Time domain analysis of the signal averaged electrocardiogram in patients with a conduction defect or a bundle branch block

Doubts have been expressed about the clinical usefulness oftime domain analysis of the signal averaged electrocardiogramin patients with prolonged QRS complex duration. We studied147 patients using a signal averaged ECG (40–250 Hz) whoseQRS complex was longer than 100 ms. A baseline electrophysiologystudy was also performed in 128 of these patients. Seventy-sevenpatients had a minor (QRS <120 and >100 ms) conductiondefect. Thirty-seven of these 77 had either induced or spontaneoussustained ventricular tachycardia (group I) and 40 had no sustainedventricular tachycardia (group II). Seventy patients had a major(QRS120 and >100 ms) conduction defect, 44 of whom had sustainedventricular tachycardia (group A). The remaining 26 withoutthis condition formed Group B. Group I compared to group IIpatients had a longer filtered QRS duration (120.8 ±14 vs 104.5 ± 9.5 ms, P<0.001), a longer low amplitudesignal duration (41 ± 12.1 vs 31 ± 12.6 ms, P<0.0001)and a lower root mean square of the last 40 ms of the filteredQRS complex (27 ± 29.8 vs 35 ± 25.3 µV,P=ns). Group A compared to group B had a longer filtered QRSduration (157.7±20.2 vs 140.7± 15.7 ms, P<0.001),a longer low amplitude signal duration (57.3 ±24.9 vs37.8 ± 20.3 ms P<0.001) and a lower root mean squareof the last 40 ms of the filtered QRS complex (14.3 ±11.2 vs 22.0 ± 10.5 1 P<0.01). Using conventionallate potential criteria, the sensitivity and specificity ofthe signal averaged ECG for the detection of sustained ventriculartachycardia patients with a minor conduction defect were 89%and 75%, respectively. The same criteria applied to patientswith a major conduction defect were sensitive (sensitivity:87%) but non-specific (specificity: 50%). However, by usingmodified late potential criteria, such as the presence of twoof any of the following three signal averaged parameters: filteredQRS duration 145 ms, low amplitude signal duration 50 ms,root mean square of the last 40 ms of the filtered QRS complex17.5µV, we derived a non-optimal but still acceptablecombination of sensitivity (68%) and specificity (73%). We concludethat traditional late potential criteria can be applied in patientswith a minor conduction defect, but modification of these criteriais necessary to derive useful clinical information for riskstratification of patients with a QRS complex duration 120ms.