Lindsay B D, Ambos H D, Schechtman K B, Arthur R M, Cain M E
Cardiovascular Division, Washington University School of Medicine, St. Louis, Missouri 63110.
J Am Coll Cardiol. 1990 Dec;16(7):1656-64. doi: 10.1016/0735-1097(90)90316-h.
Abnormalities in the fast Fourier transforms of signal-averaged electrocardiograms (ECGs) obtained during sinus rhythm appear to distinguish patients with ischemic heart disease and sustained monomorphic ventricular tachycardia from those without ventricular tachycardia. This study was performed to determine the power of frequency analysis to detect patients with a history of ventricular fibrillation, to determine the extent to which spectra of signal-averaged ECGs from patients with ischemic and nonischemic heart disease are comparable and to compare results of signal-averaged ECG analysis in patients with ventricular fibrillation with results of programmed ventricular stimulation. Signal-averaged ECGs were obtained during sinus rhythm from 60 patients with sustained ventricular tachycardia (Group I) and 34 patients with ventricular fibrillation (Group II). Results of signal-averaged ECG analysis were abnormal in 92% of patients with ventricular tachycardia and 85% of patients with ventricular fibrillation (p = NS). Abnormal spectra were detected in the signal-averaged ECGs from 90% of patients with ischemic and from 86% of patients with nonischemic heart disease (p = NS). In contrast, the results of programmed stimulation differed markedly between the two patient groups. Sustained ventricular arrhythmias were induced in 91% of the patients with ventricular tachycardia compared with only 46% of those with ventricular fibrillation (p less than 0.0001). Moreover, ventricular tachycardia was inducible in 81% of patients with ischemic heart disease compared with only 50% of those with nonischemic heart disease (p less than 0.02). Thus, abnormalities in the spectra of signal-averaged ECGs were found in the majority of patients with ventricular fibrillation and were detectable even in those whose arrhythmia was not inducible by programmed stimulation. These results broaden the potential clinical application of noninvasive interrogation of signal-averaged ECGs to include the prospective identification of patients with ischemic or nonischemic heart disease prone to ventricular tachycardia or ventricular fibrillation.
