Author + information
- Received June 6, 2016
- Revision received November 28, 2016
- Accepted November 29, 2016
- Published online March 6, 2017.
- John M. Miller, MD∗ (, )
- Vikas Kalra, MD,
- Mithilesh K. Das, MD,
- Rahul Jain, MD, MPH,
- Jason B. Garlie, MD,
- Jordan A. Brewster, MD and
- Gopi Dandamudi, MD
- Krannert Institute of Cardiology, Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana
- ↵∗Address for correspondence:
Dr. John M. Miller, Krannert Institute of Cardiology, Department of Medicine, Indiana University, 1800 North Capitol Avenue, E-488, Indianapolis, Indiana 46202.
Background Mounting evidence shows that localized sources maintain atrial fibrillation (AF). However, it is unclear in unselected “real-world” patients if sources drive persistent atrial fibrillation (PeAF), long-standing persistent atrial fibrillation (LPeAF), or paroxysmal atrial fibrillation (PAF); if right atrial sites are important; and what the long-term success of source ablation is.
Objectives The aim of this study was to analyze the role of rotors and focal sources in a large academic registry of consecutive patients undergoing source mapping for AF.
Methods One hundred seventy consecutive patients (mean age 59 ± 12 years, 79% men) with PAF (37%), PeAF (31%), or LPeAF (32%). Of these, 73 (43%) had undergone at least 1 prior ablation attempt (mean 1.9 ± 0.8; range: 1 to 4). Focal impulse and rotor modulation (FIRM) with an endocardial basket catheter was used in all cases.
Results FIRM analysis revealed sources in the right atrium in 85% of patients (1.8 ± 1.3) and in the left atrium in 90% of patients (2.0 ± 1.3). FIRM ablation terminated AF to sinus rhythm or atrial flutter or tachycardia in 59% (PAF), 37% (PeAF), and 19% (LPeAF) of patients, with 15 of 67 terminations due to right atrial ablation. On follow-up, freedom from AF after a single FIRM procedure for the entire series was 95% (PAF), 83% (PeAF), and 82% (LPeAF) at 1 year and freedom from all atrial arrhythmias was 77% (PAF), 75% (PeAF), and 57% (LPeAF).
Conclusions In the Indiana University FIRM registry, FIRM-guided ablation produced high single-procedure success, mostly in patients with nonparoxysmal AF. Data from mapping, acute terminations, and outcomes strongly support the mechanistic role of biatrial rotors and focal sources in maintaining AF in diverse populations. Randomized trials of FIRM-guided ablation and mechanistic studies to determine how rotors form, progress, and regress are needed.
Atrial fibrillation (AF) is the most common sustained arrhythmia (1), for which mechanistic uncertainty continues. Although the pandemic of AF is likely related to obesity and other comorbidities (1,2), it is unclear how they contribute to AF or how to reverse their effects (1). Because drug therapy for rate or rhythm control has modest success, ablation is increasingly used for symptom relief (3). However, even with current technology, the single-procedure success rate at 1 year for paroxysmal AF (PAF) is approximately 60% (4,5), and this rate is lower for persistent AF (PeAF) (6), results that are not improved by the ablation of complex fractionated electrograms or empirical linear ablation (6,7).
Mounting evidence suggests that localized rotational circuits (rotors) or focal sources comprise the electric substrate for AF; 1 strategy of identifying and treating these is focal impulse and rotor modulation (FIRM). Optical mapping, the gold standard for mapping AF (8), has recently been applied to atria from patients with clinical AF (9). This has shown that AF is driven by stable endocardial micro–re-entrant sources where targeted ablation terminated AF (9), with transient epicardial breakthroughs or partial re-entry. These results are similar to recent clinical reports of AF sources (10) and also reconcile less stable sources seen by epicardial mapping (11). However, rotors (12) and sources (13) are not found by classical mapping, possibly because of limitations of marking electrograms in AF (14) and/or other technical limitations (15).
The aim of the IU-FIRM (Indiana University FIRM) registry was to: 1) determine if rotors and focal sources are common in a large real-world population of unselected patients with PeAF, long-standing PeAF (LPeAF), or PAF; 2) discover whether intervention at sources supports their mechanistic role across AF phenotypes; and 3) establish the “real-world” clinical outcomes of rotor- and source-based ablation.
We studied 170 consecutive patients referred to Indiana University/Methodist Hospital from January 2012 to October 2015 for ablation of symptomatic AF for standard indications (3). Subjects were ≥18 years of age, with AF despite ≥1 antiarrhythmic drug. Patients were excluded if they were unable or refused to provide written informed consent or did not have sustained AF (>5 min) during the procedure. The population included patients with PAF, PeAF, or LPeAF by standard definitions (3). This was the first ablation procedure for most patients, although 73 (43%) had undergone at least 1 prior ablation procedure. Table 1 summarizes patient characteristics.
Each patient underwent AF mapping using a multipolar catheter inserted sequentially into both atria as previously described, followed by computational methods to interpret electrograms using repolarization and conduction dynamics to reveal sources (16). Ablation targeted the identified sources in each patient. One hundred sixty-one patients (95%) also underwent pulmonary vein isolation (PVI); the remaining 9 patients prospectively declined this. Follow-up with event and/or ambulatory monitoring or implantable device interrogation in 151 patients (89%) was used to determine clinical efficacy and establish the mechanistic impact of interventions targeting these defined mechanisms.
Electrophysiologic study was performed after discontinuing antiarrhythmic medications for 5 half-lives (2 to 6 weeks in patients on amiodarone). Electrode catheters were advanced to the coronary sinus and right atrium and then trans-septally to the left atrium. A 64-pole basket catheter (Constellation, Boston Scientific, Natick, Massachusetts [n = 44]; or FIRMap, Abbott Electrophysiology, Menlo Park, California [n = 126]) was advanced through an 8.5-F sheath to map AF in both atria in sequence. Basket sizes were selected to approximate the left atrial (LA) size on pre-procedural computed tomography or acquired electroanatomic mapping atrial “shell,” then positioned to optimize fluoroscopic coverage and electrogram signals (Figure 1). Generally, baskets sized to fit the left atrium tended to map the right atrium well.
Digital electroanatomic atrial shells were created using Carto (Biosense Webster, Diamond Bar, California). Intravenous heparin was administered as a bolus after femoral access, followed by infusion to achieve activated clotting times >350 s. Unipolar atrial electrograms from the basket catheter were filtered at 0.05 to 500 Hz and recorded at a 1-kHz sampling frequency for export from the electrophysiological recording system (Bard/Boston Scientific, Lowell, Massachusetts).
AF was the presenting rhythm in 96 patients (57%), atrial flutter in 11 (6%), and sinus rhythm in 63 (37%). In patients presenting in sinus rhythm or in whom presenting AF or atrial flutter was terminated by ablation, AF induction was attempted using rapid pacing with or without isoproterenol (5 to 15 μg/min) or epinephrine (0.05 to 0.2 μg/kg/min) as needed.
Mapping AF sources
FIRM mapping has been described elsewhere (10). Briefly, unipolar AF electrograms recorded using basket catheters are processed using algorithms to determine propagation sequences. When deflections are monophasic or noncomplex, mapping can identify rotational or focal activations as previously illustrated (10). In cases in which AF deflections are multiphasic, the use of classical rules such as dV/dt may mark deflections within repolarization (far-field) (14). In such cases, FIRM assigns local activation on the basis of physiological information, such as action potential duration restitution (17) to account for changes in refractoriness with changes in cycle length during AF, and then applies phase analysis to identify rotors. The system (RhythmView, Abbott Electrophysiology, Menlo Park, California) then generates AF propagation maps, which are subsequently correlated to corresponding basket electrode locations within the chamber.
AF propagation (FIRM) maps were analyzed in near real time for FIRM-guided ablation at sources. Electric rotors (Figure 2D) were defined as rotational activation around a phase-mapped singularity that generated peripheral disorganization, while focal impulses showed centrifugal spread of activation from an origin. Rotors and focal impulses were considered AF sources if they existed in consistent spatial regions of 2 to 8 cm2 (with some precession ) on repeated maps.
Radiofrequency (RF) energy was delivered with a 3.5-mm tip irrigated catheter (ThermoCool or SmartTouch, Biosense Webster) at 20 to 35 W. Ablation commenced guided by FIRM data in 131 patients (77%). FIRM-guided ablation followed PVI in 39 patients (23%). RF energy was delivered for approximately 30 s at any 1 location to cover the 2- to 8-cm2 area of AF sources with closely spaced RF applications (Figures 2 and 3), to achieve an electrogram amplitude decrease of >70%. The procedural endpoint was rotor elimination on repeat mapping or noninducibility of AF, if ablation terminated the arrhythmia to sinus rhythm. When AF terminated during ablation, RF application was continued to eliminate electrograms in the region designated by mapping (Figures 2 and 3). Thereafter, attempts were made to reinitiate AF as noted previously, and repeat mapping and ablation were performed if AF, atrial flutter, or tachycardia was induced. The duration of FIRM mapping and ablation, from initial basket deployment to the end of final ablation of FIRM-designated sources, was 94 ± 54 min.
Conventional ablation (PVI ) was performed in all patients but those in whom FIRM-only therapy was used or patients in whom PVI was present from prior procedures. Ablation comprised wide-area circumferential ablation of left and right pulmonary vein pairs, verifying pulmonary vein entrance and exit block using a circular mapping catheter (Lasso, Biosense Webster). Adenosine challenge or pacing along the ablation line was used to confirm isolation at procedure’s end. If atrial tachycardia or flutter was diagnosed, mapping and ablation were performed as feasible; when linear right atrial (RA) or LA ablation was used for this, confirmation of bidirectional conduction block along such lines was used. AF persisting after the completion of ablation was cardioverted.
During a 3-month post-ablation “blanking period,” antiarrhythmic medications were continued in 8% of patients, then discontinued in all but 2 patients (by patient preference). No repeat ablation procedures were performed in the blanking period. The primary endpoint was freedom from AF after a single procedure, defined as AF >30 s on ambulatory electrocardiography and/or event monitors (3) or implanted device interrogation (n = 24) following the 3-month blanking period. The secondary outcome was freedom from all atrial arrhythmias after blanking. Subjects were followed for at least 6 months (mean 15 ± 10 months) post-procedure. In patients with arrhythmia recurrence after the blanking period, continued observation, antiarrhythmic drug therapy or repeat ablation was offered (3). If ablation was performed, detailed mapping was repeated. Results presented hereafter are those of the index FIRM procedure only.
Continuous variables are summarized as mean ± SD and were compared using analysis of variance, after being verified as normally distributed by the Kolmogorov-Smirnov test. Nominal variables were compared between patient groups using chi-square tests or Fisher exact tests if expected cell frequencies were <5. Long-term outcomes were assessed, and survival analyses were conducted using the Cox proportional hazards model, censoring patients after events. Survival curves were generated using the Kaplan-Meier method and compared using log-rank tests. The proportionality assumption was deemed satisfied upon inspection of log-log plots. A p value of <0.05 was considered to indicate statistical significance.
Stable sources observed across AF classes
All patients had intraprocedural AF, of whom 96 (57%) presented in AF (Table 1). FIRM mapping revealed stable AF rotors or focal sources in all cases. Of 3.5 ± 2.1 sources per patient (median 4; interquartile range [IQR]: 3), 54% were in the left atrium (2.3 ± 1.3 per patient; median 2; IQR: 2) and 46% in the right atrium (2.1 ± 1.3 per patient; median 2; IQR: 2) (Table 2). Per patient, 85% had at least 1 RA source; 90% had LA sources. Of the 17 patients without LA sources, 5 had AF termination during RA ablation without subsequent AF inducibility, precluding LA source mapping. Most sources were rotors, with 24 (14%) being foci (16 RA, 8 LA). Mean procedure duration was 356 ± 60 min; mean fluoroscopy time was 25 ± 13 min.
Figure 2 shows AF in a 51-year-old man with PeAF and no prior ablation (Figure 2A). RA FIRM mapping showed a posterior septal rotor where ablation had no effect on the rhythm. LA FIRM mapping revealed a rotor on the anteromedial wall (Figure 2D) where ablation slowed and organized the rhythm to a rapid flutter (Figure 2B) that then terminated to sinus rhythm (Figure 2C) after a total of 3.1 min. AF was then noninducible, and per patient request, ablation concluded using FIRM-only guidance (Figure 2E) without PVI. The patient remained free of AF after 16 months following FIRM-only ablation.
AF termination with FIRM-guided ablation
Table 3 summarizes rates of AF termination from FIRM ablation alone. Most of these terminated to sinus rhythm (n = 38 [57%]) (Figure 2C) rather than typical RA flutter (n = 8; all were ablated successfully) or nonflutter atrial tachycardia (n = 21; 12 were ablated, 9 terminated with pacing or catheter manipulation and were not seen again). AF terminations by FIRM-guided ablation were more common in patients with PAF than in those with PeAF or LPeAF (59% vs. 37% vs. 19%, respectively, p < 0.001). Of note, there was no relationship between acute termination of AF (to sinus or organized atrial tachyarrhythmia) and freedom from arrhythmia recurrence: 10 of 67 (15%) with terminations had recurrent AF, atrial flutter, or atrial tachycardia versus 27 of 103 (26%) without termination (p = 0.12).
AF termination by ablating RA rotors
A specific goal of the IU-FIRM registry was to define the role of RA rotors and focal sources. Figure 3 shows termination of PeAF to sinus rhythm by RA rotor ablation alone in a 55-year-old man. The patient was free of AF at 1.5-year follow-up.
Overall, RA FIRM ablation alone terminated AF in 9 patients. Sources were located in several areas of the right atrium, with no single characteristic site.
Long-term efficacy from FIRM-guided ablation
Freedom from AF after a single-FIRM procedure at 1 year (IQR: 222 to 629 days) off medications in all but 2 patients (as noted earlier) in the IU-FIRM registry was 87% (Central Illustration). Freedom from AF was higher for PAF than for PeAF or LPeAF (95% vs. 83% vs. 82%, respectively, p < 0.01). Freedom from all atrial tachyarrhythmias after a single FIRM procedure at 1 year (IQR: 179 to 570 days) off medications in the IU-FIRM registry was 70%. The Central Illustration shows that freedom from all atrial tachyarrhythmias was higher for PAF and PeAF than for LPeAF (77% vs. 75% vs. 57%, respectively, p <0.02).
The IU-FIRM registry had an overall serious complication rate of 3.5%, including 3 cases of cardiac tamponade requiring drainage, 1 pericardial effusion, 1 case of heart block, and 1 spontaneous epidural hematoma. None were related to the basket catheter. Other complications included mild hematuria (2 cases) and 2 moderate femoral hematomas that did not require intervention. No strokes, peripheral emboli, phrenic nerve injuries, or atrioesophageal fistulae were noted. There were no procedural deaths.
The IU-FIRM registry supports the mechanistic role of localized rotors and focal sources in sustaining PeAF, LPeAF, and PAF. Ablation at FIRM-mapped sources in the left or right atrium was able to terminate AF in patients across disease classes. In combination with PVI, ablation produced 87% freedom from AF after a single FIRM procedure at 1 year in patients, mostly those with nonparoxysmal AF (almost none of whom were taking antiarrhythmic medications). In an important subset of patients who declined PVI, the ability of FIRM-only ablation to eliminate AF on long-term follow-up strongly supports the mechanistic role of rotors and sources. Thus, the IU-FIRM registry shows that clinical and mechanistic conclusions from the CONFIRM (Conventional Ablation With or Without Focal Impulse and Rotor Modulation) trial and optical mapping of human atria apply broadly to patients referred for AF ablation and that FIRM-guided ablation may improve success rates above those expected from PVI alone.
Our data support results from optical mapping of AF in human atria (9) and clinical studies (10) that human AF is sustained by localized sources. The ability of FIRM-guided ablation to terminate AF often to sinus rhythm in patients with all AF phenotypes strongly supports a mechanistic role for rotors and focal sources, as do results of FIRM-guided ablation alone, albeit in a small subset of patients.
Debate continues on AF mechanisms, yet it is increasingly clear that this may in part reflect the choice of mapping tools. Classical electrogram mapping nicely indicates the complexity of AF substrates (13), yet electrograms in AF often show far-field mixed with local activity (14,19). This may explain why classical electrogram mapping of AF rarely shows rotors, whereas optical mapping studies, including in explanted human atria (9), show AF maintained by rotational drivers or focal sources.
Localized AF sources may explain observations that AF can be terminated by focused interventions (10,20), and why extensive ablation that would be expected to limit multiple wavelets does not improve success (6,7). Of note, in preliminary studies, AF rotational sources identified by high-resolution optical mapping in explanted human hearts are also identified by basket electrodes and FIRM methods (21).
Although termination of fibrillation (especially when persistent) to an organized atrial tachyarrhythmia or sinus rhythm is proof of concept and validates the mechanistic nature of rotors and focal sources in AF maintenance, the poor correlation of termination with long-term freedom from recurrent arrhythmia is perplexing. This may reflect incomplete eradication of the entire rotor substrate in cases in which ablation terminates AF, similar to the poor correlation of long-term freedom from recurrent typical RA flutter when arrhythmia termination rather than demonstration of bidirectional cavotricuspid isthmus block is the procedural endpoint. Currently, the only means of determining that the rotor substrate has been eliminated is remapping during ongoing fibrillation and observing that the previously targeted rotor or focus has been eradicated; this is not possible if fibrillation terminates and cannot be reinitiated. Future technical advances may help resolve when the substrate has been fully treated.
Comparison with other studies of FIRM-guided ablation
The IU-FIRM registry is the largest reported experience of FIRM-guided ablation to date and validates the original CONFIRM trial (10) and smaller reports that FIRM-guided ablation can produce favorable results in unselected patients with all types of AF.
Single-procedure freedom from AF of 87% at 1 year in our study supports the 80.5% freedom from AF in the 10-center registry of FIRM-guided ablation in 78 patients (22) and the 82.4% success in the CONFIRM trial (10). These results are substantially better than the 38% freedom from AF and 21% freedom from all arrhythmias reported by Buch et al. (23) in 43 patients (one-half with PAF). This might be explained by technical factors applying FIRM or PVI (15) or challenging patients and must be reconciled. Gianni et al. (24) reported single-procedure freedom from AF or atrial tachycardia off drugs in PeAF and LPeAF of only 17% at 6 months by FIRM-guided ablation alone (i.e., without PVI). It is unclear if those results are worse than in our series because of relatively early experience with FIRM mapping and ablation by these seasoned investigators, with 29 procedures among several operators. Differences may also reflect study design, as in that trial, when FIRM organized AF to atrial tachycardias, they were not ablated but cardioverted. Conversely, in the CONFIRM trial and in this study, atrial tachycardias were targeted for ablation, which may reduce their recurrence. FIRM-guided ablation has been successful in challenging patients in other studies; Sommer et al. (25) reported 80% freedom from atrial arrhythmias in patients with mostly LPeAF, and 70% to 80% success was reported by Tomassoni et al. (26) and Rashid and Sweeney (27).
FIRM-guided versus other forms of ablation across AF populations
The relatively large population of the IU-FIRM registry provides the first opportunity to dissect FIRM-guided ablation results across types of AF, showing favorable success across all AF subgroups (Central Illustration).
In patients with nonparoxysmal AF (n = 106), the single-procedure success rate was higher than the approximate 55% freedom from AF on or off medications or approximate 40% freedom from all arrhythmias off medications in the STAR-AF2 (Substrate and Trigger Ablation for Reduction of Atrial Fibrillation Part 2) trial (6). For patients with PAF (n = 64), single-procedure success in the IU-FIRM registry (95% freedom from AF, 77% freedom from all atrial arrhythmias) was higher than the 60% to 65% freedom from AF in recent trials of contact force catheters (4,5) or the FIRE AND ICE study (comparing cryoballoon or RF ablation), in which repeat ablation was permitted in the blanking period (28).
The IU-FIRM registry also provides the first series able to detect whether FIRM-guided ablation produces a substantial rate of atrial tachycardias. The 16.5% rate of atrial tachycardias in IU-FIRM and rates found in other studies (22,25,29) are comparable to those from PVI alone in the STAR-AF2 (6), STOP-AF (Sustained Treatment of Paroxysmal Atrial Fibrillation) (30), and other trials. Because PVI was also used in most of our patients, the effect of FIRM on subsequent atrial tachycardias is not clear. Further study is needed to clarify these results.
Additional clinical studies are required to determine if the results of FIRM-guided ablation in the IU-FIRM and other recent independent studies are maintained on long-term follow-up (2 to 5 years), as in the CONFIRM trial (29). It is important to determine why the results of the IU-FIRM registry are better than those of FIRM with PVI in 2 recent studies by Buch et al. (23) in 43 patients. If differences in outcomes with our 170 patient series reflect more extensive experience in the FIRM technique, then efforts must be made to reduce the learning curve. If differences reflect study design, then treatment should be standardized to include such elements. Finally, if variations in outcome reflect different patient cohorts, then both ideal and less ideal populations must also be defined and studied.
Although IU-FIRM was not randomized, it is a relatively large registry of consecutive patients at a large academic medical center. A strength of this study is that our experience may have circumvented limitations encountered by otherwise seasoned investigators with less experience in FIRM mapping or FIRM-guided rotor elimination. The number of cases needed to achieve success with FIRM-guided ablation will vary with patient selection, operator skill, and other factors. Because this was a registry study, it is difficult to exclude case-related differences in PVI or other ablation in contributing to outcomes in some cases. However, many PVI trials included additional ablation even for PAF (4,31), such as RAAFT-2 (Radiofrequency Ablation Versus Antiarrhythmic Drugs as First-Line Therapy of Atrial Fibrillation-2) (31), which included 20% to 40% use of additional targets, including linear ablation and ablation of complex fractionated electrograms. Complete control of follow-up electrocardiographic monitoring was not possible because of the wide area of patient referrals, and thus some asymptomatic episodes of AF or other arrhythmias may have escaped detection; however, only 5 patients without known arrhythmia recurrence had no long-term monitoring. Finally, these results must be compared with those of PVI alone in a randomized clinical trial; such trials are under way.
The IU-FIRM registry is the largest reported series of FIRM-guided ablation for unselected patients with a range of AF presentations, mostly nonparoxysmal AF. We found that FIRM-guided ablation at AF substrates produced single-procedure success rates higher than those from PVI alone or with traditional forms of substrate ablation. Mechanistically, the ability of targeted ablation at localized rotors and focal sources to terminate AF and eliminate AF on long-term follow-up supports the role of sources in the left and right atrium in maintaining AF. These data strongly motivate randomized controlled trials of FIRM-guided ablation.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: Use of FIRM technology to localize areas of atrial myocardium responsible for maintenance of AF results in greater freedom from recurrent AF than PVI alone.
TRANSLATIONAL OUTLOOK: Further studies are needed to more clearly define the role of FIRM-guided AF ablation and to identify characteristics of patients most likely to gain incremental benefit from this approach.
The authors thank Susan Straka, RN, for coordinator support and Tina Baykaner, MD, MPH, for statistical support.
This work was supported by institutional funds. Dr. Miller has received honoraria from Medtronic, St. Jude Medical, Biotronik, Biosense Webster, and Boston Scientific; and has been a scientific advisor to Abbott/Topera (modest, <$10,000). Dr. Dandamudi has received honoraria from Medtronic and Biosense Webster. Dr. Jain has received honoraria from Biosense Webster. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- focal impulse and rotor modulation
- interquartile range
- left atrial
- long-standing persistent atrial fibrillation
- paroxysmal atrial fibrillation
- persistent atrial fibrillation
- pulmonary vein isolation
- right atrial
- Received June 6, 2016.
- Revision received November 28, 2016.
- Accepted November 29, 2016.
- American College of Cardiology Foundation
- January C.T.,
- Wann L.S.,
- Alpert J.S.,
- et al.
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.
- Natale A.,
- Reddy V.Y.,
- et al.
- Reddy V.Y.,
- Dukkipati S.R.,
- Neuzil P.,
- et al.
- Wong K.C.,
- Paisey J.R.,
- Sopher M.,
- et al.
- Efimov I.R.,
- Nikolski V.P.,
- Salama G.
- Hansen B.J.,
- Zhao J.,
- Csepe T.A.,
- et al.
- Narayan S.M.,
- Krummen D.E.,
- Shivkumar K.,
- Clopton P.,
- Rappel W.-J.,
- Miller J.
- Haïssaguerre M.,
- Hocini M.,
- Denis A.,
- et al.
- Lee S.,
- Sahadevan J.,
- Khrestian C.M.,
- Cakulev I.,
- Markowitz A.,
- Waldo A.L.
- Allessie M.A.,
- de Groot N.M.,
- Houben R.P.,
- et al.
- Narayan S.M.,
- Wright M.,
- Derval N.,
- et al.
- Jalife J.,
- Filgueiras Rama D.,
- Berenfeld O.
- Narayan S.M.,
- Franz M.R.,
- Clopton P.,
- Pruvot E.J.,
- Krummen D.E.
- Narayan S.M.,
- Shivkumar K.,
- Krummen D.E.,
- Miller J.M.,
- Rappel W.J.
- Narayan S.M.,
- Zaman J.A.
- Shivkumar K.,
- Ellenbogen K.A.,
- Hummel J.D.,
- Miller J.M.,
- Steinberg J.S.
- Hansen B.J.,
- Briggs C.,
- Moore B.T.
- Tomassoni G.,
- Duggal S.,
- Muir M.,
- et al.
- Rashid H.,
- Sweeney A.
- Narayan S.M.,
- Baykaner T.,
- Clopton P.,
- et al.
- Packer D.L.,
- Kowal R.C.,
- Wheelan K.R.,
- et al.