Effect of spatial resolution and filtering on mapping cardiac fibrillation

Abstract

Background: Endocardial mapping tools use variable interelectrode resolution, whereas body surface mapping tools use narrow bandpass filtering (BPF) to map fibrillatory mechanisms established by high-resolution optical imaging.

Objective: The purpose of this study was to study the effect of resolution and BPF on the underlying mechanism being mapped.

Methods: Hearts from 14 healthy New Zealand white rabbits were Langendorff perfused. We studied the effect of spatial resolution and BPF on the location and characterization of rotors by comparing phase singularities detected by high-resolution unfiltered optical maps and of fibrillating myocardium with decimated and filtered maps with simulated electrode spacing of 2, 5, and 8 mm.

Results: As we decimated the maps with 2-mm, 5-mm, and 8-mm interelectrode spacing, the mean ( ± SD) number of rotors detected decreased from 10.2 ± 9.6, 1.6 ± 3.2, and 0.2 ± 0.5, respectively. Lowering the resolution led to synthesized pseudo-rotors that may be inappropriately identified. Applying a BPF led to fewer mean phase singularities detected (248 ± 207 vs 333 ± 130; P<.01), giving the appearance of pseudo-spatial stability measured as translation index (with BPF 3.6 ± 0.4 mm vs 4.0 ± 0.5 mm without BPF; P<.01) and pseudo-temporal stability with longer duration (70.0 ± 17.6 ms in BPF maps vs 44.1 ± 6.6 ms in unfiltered maps; P<.001) than true underlying fibrillating myocardium mapped.

Conclusion: Electrode resolution and BPF of electrograms can result in distortion of the underlying electrophysiology of fibrillation. Newer mapping techniques need to demonstrate sensitivity analysis to quantify the degree of distortion before clinical use to avoid inaccurate electrophysiologic interpretation.

Keywords: Atrial fibrillation; Electrode resolution; Multielectrode mapping; Optical mapping; Rotor; Signal filters; Ventricular fibrillation.