Use of the signal-averaged electrocardiogram for predicting inducible ventricular tachycardia in patients with unexplained syncope: relation to clinical variables in a multivariate analysis

Abstract

Objectives: The purpose of this study was to determine the predictors of electrically induced ventricular tachycardia in a large sample of patients with unexplained syncope and to examine the value of the signal-averaged electrocardiogram (ECG) in those patient subsets with varying pretest probability of ventricular tachycardia.

Background: In patients with unexplained syncope, electrophysiologic study can provide important diagnostic information, such as inducibility of ventricular tachycardia. The signal-averaged ECG can predict inducible ventricular tachycardia, but its utility has not been prospectively studied in a large group of patients with unexplained syncope.

Methods: At six hospitals, 189 consecutive patients with unexplained syncope underwent signal-averaged ECG and electrophysiologic studies.

Results: Ventricular tachycardia was induced in 28 patients (15%). Univariate predictors of ventricular tachycardia included history of previous myocardial infarction, reduced left ventricular ejection fraction and abnormal signal-averaged ECG results. The signal-averaged ECG was the most sensitive test but had poor specificity. By multivariate analysis, the signal-averaged ECG and history of previous myocardial infarction were independently predictive. The risk of ventricular tachycardia increased 17-fold in patients with a previous myocardial infarction who also had an abnormal signal-averaged ECG. In patients with no history of previous myocardial infarction, no additional testing was useful in identifying those at risk for inducible ventricular tachycardia.

Conclusions: The signal-averaged ECG was the most sensitive noninvasive test available to predict sustained ventricular tachycardia at electrophysiologic study but was false positive in many patients. A history of previous myocardial infarction followed by the signal-averaged ECG was the most efficient screening process for predicting electrically induced ventricular tachycardia.