THE SIGNAL-AVERAGED ELECTROCARDIOGRAM IS OF LIMITED VALUE IN PATIENTSWITH BUNDLE-BRANCH BLOCK AND DILATED CARDIOMYOPATHY IN PREDICTING INDUCIBLE VENTRICULAR-TACHYCARDIA OR DEATH

An abnormal signal-averaged electrocardiogram (SAECG) has predictive v alue for arrhythmic events in patients with idiopathic dilated cardiom yopathy and a normal conduction. The purpose of this study was to inve stigate whether the presence of a complete bundle branch block (BBB) a ffects prognostic information of the SAECG. We prospectively obtained SAECGs in 128 patients with idiopathic dilated cardiomyopathy. Forty-t hree of them had BBB and 85 had a normal QRS duration. According to th eir clinical history and results of ventricular programmed stimulation , patients were divided into 4 groups : (1) group IA with BBB and vent ricular tachycardia (VT) (n = 18); (2) group IB with BBB but without V T (n = 25); (3) group IIA without BBB but with VT (n = 40); (4) group IIB without BBB and without VT (n = 45). Patients were compared with 1 29 patients without heart disease and without VT. Fifty-seven of them had BBB (group III) and 72 had normal conduction (group IV). The filte red QRS duration was longer in group IB than in group III (175 +/- 21 vs 149 +/- 16 ms, p <0.001), and in group IIB than in group IV (111 +/ - 19 vs 96 +/- 12 ms, p <0.05). QRS duration was similar in groups IA and IB (176 +/- 24 vs 175 +/- 21 ms) but longer in group IIA than in g roup IIB (131 +/- 24 vs 111 +/- 19 ms, p <0.001). The few-amplitude si gnal duration (LAS) and the root-mean-square voltage (RMS) of the last 40 ms of the filtered QRS did not differ between groups IB and III an d IA and IB. LAS and RMS were, respectively, longer (44 +/- 20 vs 31 /- 13 ms, P <0.01) and lower (21 +/- 20 vs 43 +/- 33 mu V, p <0.001) i n groups IIA and IIB. In groups IA and IB the combination of 2 of the 3 available criteria: QRS duration >170 ms, RMS <20 mu V, LAS >45 ms l ead up to the best overall statistical result, with a sensitivity and specificity of 78% and 56%, respectively. in groups IIA and IIB, using conventional late potential criteria, the sensitivity and specificity of the SAECG for VT detection were 65% and 73%, respectively. The ris k of sudden death was not predicted by the SAECG, and total cardiac mo rtality was only dependent on left ventricular election fraction. In c onclusion, QRS duration was prolonged in all of the patients with a di lated cardiomyopathy compared with those without heart disease, BBB di d not change the sensitivity but decreased the specificity of the SAEC G to predict any VT risk in dilated cardiomyopathy. The risk of sudden death and total cardiac mortality could not be predicted by the SAECG . (C) 1997 by Excerpta Medica, Inc.