A method for quantifying recurrent patterns of local wavefront direction during atrial fibrillation. (1st October 2017)

• Record Type: Journal Article
• Authors: Hummel, James P.; Baher, Alex; Buck, Ben; Fanarjian, Manuel; Webber, Charles L.; Akar, Joseph G.
• Is Part Of: Computers in biology and medicine. Volume 89(2017)
• Page Start: 497
• Page End: 504
• Publication Date: 2017-10-01
• Subjects: Atrial fibrillation -- Spiral wave -- Multiple wavelet reentry -- Wavefront direction -- Nonlinear dynamics -- Recurrence quantification analysis -- Computer modeling; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2017.08.027
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Introduction: Spiral wave reentry is a potential mechanism of atrial fibrillation (AF), but is difficult to differentiate clinically from multiple wavelet breakup using standard bipolar recordings. We developed a new methodology using bipolar recordings to estimate the direction of local activation wavefronts during AF by calculating the electrogram conformation (Egm-C). We subsequently used recurrence quantification analysis (RQA) of Egm-C to differentiate regions of spiral wave reentry from wavelet breakup. Methods: A 2D computer simulation was created with regions containing a stable spiral wave and also regions of wavebreak. A grid of 40 × 40 unipolar electrodes was superimposed. At each site, the actual wavefront direction (WD) was determined by comparing relative activation timings of the local intracellular recordings, and the estimated wavefront direction (Egm-C) was determined from the morphology of the local bipolar electrogram. RQA of Egm-C was compared to RQA of actual WD in order to differentiate AF mechanisms. Results: RQA of actual WD and Egm-C both distinguished regions of spiral wave reentry from wavelet breakup with high correlation between the two methods (recurrence rate, r = 0.96; determinism, r = 0.61; line max, r = 0.95; entropy, r = 0.84; p < 0.001 for all). In areas of stable spiral wave reentry, the recurrence plots of both Egm-C and actual WD demonstrated stable, periodic dynamics, while regions of wavelet breakup demonstrated chaotic behavior largely devoid of repetitive activation patterns. Conclusion: Calculation of Egm-C allows RQA to be performed on bipolar electrograms during AF and differentiates regions of spiral wave reentry from multiple wavelet breakup. Highlights: A method for discretization of electrograms during AF for recurrence analysis is described. This method yields similar information to a series of local wavefront activation directions. RQA of this discrete series differentiates spiral wave reentry from multiple wavelet reentry.