Related Experiment Videos
Comparison of time domain and frequency domain variables from the signal-averaged electrocardiogram: a multivariable
S J Worley1, D B Mark, W M Smith
1Department of Medicine, Duke University Medical Center, Durham, North Carolina.
Insights
Signal-averaged electrocardiogram (ECG) analysis effectively identifies ventricular tachycardia risk in myocardial infarction patients. Total filtered QRS duration is the key predictor, distinguishing patients with and without ventricular tachycardia.
Area of Science:
- Cardiology
- Signal Processing
- Medical Diagnostics
Background:
- Myocardial infarction (MI) survivors are at risk for ventricular tachycardia (VT).
- Signal-averaged electrocardiogram (SAECG) analysis aids in risk stratification.
- Differentiating between time and frequency domain analyses for VT prediction is crucial.
Purpose of the Study:
- To compare the diagnostic value of unprocessed, time-domain, and frequency-domain SAECG analyses.
- To identify the most effective SAECG parameters for predicting VT in MI patients.
- To assess the utility of SAECG in distinguishing VT risk among MI patients.
Main Methods:
- SAECG was performed on 36 patients with MI and VT, 29 with MI but no VT, and 23 normal subjects.
- Frequency domain analysis used Fast Fourier Transform (FFT) area ratios over various intervals.
- Time domain analysis measured total QRS duration, late potential duration, and terminal QRS voltage.
Main Results:
- Seven SAECG variables differed significantly between MI patients with and without VT.
- FFT area ratios and all time-domain variables showed significant differences in univariate analysis.
- Total filtered QRS duration (time domain) was the sole independent predictor differentiating MI patients with and without VT in multivariate analysis.
Conclusions:
- Total filtered QRS duration from SAECG is a powerful independent predictor of VT in MI patients.
- Frequency domain analysis (FFT) shows promise but requires further validation.
- SAECG, particularly time-domain measures like QRS duration, is vital for VT risk stratification post-MI.
Abstract:
The relative values of the unprocessed signal-averaged electrocardiogram (ECG) and time domain analysis and frequency domain analysis of the signal-averaged ECG were compared in 36 patients with sustained monomorphic ventricular tachycardia and a remote myocardial infarction, in 29 asymptomatic patients with a remote myocardial infarction and in 23 normal subjects. Area ratios of the energy spectra derived from fast Fourier transform analysis were calculated using six separate 140 ms intervals starting at 0, 40, 50 and 60 ms after QRS onset; 40 and 50 ms before QRS end and a variable length interval starting 40 ms before QRS end and extending to the T wave. Total filtered QRS duration, late potential duration and root mean square voltage of the terminal QRS complex were measured from the filtered vector magnitude signal-averaged ECG. The total QRS duration was also measured from the X, Y, Z leads of the unfiltered signal-averaged ECG. Seven variables were significantly different in univariate tests between myocardial infarction patients with and without ventricular tachycardia: three fast Fourier transform area ratios with the sampling interval starting at 1) QRS onset (p = 0.007), 2) 40 ms after QRS onset (p = 0.02), and 3) 60 ms after QRS onset (p less than 0.0001); and all four time domain variables at 1) total filtered QRS duration (p less than 0.0001), 2) late potential duration (p = 0.0001), 3) root mean square terminal QRS voltage (p = 0.0001), and 4) QRS duration from the unprocessed signal-averaged ECG (p less than 0.0001). Of these seven variables, only the fast Fourier transform area ratio starting at QRS onset was significantly different between patients with myocardial infarction without ventricular tachycardia and normal subjects. In multi-variable analysis, the total filtered vector magnitude QRS duration, a time domain variable that includes the late potential, was the only independent factor that separated patients with myocardial infarction with and without associated ventricular tachycardia.