This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gomes, J. A. Articles by Hafley, G. E. Search for Related Content PubMed PubMed Citation Articles by Gomes, J. A. Articles by Hafley, G. E. Related Collections Electrophysiology Electrocardiology Arrhythmias, clinical electrophysiology, drugs

(Circulation. 2001;104:436.)
© 2001 American Heart Association, Inc.

Clinical Investigation and Reports

Prediction of Long-Term Outcomes by Signal-Averaged Electrocardiography in Patients With Unsustained Ventricular Tachycardia, Coronary Artery Disease, and Left Ventricular Dysfunction

J. Anthony Gomes, MD; Michael E. Cain, MD; Alfred E. Buxton, MD; Mark E. Josephson, MD; Kerry L. Lee, PhD; Gail E. Hafley, MS

From Mount Sinai School of Medicine of New York University (J.A.G.), New York, NY; Washington University School of Medicine (M.E.C.), St Louis, Mo; Rhode Island Hospital and Brown University School of Medicine (A.E.B.), Providence, RI; Beth Israel Hospital (M.E.J.), Boston, Mass; and Duke Clinical Research Institute (K.L.L., G.E.H.), Durham, NC.

Correspondence to Dr J. Anthony Gomes, Box 1054, The Mount Sinai Medical Center, One Gustave L. Levy Place, New York, NY 10029. E-mail anthony.gomes{at}mountsinai.org

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Background— An abnormal signal-averaged ECG (SAECG) is a noninvasive marker of the substrate of sustained ventricular tachycardia after myocardial infarction. We assessed its prognostic ability in patients with asymptomatic unsustained ventricular tachycardia, coronary artery disease, and left ventricular dysfunction.

Methods and Results— A blinded core laboratory analyzed SAECG tracings from 1925 patients in a multicenter trial. Cox proportional hazards modeling was used to examine individual and joint relations between SAECG variables and arrhythmic death or cardiac arrest (primary end point), cardiac death, and total mortality. We also assessed the prognostic utility of SAECG at different levels of ejection fraction (EF). A filtered QRS duration >114 ms (abnormal SAECG) independently predicted the primary end point and cardiac death, independent of clinical variables, cardioverter-defibrillator implantation, and antiarrhythmic drug therapy. With an abnormal SAECG, the 5-year rates of the primary end point (28% versus 17%, P=0.0001), cardiac death (37% versus 25%, P=0.0001), and total mortality (43% versus 35%, P=0.0001) were significantly higher. The combination of EF <30% and abnormal SAECG identified a particularly high-risk subset that constituted 21% of the total population. Thirty-six percent and 44% of patients with this combination succumbed to arrhythmic and cardiac death, respectively.

Conclusions— SAECG is a powerful predictor of poor outcomes in this population. The noninvasive combination of an abnormal SAECG and reduced EF may have utility in selecting high-risk patients for intervention.

Key Words: electrocardiography • arrhythmia • prognosis • mortality

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
The signal-averaged ECG (SAECG)¹ is a highly amplified and signal-processed ECG. Unlike standard tracings, SAECGs can detect microvolt-level electrical potentials in the terminal QRS complex, known as late potentials.² These arise from scarred myocardium, which can be the source of reentrant malignant ventricular arrhythmias.³ An abnormal duration of the QRS complex on SAECG or the presence of late potentials has been correlated with the inducibility of ventricular tachycardia (VT).^4,5 Its prognostic value for arrhythmic events after myocardial infarction (MI)^6–11 has been established. However, its role in long-term prognostication in patients with unsustained VT, coronary artery disease (CAD), and left ventricular dysfunction has not been studied. These patients have a long-term mortality rate of 20% over a 4- to 5-year period, of which at least one third reflects sudden death.^12–14

The purpose of the present study was to assess the relation between SAECG and long-term outcomes in patients with unsustained VT, CAD, and left ventricular dysfunction and to define an algorithm that would identify patients at high risk for arrhythmic and/or cardiac death. Patients enrolled in the Multicenter Unsustained Tachycardia Trial (MUSTT) are the subject of the present report.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Patients
Details of patient enrollment and the protocol have been described.¹⁵ Of 2202 patients enrolled at 85 sites in the United States and Canada from November 1, 1990, to October 31, 1996, baseline SAECGs were obtained in 1925 patients. The SAECGs of 1268 patients (66%) qualified for this analysis. We excluded the tracings of 612 patients (32%) with bundle-branch block or intraventricular conduction defects (>120 ms), 30 patients (1.6%) with ventricular-paced rhythm, and 15 patients (0.8%) with unreadable tracings. Of the 1268 patients, 364 had inducible sustained VT and were randomized into the trial: 193 patients to no antiarrhythmic therapy and 171 patients to electrophysiologically (EPS)-guided antiarrhythmic therapy. The remaining 904 patients were followed in a registry. All tracings were obtained with commercially available systems and analyzed at core laboratories (Mount Sinai Medical Center and Washington University) without knowledge of treatments or outcomes or whether sustained VT was inducible. In all patients, SAECG quantitative variables¹ were processed at 40 to 250 Hz and included filtered QRS duration (fQRS), the root-mean-square (RMS) voltage of the terminal 40 ms of the QRS complex, and duration of low-amplitude signals (<40 µV). Variables were measured with computer-derived algorithms.

End Points
The primary end point of the trial was cardiac arrest or death from arrhythmia. Secondary end points were cardiac death and total mortality. Deaths were classified by a committee using a modified Hinkle-Thaler system,¹⁶ independent of the SAECG core laboratories. Arrhythmic deaths included the following: unwitnessed death in patients in their usual state of health when last seen, witnessed and instantaneous death, nonsudden death due to incessant tachycardia, death due to cardiac arrest sequelae, death due to antiarrhythmic drug toxicity, and death due to complications of implantable defibrillators. Deaths from heart failure or cardiogenic shock were not considered arrhythmic. Narrative descriptions of events and hospital records were edited by the data-coordinating center to ensure that outcomes were appropriately classified without the knowledge of treatment assignments or inducibility of sustained VT. The median length of follow-up in these 1268 patients was 44 months (25th and 75th percentiles: 29 and 60 months).

Statistical Analysis
Cox proportional hazards modeling¹⁷ was used to examine individual and joint relationships between the SAECG variables and the outcomes of time to arrhythmic death or cardiac arrest and time to cardiac death. We used a flexible model-fitting approach with cubic splines (polynomials) to characterize possible nonlinear relations between the continuous SAECG variables and the outcome.^18,19

To assess the prognostic content of a "normal" versus "abnormal" SAECG, we also prospectively dichotomized the SAECG variables at standard cut points.^1,20 The relative prognostic significance of several definitions of an abnormal test was assessed and compared by using the Cox model.

We also performed covariate-adjusted assessments of the strength of the relations of the SAECG variables with the primary and secondary outcomes, adjusting for age, race, sex, ejection fraction (EF), prior MI, prior thrombolysis, prior bypass surgery, time from MI to enrollment, severity of CAD, inducibility status at the baseline EPS, initial antiarrhythmic drug treatment, and initial implantable cardioverter-defibrillator (ICD) use.

To illustrate the ability of the SAECG to stratify risk, we divided patients by fQRS (>114 ms [abnormal] versus 114 [normal]) and generated Kaplan-Meier survival curves for each outcome. We also generated Kaplan-Meier curves for these categories of fQRS combined with an EF <30% versus 30%.

We assessed the consistency of the prognostic relations of the SAECG across the spectrum of patients in this population, whether they were treated with antiarrhythmic therapy or untreated, by reporting 5-year Kaplan-Meier²¹ event rates and Cox hazard ratios (with 95% CIs) by normal or abnormal fQRS.

	Results

Top Abstract Introduction Methods Results Discussion References

 
There were 230 cardiac arrests or deaths from arrhythmia (18%), 341 cardiac deaths (27%), and 457 total deaths (36%) among the 1268 patients. All patients had CAD; 90% had had a previous MI. The median EF was 30% (25th and 75th percentiles: 25% and 35%).

Significance of SAECG
To determine the univariate significance of SAECG, we treated its quantitative values as continuous and dichotomized variables. Filtered QRS duration (P<0.001) and the RMS voltage in the terminal 40 ms of the QRS complex (P=0.001) were highly significant when assessed as continuous variables for the primary end point and for cardiac death. For fQRS values of 125 ms, survival decreased approximately linearly with increasing fQRS. Survival did not decrease further at fQRS >120 ms. The relation of the RMS voltage with survival was more complex; survival was lower with low voltages (<20 µV), improved at intermediate levels, and decreased again at high voltages. Filtered QRS duration was most strongly related to both arrhythmic and cardiac death.

When SAECG variables were dichotomized, an fQRS cut point of 114 ms was the most significant prognostically compared with cut points of 105, 110, and 120 ms, with the RMS voltage, and with the duration of low-amplitude signals. The RMS voltage cut point of 20 µV was significant for cardiac death (P=0.023) but not for arrhythmic death; voltage dichotomized at 15, 25, and 30 µV was not significant for either outcome. Duration of low-amplitude signals dichotomized at 38 ms likewise was significant for cardiac death (P=0.04) but not for arrhythmic death; cut points at 26, 32, and 44 ms were significant for neither outcome. When an abnormal SAECG was defined as any single abnormal variable (with standard cut points used for each variable) or as abnormal fQRS and RMS voltage,¹⁰ it was highly significant for both end points. However, an fQRS >114 ms was considerably more significant alone than were other definitions of an abnormal test. Relations were stronger for cardiac death than for arrhythmic death.

In multivariable analysis of SAECG quantitative values (Table 1), only fQRS and RMS voltage significantly predicted survival when assessed as continuous variables. When the SAECG variables were dichotomized, fQRS was the only significant predictor. Thus, to illustrate the prognostic significance of the SAECG, we defined an abnormal SAECG as fQRS >114 ms.

View this table: [in this window] [in a new window]   Table 1. Multivariable SAECG Predictors of Outcome

The SAECG variables remained significant predictors after adjustment for prognostic clinical and treatment factors (Table 2). The prognostic relations of the SAECG variables were consistent across the spectrum of study patients (Table 3). The hazard ratio remained consistent among treated and untreated patients, including those who underwent ICD implantation. Although the latter group constituted only 89 patients, 1 (3%) of 34 patients with a normal SAECG suffered cardiac death versus 10 (18%) of 55 patients with an abnormal SAECG.

View this table: [in this window] [in a new window]   Table 2. Adjusted Significance of SAECG Variables

View this table: [in this window] [in a new window]   Table 3. Events in Subgroups

Significantly more patients with an abnormal SAECG (553 patients [44%]) were male or white, had MI remote from the time of enrollment, had prior bypass surgery, had more severe CAD, and had induction of sustained VT (Table 4). Significantly more patients with a normal SAECG had had prior thrombolysis. Figures 1 and 2 show Kaplan-Meier curves for arrhythmic death or cardiac arrest and cardiac death for patients with a normal versus abnormal SAECG. Similar curves were seen for total mortality.

View this table: [in this window] [in a new window]   Table 4. Baseline Characteristics

View larger version (11K): [in this window] [in a new window]   Figure 1. Kaplan-Meier estimates of arrhythmic death or cardiac arrest by SAECG result. P<0.001 between groups.

View larger version (11K): [in this window] [in a new window]   Figure 2. Kaplan-Meier estimates of cardiac death by SAECG result. P<0.001 between groups.

Defining a High-Risk Subgroup
EF, dichotomized at the median value of 30% (relative to <20, 25, 25 to 30, and >30), was also a powerful independent predictor of the primary end point. We combined EF with fQRS to determine whether this would define a high-risk group of patients. Patients with an EF <30% and fQRS >114 ms (n=260, 21% of the population) had significantly higher rates of the primary end point at 2 and 5 years than did patients in any other subgroup defined by these 2 variables (Figure 3). Patients with an EF 30% and fQRS 114 ms had the lowest event rates, whereas those with an EF <30% and fQRS 114 ms and those with an EF of 30% and fQRS >114 ms had intermediate rates. Even in patients with an EF 30%, those with fQRS >114 ms had significantly more events than did those with a value 114 ms. We noted similar results for cardiac death (Figure 4).

View larger version (14K): [in this window] [in a new window]   Figure 3. Kaplan-Meier estimates of arrhythmic death or cardiac arrest by SAECG result and EF. Two-year and 5-year event rates for patients with EF <30% and fQRS >114 ms were 17% and 36%, respectively; for patients with EF <30% and fQRS 114 ms, they were 10% and 23%, respectively; for patients with EF 30% and fQRS >114 ms, they were 11% and 22% respectively; and for patients with EF 30% and fQRS 114 ms, they were 6% and 13%, respectively. Differences between those with EF <30% and fQRS >114 ms compared with those with fQRS 114 ms was highly significant (P=0.0001). Difference between those with EF 30% and fQRS >114 ms compared with those with fQRS 114 ms was also significant (P=0.01).

View larger version (15K): [in this window] [in a new window]   Figure 4. Kaplan-Meier estimates of cardiac death by SAECG result and EF. Two-year and 5-year event rates for patients with EF <30% and fQRS >114 ms were 22% and 44%, respectively; for patients with EF <30% and fQRS 114 ms, they were 13% and 34%, respectively; for patients with EF 30% and fQRS >114 ms, they were 16% and 30%, respectively; and for patients with EF 30% and fQRS 114 ms, they were 8% and 20%, respectively. Difference between those with EF <30% and fQRS >114 ms compared with those with fQRS 114 ms was highly significant (P=0.0001). Difference between those with EF 30% and fQRS >114 ms compared with those with fQRS 114 ms was highly significant (P=0.002) as well.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
The MUSTT study showed that patients with CAD, left ventricular dysfunction, and unsustained VT and with sustained VT induced by EPS have a high rate of arrhythmic death or cardiac arrest.²² Whereas the 2- and 5-year rates of arrhythmic death or cardiac arrest were 18% and 32%, respectively, for patients assigned to no antiarrhythmic therapy, they were 12% and 25%, respectively, for patients assigned to EPS-guided therapy.²² The present study, like the Multicenter Automatic Defibrillator Implantation Trial (MADIT),²³ also showed a significant survival benefit in patients who received implantable defibrillators. On the other hand, no survival benefit was noted in patients who received EPS-guided antiarrhythmic drug therapy. Additionally, although inducible patients had significantly higher rates of arrhythmic death (32% versus 24% for noninducible patients) and cardiac death (48% versus 44% for noninducible patients), the long-term mortality in noninducible patients remained high.²⁴ The aforementioned approach in the MUSTT trial had presupposed that EPS is the optimal test to predict outcome and define therapy. The results of the present study suggest that the test is more beneficial for risk stratification, because EPS-guided antiarrhythmic serial drug testing alone was of no benefit. Simple, cost-effective, noninvasive tests that could identify patients at the highest risk for arrhythmic or cardiac mortality would have considerable clinical relevance.

In this prospective multicenter study, the fQRS relative to other SAECG variables independently predicted the primary end point of arrhythmic death or cardiac arrest and cardiac death alone in patients with CAD, unsustained VT, and left ventricular dysfunction. In addition, patients with an fQRS >114 ms and EF <30% had substantially greater event rates at 2 and 5 years for the primary end point of arrhythmic death or cardiac arrest, cardiac death, and total mortality than did patients with an fQRS 114 ms and EF 30%. Thirty-six percent of patients who had a combination of abnormal SAECG and an EF <30% died of the primary end point, whereas 87% of those without the abnormalities survived the primary end point. Moreover, 44% of patients with this combination suffered cardiac death. The total mortality in patients with this combination was 50%. These figures were higher than those for the primary end point.

Small prospective and retrospective studies have invariably shown that SAECG can predict the inducibility of sustained VT in patients with CAD and unsustained VT.^25–27 However, many patients in these studies were symptomatic, with syncope or dizzy spells. Moreover, none assessed the ability of the SAECG to predict long-term outcome. The present study shows that significantly more patients with an abnormal SAECG have inducible sustained VT, monomorphic and overall.

The observation that SAECG was not a specific predictor of arrhythmic death but was a more powerful predictor of cardiac death is intriguing, inasmuch as the original hypothesis was that SAECG, a determinant of heterogeneous myocardial depolarization, would predict only arrhythmic events. The reason for this observation is unclear. However, it is tempting to speculate that the longer the fQRS, the greater the myocardial scarring, resulting in heterogeneous propagation not only in the terminal depolarization wavefront (late potentials) but perhaps also in the initial and midportions of the depolarization wavefront. Furthermore, a prolonged fQRS may reflect an early stage in the development of overt widening of the QRS complex secondary to extensive scar. These extensive abnormalities could, in turn, predispose patients to sudden as well as nonsudden cardiac death. It may explain the poorer predictive abilities of the other 2 SAECG variables. Another explanation is that the observation may reflect more cardiac deaths relative to arrhythmic or sudden cardiac death. Furthermore, the definition of sudden or nonsudden cardiac death may be subject to error.

An important new observation from the present study is that the SAECG remained a powerful predictor of survival whether or not the patients received treatment, were inducible, or received an ICD. The obvious implication is that if the SAECG is abnormal, the prognosis for survival from cardiac death and total mortality is worse, irrespective of treatment, including ICDs. This would suggest that the fQRS is valuable in long-term prognostication. On the other hand, ICD use was strongly protective from sudden cardiac death, irrespective of abnormalities in the SAECG. However, patients with an abnormal SAECG had a lower long-term survival rate and ultimately succumbed to nonsudden cardiac death.

The CABG-Patch study²⁸ enrolled patients with an EF 35% and abnormal SAECG. The patients randomized to ICD showed no survival benefit compared with patients not receiving ICD. That study randomized patients in the operating room after bypass surgery. Moreover, they did not define SAECG solely by abnormal fQRS, as done in the present study. Revascularization may have altered the immediate prognosis irrespective of the SAECG and EF; on the other hand, in many patients, EF may have improved after revascularization, or the substrate for arrhythmogenesis could have been substantially altered. In our population, 59% of the patients underwent bypass surgery but long before recruitment into the study (median 14 months before enrollment). Further analysis of the CABG-Patch study showed that ICD use reduced arrhythmic death by 45%, although total mortality was not reduced because of an increase in nonarrhythmic deaths.²⁹ This observation indirectly supports a role for the SAECG in risk stratification for arrhythmic death.

Of note, we excluded MUSTT patients with intraventricular conduction defects and bundle-branch blocks because no acceptable standard definitions exist for abnormal quantitative SAECG variables. Thus, these results apply only to patients with a QRS complex <120 ms on standard ECG. We included patients who were treated and untreated and were inducible and noninducible; treatment differences could have influenced the results of the present study. This appears unlikely because the SAECG predicted survival across all groups of patients.

Patients who have CAD, unsustained VT, an EF <40%, and a prolonged fQRS, without bundle-branch block or intraventricular conduction defects, have significantly higher rates of arrhythmic deaths or cardiac arrest, cardiac mortality, and all-cause mortality than do patients with a normal SAECG. The highest mortality at 2 years and 5 years occurred in patients with an EF <30% and fQRS >114 ms. This noninvasive algorithm using SAECG and EF may be valuable in the selection of high-risk patients for intervention and for long-term prognostication.

	Acknowledgments

 
This study was supported by grants UO1 HL-45700 and -45726 from the National Heart, Lung, and Blood Institute, C.R. Bard, Berlex Laboratories, Boehringer-Ingelheim Pharmaceuticals, Guidant Corp, Knoll Pharmaceuticals, Medtronic, Searle, Ventritex-St. Jude Medical, Wyeth-Ayerst Laboratories, and Merck and Co.

	Footnotes

 
Presented in part at the 72nd Scientific Sessions of the American Heart Association, Atlanta, Ga, November 7–10, 1999.

Received February 13, 2001; revision received May 8, 2001; accepted May 10, 2001.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Cain ME, Anderson JL, Ansdorf MF, et al. ACC Expert Consensus Document: signal-averaged electrocardiography. J Am Coll Cardiol. 1996; 27: 238–249.[Medline] [Order article via Infotrieve]
   
2. Berbari BJ, Scherlag RJ, Hope RR, et al. Recordings from body surface of arrhythmogenic ventricular activity during the ST segment. Am J Cardiol. 1978; 41: 697–702.[Medline] [Order article via Infotrieve]
   
3. Simson MB. Use of signals in the terminal QRS complex to identify patients with recurrent ventricular tachycardia after myocardial infarction. Circulation. 1981; 64: 235–242.[Abstract/Free Full Text]
   
4. Kanovsky MS, Falcone RA, Dresden CA, et al. Identification of patients with ventricular tachycardia after myocardial infarction: signal averaged electrocardiogram, Holter monitoring and cardiac catheterization. Circulation. 1984; 79: 264–270.
   
5. Denes P, Santarelli P, Hauser RG, et al. Quantitative analysis of the high-frequency components of the terminal portion of the body surface QRS in normal subjects and in patients with ventricular tachycardia. Circulation. 1983; 67: 1129–1138.[Abstract/Free Full Text]
   
6. Breithardt G, Becker R, Seipel L, et al. Non-invasive detection of late potentials in man: a new marker for ventricular tachycardia. Eur Heart J. 1981; 2: 1–11.[Abstract/Free Full Text]
   
7. Gomes JA, Mehra R, Barreca P, et al. Quantitative analysis of the high frequency components of the signal-averaged QRS complex in acute myocardial infarction. Circulation. 1985; 72: 102–111.
   
8. Gomes JA, Winters SL, Stewart D, et al. A new noninvasive index to predict sustained ventricular tachycardia and sudden death in the first year after myocardial infarction: based on the signal-averaged electrocardiogram, radionuclide ejection fraction and Holter monitoring. J Am Coll Cardiol. 1987; 10: 349–357.[Abstract]
   
9. Kuchar DL, Thorburn CW, Sammel L. Prediction of serial arrhythmic events after myocardial infarction: signal-averaged electrocardiogram, Holter monitoring and radionuclide ventriculography. J Am Coll Cardiol. 1987; 9: 531–538.[Abstract]
   
10. Gomes JA, Winters SL, Martinson M, et al. The prognostic significance of quantitative signal-averaged variables relative to clinical variables, site of myocardial infarction, ejection fraction and ventricular premature beats: a prospective study. J Am Coll Cardiol. 1988; 13: 377–384.
    
11. El-Sherif N, Denes P, Katz R, et al. Definition of the best prediction criteria of the time-domain signal-averaged electrocardiogram for serious arrhythmic events in the post-infarction period. J Am Coll Cardiol. 1995; 25: 908–914.[Abstract]
    
12. Simoons ML, Vos M, Tijssen JGP, et al. Long-term benefit of early thrombolytic therapy in patients with acute myocardial infarction: 5 year follow-up of a trial conducted at the Interuniversity Cardiology Institute of the Netherlands. J Am Coll Cardiol. 1989; 14: 1609–1615.[Abstract]
    
13. Rouleau JL, Talajic M, Sussex B, et al. Myocardial infarction patients in the 1990s: their risk factors, stratification and survival in Canada: the Canadian Assessment of myocardial Infarction (CAMI) Study. J Am Coll Cardiol. 1996; 27: 1119–1127.[Abstract]
    
14. Tavazzi L, Volpi A. Remarks about postinfarction prognosis in light of the experience with the Gruppo Italiano per lo Studio della Sopravvivenza nell’ Infarto Miocardico (GISSI) trials. Circulation. 1997; 95: 1341–1345.[Free Full Text]
    
15. Buxton AE, Fisher JD, Josephson ME, et al. Prevention of sudden death in patients with coronary artery disease: the Multicenter Unsustained Tachycardia Trial (MUSTT). Prog Cardiovasc Dis. 1993; 36: 215–226.[Medline] [Order article via Infotrieve]
    
16. Hinkle LE Jr, Thaler HT. Clinical classification of cardiac deaths. Circulation. 1982; 65: 457–464.[Abstract/Free Full Text]
    
17. Cox DR. Regression models and life-tables. J R Stat Soc (B). 1972; 34: 187–220.
    
18. Smith PL. Splines as a useful and convenient statistical tool. Am Stat. 1979; 33: 57–62.
    
19. Harrell FE Jr, Lee KL, Pollack BG. Regression models in clinical studies: determining relationships between predictors and response. J Natl Cancer Inst. 1988; 80: 1198–1202.[Abstract/Free Full Text]
    
20. Gomes JA, Winters SL, Stewart D, et al. Optimal band pass filters for time domain analysis of the signal-averaged electrocardiogram. Am J Cardiol. 1987; 60: 1200–1298.
    
21. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958; 53: 457–481.
    
22. Buxton AE, Lee KL, Fisher JD, et al. A randomized study of the prevention of sudden death in patients with coronary artery disease. N Engl J Med. 1999; 341: 1882–1890.[Abstract/Free Full Text]
    
23. Moss AJ, Hall WJ, Cannom DS, et al. Improved survival with an implantable defibrillator in patients with coronary artery disease at high risk for ventricular arrhythmia. N Engl J Med. 1996; 335: 1933–1940.[Abstract/Free Full Text]
    
24. Buxton AE, Lee KL, DiCarlo L, et al. Electrophysiologic testing to identify patients with coronary artery disease who are at risk for sudden death. N Engl J Med. 2000; 342: 1937–1945.[Abstract/Free Full Text]
    
25. Buxton AE, Simson MB, Falcone RA, et al. Results of signal-averaged electrocardiography and electrophysiological study in patients with nonsustained ventricular tachycardia after healing of acute myocardial infarction and the usefulness of the signal-averaged electrocardiogram. Am J Cardiol. 1987; 60: 80–85.[Medline] [Order article via Infotrieve]
    
26. Turitto G, Fontaine JM, Ursell SN, et al. Value of the signal-averaged electrocardiogram as a predictor of the results of programmed stimulation in nonsustained ventricular tachycardia. Am J Cardiol. 1988; 61: 1272–1278.[Medline] [Order article via Infotrieve]
    
27. Winters SL, Ip J, Deshmukh P, et al. Determinants of induction of ventricular tachycardia in nonsustained ventricular tachycardia after myocardial infarction and the usefulness of the signal-averaged electrocardiogram. Am J Cardiol. 1993; 72: 1281–1285.[Medline] [Order article via Infotrieve]
    
28. Bigger JT Jr. Prophylactic use of implantable defibrillators in patients at high risk for ventricular arrhythmias after coronary-artery bypass graft surgery. N Engl J Med. 1997; 337: 1569–1575.[Abstract/Free Full Text]
    
29. Bigger JT Jr, Whang W, Rotman JN, et al. Mechanisms of death in the CABG Patch trial: a randomized trial of implantable cardiac defibrillator prophylaxis in patients at high risk of death after coronary artery bypass graft surgery. Circulation. 1999; 99: 1416–1421.[Abstract/Free Full Text]
    

This article has been cited by other articles:

				J. J. Goldberger, M. E. Cain, S. H. Hohnloser, A. H. Kadish, B. P. Knight, M. S. Lauer, B. J. Maron, R. L. Page, R. S. Passman, D. Siscovick, et al. American Heart Association/American College of Cardiology Foundation/Heart Rhythm Society Scientific Statement on Noninvasive Risk Stratification Techniques for Identifying Patients at Risk for Sudden Cardiac Death: A Scientific Statement From the American Heart Association Council on Clinical Cardiology Committee on Electrocardiography and Arrhythmias and Council on Epidemiology and Prevention Circulation, September 30, 2008; 118(14): 1497 - 1518. [Full Text] [PDF]

				L. Ding, W. Hua, H. Niu, K. Chen, and S. Zhang Primary prevention of sudden cardiac death using implantable cardioverter defibrillators Europace, June 16, 2008; (2008) eun150v1. [Abstract] [Full Text] [PDF]

				W. Jung and B. Schumacher What is the role of risk stratification for sudden death in the defibrillator era? Eur. Heart J. Suppl., December 1, 2007; 9(suppl_I): I59 - I65. [Abstract] [Full Text] [PDF]

				H. Kioka, T. Yamada, T. Mine, T. Morita, Y. Tsukamoto, S. Tamaki, M. Masuda, K. Okuda, M. Hori, and M. Fukunami Prediction of sudden death in patients with mild-to-moderate chronic heart failure by using cardiac iodine-123 metaiodobenzylguanidine imaging Heart, October 1, 2007; 93(10): 1213 - 1218. [Abstract] [Full Text] [PDF]

				S. M. Narayan T-Wave Alternans and the Susceptibility to Ventricular Arrhythmias J. Am. Coll. Cardiol., January 17, 2006; 47(2): 269 - 281. [Abstract] [Full Text] [PDF]

				A. K. Gehi, R. H. Stein, L. D. Metz, and J. A. Gomes Microvolt T-Wave Alternans for the Risk Stratification of Ventricular Tachyarrhythmic Events: A Meta-Analysis J. Am. Coll. Cardiol., July 5, 2005; 46(1): 75 - 82. [Abstract] [Full Text] [PDF]

				R. E Lane, M. R Cowie, and A. W C Chow Prediction and prevention of sudden cardiac death in heart failure Heart, May 1, 2005; 91(5): 674 - 680. [Full Text] [PDF]

				A. Bauer, P. Guzik, P. Barthel, R. Schneider, K. Ulm, M. A. Watanabe, and G. Schmidt Reduced prognostic power of ventricular late potentials in post-infarction patients of the reperfusion era Eur. Heart J., April 2, 2005; 26(8): 755 - 761. [Abstract] [Full Text] [PDF]

				K. A. Gatzoulis, G. K. Andrikopoulos, T. Apostolopoulos, E. Sotiropoulos, G. Zervopoulos, J. Antoniou, S. Brili, and C. I. Stefanadis Electrical storm is an independent predictor of adverse long-term outcome in the era of implantable defibrillator therapy Europace, January 1, 2005; 7(2): 184 - 192. [Abstract] [Full Text] [PDF]

				M. Josephson and H. J.J. Wellens Implantable Defibrillators and Sudden Cardiac Death Circulation, June 8, 2004; 109(22): 2685 - 2691. [Full Text] [PDF]

				M. R. Reynolds and M. E. Josephson MADIT II (Second Multicenter Automated Defibrillator Implantation Trial) Debate: Risk Stratification, Costs, and Public Policy Circulation, October 14, 2003; 108(15): 1779 - 1783. [Full Text] [PDF]

				H. V. Huikuri, T. H. Makikallio, M. J. P. Raatikainen, J. Perkiomaki, A. Castellanos, and R. J. Myerburg Prediction of Sudden Cardiac Death: Appraisal of the Studies and Methods Assessing the Risk of Sudden Arrhythmic Death Circulation, July 8, 2003; 108(1): 110 - 115. [Full Text] [PDF]

				W. G. Stevenson and L. M. Epstein Predicting Sudden Death Risk for Heart Failure Patients in the Implantable Cardioverter-Defibrillator Age Circulation, February 4, 2003; 107(4): 514 - 516. [Full Text] [PDF]

				Signal-Averaged ECG as Outcome Predictor for Unsustained VT? Not Yet. Journal Watch Cardiology, October 12, 2001; 2001(1012): 7 - 7. [Full Text]

				J. T. Bigger Expanding Indications for Implantable Cardiac Defibrillators N. Engl. J. Med., March 21, 2002; 346(12): 931 - 933. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gomes, J. A. Articles by Hafley, G. E. Search for Related Content PubMed PubMed Citation Articles by Gomes, J. A. Articles by Hafley, G. E. Related Collections Electrophysiology Electrocardiology Arrhythmias, clinical electrophysiology, drugs