Elsevier

Heart Rhythm

Volume 15, Issue 2, February 2018, Pages 182-192
Heart Rhythm

Clinical
Atrial Fibrillation
Absence of rotational activity detected using 2-dimensional phase mapping in the corresponding 3-dimensional phase maps in human persistent atrial fibrillation

https://doi.org/10.1016/j.hrthm.2017.09.010Get rights and content

Background

Current phase mapping systems for atrial fibrillation create 2-dimensional (2D) maps. This process may affect the accurate detection of rotors. We developed a 3-dimensional (3D) phase mapping technique that uses the 3D locations of basket electrodes to project phase onto patient-specific left atrial 3D surface anatomy.

Objective

We sought to determine whether rotors detected in 2D phase maps were present at the corresponding time segments and anatomical locations in 3D phase maps.

Methods

One-minute left atrial atrial fibrillation recordings were obtained in 14 patients using the basket catheter and analyzed off-line. Using the same phase values, 2D and 3D phase maps were created. Analysis involved determining the dominant propagation patterns in 2D phase maps and evaluating the presence of rotors detected in 2D phase maps in the corresponding 3D phase maps.

Results

Using 2D phase mapping, the dominant propagation pattern was single wavefront (36.6%) followed by focal activation (34.0%), disorganized activity (23.7%), rotors (3.3%), and multiple wavefronts (2.4%). Ten transient rotors were observed in 9 of 14 patients (64%). The mean rotor duration was 1.1 ± 0.7 seconds. None of the 10 rotors observed in 2D phase maps were seen at the corresponding time segments and anatomical locations in 3D phase maps; 4 of 10 corresponded with single wavefronts in 3D phase maps, 2 of 10 with 2 simultaneous wavefronts, 1 of 10 with disorganized activity, and in 3 of 10 there was no coverage by the basket catheter at the corresponding 3D anatomical location.

Conclusion

Rotors detected in 2D phase maps were not observed in the corresponding 3D phase maps. These findings may have implications for current systems that use 2D phase mapping.

Introduction

The mechanism of persistent atrial fibrillation (AF) remains incompletely understood. Various investigators using different mapping techniques have observed different dominant activation patterns.1, 2, 3, 4 Recent attention has focused on the role of rotors, with both basic5 and clinical1, 2 studies suggesting that rotors may act as drivers for persistent AF. For example, recent work using 2-dimensional (2D) phase mapping and the Constellation basket catheter to endocardially map AF identified a high prevalence of sustained rotors.1 Catheter ablation targeted at the center of these rotors was reported to be successful in terminating AF.6 While these promising findings have been reproduced in other laboratories,7 they have not been observed uniformly.8, 9

One of the limitations of the current version of this mapping system has been the use of an idealized 2D representation of the atria that may explain the discrepancy between these studies. The current 2D animation assumes that the electrodes of the basket catheter are evenly distributed within the atria, and as such phase values are animated over a uniform 8 × 8 grid of evenly spaced electrodes in 2D phase space. As observed by Allessie et al,10 these 2D representations do not appear to take into account the effect of π in relating the diameter to the circumference of the spherically shaped basket catheter or the relative position of the basket electrodes within the atria or the patient's actual 3-dimensional (3D) left atrial (LA) anatomy. These spatial assumptions may lead to errors in phase animation, as the complexity and anatomical variability of the 3D LA are not taken into consideration. We recently developed a novel mapping technique whereby phase is projected onto patient-specific 3D LA geometry, which takes into account the relative positions of the basket electrodes and the complex curved surfaces of the LA in both the interpolation and the visualization of the data.11

Using this technology, we sought to determine whether rotational activity observed using 2D phase mapping was present at the corresponding time segments and anatomical locations during 3D phase mapping given the above-mentioned limitations of 2D phase mapping.

Section snippets

Patient population

Fourteen patients with persistent AF referred for catheter ablation were studied. Persistent AF was defined as continuous AF sustained for >7 days. All patients gave written informed consent, and the study protocol was approved by the local ethics committee.

Electrophysiology study

Antiarrhythmic medications were discontinued for 5 half-lives before the procedure. All patients underwent cardiac computerized tomography (CT). All cases were performed under general anesthesia. AF was induced in patients in sinus rhythm at

Baseline characteristics

Fourteen patients underwent LA mapping using the basket catheter (Table 1). The mean age was 62 ± 7 years. The mean CHA2DS2-VASc score was 1.4 ± 1. The 38-mm basket catheter was used in 1 patient (8%), the 48-mm catheter in 10 (71%), and the 60-mm catheter in 3 (21%) patients. In 7 of 14 patients (50%), a steerable sheath was used. Two of 14 patients (14%) required induction of AF with the remaining patients in spontaneous AF at the time of the procedure.

LA coverage by the basket catheter

The overall mean percentage of the LA

Discussion

The main findings of this study are as follows:

  • 1.

    Transient rotors observed in 2D phase maps were not observed at the same time segments and anatomical locations in the corresponding 3D phase maps.

  • 2.

    Wavefronts (single and multiple) and disorganized activity were observed in 3D phase maps at 2D phase detected rotor sites.

  • 3.

    In some cases, the anatomical sites at which rotors were detected in 2D phase maps were found to have no basket coverage when using the actual electrode location in 3D anatomical

Conclusion

Rotational activity observed using 2D phase mapping was not seen at the corresponding time segments and anatomical locations in 3D phase maps. These findings suggest that the multiple assumptions inherent in a regular 2D 8 × 8 grid representation of the basket catheter leads to misleading patterns on phase maps. These observations are likely to have significant implications for the use of 2D phase mapping to identify and ablate rotors.

References (20)

There are more references available in the full text version of this article.

Cited by (20)

  • Site-Specific Epicardium-to-Endocardium Dissociation of Electrical Activation in a Swine Model of Atrial Fibrillation

    2020, JACC: Clinical Electrophysiology
    Citation Excerpt :

    We were primarily interested in the differences in endocardial versus epicardial activation times at multiple paired locations simultaneously rather than in the presence of phase singularities that are typically demonstrated by phase mapping, but we chose to analyze the signal phase of the recorded signal rather than attempt to annotate each discrete activation given the previously demonstrated difficulty in doing so in the presence of complex fractionated signals (15). The sinusoidal recomposition method for instantaneous phase reconstruction previously validated by Kuklik et al. (16) was used for phase analysis and has previously been described in depth and used in clinical mapping studies (17,18). It has been demonstrated that the zero-crossing point in the transformed phase signal, at which there is phase inversion and transition from –π to π, corresponds closely to the local moment of activation as identified by the steepest portion (maximal –dV/dt) of the unipolar voltage signal, so this study would be expected to demonstrate similar results whether one uses phase or activation time.

  • New Findings in Atrial Fibrillation Mechanisms

    2019, Cardiac Electrophysiology Clinics
    Citation Excerpt :

    However, recent systematic reviews and meta-analyses of AF rotor and driver ablation have shown high variability in success rates that seem to be nonsuperior to conventional pulmonary vein ablation alone, whereas acute termination of AF to sinus rhythm or atrial tachycardia rate was less than 40%.25–27 The discrepancy of these findings suggest several challenges with the FIRM-guided approach that may require a steeper operator-dependent learning curve, including electrode contact or coverage, low-density mapping, and analytical assumptions that the electrodes are evenly spaced over a 2-dimensional grid as opposed to the actual spread in 3-dimensional spatial orientation.28–30 Larger, multicenter, prospectiv randomized, controlled trials are needed to delineate its utility in our armamentarium against AF.

  • A novel method for the prediction of focal wavefront origins in cardiac arrhythmias

    2018, Computers in Biology and Medicine
    Citation Excerpt :

    The feasibility analysis was done by constructing prediction curves in two dimensions and projecting them onto a three-dimensional surface. Although EAM projections and sampled points were chosen to minimize chamber curvature, there is an inherent limitation in using planar projections in a three-dimensional structure which may result in inaccurate results [6]. Therefore, any conclusions about the accuracy of this method based on these data should be approached cautiously.

View all citing articles on Scopus

Dr Pathik is supported by the Postgraduate Research Scholarship from the National Health and Medical Research Council of Australia (NHMRC) and the National Heart Foundation of Australia. Dr Lee is supported by a NHMRC Early Career Fellowship. Dr Sanders, Dr Kistler, and Dr Kalman are supported by NHMRC Practitioner Fellowships.

View full text