Value of the signal-averaged electrocardiogram as a predictor of sudden death in myocardial infarction and dilated cardiomyopathy.

To clarify the prognostic significance of signal averaged electrocardiogram (SAE), 100 patients with old myocardial infarction (OMI) and 54 patients with dilated cardiomyopathy (DCM) were studied. Late potentials (LPs) were detected in 31 patients with OMI and in 21 patients with DCM. During a mean follow up of 18 months (3 to 60) in OMI and 28 months (3 to 71) in DCM, 29 patients died. Fifteen patients died suddenly (8 in OMI, 7 in DCM). In OMI, the sensitivity (Se), specificity (Sp), predictive accuracy (PA) of LPs for sudden death were 75%, 72%, and 73%, respectively. The presence of either LPs or prolonged filtered QRS (f-QRS) predicted sudden death with a high Se, and the presence of both LPs and prolonged f-QRS predicted with high Sp and PA. In DCM, Se, Sp, and PA of LPs were lower than those in OMI (Se; 71%, Sp; 66%, PA; 67%). A life table analysis showed that the probability of remaining free from sudden death was significantly lower in patients with LPs than those without them in OMI, but no significant difference was observed between those with and without LPs in DCM. Patients with either LPs or prolonged f-QRS, however, had a significantly higher probability of sudden death in both diseases and no patient with normal SAE died suddenly. SAE was also useful in separating high risk patients in either normal or low cardiac index group in both diseases. Ventricular tachycardia (VT) and % fractional shortening in OMI and only VT in DCM were also useful predictors among other parameters. In conclusion, SAE provides useful information in a noninvasive method to identify patients at risk of sudden death, and patients with normal SAE have a low risk of sudden death in OMI and DCM.