arrest (primary outcome) among patients with impaired autonomic tone, measured with heart rate turbulence + abnormal repolarization alternans + ejection fraction <50% versus remaining patients. Left panel indicates TWA results with the Spectral Method during exercise; right panel indicates results of MMA analysis during recovery from exercise. Patients at risk are shown at selected time points. (C) TWA sub-study of SCD-HeFT (Sudden Cardiac Death in Heart Failure Trial) (61). Comparison of primary event rates for microvolt T-wave alternans (MTWA) in non-negative (dashed line) and negative (solid line) patients (n = 490). No significant difference between event rates was found for the 2 groups. Reprinted, with permission, from Gold et al. (61). (D) Ambulatory electrocardiogram (ECG) TWA study (71). Event-free survival from cardiac mortality in ischemic (left) and nonischemic (right) subgroups on the basis of maximum TWA voltage by the Modified Moving Average (MMA) method. CI = confidence interval.

Figure 7. Beta-Blocker TWA Meta-Analysis

Association between microvolt TWA analyzed by the Spectral Method and ventricular arrhythmic events, stratified by screening protocol discontinuation of beta-blocker therapy. Washout of beta-adrenergic blockade impaired predictivity. Risk ratio (RR): 12.44 reflects 0.5 correction factor; therefore, no 95% CI is depicted. *Study estimates for meta-analysis were hazard or risk ratios. **No events in microvolt TWA negative group (p = 0.04). ***Test for heterogeneity: p = 0.025. Abbreviations as in Table 1. Reprinted, with permission, from Chan et al. (83).

Figure 8. The REFINE/FINCAVAS TWA Study

Rates of cardiovascular death (orange bars) and total mortality (blue bars) at 4 years, by quintile of TWA magnitude with the MMA method (67). FINCAVAS = Finnish Cardiovascular Study; other abbreviations as in Figure 6.