Assessment of the Effect of Fibrillatory Waves in the Analysis of Spatial Heterogeneity of Ventricular Repolarization

In providing proper treatment for Atrial Fibrillation (AF) patients, the clinician must establish the pattern of arrhythmia, determine associated symptoms, and asses for underlying comorbidities in order to define short- and long-term management strategies. To do so, this paper describes a preliminary study in which a novel ECG-based estimator of the standard deviation of ventricular myocytes' repolarization times (sθv ) called the V-index was computed in simulated data, assessing whether the presence of f-waves may corrupt the V-index. The analysis was performed in simulated ECGs generated by combining synthetic bundles of 300 T-waves with seven different values of sθv, with QRS template and 50 f-waves extracted from real data. The results show a strong correlation between the calculated target V-index (tV-index) and the sθv (r = 0.99, p-value < 0.001). They also show the effect of the f-waves in the computation of the V-index. The V-index calculated with f-waves (fV-index) and after the f-waves were removed (cV-index) had highly significant differences (p-value < 0.001) and significant differences (p-value < 0.05) with the tV-index for values below 53.3 ms, which proves the corruptive effect the f-waves have on the computation of the V-index and stresses the importance of removing them before attempting further analysis.