Biatrial, 3-Dimensional Mapping of
Human Atrial Fibrillation: Methodology and Clinical Observations
Nicholas D. Skadsberg, PhD aRangadham Nagarakanti, MD b
Sanjeev Saksena, MD, FHRS, FACC, FESC c,d
Delivery, Cardiac Rhythm Disease Management, Medtronic, Inc., Minneapolis, Minnesota,
Institute for Medical Research, Wynnewood, Pennsylvania,
Research Foundation, Warren, New Jersey, and
of Medicine Robert Wood Johnson Medical School, New Brunswick, New Jersey,
for Correspondence: Sanjeev Saksena, MD, Medical Director, Electrophysiology Research Foundation, 161
Washington Valley Rd, Suite 201, Warren, New Jersey, 07059, USA.
Short Title – Mapping of Atrial Fibrillation
Atrial fibrillation (AF), the most
common arrhythmia in clinical practice, accounts for nearly one third of all
hospitalizations for cardiac rhythm disturbances. Consequently, this has
stimulated intense investigative interest in the development of effective
therapeutic options. However, the electrophysiologic (EP) mechanisms of this
arrhythmia have been long debated and remain unclear. This has limited the
development of effective management strategies. Previous studies have shown
the progressive remodeling associated with AF, initially believed to be
functional and electrical in nature, now has structural and contractile impact . It is increasingly clear that the latter two processes play an
increasingly important role in the recurrence and persistence of AF [2-4]. In an effort to clarify AF
mechanisms, numerous experimental models have been developed. Their
relationship to human mechanisms remains poorly defined. Direct mapping of
human AF has been attempted but is still in its evolution. It is the purpose of
this commentary to review existing mapping techniques and propose a new
approach for mapping of human AF.
In the last few years, there have
been several successful techniques reported for the regional catheter mapping
of discrete focal sources of AF in specific atrial regions [5-7]. These methods generally consist of placing multipolar catheters in
these regions, characterization of AF using activation maps, and assessing
tachycardia responses to pacing and other maneuvers. These methods have proven
most suitable in specific atrial regions (e.g., pulmonary veins, lateral right
atrium (RA), or superior vena cava (SVC)), but are less suitable for global
biatrial mapping due to multiple activation patterns and electrogram variations.
Obtaining global mapping data to
characterize a complex rhythm often dictates the placement of multiple
catheters at numerous locations within the chamber of interest, which in turn
requires a considerable amount of time. As a result, the patient and medical
staff are exposed to extensive use of fluoroscopy for catheter placement and
navigation. Most of these mapping techniques have to record from multiple sites,
therefore performing a complete map requires sequential mapping over several
cardiac cycles to allow for the recording of the complete EP substrate. This
approach cannot account for the variability observed in electrocardiographic
(ECG) patterns of human AF and as a result, may provide an incomplete
explanation of mechanisms (Figure 1).
Figure 1:Electrocardiographic recordings in a patient with paroxysmal atrial fibrillation showing variable ECG patterns in a single patient over the course of 1 day. Note the onset of the first episode (top left panel labeled 9:00 am) shows organized flutter or coarse fibrillation waves consistent with an organized tachycardia. These are also seen at termination on the top right panel (labeled 10:31 am). The second episode (middle panel, labeled 2:12 pm) shows fine fibrillatory waves, with a distinct pattern from the first event. The third event ( bottom left panel, labeled 3:05 pm ) shows typical saw toothed flutter waves consistent with type 1 flutter. The final event ( bottom right panel, labeled 7:43 pm) shows flutter waves with a different morphology and rate from the previous event, consistent with another flutter morphology.
In an effort to
overcome the limitations associated with conventional contact catheter mapping
techniques, considerable advances have been made to achieve 3-dimensional (3D) electroanatomic
mapping methodologies. Several mapping systems have been developed that can be
used independently or in a complementary role to conventional electrode
catheter mapping techniques. As a result, all of these approaches have
potential to reduce fluoroscopic imaging required to perform the procedure.
These techniques can broadly be categorized into two primary technology
categories, each possessing their own unique advantages and disadvantages.
technologies comprise the first category, termed “sequential” mapping systems,
and include: (a) Electroanatomical mapping, commonly performed using the CARTOTM
system (Biosense Webster, USA), and (b) the LocaLisa® system
(Medtronic, Inc., USA). Common to each system in this category is the
capability to collect 3D images as well as their respective electrograms in the
target cardiac chamber and collate them in a virtual 3D reconstruction of the chamber’s
mapping systems that permit single beat global chamber activation mapping
represent another approach. This can be performed by either basket contact
catheter or non-contact catheter array mapping (NCM). In this category, the
systems allow for the recording of global electrogram data from a cardiac
chamber so that the activation map can be characterized in a single beat. Basket
or mesh contact catheter mapping necessitates electrode contact with the
chamber’s walls in order to obtain electrograms.
HD Mesh Ablation System (Bard Electrophysiology, Lowell, MA, USA) comprises a 36 pole high density mapping and ablation catheter which has a braided, expandable
mesh electrode configuration mounted onto a non-steerable 8 French (Fr) shaft.
The HD Mesh Ablator catheter is introduced via the 9 Fr Channel steerable
transseptal sheath into the LA and directed to the desired cardiac chamber.
The expandable mesh electrode is deployed and un-deployed by using a sliding
mechanism on the catheter handle. Pulsed radio frequency (RF) energy is
delivered to the HD Mesh catheter electrodes to create the desired lesion
geometry. A surface ECG and endocardial electrograms (EGM) are used for
monitoring. Currently, this technique is being used for pulmonary vein
isolation. After selective venography to identify the location and orientation
of the pulmonary vein ostium, lesions are created circumferentially by
delivering pulsed RF energy via the HD Mesh ablator catheter using EGM guidance
until electrical isolation of the veins is observed. If the vein cannot be
isolated with the catheter alone, EGM guided adjunctive ablation with a distal
tip ablation catheter can be performed .
NCM mapping employs
a 64 electrode balloon catheter array in the blood pool of the chamber of
interest, constructs a virtual contour geometry of the heart using catheter
navigation and obtains over 3,000 virtual electrograms derived using the
Laplace equation. These are then amalgamated to reconstruct the anatomy and EP wavefronts
in the chamber of interest. These approaches can place the activation and
anatomic constructs of the cardiac chamber in a single electrophysiologic 3D
map, permit real-time wavefront propagation analysis, and allow for beat-to-beat
In 1995, we initiated the use of simulataneous biatrial catheter mapping
for study of human AF [9, 10]. The methodology provided important insights
suggesting organization and diversity in mechanisms of AF initiation. More
recently, 3D NCM has provided the opportunity for high-resolution mapping on a
beat-to-beat basis. We have integrated these two approaches to develop a novel practical,
rapid and reliable method to map human AF. This report describes our
methodology and summarizes some of our recent findings with respect to biatrial
and 3D NCM in spontaneous AF patients.
A. Biatrial Catheter Mapping
A duodecapolar catheter is placed
along the RA at the lateral RA free wall (Crista Terminalis; CT), and interatrial
septum (IAS). Decapolar catheters are placed in the coronary sinus (CS), and in
the left atrium (LA) via a patent foramen ovale or in the left pulmonary artery
(LPA) under local anesthesia. The His bundle electrogram location serves as an
anatomic landmark. The spacing between
successive pairs of bipoles of the decapolar or duo-decapolar (halo) catheter
located at the IAS or CT is usually 5 mm and each electrode pair has an
interelectrode distance of 2 mm. A decapolar catheter with a 2-mm
interelectrode distance and electrode spacing at 8 mm is used in the CS
endocardially in the LA, or epicardially via the LPA. Bipolar recording sites include
high, mid and low
lateral RA, high, mid and low IAS, proximal and distal His bundle, proximal to
distal CS ostium, proximal, and mid and distal LPA for superior LA recordings.
Additional recordings are obtained from the superior LA at the septal,
mid-superior and lateral appendageal regions via a patent foramen ovale if
present or via the LPA. The LPA recordings used have
been previously validated with endocardial LA recordings. Bipolar electrograms are amplified and filtered
between 30 and 100 Hz. Throughout the analysis, local intervals are measured
between the onset of high frequency deflections of the successive EGMs. All
intracardiac recordings and 12 surface ECG leads are simultaneously recorded on
hard copy at paper speeds of 100 to 200 mm/s. Figure 2
demonstrates the onset of spontaneous AF which is analyzable on a beat-to-beat
basis. We have
previously demonstrated the stability and reliability of multiple catheters for
biatrial mapping . The reproducibility of
triggering beats sites with this mapping methodology has also been previously
reported [9, 10].
Figure 3 shows an organized, single activation
wavefront in the left atrium which then changes cycle length, activation
pattern, and appearance of fragmented electrograms on the right atrium.
Figure 2:Integrated recording of right and left atrial electrograms obtained using biatrial catheter electrode array with three dimensional non contact map of left atrium which permits beat to beat biatrial and high resolution left atrial mapping. Organized electrogram activity is noted at AF onset from the left atrium.
Figure 3:Demonstration of an organized single activation wavefront on the left side of the panel which then changes cycle length, activation pattern and with appearance of fragmented electrograms on the right side.
B. Three-Dimensional Non-Contact Balloon Catheter Mapping
An endocardial balloon electrode (EnSitea, St. Jude Medical, St. Paul, MN) is
advanced through a special sheath into the RA via the right femoral vein. A
typical RA placement can be seen in Figure 4. In patients with a patent foramen ovale, it was
placed first in the LA and later relocated to the RA for biatrial mapping. 3D
mapping of AF in the RA or LA or both was performed. The 7.5 mL balloon
is deployed by injection of a 50/50 mixture of saline and contrast media fully expanding the multi electrode array (MEA)
containing the 64 electrodes. Under
fluoroscopic guidance, a 3D contour in the
RA is developed using EGMs and anatomically identifiable landmarks in the
lateral RA, such as the orifices of the great veins, the entire ring of the
tricuspid annulus, CS ostium, tricuspid valve - inferior vena cava isthmus
sites, the lateral RA from superior to inferior locations and the IAS at
multiple sites along a cranio-caudal axis. Similarly, in the LA, we identified
mitral valve annulus and its posterior and anterior aspects from the free wall
to the septal aspects, orifices of the four pulmonary veins, superior to
inferior lateral LA, the LA appendage and the foramen ovale. Virtual
electrogram recordings from over 3,300 endocardial sites and digitized biatrial
catheter electrograms were simultaneously recorded for beat-to-beat analysis of
spontaneous atrial premature beats (APBs) (regional origin) and atrial
tachyarrhythmias (ATs). Figure 5 demonstrates an
example of an upper loop RA flutter recorded during a spontaneous AF episode in
a patient with a mitral valve prosthesis and persistent AF.
Figure 4:Position of the multi electrode array balloon catheter in the right atrium and biatrial contract catheters in a right anterior oblique fluoroscopic view.
Abbreviations: CS = Coronary Sinus, HB = His bundle, LPA = Left Pulmonary Artery, MEA = Multielectrode Array.
Figure 5:Non Contact Map example of an upper loop RA flutter recorded during a spontaneous AF episode in a patient with a mitral valve prothesis and persistent AF. The head of the wavefront is shown with a white zone and the development of linear lesions (labeled L) to form a right atrial compartmentalization lesion set for interruption of the flutter.
C. Mapping of Spontaneous and
This mapping approach has been employed
in our studies to map both induced and spontaneous AF [9, 10]. In both types of AF, organized electrogram activation
patterns have been documented. However, induced AF is most often initiated by
local intra-atrial reentry near the pacing site. In contrast, spontaneous AF
can be initiated by different premature beats that may be significantly distant
from the onset tachycardia. These premature beats may differ spatially in
location but this may appreciate only the 3D NCM rather than the surface ECG or
EGM configuration (Figure 6).
Simultaneous biatrial and NCM offers the ability to immediately identify differing sites of origin of APBs recorded in a single procedure in a high resolution three dimensional location . The RA three dimensional NCM contour is shown with the body torso orientation on the top right of each panel and selected ECG recordings and electrograms from the biatrial contact catheter map and virtual recordings from the non contact map are shown at the bottom. Note that the catheter electrograms show the differing atrial activation patterns while the P-wave morphology may show modest or very subtle changes.
Abbreviations: APB = atrial premature beat, Inf = inferior, Lat = lateral, RA = right atrium, SLA = superior left atrium.
spontaneous AF is clearly the clinically relevant event. However, the
spontaneous event may not occur during the course of an EP study and has to be stimulated.
Adminstration of large doses of isoproterenol is often used for this purpose,
but we have observed that induced AF events can promote spontaneous AF events either
by evolving into that rhythm, after spontaneous termination of the induced AF,
or after cardioversion of the induced AF. In the event that the patient arrives
in persistent AF, EGM recordings may show simultaneous tachycardias and NCM is
employed to delineate the substrate involved. It is our experience that this
extent of rotor mapping is rarely feasible with regional atrial mapping with a
single or even two multipolar electrode catheters. Cardioversion of the
persistent AF event often results in unmasking triggers and early AF recurrences
after cardioversion identify the mechanisms of AF onset and contributing rotors.
Evolution of AF over the next few minutes in the study may often clarify
additional tachycardia rotors that develop and maintain persistent AF.
It is from these data
obtained in such procedures that we have proposed multiple triggers exist in
human AF and that multiple ATs are present in both paroxysmal and persistent AF
and these constitute rotors that initiate or maintain AF episodes. AF
progression is most likely related to the development of a biatrial substrate
for the arrhythmia and proliferation of these rotors that result in persistence
of an individual event. In the following section, we describe some of our
findings in different AF presentations and disease states.
D. Clinical Observations with
In this section, we summarize
some of our novel observations from this novel approach to AF mapping. For our
initial studies, we selected consecutive patients meeting the following study
inclusion criteria: (1) Presence of symptomatic, recurrent AF, failed at least
2 antiarrhythmic drug trials, and were undergoing a clinically indicated EP
study for definition of mechanisms, atrioventricular conduction and the treatment
of associated tachyarrhythmias, and (2) In all patients a written informed
consent was obtained for the EP testing and mapping procedure.
In these patients, we
systematically performed the following analyses:
(1)Identified spontaneous APBs over a period of observation for regional
origin and coupling intervals regardless of their involvement in initiating an
episode of AF,
(2)Examined spontaneous AF onset, for tachyarrhythmia mechanism, regional
origin, cycle length and termination modes,
(3)Compared activation, evolution and termination patterns of AF in
individual patients whether arising from the RA or LA, and
(4)Compared these patterns and their relationship to the presence of
structural heart disease.
A total of 76 patients, mean age 63 + 11
years (68% Male) were included in this analysis (Table 1).
The study population had structural heart disease in the majority of patients (79%)
and was mostly in New York Heart Association Class I or II. Further analyses
were conducted to permit study of different AF populations, specifically
paroxysmal AF patients (Gp.1, n=20) compared to persistent AF patients (Gp.2, n=56).
Table 1:Patient demographics in total population and subgroups (Group 1 = paroxysmal AF, Group 2 = Persistent/Permanent AF).
HTN = hypertension, CAD = coronary artery disease, DCM = dilated cardiomyopathy, VHD = valvular heart disease, ASD = atrial septal defect.
Mapping of Spontaneous APBs
and Spontaneous AF
66 patients demonstrated spontaneous
APBs in the RA (n=90) and LA (n=41) with similar regional distributions
regardless of structural heart disease status. 42 patients (64%) had > 2
disparate regional origins and biatrial regional foci demonstrated equal
frequency in Gp.1 and 2 patients (41% and 43%, respectively; see Figure
7 (a)). The regional distribution by patients showed the predominant
region of origin as being the IAS in the overall population (40 patients or 61%) and
Gp.2 patients (30 patients or 65%). In Gp.1 patients, the predominant regions of origin was
equally observed both in the IAS and the superior LA (10 patients or 59%,
respectively). Figure 8 demonstrates the ability of
simultaneous biatrial and NCM to identify differing sites of origin of five
APBs recorded in a single patient in rapid sequence in a single procedure.
Figure 7:Summary of the (a) region of origin of the atrial premature beatsin the left figure panel, and (b) atrial tachyarrhythmias in the patient population in the right figure panel as demonstrated by biatrial and non-contact mapping.
Figure 8:Simultaneous biatrial and NCM to identify differing sites of origin of five APBs recorded in a single patient in rapid sequence in a single procedure. The RA three-dimensional NCM contour is shown with the body torso orientation on the top right of each panel and selected ECG recordings, electrograms from the biatrial contact catheter map and virtual recordings from the NCM are shown on the right. The site of origin of the APB is shown with a red asterisk and can be spatially related using the body torso orientation. The left column shows APBs arising in the RA at three differing locations. The right column shows catheter electrograms showing LA onset of two different APBs with widely disparate RA breakthrough locations and activation patterns.
Abbreviations: SVC = superior vena cava, IVC = inferior vena cava; CT 1-8 = Lateral RA; HLRA, MLRA, LLRA = high, middle, and low lateral RA, respectively; HCT = high lateral RA; LCT = low lateral RA; HIAS = high interatrial septum; CT 11-20 = septal locations; CSpx = coronary sinus ostium; CS dis = coronary sinus distal; LLPA and LPA = left pulmonary artery. The numbers represent virtual electrogram recording sites.
56 patients (74%) demonstrated 93 spontaneous episodes of AF as
documented by simultaneous ECG recordings during the mapping procedure. 61% of
patients had a right-sided origin and 39% has a left-sided origin of the AT.
It was observed that the regional distribution was more extensive in
Gp.2 patients compared with Gp.1 patients (see Figure 7 (b)). Isthmus
dependent RA counterclockwise flutters were most common in the overall
population and Gp.1 patients (48% and 58%, respectively, while nonisthmus dependent
RA flutter was most common in Gp.2 patients (47%). Overall, focal or macroreentrant left sided ATs were seen
in 10 patients (18%), and macroreentrant LA flutter in 14 patients (25%).
In patients with triggers localized to
the superior LA or superior pulmonary veins and in those localized to the
inferior LA or pulmonary veins, there was a single centrifugal wavefront, as
discerned from both contact and NCM recordings. RA or septal triggers
resulted in focal origin with centrifugal propagation (Figure
9). During RA propagation, a single dominant wavefront activated the RA.
After AF onset, a macroreentrant wavefront with a head to tail relationship was
seen. There was no period of electrical quiescence noted. Figure
10 demonstrates a cascade of ATs initiated by a left atrial APB probably
arising in a superior pulmonary vein. The APB shows a breakthrough into the RA
septum (top panel) and is followed by a left sided AT (middle panel) with a
similar RA breakthrough and finally induces a counterclockwise RA flutter
Figure 9: Biatrial and right atrial noncontact high resolution map of the onset of AF from the right side of the interatrial septum. Body torso orientation is shown in the inset, with a left lateral view of the right atrium. The red asterisk shows the onset of the AF wavefront in a single beat.
Figure 10: Noncontact mapping demonstrates a cascade of ATs initiated by a left atrial APB probably arising in a superior pulmonary vein. The APB shows a breakthrough into the RA septum (top panel) and is followed by a left sided AT (middle panel) with a similar RA breakthrough and finally induces a counterclockwise RA flutter (lower panel). The biatrial catheter electogram map and virtual electrograms are shown.
Abbreviations: LPV = left pulmonary vein, RA = right atrium, Tach = tachycardia
There are several major findings to report from these early
(1)There is a broad distribution
of triggering beats in the RA, septum and LA/pulmonary veins in a
drug-refractory AF population from paroxysmal to persistent/permanent patients.
(2)Both Gp. 1 and Gp.2 patients
demonstrated equal frequency of biatrial regional foci and distribution;
commonly with multiple regional sources of APBs in an individual patient. The
reproducible origin of APBs and ATs in this study argues against catheter-mediated
initiation of the arrhythmias.
(3)At sustained AF onset, focal
triggers initiate an organized tachyarrhythmia, which may be focal or reentrant
in nature and may localize to an adjoining atrial region or recruit a larger
(4)In patients with paroxysmal AF or
in those without structural heart disease, LA tachycardias and RA flutter are
commonly present while a much broader spectrum of tachyarrhythmias are seen in
patients with persistent/permanent AF or in the presence of structural heart disease.
Implications of biatrial mapping for mechanisms
In the early twentieth century, several
different mechanisms were proposed including Lewis who propounded the theory
that AF was “an impure flutter with fibrillatory conduction” . It was in 1959 that Moe
and Abildskov replaced this concept with the multiple wavelet hypothesis based
on computer simulations of AF . Available animal models
studied by Allessie and coworkers corroborated the concept in the multiple
wavelet model and they described the basis of AF as random functional reentry,
with wavelength being critical to defining the reentrant circuit . Subsequently, their
studies have shown that persistence of sustained AF results in functional
electrical remodeling in the atrium .
Human studies have provided
vignettes from different sources as to clinical mechanisms. Attuel and
coworkers have shown the importance of electrophysiologic changes in the
substrate in AF patients . Intraoperative mapping
studies of Cox and Schuessler indicated evidence for macroreentrant circuits in
human AF . Focal triggers of AF
originating from pulmonary veins were identified and have been studied
extensively . Considerably less focus
has been paid to other regions of AF origin. While some investigators have
described superior vena caval triggers, other sites or arrhythmias triggering
AF have been poorly defined .
catheter mapping of human AF has been limited to EGM recordings in specific
regions of interest (e.g., pulmonary veins) and regional mapping has been
limited to intraoperative studies [17, 18]. A major limitation of the regional catheter
method has been infrequent and sporadic recording of spontaneous sustained AF,
requiring repetitious, stable rhythms. Absence of regional or global mapping
limits the trigger analysis to one or few regions of interest and provides
limited insight into the behavior of both atria during initiation or
perpetuation of AF. Haisseguire et al., used Lasso catheters for the
focal mapping of pulmonary veins, where rapidly firing foci play an important
role in the initiation and maintenance of AF . Other investigators have employed duodecapolar Halo
catheters placed around the tricuspid annulus, however, it is difficult to
obtain regional mapping information. These techniques require catheter
manipulation and stabilization to record from multiple sites. As a direct
consequent, the changing nature of electrical activation patterns in AF pose a
severe limitation to sequential mapping methods. Our approach overcomes many of these issues and permits the recording of
important biatrial behavior in the vast majority of patients in the study (90%).
Biatrial Contact and NCM Mapping
Simultaneous biatrial contact and
high-resolution NCM has been able to
mechanistic information on triggers (atrial premature beats (APBs), initiation
and evolution of spontaneous AF, and ATs,
(b)Permit study of
different AF populations,
(c)Provide a global
view of the EP behavior of both atria. Used simultaneously, we demonstrated
that these two techniques provide extensive mechanistic information on
triggers, initiation, and evolution of human AF in contrast to currently
employed techniques. We can suggest from this approach that it progression of
AF from paroxysmal to persistent and permanent forms is characterized, not by
change in triggering APB distribution, but by the more extensive distribution
of ATs. The varying frequency of AF recurrences with single therapeutic
interventions in AF including RA maze procedures, focal pulmonary vein ablation
or pacing therapy and inadequate long term rhythm control as seen in clinical
trials may be due to multiple organized tachycardias and triggers [19-22]. Our findings support extensive
biatrial interventions or a hybrid therapy approach for long-term suppression
of AF [23-25], and
(d)Allows for tailored
linear ablation procedures to be designed. Figure 11 demonstrates
a left atrial isolation with linear lesions including a roof line and a mitral
Figure 11:Example of a tailored left atrial compartmentalization procedure with linear lesions including bilateral pulmonary vein antral isolation, a roof line and a mitral isthmus line. The wavefront seen is a tachycardia arising from a superior pulmonary vein that is entrapped in a posterior compartment.
There are technical limitations of
NCM including a loss of resolution beyond a 4 cm radius of the balloon
electrode and pulmonary artery recordings that cannot be equated with detailed
endocardial LA catheter and balloon mapping that would allow for more detailed
LA localization. The NCM system employs contact catheters to create a 3D
representation of the endocardial anatomy of the patient’s chamber of interest,
however to provide the most accurate representation of the patient’s true
anatomy one must have the ability to import a cardiac CT/MRI into the mapping
system for navigation purposes.
The methodologies described in this
report provide a clinically relevant, global atrial mapping technique to help
classify AF initiation and maintenance in a given patient and provide targets
for interventional therapeutic procedures. In large part, owing to the ability
of the technique to record widely disparate triggers and ATs in a rapid and
reliable fashion. Simultaneous biatrial and NCM permits successful AF mapping
in different AF populations and demonstrates a biatrial spectrum of spontaneous
triggers and organized monomorphic tachycardias with multiple or biatrial
locations. Evolution into sustained AF can be studied with global biatrial
contact and NCM in humans. These methodologies bridge the divide between
observations obtained with regional catheter recordings and intraoperative
findings in humans.
Allessie M, Ausma J, Schotten U. Electrical, contractile and structural remodeling during atrial fibrillation. Cardiovasc Res 2002;54:230-246.
Shinagawa K, Shi YF, Tardif JC, Leung T-K, Nattel S. Dynamic nature of atrial fibrillation substrate during development and reversal of heart failure in dogs. Circulation. 2002; 105: 2672-78.
Mihm MJ, Yu F, Carnes CA, Reiser PJ, McCarthy PM, Van Wagoner DR, Bauer JA. Impaired myofibrillar energetics and oxidative injury during human atrial fibrillation. Circulation. 2001; 104: 174-80.
Saksena S, Hettrick DA, Koehler JL, Grammatico A, Padeletti L. Progression of paroxysmal atrial fibrillation to persistent atrial fibrillation in patients with bradyarrhythmias. Am Heart J, 2007. 154(5): p. 884-92.
Haissaguerre M, Jais P, Shah DC, Gencel L, Pradeau V, Garriques S, Chouairi S, Hocini M, Le Metayer P, Roudant R, Clementy J. Right and left atrial radiofrequency catheter therapy of paroxysmal atrial fibrillation. J Cardiovasc Electrophysiol. 1996; 7: 1132-44.
Haissaguerre M, Jais P, Shah DC, Takahashi A, Hocini M, Quiniou S, Garriques S, Le Mouroux A, Le Metayer P, Clementy J. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med. 1998; 339(10): 659-66.
N WS, Tai CT, Hseih CF, Lin YK, Tsao HM, Huang JL, Yu WC, Yang SP, Ding YA, Change MS, Chen SA. Catheter ablation of paroxysmal atrial fibrillation initiated by non-pulmonary vein ectopy. Circulation. 2003; 107: 3176-83.
Mansour M, Forleo GB, Pappalardo A, Heist EK, Avella A, Laurenzi F, De Giralmo P, Bencardino G, Dello Russo A, Mantica M, Ruskin JN, Tondo C. Initial experience with the Mesh catheter for pulmonary vein isolation in patients with paroxysmal atrial fibrillation. Heart Rhythm. 2008; 5(11): 1510-6.
Saksena S, Giorgberidge I, Mehra R, Hill M, Prakash A, Krol RB, Mathew P. Electrophysiology and endocardial mapping of induced atrial fibrillation in patients with spontaneous atrial fibrillation. Am J Cardiol, 1999. 83(2): p. 187-93.
Saksena S, Prakash A, Krol RB, Shankar A. Regional endocardial mapping of spontaneous and induced atrial fibrillation in patients with heart disease and refractory atrial fibrillation. Am J Cardiol, 1999. 84(8): p. 880-9.
Lewis T. The Mechanism and Graphic Registration of the Heart Beat. 3rd ed. 1925, London: Shaw & Sons. 319-374.
Moe GK, Abildskov JA. Atrial fibrillation as a self-sustaining arrhythmia independent of focal discharge. Am Heart J, 1959. 58(1): p. 59-70.
Rensma PL, Allessie MA, Lammers WJ, Bonke FI, Schalij MJ. Length of excitation wave and susceptibility to reentrant atrial arrhythmias in normal conscious dogs. Circ Res, 1988. 62(2): p. 395-410.
Wijffels M.C, Kirchhof CJ, Dorland R, Allessie MA. Atrial fibrillation begets atrial fibrillation. A study in awake chronically instrumented goats. Circulation, 1995. 92(7): p. 1954-68.
Attuel P, Childers R, Cauchemez B, Poveda J, Mugica J, Coumel P. Failure in the rate adaptation of the atrial refractory period: its relationship to vulnerability. Int J Cardiol, 1982. 2(2): p. 179-97.
Cox JL, Canavan TE, Schuessler RB, Cain ME, Lindsay BD, Stone C, Smith PK, Corr PB, Boineau JP. The surgical treatment of atrial fibrillation. II. Intraoperative electrophysiologic mapping and description of the electrophysiologic basis of atrial flutter and atrial fibrillation. J Thorac Cardiovasc Surg, 1991. 101(3): p. 406-26.
Tsai CF, Tai CT, Hsieh MH, Lin WS, Yu WC, Ueng KC, Ding YA, Chang MS, Chen SA. Initiation of atrial fibrillation by ectopic beats originating from the superior vena cava: electrophysiological characteristics and results of radiofrequency ablation. Circulation, 2000. 102(1): p. 67-74.
Konings KT, Kirchhof CJ, Smeets JR, Wellens HJ, Penn OC, Allessie MA. High-density mapping of electrically induced atrial fibrillation in humans. Circulation, 1994. 89(4): p. 1665-80.
Garg A, Finneran W, Mollerus M, Birgersdotter-Green U, Fujimura O, Tone L, Feld GK. Right atrial compartmentalization using radiofrequency catheter ablation for management of patients with refractory atrial fibrillation. J Cardiovasc Electrophysiol, 1999. 10(6): p. 763-71.
Gerstenfeld EP, Callans DJ, Dixit S, Zado E, Marchlinski FE. Clinical outcome after radiofrequency catheter ablation of focal atrial fibrillation triggers. J Cardiovasc Electrophysiol, 2001. 12(8): p. 900-8.
Gillis AM, Wyse DG, Connolly SJ, Dubuc M, Philippon F, Yee R, Lacombe P, Rose MS, Kerr CD. Atrial pacing periablation for prevention of paroxysmal atrial fibrillation. Circulation, 1999. 99(19): p. 2553-8.
Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, Rosenberg Y, Schron EB, Kellen JC, Greene HL, Mickel MC, Dalquist JE, Corley SD for Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) Investigators. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med, 2002. 347(23): p. 1825-33.
Madan. N, Saksena S. Long-term rhythm control of drug-refractory atrial fibrillation with "hybrid therapy" incorporating dual-site right atrial pacing, antiarrhythmic drugs, and right atrial ablation. Am J Cardiol, 2004. 93(5): p. 569-75.
Saksena S, Madan N. Hybrid therapy of atrial fibrillation: algorithms and outcome. J Interv Card Electrophysiol, 2003. 9(2): p. 235-47.
Swartz J, Pellersels G, and SilversJ. A catheter-based curative approach to atrial fibrillation in humans (abstract). Circulation, 1994. 90: p. I-335.