Correlation between inducibility of sustained ventricular tachycardia and QRS duration

Some studies provide a link between the width of QRS complexes and late potentials occurring at the end of the QRS complex in signal-averaged recordings. The purpose of this study was to compare three methods of QRS duration measurement: the conventional 12 lead ECG, the Frank vectorcardiogram (VCG) and the signal-averaged electrocardiogram. The recordings were made at a similar time in 121 consecutive patients with the Cardionics PC-based system (ECG and VCG) and the Cardionics high resolution ECG, based on methods described by Simson. Patients with bundle branch block were excluded. All patients had presented a myocardial infarction and were studied either for spontaneous ventricular arrhythmias or systematically 3 to 6 weeks after an acute myocardial infarction. The signal-averaged ECG and VCG QRS durations were similar in 41 patients without inducible ventricular arrhythmias and with normal signal-averaged ECG but were longer (P < 0.001) than the conventional ECG QRS duration. In 36 patients with spontaneous and inducible ventricular tachyarrhythmias, the QRS duration was significantly longer on signal-averaged ECG than on VCG (P < 0.05) and longer on VCG than on conventional ECG (P < 0.05). The QRS duration was also significantly (P < 0.001) longer with the three techniques in patients with spontaneous ventricular tachycardia (VT) than in patients without spontaneous and inducible VT. A QRS duration on VCG > or = 110 ms and on conventional ECG > or = 100 ms had a sensitivity of 93% and 77% and a specificity of 83% and 85% respectively for predicting an abnormal signal-averaged ECG. In conclusion, the measurement of QRS duration with the conventional ECG, VCG or the signal-averaged ECG could be a simple method to detect the patients with myocardial infarction prone to VT.