Combination of QT Variability and Signalaveraged Electrocardiography in Association With Ventricular Tachycardia in Postinfarction Patients

Transcription

1 Journal of Electrocardiology Vol. 36 No Combination of QT Variability and Signalaveraged Electrocardiography in Association With Ventricular Tachycardia in Postinfarction Patients Gulmira Kudaiberdieva, MD, FESC, Bulent Gorenek, MD, Omer Goktekin, MD, Yuksel Cavusoglu, MD, Alpaslan Birdane, MD, Ahmet Unalir, MD, Necmi Ata, MD, and Bilgin Timuralp, MD Abstract: The authors investigate incidence of ventricular tachycardia/ventricular fibrillation (VT/VF) in relationship with combination of noninvasive arrhythmia risk markers as left ventricular ejection fraction (LVEF), late potentials (LP), and QT variability index (QTVI) and compare the utility of their combination in association with sustained ventricular arrhythmias in patients after myocardial infarction (MI). Fifty-four patients with old MI, among them 27 with documented spontaneous sustained VT/VF entered the study. All of them underwent evaluation for arrhythmias and noninvasive risk stratificaton. Logistic regression analysis demonstrated that the highest association with ventricular tachyarrhythmia had combination of LP and increased QTVI (13.8, P.0002), followed then by combination of LVEF and LP (12.2, P.0005), LP alone (P.001), QTVI (P.002) and LVEF (P.003) alone and age (P.01). After stepwise regression analysis showed that the model including association of LP and QTVI, age and EF is the best one for delineating patients having the risk of ventricular tachyarrhythmia development. In conclusion, patients with combination of positive LP and increased QTVI after MI have high likelihood for development of serious sustained arrhythmia. Key words: Ventricular tachycardia, QT variability, late potentials, ejection fraction, myocardial infarction. Identification and selection of patients with the highest risk for arrhythmic and sudden death are From the Osmangazi University, Medical Faculty, Cardiology Department, Eskisehir, Turkey. Reprint requests: Gulmira Kudaiberdieva, MD, Beyazevler Mah. 26 Sokak, Irem Apt, Kat 1, No 2, Adana, Turkey; gulmira_kudaiberdieva@hotmail.com. Copyright 2003, Elsevier Science (USA). All rights reserved /03/ $35.00/0 doi: /jelc still considered as the main problem of management of patients after myocardial infarction (MI). Several noninvasive markers for arrhythmic events as reduced heart rate variability (HRV), abnormal baroreflex sensitivity, presence of late potentials (LP) on signal-averaged electrocardiogram (SAECG), T wave alternans and QT dispersion are used in risk stratification of patients after MI (1-10). Their predictive accuracy for sudden or arrhythmic death alone or in combination with ejection fraction, nonsustained 17

2 18 Journal of Electrocardiology Vol. 36 No. 1 January 2003 ventricular tachycardia and ventricular extrasystolia on Holter ECG recordings has been evaluated in prospective studies with reported different predictive values (2,4,11-14). On the other hand left ventricular (LV) dysfunction is a well established predictor of sudden and arrhythmic death in patients after MI, specially those with values 30% had the most strong benefit from automated implantable cardioverter defibrillators therapy (15-17). Recently, QT interval variability index (QTVI) (18,19) has been found to have high prognostic value in prediction of sudden death in structural heart disease population (20). Assuming (LV) dysfunction and scar formation with slowing of conduction between viable and non-viable tissues, as well transient factors as ischemia in patients after MI may predispose to arrhythmic events (21-23), the combination of LV dysfunction index with markers of abnormal conduction and repolarization seems to be attractive approach in determining the risk for arrhythmias in patients after MI. We aimed to investigate incidence of ventricular tachycardia/ventricular fibrillation (VT/VF) in relationship with combination of noninvasive arrhythmia risk markers as LVEF, SAECG, and QTVI and to compare the utility of their combination in association with sustained ventricular arrhythmias patients after myocardial infarction. Material and Methods Fifty-four patients with old MI, among them 46 men and 8 women without signs of bundle branch block and atrial fibrillation underwent arrhythmia evaluation in our clinic entered the study. Clinical examinations, chest x-ray, 12-lead electrocardiogram (ECG) and 24 hours ECG monitoring were performed in all patients. Two-dimensional echocardiography was accomplished using 3.75 MHz transducer (Acuson, Sequoia C256) due to recommendations of the American Society of Echocardiography (25) and apical two-chamber views were used for measurement of LV volumes (area-length method) and ejection fraction. Thirty patients with syncope, history of sudden death, sustained and nonsustained ventricular tachycardia underwent programmed ventricular stimulation from right ventricular apex and right ventricular outflow tract using one to triple premature ventricular stimuli with and without isoproterenol infusion until induction of clinical ventricular tachycardia. Coronary angiography was done by Judkins technique with estimation of the mean number of stenosed vessels (arterial narrowing of 50% in planar and 75% cross-sectional views were accepted as significant) and infarct-related artery (IRA) patency in all patients. All patients underwent short-term high-resolution ECG recordings using Kardiosis ard-lp PC based high resolution system with further analysis of SAECG, HRV and QT variability. Bipolar Frank X, Y, Z derivations ( Hz) were recorded and sampled at rate of 1000 samples per second and digitized using 12 bit A/D converter. Each recording lasted for 7 minutes and data were stored in disk for postprocessing analysis. Signal-averaged ECG analysis with estimation of LP was done by time-domain measurements of the vector magnitude in X, Y, Z leads after high-pass filtering ( Hz), amplification and averaging of QRS complexes. The following parameters were assessed: mean filtered QRS duration (fqrs), the root mean square voltage of the terminal (40 ms) portion of the filtered QRS complex (RMS 40) and duration of the low-amplitude ( 40 V) signal in the terminal filtered QRS complex (LAS40). The LP were accepted as positive when RMS40 20 V and LAS40 38 ms (26). The acceptable mean noise level was 0.7 V. QT variability index. Time and frequency domain analysis of heart rate and QT interval variability was done after extracting of RR and QT tachograms from the signal. The detection of R wave was performed after visual selection of R wave sample and then by cross correlation template matching algorithm method of the consecutive R waves. Detection of T waves and Q waves for QT variability analysis was performed using similar algorithm from the leads with the maximum T wave amplitude. End of the T wave was selected as the intercept point of the T wave slope met the TP baseline at the steepest angle. For Q wave in the same lead the onset of QRS was considered. Then RR and QT duration sequences were obtained and interpolated by linear interpolation with further construction of one second equally spaced RR and QT tachograms. To obtain power spectrums a Fast Fourier transformation was used and following time and spectral domain measures were extracted: mean RR and QT intervals durations, standard deviation of normal to normal adjacent RR intervals (SDNN) and QT intervals, total variance of RR and QT variability s, total powers under spectral RR and QT variability s curves. Then QTVI was calculated with a formula

3 QT Variability Data Potentials, Ventricular Tachycardia after MI Kudaiberdieva et al. 19 Table 1. Clinical Characteristics Parameters Group I Group II P Age, years Gender Male, % 85.2 (23) 85.2 (23) NS Female, % 14.8 (4) 14.8 (4) Smoking, % 51.8 (14) 70.4 (19) NS Diabetes, % 29.6 (8) 7.4 (2).07 Hypertension, % 70.3 (19) 59.3 (16) NS Presentation Syncope, % 59.3 (16) 22.2 (6).001 Palpitations, % 18.5 (5) 11.1 (3) History of sudden death, % 14.8 (4) - Monomorphic VT, % 71.4 (20) - Polymorphic VT/VF, % 28.5 (7) - NSVT, % (2) Treatment Beta-blockers, % 14.8 (4) 25.9 (7) NS ACE inhibitors, % 44.4 (12) 51.8 (14) NS Class III antiarrhythmic drugs, % 51.9 (14) 15.4 (4).06 CAD extent NS IRA occlusion, % 51.9 (14) 48.1 (13) NS PVS 23 7 VT, % 82.6 (19) - VT, % 17.3 (4) 100 EF, % MRR, ms NS SDNN, ms MQTc, ms NS fqrs, ms LAS40, ms RMS40 ( V) QTVI ACE, angiotensin-converting enzyme; CAD, coronary artery disease; EF, ejection fraction; fqrs, filtered QRS duration; IRA, infarct-related artery; LAS40, duration of the low-amplitude ( 40 V) signal in the terminal filtered QRS complex; RMS40, the root mean square voltage of the terminal (40 ms) portion of the filtered QRS complex; LP, late potential; MQTc, mean corrected QT duration; MRR, mean RR interval; NS, nonsignificant; NSVT, nonsustained ventricular tachycardia; QTVI, QT variability index; VEB, ventricular ectopic beats; VT/VF, ventricular tachycardia/ventricular fibrillation. proposed by Berger et al. (17,18). QT interval correction was made by using the Bazett formula. Patients groups. According with clinical records and arrhythmia evaluation all patients were divided into 2 groups: group I, 27 patients with clinically documented episodes of spontaneous sustained VT/VF and group II, 27 patients without sustained VT/VF.. Patients were further subdivided into subgroups after dichotomization of noninvasive arrhythmia markers: LV EF 35% and 35% (16,17), presence or absence of LP on SAECG (26), SDNN 50ms and 50 ms, QTVI 0.5 and 0.5 (20), combination of presence of LP with EF 35%, combination of QTVI 0.5 and presence of LP. Statistical analysis was performed using Chisquare test for comparison of incidence of dichotomized values of noninvasive arrhythmia markers and unpaired students t test was applied for assessment of differences in quantitative variables (SPSS for Windows 10.0). To assess the value of combination of noninvasive arrhythmia markers in association with sustained ventricular arrhythmias, we used adjusted logistic stepwise regression analysis, where presence (encoded as 1) or absence (encoded as 0) of sustained VT was accepted as independent variable. While following parameters showing the statistically significant differences based on descriptive analysis as age and dichotomized noninvasive arrhythmia markers and their combinations were considered as dependent variables. Further forward stepwise regression analysis was attempted for selection of the best model, where variables with the significant P value were included into the model step-by-step until no more significant associates remain. Results As can be seen from Table 1 patients without VT/VF were older than those without arrhythmia (P.001); however, groups did not differ as regards

4 20 Journal of Electrocardiology Vol. 36 No. 1 January 2003 Table 2. Incidence of Positive and Negative Results of Noninvasive Arrhythmia Markers in Patients With and Without Sustained Ventricular Arrhythmias Parameters VT/VF Negative VT/VF Positive P LP ( ) (26), % 26.9 (7) 73.1 (19).001 LR 11.0, P.001 QTVI 0,5 (21), % 23.8 (5) 76.2 (16).002 LR 9.7, P.002 SDNN 50 ms (46) 47.8 (22) 52.2 (24) NS LR 0.54, NS EF 35%, (18) 22.2 (4) 77.8 (14).004 LR 8.6, P.003 EF 35% LP ( ) (10), % (10).0001 LR 16.1, P.0001 LP ( ) QTVI 0.5 (11), % (11).0001 LR 18.0, P.0001 EF, ejection fraction; LP, late potential; VT/VF, ventricular tachycardia/ventricular fibrillation; QTVI, QT variability index. to gender, risk factors of coronary artery disease and treatment, although class III antiarrhythmic agents were taken slightly more often by patients with VT/VF (P.05). Patients of group I had more often presented with syncope and palpitations on admission and 4 patients had history of aborted sudden death. Twenty patients of group I had monomorphic VT and 7 patients had polymorphic VT/VF during clinical observation and clinical VT/VF was induced during programmed ventricular stimulation in 19 of them. Patients did not differ as regards to mean number of stenosed vessels and incidence of IRA occlusion. Left ventricular ejection fraction was markedly lower (P.0001), SDNN was reduced (P.05), duration of fqrs and LAS40 were longer and QTVI was higher in patients with VT/VF (P.001, P.01 and P.001, respectively). There were no statistically significant differences in mean RR and QTc interval duration s between groups. Analysis of incidence of positive and negative results of noninvasive arrhythmia markers showed (Table 2) that LP were recorded in 26 patients and VT/VF were found in 73.1% of them (P.001). Increased QTVI ( 0.5) was found in 21 patients and VT/VF was documented in 76.2% of them (P.002). Reduced EF 35% was registered in 18 patients and 77.8% of them had sustained arrhythmia (P.001). Reduced SDNN was present in 46 patients and arrhythmia incidence was only 52.2% (P.05). Reduced LVEF and LP were found in 10 patients and all of them had sustained ventricular arrhythmia (P.0001). Eleven patients had both positive LP and increased QTVI and all of them had VT/VF (P.0001). Logistic regression analysis (Table 3) demonstrated that the highest odds ratio for having ventricular tachyarrhythmia had combination of LP and increased QTVI (15.4, P.0009), followed then by combination of LVEF and LP (12.2, P.0005), QTVI (P.0009), LVEF (P.001) alone, LP alone (P.002), and age (P.01). Forward stepwise regression analysis (Table 4) showed that the best model was those included combination of QTVI and LP (P.00001), age (P.005) and LVEF (P.01). Discussion Our study shows that patients presented with sustained VT/VF were older, the LVEF was lower and duration of filtered QRS and LAS40 were longer, while temporal liability of ventricular repolarization was significantly higher than in those without sustained arrhythmias. Analysis of the incidence of spontaneous sustained VT/VF according with the results of noninvasive risk stratification showed that when arrhythmia risk predictors were used alone the VT/VF incidence Table 3. Adjusted Logistic Regression Analysis Data and Odds Ratios of Noninvasive Arrhythmia Markers in Association With Ventricular Tachycardia/Ventricular Fibrillation Variable Odds Ratio P QTVI LP SDNN, ms 0.75 NS LVEF, % QTVI ( 0.5) LP ( ) EF ( 35%) LP ( ) Age, years LVEF, left ventricular ejection fraction; LP, late potential; QTVI, QT variability index; SDNN, standard deviation of normal to normal RR intervals.

5 QT Variability Data Potentials, Ventricular Tachycardia after MI Kudaiberdieva et al. 21 Table 4. Forward Stepwize Regression Analysis Data Variable 2 Log LR P QTVI ( 0.5) LP ( ) LVEF, % Age, years LVEF, left ventricular ejection fraction, LP, late potential; QTVI, QT variability index. ranged between 55.2% to 77.8% with the highest rate of serious arrhythmia in those with depressed LVEF. However, when the logistic stepwise regression analysis was used the odds ratio for association with ventricular tachycardia was the highest for combination of increased QT variability and late potentials, which appeared to be further strengthened when LVEF and age were added into the model. These mean that patients with late potentials, increased QT variability and LV dysfunction are at the highest risk for having ventricular tachycardia (Fig. 1). Late Potentials Our results on 73.1% incidence of LP in post-mi patients with sustained ventricular arrhythmias are close to the reported in previous studies, where LPs were detected in 73% to 100% of patients with sustained VT and only in 7% to 15% of those without (27). Zimmerman et al. (28) have also established that the predictors for LP in post-mi patients were LVEF 40% and occluded IRA. In their study the presence of ventricular LP, the total filtered QRS duration, the LAS-40 interval, older age and an occluded infarct-related artery were the only variables significantly associated with the occurrence of serious arrhythmic events during 70 months follow-up period (28). Recent studies on combination of LVEF 30% with LP showed that use of both markers could identify patients (of 260 patients - 36% had arrhythmic death and 44% cardiac death) at the risk for arrhythmic death during long-term follow-up period (11), being in accordance with our results showed the high pre- Fig. 1. Late potentials on the signal-averaged ECG (top left), reduced heart rate variability and increased QT variability tachograms (bottom left) in a patient with left ventricular dysfunction (EF 20%) postero-basal wall aneurysm, apical segments akinesia with thrombus (top right) and sustained ventricular tachycardia (bottom right).

6 22 Journal of Electrocardiology Vol. 36 No. 1 January 2003 dictive value of combination of LVEF and LP (P.0005). QT Interval Variability Index Our findings on increased QTVI in patients with sustained arrhythmias after MI are in agreement with the study of Atiga et al. (20) who have found that temporal lability of ventricular repolarization is increased in patients with sustained VT, most of them were with CAD, as compared with controls without arrhythmias and healthy subjects. Moreover, authors established that QTVI might work better than HRV and TWA in stratification for sudden death and was higher in those with VF. QTVI has been established to be enhanced in patients with ischemic and dilated cardiomyopathy, as well was found to correlate with other risk factors for sudden death in hypertrophic cardiomyopathy (18,19). Increased QTVI was found in 76.2% of patients with VT/VF in our study. Interestingly that combination of LP and QTVI was associated with much higher association with VT/VF (P.0002), than other risk stratifiers used alone or in combination. This association further has become stronger when accompanied by advanced age and lower left ventricular contractility. These findings on the combination of delayed conduction and repolarization alone better predict sustained VT/VF are in agreement with other study, that assessed predictive value of LP and other repolarization marker TWA in stratification of patients for arrhythmic events (13). Authors (13) have found that combination of LP and TWA had the highest predictive accuracy for arrhythmic events during 13 months of follow-up period 85%, independently of LVEF. It should be noted that assessment of repolarization abnormalities by TWA needs exercise testing or atrial pacing for abolishment of heart rate effects, while QTVI seems to be more easier to apply. Our findings support the opinion that combination of both depolarization and repolarization markers may increase susceptibility for VT/VF in post-mi patients (13). Our results may be explained by fact that MI leading to anatomical and electrical remodeling, namely LV dysfunction and inhomogenity of conduction between viable and non-viable tissues forms the substrate for arrhythmias and reentry (21,22). At the same time spatial and temporal heterogeneity of repolarization may predispose to arrhythmias (23,24). Experimental studies have documented that beta-to-beat variability of ventricular action potential duration was dependent on stochastic behavior of ionic channels and electrotonic interaction between cells (29). So, we could speculate that MI causing disruption of these electrotonic connections between cells as well ischemia leading to challenges in ionic gradients may lead to temporal dispersion of refractoriness and increase temporal lability of repolarization. Therefore, patients with MI and LV dysfunction, having both depolarization and repolarization abnormalities are prone to develop sustained ventricular arrhythmias. Our findings may assist in pre-selection of patients for risk stratification after MI (14) and could be applied as the first step strategy with bedside simple test with high-resolution ECG acquisition and assessment of LP and QTVI. Limitations of the study. The main limitations of the study are the small number of patients and slightly higher use of class III antiarrhythmic drugs in group of patients with ventricular tachycardia, however the QTc values did not differ between groups. Second, one could argue that control group also consists of patients with nonsustained VT that have also been shown to associate with the risk for sudden death. Although programmed ventricular stimulation failed to induce ventricular tachycardia in these patients, the value of QTVI in risk assessment in the latter category of patients needs further elucidation. It should also be emphasized that we did not include patients with bundle brunch block and atrial fibrillation, both of which have been associated with increased mortality in patients after myocardial infarction. In conclusion, patients with combination of positive LP or QTVI after MI have high likelihood for development of serious sustained arrhythmia. Simple bedside ECG recordings with further analysis of LP and QTVI may be applied as the first step strategy for identifying patients at risk for arrhythmia in patients after MI. References 1. Kleiger RE, Miller JP, Bigger JT, Moss AJ, and the Multicenter Post-Infarction Research Group. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol. 59:256, Farrell TG, Bashir Y, Cripps T, et al: Risk stratification for arrhythmic events in postinfarction patients based on heart rate variability, ambulatory electrocardio-

7 QT Variability Data Potentials, Ventricular Tachycardia after MI Kudaiberdieva et al. 23 graphic variables and the signal-averaged electrocardiogram. J Am Coll Cardiol. 18:687, Zuanetti G, Neilson JMM, Latini R, Santoro E, Maggioni AP, Ewing DJ, on Behalf of GISSI-2 Investigators. Prognostic Significance of Heart Rate Variability in Post-Myocardial Infarction Patients in the Fibrinolytic Era. The GISSI-2 Results. Circulation. 94:432, La Rovere MT, Bigger JT Jr, Marcus FI, Mortara A, Schwartz PJ, for the ATRAMI (Autonomic Tone and Reflexes After Myocardial Infarction) Investigators. Baroreflex sensitivity and heart-rate variability in prediction of total cardiac mortality after myocardial infarction. Lancet 478:351, El-Sherif N, Denes P, Katz R, et al: for the Cardiac Arrhythmia Suppression Trial/Signal-Averaged Electrocardiogram (CAST/SAECG) substudy investigators. Definition of the best prediction criteria of the time domain signal-averaged electrocardiogram for serious arrhythmic events in the postinfarction period. J Am Coll Cardiol 25:908, Kuchar DL, Thorburn CW, Sammel NL: Prediction of serious arrhythmic events after myocardial infarction: signal-averaged electrocardiogram, Holter monitoring and radionuclide ventriculography. J Am Coll Cardiol 9:531, 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 signal-averaged electrocardiogram, radionuclide ejection fraction and Holter monitoring. J Am Coll Cardiol 10:349, Armoundas AA, Rosenbaum DS, Ruskin JN, et al: Prognostic significance of electrical alternans versus signal-averaged electrocardiography in predicting the outcome of electrophysiological testing and arrhythmia-free survival. Heart 80:251, Hohnloser SH, Klingenheben T, Li YG, et al: T wave alternans as a predictor of recurrent ventricular tachyarrhythmias in ICD recipients: Prospective comparison with conventional risk markers. J Cardiovasc Electrophysiol 9:1258, Lee KW, Okin PM, Kligfield P, et al: Precordial QT dispersion and inducible ventricular tachycardia. Am Heart J 134:1005, Gomes JA, Cain ME, Buxton AE, et al: Prediction of long-term outcomes by signal-averaged electrocardiography in patients with unsustained ventricular tachycardia, coronary artery disease, and left ventricular dysfunction. Circulation 104:436, Buxton AE, Hafley GE, Michael H. Lehmann ME, et al, for the Multicenter Unsustained Tachycardia Trial (MUSTT) Investigators. Prediction of sustained ventricular tachycardia inducible by programmed stimulation in patients with coronary artery disease. Utility of Clinical Variables. Circulation 99:1843, Ikeda T, Sakata T, Takami M, et al : Combined assessment of T-wave alternans and late potentials used to predict arrhythmic events after myocardial infarction. A prospective study. J Am Coll Cardiol 35:722, Schmitt C, Barthel P, Ndrepepa C, et al: Value of programmed ventricular stimulation for prophylactic ınternal cardioverter defibrillator implantation in postinfarction patients preselected by noninvasive risk stratifiers. J Am Coll Cardiol 37:1901, Moss AJ, Zareba W, Hall WJ, et al: The Multicenter Automatic Defibrillator Implantation Trial II Investigators Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 346:877, Moss AJ, Hall WJ, Cannom DS, et al: For the MADIT trial investigators. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmia. N Engl J Med 335:1933, Domanski MJ, Sakseena S, Epstein AE, for the AVID Investigators: Relative effectiveness of the implantable cardioverter-defibrillator and antiarrhythmic drugs in patients with varying degrees of left ventricular dysfunction who have survived malignant ventricular arrhythmias. J Am Coll Cardiol 34:1090, Berger R, Kasper E, Baughman KL, Marban E, et al: Beat-to-beat QT interval variability. Novel evidence for repolarization lability in dilated cardiomyopathy. Circulation 96:1557, Atiga WL, Fananapazir L, McAreavey D, et al: Temporal repolarization lability in hypertrophic cardiomyopathy caused by b-myosin heavy-chain gene mutations. Circulation 101:1237, Atiga WL, Calkins H, Lawrence JH, et al: Beat-to-beat repolarization lability identifies patients at risk for sudden cardiac death. J Cardiovasc Electrophysiol 9:899, Gaudron P, Kugler I, Hu K, et al: Time course of cardiac structural, functional and electrical changes in asymptomatic patients after myocardial ınfarction: their ınter-relation and prognostic ımpact. J Am Coll Cardiol 38:33, Gardner PI, Ursell PC, Fenoglio JJ, Wit AL: Electrophysiologic and anatomic basis for fractionated electrograms recorded from healed myocardial infarcts. Circulation 72:596, Adamson PB, Vanoli E: Early autonomic and repolarization abnormalities contribute to lethal arrhythmias in chronic ıschemic heart failure. characteristics of a novel heart failure model in dogs with postmyocardial ınfarction left ventricular dysfunction. J Am Coll Cardiol 37:1741, Viswanathan PC, Shaw RM, Rudy Y: Effects of Ikr and Iks heterogeneity on action potential duration and its rate dependence: a simulation study. Circulation 99:2466, Cheitlin MD, Alpert JS, Armstrong WF, et al: ACC/ AHA guidelines for the clinical application of echocardiography: a report of the American College of Cardiology/American Heart Association Task Force

8 24 Journal of Electrocardiology Vol. 36 No. 1 January 2003 on Practice Guidelines (Committee on Clinical Application of Echocardiography). Circulation 95:1686, Breithardt G, Cain ME, El-Sherif N, et al: Standards for analysis of ventricular late potentials using highresolution or signal-averaged electrocardiography: A statement by a task force committee of the European Society of Cardiology, the American Heart Association and the American College of Cardiology. J Am Coll Cardiol 17:999, Borggrefe M, Fetsch T, Martinez-Rubio A, et al: Prediction of arrhythmia risk based on signal-averaged ECG in postinfarction patients. PACE 2566: 20 (Part II) :2566, Zimmermann A, Sentici A, Adamec R, et al: Longterm prognostic significance of ventricular late potentials after a first acute myocardial infarction. Am Heart J 134:1019, Zaniboni M, Pollard AE, Yang L, et al: Beat-to-beat repolarization variability in ventricular myocytes and its suppression by electrical coupling. Am J Physiol (Heart Circ Physiol) H677: 278:H677, 2000