[Late ventricular potentials in the post-infarct patient. A follow-up at 4 years]

We performed a prospective study to evaluate the prognostic significance and the natural history of late ventricular potentials (LPs) in 209 patients (165 males and 44 females; mean age 59.8 +/- 10 years) who survived acute myocardial infarction. Signal-averaged electrocardiograms (SA-ECGs) were performed before hospital discharge (16 +/- 5 days) and after four years (mean follow-up 42 +/- 7 months). SA-ECGs were processed using a 40 Hz high-pass bidirectional filter. Duration of "filtered" QRS (normal value < 120 msec), duration of the low-amplitude signals (n.v. < 39 msec) and last 40 msec voltage of the QRS complex (n.v. > 20 microV) were measured. LPs were defined as the presence of two or more abnormal values. In addition, 24-hour Holter monitoring was performed in all patients, and left ventricular ejection fraction (LVEF) was determined by scintigraphy in 120 (57.4%). Sixty patients (28.7%) had LPs before hospital discharge (group 1), and 149 (71.3%) had normal SA-ECGs (group 2). During the follow-up period there were 10 arrhythmic events, 7 of which were sudden deaths, and three cases of sustained ventricular tachycardia. SA-ECG was repeated in 141 patients (68%). The mean values of SA-ECG's parameters did not change significantly between the two controls, and the correlation was good for all of them. Despite this, spontaneous normalization of SA-ECGs occurred in 21 patients (60%) and the subsequent appearance of LPs was seen in 13 (12%); in these latter, the SA-ECG's parameters measured before hospital discharge were "borderline" and significantly different from those who did not change. The sensitivity of SA-ECG as a predictor of arrhythmic events was 80% and the specificity 74%. Patients with arrhythmic events had a longer filtered QRS (126 +/- 33 vs 103 +/- 12 msec; p < 0.001), longer duration of the low-amplitude signals (57 +/- 23 vs 32 +/- 11 msec; p < 0.001), lower voltages (17 +/- 8 vs 36 +/- 24 microV; p < 0.001), and, moreover, higher peak CK values, lower LVEF and higher value of Lown modified class. In conclusion, SA-ECG confirms its value in identifying patients at risk of arrhythmic events after myocardial infarction. SA-ECG recordings taken before the discharge can be used to predict serial changes during follow-up.