Signal Averaging in Patients withCoronary Artery Disease:

Abstract

In patients who have survived acute myocardial infarction, the presence of ventricular late potentials using the high resolution signal-averaged ECG indicates areas of slow conduction and delayed activation that may potentially serve as a substrate for malignant ventricular arrhythmias. Although detection of late potentials is technique specific, signal-averaged analysis in the time or frequency domain may be a useful index for risk stratification with regard to ventricular tachycardia or sudden cardiac death. The sensitivity and specificity of late potentials for this purpose may be enhanced by combination with other variables, such as left ventricular ejection fraction and presence of complex ventricular ectopy. Therefore, the presence of ventricular late potentials in postmyocardial infarction patients, particularly in those patients with impaired left ventricular function, identifies those patients who are at high risk of malignant ventricular tachyarrhythmias. However, the strategies for prevention of serious arrhythmia complications during follow-up need to be established. The negative predictive value of late potentials is very high. Thus, the absence of late potentials indicates a low propensity to sustained ventricular tachycardia or sudden death, even in the presence of complex ventricular ectopy. Interventions may therefore not be necessary or should even be avoided. The incidence of late potentials in patients with spontaneous or induced ventricular fibrillation is lower and, if present, less pronounced than in those with sustained monomorphic ventricular tachycardia. This presumably is due to a lower degree of conduction delay, which serves as a substrate for reentry. Therefore, the ability of the signal-averaged ECG to predict a propensity to ventricular fibrillation is limited. Despite these limitations, the signal-averaged ECG may be used as a risk predictor in evaluation of patients after myocardial infarction. Unfortunately, at least as far as time domain analysis is concerned, it cannot be used as an efficacy predictor for response to pharmacologic interventions. Further studies will determine whether other modes of signal-averaged analysis can predict the response to drugs.