BACKGROUND: The present study investigates spatial properties of atrial fibrillation (AF) by analyzing vectorcardiogram loops synthesized from 12-lead electrocardiograms (ECGs). METHODS: After atrial signal extraction, spatial properties are characterized through analysis of successive, fixed-length signal segments and expressed in loop orientation, that is, azimuth and elevation, as well as in loop morphology, that is, planarity and planar geometry. It is hypothesized that more organized AF, expressed by a lower AF frequency, is associated with decreased variability in loop morphology. Atrial fibrillation frequency is determined using spectral analysis. RESULTS: Twenty-six patients with chronic AF were analyzed using 60-second ECG recordings. Loop orientation was similar when determined from either entire 60- or 1-second segments. For 1-second segments, the correlation between AF frequency and the parameters planarity and planar geometry were 0.608 (P < .001) and 0.543 (P < .005), respectively. CONCLUSIONS: Quantification of AF organization based on AF frequency and spatial characteristics from the ECG is possible. The results suggested a relatively weak coupling between loop morphology and AF frequency when determined from the surface ECG.
BACKGROUND: The present study investigates spatial properties of atrial fibrillation (AF) by analyzing vectorcardiogram loops synthesized from 12-lead electrocardiograms (ECGs). METHODS: After atrial signal extraction, spatial properties are characterized through analysis of successive, fixed-length signal segments and expressed in loop orientation, that is, azimuth and elevation, as well as in loop morphology, that is, planarity and planar geometry. It is hypothesized that more organized AF, expressed by a lower AF frequency, is associated with decreased variability in loop morphology. Atrial fibrillation frequency is determined using spectral analysis. RESULTS: Twenty-six patients with chronic AF were analyzed using 60-second ECG recordings. Loop orientation was similar when determined from either entire 60- or 1-second segments. For 1-second segments, the correlation between AF frequency and the parameters planarity and planar geometry were 0.608 (P < .001) and 0.543 (P < .005), respectively. CONCLUSIONS: Quantification of AF organization based on AF frequency and spatial characteristics from the ECG is possible. The results suggested a relatively weak coupling between loop morphology and AF frequency when determined from the surface ECG.