Response to Letters Regarding Article, “Role of Microvolt T-Wave Alternans in Assessment of Arrhythmia Vulnerability Among Patients With Heart Failure and Systolic Dysfunction: Primary Results From the T-Wave Alternans Sudden Cardiac Death in Heart Failure Trial Substudy”
We thank the authors who responded to our article.1 We value their interest and insightful comments on the primary results of the T-wave Alternans Sudden Cardiac Death in Heart Failure Trial (TWA SCD-HeFT) substudy. Madias questions whether a single test of microvolt TWA is useful for a changing substrate. We certainly agree that patients with dilated cardiomyopathy (ischemic or nonischemic) have a changing substrate over time. Despite this limitation, the SCD-HeFT main trial showed that a single assessment of left ventricular ejection fraction and heart failure functional status was sufficient to identify a cohort that showed a mortality reduction with implantable cardioverter defibrillator (ICD) placement. Although the traditional Kaplan-Meier method was prospectively used to assess differences among groups, it is suggested that repeating measurements every 6 months may be more useful. This strategy was not tested in the SCD-HeFT study, so we cannot comment on it. To apply this approach for TWA would dictate similar approaches by every other test we use in cardiology, and this is not normal practice and proved unnecessary in SCD-HeFT. Madias also suggests that treating TWA amplitude as a continuum rather than a dichotomous variable could be more predictive. We used the validated methodology of alternans amplitude, duration, and heart rate achieved to define tests as positive, negative, or indeterminate by the manufacturer and the Food and Drug Administration. This standard has been used in all large prospective studies of TWA. Unfortunately, changing the predefined threshold value (1.9 μV) for 490 traces repeatedly to measure TWA as a continuum is not possible with the software provided by the manufacturer and used in this study.
De Ferrari and Verrier raise the concern that including ICD shocks as an end point may have decreased TWA’s predictive value. We addressed this point with 2 additional analyses. First, we evaluated the predictive value of TWA in the placebo and amiodarone arms of SCD-HeFT, eliminating subjects with ICDs. The hazard ratios for the primary end point were 1.251 (95% CI, 0.539 to 2.903), positive versus negative, and 0.98 (95% CI, 0.447 to 2.153), nonnegative versus negative. Second, we eliminated ICD shocks as a component of the primary end point from the full cohort of 490 patients. This analysis revealed that TWA still had no significant predictive value with a hazard ratio of 1.039 (95% CI, 0.542 to 1.992), positive versus negative, and 1.016 (95% CI, 0.563 to 1.833), nonnegative versus negative. De Ferrari and Verrier also point out that some ICD shocks were also classified as sustained VT. It should be noted that prolonged-detection, high-energy, shock-only therapy and older-generation devices with longer charge times were used in this trial. Therefore, short runs of nonsustained VT seldom received therapy. Any such shocks were considered inappropriate and were not included in this analysis. However, it is likely that many shocks were delivered before the arrhythmia lasted 30 seconds. The classification of arrhythmias and shocks was adjudicated by a blinded events committee, and these results were completed and “locked” in the database before any TWA analysis was performed. De Ferrari and Verrier also question the lack of consecutive patients and the high indeterminate rate. As explained in the article, this substudy cohort was similar clinically to the full study group in SCD-HeFT. Moreover, we intentionally included many study sites and did not exclude patients unlikely to complete the study so that our data would more closely represent general clinical practice and not specialized centers more experienced in this test. This approach actually represents a more realistic assessment of the test’s value than when used only by those interested in this type of equipment. Finally, the authors suggest that TWA better predicts arrhythmic events among patients with nonischemic cardiomyopathy. Although the predictive value was better in our study for nonischemic patients, this result still did not reach statistical significance (hazard ratio 1.67, P=0.21, for TWA versus non-TWA) for the 250 subjects included in this analysis.
In summary, despite the suggestions made by these authors, the data still strongly support the conclusion that TWA has little value in aiding clinical decisions about ICD therapy among patients with symptomatic heart failure and left ventricular systolic dysfunction.
Source of Funding
This study was supported by the National Institutes of Health and Cambridge Heart Inc, Bedford, Mass.
Dr Gold has received research grant support from Cambridge Heart and has served as a consultant for Medtronic, Boston Scientific, and St. Jude Medical. Dr Costantini has served as a consultant for Boston Scientific and St. Jude Medical. Dr Poole has received research grant support from HAT and Biotronik and is on the Speakers Bureau for Boston Scientific, Sorin, Medtronic, and Biotronik Dr Mark has received research grant support from and has served as a consultant for Medtronic. Dr Lee has received research grant support from Cambridge Heart and honoraria from Medtronic. Dr Bardy has received research grant support from the National Institutes of Health, Medtronic, Philips, and Laerdal, has an ownership interest in Cameron Health, and has served as a consultant for Philips and Boston Scientific. He is employed by the Seattle Institute for Cardiac Research. The remaining authors report no conflicts.
Gold MR, Ip JH, Costantini O, Poole JE, McNulty S, Mark DB, Lee KL, Bardy GH. Role of microvolt T-wave alternans in assessment of arrhythmia vulnerability among patients with heart failure and systolic dysfunction: primary results from the T-Wave Alternans Sudden Cardiac Death in Heart Failure Trial substudy. Circulation. 2008; 118: 2022–2028.