Elsevier

Journal of Electrocardiology

Volume 46, Issue 6, November–December 2013, Pages 480-486
Journal of Electrocardiology

Prognostic value of average T-wave alternans and QT variability for cardiac events in MADIT-II patients

https://doi.org/10.1016/j.jelectrocard.2013.08.004Get rights and content

Abstract

Background

Identifying which patients might benefit the most from ICD therapy remains challenging. We hypothesize that increased T-wave alternans (TWA) and QT variability (QTV) provide complementary information for predicting appropriate ICD therapy in patients with previous myocardial infarction and reduced ejection fraction.

Methods

We analyzed 10-min resting ECGs from MADIT-II patients with baseline heart rate > 80 beats/min. TWA indices IAA and IAA90 were computed with the multilead Laplacian Likelihood ratio method. QTV indices QTVN and QTVI were measured using a standard approach. Cox proportional hazard models were adjusted considering appropriate ICD therapy and sudden cardiac death (SCD) as endpoints.

Results

TWA and QTV were measured in 175 patients. Neither QTV nor TWA predicted SCD. Appropriate ICD therapy was predicted by combining IAA90 and QTVN after adjusting for relevant correlates.

Conclusion

Increased TWA and QTV are independent predictors of appropriate ICD therapy in MADIT-II patients with elevated heart rate at baseline.

Introduction

Cardiovascular diseases are the major cause of death in adults worldwide. The presence of electrical instability and the presence of severely depressed left ventricular function have been studied as markers for an arrhythmic death in many clinical trials for primary prevention of sudden cardiac death (SCD): MADIT I1 and II,2 MUSTT,3 SCD-HeFT,4 DINAMIT,5 and more recently in MADIT-CRT.6 Most of these trials have shown that prophylactic therapy with an implantable cardioverter defibrillator (ICD) significantly reduces overall mortality in post-infarction patients with severe left ventricular dysfunction (LVEF ≤ 30%). In MADIT-II,2 defibrillator implantation was associated with a significant improvement in survival compared to medication therapy. Patients treated only with medication presented a 78% survival rate after two years whereas patients treated with an ICD and medication had an 84% survival. Yet, only a small fraction of patients with ICDs actually receive life saving therapy from the devices. Therefore, clinicians lack risk markers which identify patients at a higher risk of experiencing ventricular tachycardia (VT) or ventricular fibrillation (VF) requiring an ICD shock, so that prophylactic ICD therapy can be selectively applied only to those patients who will benefit the most from it.

Electrocardiographic markers of the presence of myocardial vulnerability, such as the frequency of ventricular ectopic beats and the presence of sustained/non-sustained ventricular arrhythmias, help the clinicians to assess the presence of electrical susceptibility to life-threatening arrhythmias. Generally, they are considered in combination with New York Heart Association (NYHA) class and left ventricular ejection fraction (LVEF), which reflect the loss of cardiac reserve and adversely affect the prognosis of patients with structural heart disease.7

In this work, we propose to investigate electrocardiographic markers of repolarization instability in combination with relevant clinical markers in order to predict ICD therapies and SCD in MADIT-II patients. We hypothesize that our new method to measure T-wave alternans (TWA), which successfully predicts SCD in patient with chronic heart failure,8 could help identifying the patients who would benefit the most from ICD implantation. In the previous work8 we studied the prognostic value of different measures of average and maximum TWA activity over heart rates ranging from 60 to 110 beats/min, and found two indices which predicted SCD: the index of average alternans (IAA) and the average alternans activity in the heart rate (HR) range of 80–90 beats/min (IAA90). The aim of the present study is to validate those findings and to combine the measurements of TWA with measures of QT variability (QTV) in order to evaluate their complementarity in MADIT-II patients.

Section snippets

Study population

The study population consisted of patients enrolled in the MADIT-II trial.2 Each patient had a history of myocardial infarction at least 30 days prior to enrollment, and LVEF≤30%. No other risk stratifier was used at enrollment. Patients were randomized to either ICD therapy or conventional medication therapy.

Holter ECGs were recorded for 10 minutes at rest in supine position at the time of enrollment in the study for 902 of the 1232 enrolled patients. For this study, we excluded patients with

Study population

Among the 902 patients with at least one Holter recording in the overall MADIT-II population, 46 patients experienced AF. After excluding these patients, and the patients with HR < 80 beats/min during the whole ECG (n=518), the final study population consisted of 338 subjects (Fig. 2). The average follow-up of patients was 1.8 ± 1.03 years. IAA could not be computed in 16 patients because all segments of their ECGs were rejected for automatic analysis by the TWA algorithm, due to unstable HR,

Discussion

The prognostic value of TWA has been subject to extensive research.[14], [15] In MADIT-II-like patients, TWA identified a low-risk group unlikely to benefit from ICD therapy,16 although there is conflicting evidence suggesting that in ICD-treated patients the risk of VT/VF events does not differ according to TWA classification.17 Such discrepancy has been addressed by recent meta-analyses,[14], [15] which questioned the suitability of appropriate ICD therapy as a surrogate for SCD in clinical

Conclusion

Increased TWA measured in rest ECGs was found to predict appropriate ICD therapy in patients with elevated resting HR from the ICD arm of the MADIT-II trial. IAA90 and QTVN were found to be complementary and both associated with an increased probability of arrhythmic events. Our results suggest that MADIT-II –type patients with elevated resting HR could be pre-selected for ICD implantation based on a 10-minute ECG test.

Acknowledgment

This work was supported by the Spanish Government (MINECO), and UE (FEDER) under project TEC2010-21703-C03-02; and from European Social Fund and Aragon Government (T30). CIBER-BBN is financed by Instituto de Salud Carlos III through the European Regional Development Fund (Spain). WZ and AJM have received research support from Boston Scientific.

References (23)

Cited by (12)

View all citing articles on Scopus
View full text