Analysis of T-wave alternans using the dominant T-wave paradigm

The dominant T wave (DTW) reflects the derivative of the repolarization phase of the transmembrane potential of myocytes. T-wave alternans (TWA) is defined as an alteration of this repolarization that repeats every other beat. We investigate if the DTW can offer new insight on TWA. We first proved that the DTW estimate obtained through singular value decomposition is optimal, because it minimizes the norm of the residuals. Then we suggested an optimal estimate of the vector of lead factors, in the case in which the DTW is given. Finally, we derived a mathematical relationship between observable TWA on electrocardiogram and DTW morphology. The relationship depends on the slope of the repolarization phase of the myocytes' transmembrane potentials and on the dispersion of the repolarization times. Based on this finding, a new index meant to quantify TWA was defined and termed amplitude of dominant T-wave alternans (ADTWA). A preliminary validation of the index was performed using the synthetic records contained in the Computers in Cardiology 2008 data set. They were obtained from 5 electrocardiogram models to which TWA was added at different extents. We found a linear relationship between the TWA amplitude and the ADTWA metric (R(2) = 0.9898 ± 0.100 across all models). Moreover, the root mean square error between actual and estimated TWA amplitudes was 10.9 μV (ADTWA) vs 12.9 μV obtained with the classical spectral method.