Frequency and significance of occult late potentials on the signal-averaged electrocardiogram in sustained ventricular tachycardia after healing of acute myocardial infarction

The quantitative and morphologic characteristics and significance of late potentials on the signal-averaged electrocardiographic QRS complex remain unknown. To assess this, the signal-averaged electrocardiogram of 48 patients (mean age ± standard deviation 62 ± 9 years) with sustained ventricular tachycardia (VT) after healing of acute myocardial infarction and late potentials were analyzed. Late potentials could be classified into 3 morphologic subtypes: type I late potentials (19 patients, 40%) occurred in the terminal 40 ms of the QRS complex; type II late potentials (16 patients, 33%) started before the end of the QRS complex and extended 30 ± 17 ms into the ST segment; type III late potentials (13 patients, 27%) started after the end of the QRS complex in the ST segment and ended 67 ± 27 ms after the end of the QRS complex. The amplitude of the late potentials in type III, when compared with types I and II, was significantly lower, whereas the QRS duration on the electrocardiogram in type I, when compared with types II and III, was significantly longer. Computer algorithm based on noise failed to identify most type III late potentials. No difference was noted in age, sex, site of the myocardial infarction, and rate of induced VT among the 3 types. It is concluded that (1) morphologic types of late potentials are likely a function of anatomic and geometric differences with resultant differences in conduction; (2) occult late potentials (type III) are seen in 27% of patients with sustained VT after healing of acute myocardial infarction and likely reflect activation of a smaller mass of muscle bundles with a smaller extracellular field, as compared with type I and II late potentials; and (3) because type III late potentials are often missed by computer algorithms based on noise, they are better suited for qualitative analysis.