ABNORMAL SIGNAL-AVERAGED ELECTROCARDIOGRAM IN PATIENTS WITH DUCHENNE MUSCULAR-DYSTROPHY - COMPARISON OF TIME AND FREQUENCY-DOMAIN ANALYSES FROM THE SIGNAL-AVERAGED ELECTROCARDIOGRAM

The clinical significance of an abnormal signal-averaged electrocardiogram (SA-ECG) in time- and frequency-domain analyses was assessed in patients with Duchenne muscular dystrophy (DMD). Twenty-four DMD patients and 20 age-matched healthy volunteers were studied. The SA-ECG was recorded by time-domain signal processing using the vector-magnitude method and frequency-domain signal processing using the fast-Fourier transform. Abnormal SA-ECGs were based on comparison with controls and eight abnormal SA-ECGs were detected among 24 DMD patients, seven by frequency-domain analysis and four by time-domain analysis. The end-diastolic left ventricular volume was larger in the patients with abnormal SA-ECG than in those with normal SA-ECG (136 +/- 32 ml vs. 77 +/- 43 ml, p < 0.01). Compared with patients with normal SA-ECGs, the DMD patients with abnormal SA-ECGs had a lower left ventricular (LV) ejection fraction (54 6 vs. 61 +/- 8%, p < 0.05), as an index of LV contractility, and less LV posterior wall velocity (46 +/- 9 mm/min vs. 62 +/- 16 mm/min, p <0.01), as an index of LV relaxation. The sensitivity and specificity of frequency-domain analysis for predicting malignant ventricular arrhythmias was 60 and 85%, respectively, compared with 30 and 92%, respectively, for time-domain analysis. The combination of time- and frequency-domain analyses can facilitate identification of DMD patients with ventricular tachyarrhythmias associated with depression of cardiac performance.