Author + information
- Gery Tomassoni, MD∗ ()
- ↵∗Address for correspondence:
Dr. Gery Tomassoni, Baptist Health Lexington, 1720 Nicholasville Road, Suite 601, Lexington, Kentucky 40503.
Newer mapping techniques have identified the presence of patient-specific recurrent rotational organized activity or rotors during atrial fibrillation (AF) (1–4). The more commonly used technique includes biatrial endocardial and computational spatiotemporal phase mapping during AF to identify and guide targeted ablation of both stable rotors and focal sources (3,4). With the use of a 64-electrode basket catheter, multiple studies have confirmed that rotors are a part of the electrophysiological AF substrate and that focal impulse and rotor modulation-guided ablation (FIRM) (Rhythm View, Topera Medical, Abbott, Abbott Park, Illinois) may improve success rates in patients with AF (4,5). However, controversy exists over whether rotors are critical for the maintenance of AF or are merely a coincidental collision of wandering atrial activation wavefronts (6). In addition, because the exact computational method for analyzing the FIRM AF recordings has not been delineated, there has been skepticism about the validity of FIRM-derived rotor sites (7). There is a need, therefore, for additional information to provide further insight into the role and endocardial signal analysis of rotors in human AF.
In this issue of JACC: Clinical Electrophysiology, Daoud et al. (8) present their initial results using an activation based mapping technique to identify both right atrial (RA) and left atrial (LA) rotational repetitive activation patterns (RAPs) seen in human paroxysmal and persistent AF and atrial flutter. Using a 64-electrode basket catheter (Constellation, Boston Scientific, Natick, Massachusetts), AF endocardial recordings were analyzed offline with a novel software (CARTOFINDER [CF], Biosense-Webster, Diamond Bar, California) and a 3-dimensional electroanatomic mapping system (CARTO, Biosense-Webster) in patients with AF and/or atrial flutter. The study consisted of 2 phases.
In phase I (n = 20), development of the CF software and analysis of sequential biatrial recordings (two 1-min sampling in each atrium) were performed before pulmonary vein (PV) isolation. Additional recordings were performed in each atrium after PV isolation. From the recordings of the first 6 patients, the CF software was developed to identify rotational RAP. In the remaining 14 patients, including both paroxysmal and persistent AF, 3-dimensional CF maps were created and assessed by 4 blinded electrophysiologists (EPs). In phase II, validation of the electrode activation events and the CF algorithm was performed in 5 separate patients with RA isthmus-dependent atrial flutter.
Rotational RAP were successfully identified in 86% of AF patients (12 of 14) with an 82% reproducibility and were correctly identified by a group of 4 blinded EPs 95% of the time (122 of 128 maps). In addition, biatrial mapping after PV isolation demonstrated a 45% reduction in the amount of RAP. The authors concluded that the CF software in combination with a basket catheter and a 3-dimensional mapping system in human AF resulted in the following: 1) successful identification of biatrial RAP during both paroxysmal and persistent AF; 2) high reproducibility in finding the same RAP at different times; 3) PV isolation resulted in a significant reduction of both LA and RA RAP; and 4) EPs who were first time users of CF could readily identify rotational RAP. In phase II, the reproducibility of identifying the single RAP in the atrial flutter patients was 100%.
The findings from this study show many similarities between rotational RAP and rotor data previously reported from FIRM trials (3–5). The presence (44% in RA, 56% in LA), distribution, and numbers of RAP (2.9 RAP per patients with 1.3 RA and 1.6 LA) were similar to rotor data, again with a significant number of RAP seen outside the PVs (3–5). Minimal meandering of the RAP center was reported as with the core of the rotor (3). The reproducibility of the basket catheter to identify RAP at similar locations despite different time segments during AF was also present (3). Finally, the biatrial effect of RFA on the frequency and total reduction of RAP was consistent with FIRM data (4,5).
However, important differences are noted. First, compared with published FIRM results, RAP duration was quite shorter (12 ± 10 cycles; range 5 to 56 cycles) with overall less stable and persistent sources (9). Second, the validation phase of CF software was limited to RA isthmus dependent flutter alone, where endocardial basket catheter signal quality is very high and there is a low presence of complicated multiphasic unipolar signals compared with AF. Correct annotation of these signals can be quite challenging with incorrect signal annotation site resulting in poor RAP identification. More important, however, are the differences in endocardial signal analysis between CF rotational RAP and RhythmView (Topera Medical) rotor maps. By this author’s assessment, each software seems to use its own signal quality autocorrelation algorithm followed by subtraction of far-field ventricular signals and then construction of electrogram timing windows. In addition, detection of atrial activations using both bipolar and unipolar information from neighboring electrodes is performed to a certain extent. The difference is that CF uses a wavelet analysis of unipolar signals to identify the site of signal annotations, whereas RhythmView constructs a fixed amplitude sawtooth wave train for each channel using a linear ramp waveform to estimate the cardiac action potential phase. The CF software displays the data as a 3-dimensional colored propagation map (activation timing) with the leading edge of activation in red. The RhythmView FIRM map (activation timing and phase information) is displayed on a 2-dimensional animation grid with white representing the start of activation and shades of grey representing the state of the action potential phase. Based on this assessment, it seems that both the quality and interpretation of the endocardial basket catheter unipolar signals are critical for correct and precise RAP identification. Unfortunately, identification of rotors from present-day 64-electrode basket catheters can be very difficult because of low signal quality, low spatial sampling density of electrodes, poor atrial wall contact, and insufficient atrial coverage (7,10). Newer basket catheter development with an increase in electrode numbers, shorter inter-electrode distances, and better atrial coverage may alleviate this concern.
The concept for AF RAP or rotor identification using software already built into a conventional 3-dimensional mapping system is attractive for many reasons. First, voltage information would be immediately available to evaluate potential areas of scar/fibrosis. The location of certain anatomic structures that may be prone to sustain rotors can be easily identified. Finally, once online analysis occurs, direct local electrogram assessment and RAP/rotor visualization on a pre-existing atrial map may in turn allow for better RFA targeting. Although the results of this study provide no commentary on either the mechanism of RAP or whether RAP elimination can improve ablation success, the question is whether or not CF can truly identify rotational RAP or rotor sources. Based on the preliminary data, it seems that the answer is yes. This study adds to the growing body of literature regarding the presence of both RA and LA rotational activation patterns in both paroxysmal and persistent AF. It is also important to note that RAP visual identification was readily achieved by EPs. However, this is the first step in a long process for this new technology. More extensive validation testing of the algorithms are required in all types of AF, because a robust and highly accurate annotation algorithm is critical to analyze intracardiac signals correctly in areas of scar and functional block. Mechanistic- and ablative-guided strategies are absolutely needed. It will be of great interest to follow this technology as more results become available in the future.
↵∗ Editorials published in JACC: Clinical Electrophysiology reflect the views of the authors and do not necessarily represent the views of JACC: Clinical Electrophysiology or the American College of Cardiology.
Dr. Tomassoni has received consulting fees/honoraria from Topera/Abbott, Stereotaxis, Biosense Webster, St. Jude Medical, Boston Scientific, Medtronic, Pfizer, and Atricure.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors' institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- 2017 American College of Cardiology Foundation
- Haissaguerre M.,
- Hocini M.,
- Denis A.,
- et al.
- Narayan S.M.,
- Krummen D.E.,
- Shivkumar K.,
- Clopton P.,
- Rappel W.J.,
- Miller J.M.
- Benharash P.,
- Buch E.,
- Frank P.,
- et al.
- Daoud E.G.,
- Zeidan Z.,
- Hummel J.D.,
- et al.
- Kuklik P.,
- Zeemering S.,
- van Hunnik A.,
- et al.