Objectives: This study investigated prediction of arrhythmic events by the signal-averaged electrocardiogram (ECG) and programmed stimulation in patients with nonischemic dilated cardiomyopathy.
Background: Risk stratification in patients with nonischemic dilated cardiomyopathy remains controversial.
Methods: Eighty patients with nonischemic dilated cardiomyopathy and spontaneous nonsustained ventricular tachycardia underwent signal-averaged electrocardiography (both time-domain and spectral turbulence analysis) and programmed stimulation. All patients were followed up for a mean of 22 +/- 26 months.
Results: Sustained monomorphic ventricular tachycardia was induced in 10 patients (13%), who all received amiodarone. The remaining 70 patients were followed up without antiarrhythmic therapy. Of the 80 patients, 15% had abnormal findings on the time-domain signal-averaged ECG, and 39% had abnormal findings on spectral turbulence analysis. Time-domain signal-averaged electrocardiography had a better predictive accuracy for induced ventricular tachycardia than spectral turbulence analysis (88% vs. 66%, p < 0.01). During follow-up, there were 9 arrhythmic events (5 sudden deaths, 4 spontaneous ventricular tachycardia/fibrillation) and 10 nonsudden cardiac deaths. Cox regression analysis showed that no variables predicted arrhythmic events or total cardiac deaths. The 2-year actuarial survival free of arrhythmic events was similar in patients with or without abnormal findings on the signal-averaged ECG or induced ventricular tachycardia.
Conclusions: In patients with nonischemic dilated cardiomyopathy, 1) there is a strong correlation between abnormal findings on the time-domain signal-averaged ECG and induced ventricular tachycardia, but both findings are uncommon; and 2) normal findings on the signal-averaged ECG, as well as failure to induce ventricular tachycardia, do not imply a benign outcome.