Focal Atrial Tachycardia
New Insight From Noncontact Mapping and Catheter Ablation

Satoshi Higa, MD; Ching-Tai Tai, MD; Yenn-Jiang Lin, MD; Tu-Ying Liu, MD; Pi-Chang Lee, MD;Jin-Long Huang, MD; Ming-Hsiung Hsieh, MD; Yoga Yuniadi, MD; Bien-Hsien Huang, MD;Shih-Huang Lee, MD; Kwo-Chang Ueng, MD; Yu-An Ding, MD; Shih-Ann Chen, MD

Background ¡ª This study investigated the electrophysiologic characteristics, atrial activation pattern, and effects of radiofrequency (RF) catheter ablation guided by noncontact mapping system in patients with focal atrial tachycardia (AT).

Methods and Results ¡ª In 13 patients with 14 focal ATs, noncontact mapping system was used to map and guide ablation of AT. AT origins were in the crista terminalis (n
8), right atrial (RA) free wall (n
3), Koch triangle (n
1), anterior portion of RA¨Cinferior vena cava junction (n
1), and superior portion of tricuspid annulus (n
1); breakout sites were in the crista terminalis (n
5), RA free wall (n
5), middle cavotricuspid isthmus (n
2), and RA¨Csuperior vena cava junction (n
2). ATs arose from the focal origins (11 ATs inside or at the border of low-voltage zone), with preferential conduction, breakout, and spread to the whole atrium. After applications of RF energy on the earliest activation site or the proximal portion of preferential conduction from AT origin, 13 ATs were eliminated without complication. During the follow-up period (85 months), 11 (91.7%) of the 12 patients with successful ablation were free of focal ATs.

Conclusions ¡ª Focal AT originates from a small area and spreads out to the whole atrium through a preferential conduction. Application of RF energy guided by noncontact mapping system was effective and safe in eliminating focal AT. (Circulation. 2004;109:84-91.)

Key Words: mapping tachycardia ablation

Most focal atrial tachycardias (ATs) can be successfully eliminated by catheter ablation, with low complication and recurrence rate.1¨C5 Although the details of electrophysi-ologic
mechanisms and electropharmacological characteris-tics of focal AT have been demonstrated, the knowledge about activation pattern of focal AT is very limited.4,5 Recently, oncontact mapping system has been demonstrated to facilitate the identification of ectopic focus, because it is able to reconstruct the precise geometry in the atrium and ventricle and localize the ectopic beats.6,7
The purposes of this study were to demonstrate the electrophysiologic characteristics, activation patterns, and results of catheter ablation in patients with focal AT using noncontact mapping system.

Methods
Patient Characteristics
This study included 13 consecutive patients (7 men and 6 women;age 4523 years) with clinically documented AT who were referred for electrophysiologic study and catheter ablation guided by the noncontact mapping system. Two patients had cardiomyopathy, and
1 patient had coronary artery disease (Table 1).

Catheter Position and Electrophysiologic Study
Informed written consent was obtained from all patients. As de-scribed previously, the patients were studied in the postabsorptive, nonsedated state.4,5 All antiarrhythmic drugs were discontinued for at least 5 half-lives before the study. A 7F, deflectable, decapolar catheter with 2-mm interelectrode distance and 5-mm space between each electrode pair was also inserted into the coronary sinus via the internal jugular vein. The position of the proximal electrode pair at the ostium of the coronary sinus was confirmed with contrast injection. A 9F sheath placed in the left femoral vein was used to introduce the noncontact mapping catheter.
Rapid right atrial stimulation (paced cycle length from 600 ms until 2:1 capture was noted) and right atrial extrastimuli (single or double) were used for induction and termination of AT, and they were repeated 2 to 4 times to ensure reproducibility of the responses.If programmed electrical stimulation failed to induce AT, isoproter-enol (at graded dosages from 1 to 4
g/min) was infused intrave-nously until AT developed or the sinus rate increased to 20% above the resting value. The ectrophysiologic criteria used for diagnosis of focal AT have been reported.1¨C5.

Received May 7, 2003; de novo received July 21, 2003; revision received September 29, 2003; accepted September 30, 2003.
From the Division of Cardiology, Department of Medicine, National Yang-Ming, University School of Medicine, Taipei Veterans General Hospital,Taiwan. Dr Higa is a research fellow from Okinawa University, Okinawa, Japan.
Dr Chen has received research support (including free catheters and software analysis) from ESI Co.Correspondence to Shih-Ann Chen, MD, Division of Cardiology, Department of Medicine, Taipei Veterans General Hospital, 201, Sec. 2, Shih-Pai Road, Taipei, Taiwan. E-mail epsachen@ms41.hinet.net© 2004 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org DOI: 10.1161/01.CIR.0000109481.73788.2E

Response to Adenosine
In 6 patients (cases 2, 3, 8, 11, 12, and 13), an intravenous bolus of adenosine (3 to 12 mg) was given to observe the effects of adenosine on terminating AT.

Noncontact Mapping System
The noncontact mapping system (EnSite 3000 with Precision Soft-ware,Endocardial Solutions) has been described in detail previous-ly.6¨C10 In brief, the system consists of a noncontact catheter (9F) with a multielectrode array (MEA) surrounding a 7.5-mL balloon
mounted at the distal end. Raw data detected by the MEA are transferred to a silicon graphics workstation via a digitalized ampli-fier system.
The MEA catheter was deployed over a 0.035-inch guide wire,which had been advanced to the superior vena cava (SVC) (Figure 1). It is used to construct a 3D computer model of the virtual endocardium, providing a geometry matrix for the inverse solution.The system is able to reconstruct more than 3000 unipolar electro-grams simultaneously and superimpose them onto the virtual endo-cardium,producing isopotential maps with a color range represent-ing voltage amplitude.
During review of the recorded data, we always began analysis with a default high-pass filter setting of 2 Hz to preserve components of slow conduction on the isopotential map. Color settings were adjusted so that the color range matched 1 to 1 with the millivolt
range of the electrogram deflection of interest. We also interactively placed virtual electrodes on the map color contours to analyze the corresponding noncontact unipolar electrograms. Occasionally, con-duction of activation wavefront was sufficiently slow that we moved the high-pass filter down to 1.0 to 0.5 Hz.

Definitions
Origin of AT was defined as the earliest site showing a single spot of isopotential map and a QS pattern of noncontact unipolar electrogram. Breakout site of AT was the earliest site that showed an rS pattern with sudden increase of peak negative potential of noncontact unipolar electrogram after AT depolarized. The prefer-ential conduction was the initial direction of depolarization away from an origin. Double potentials were noncontact unipolar atrial electrograms with 2 discrete deflections per beat either separated by
an isoelectric baseline or a low amplitude interval.10 Low voltage zone (LVZ) was an area with 30% amplitude of noncontact unipolar electrogram peak negative potential.

Validation of Noncontact Electrogram
Contact electrograms were recorded from 99 randomly chosen locations around the chamber during sinus rhythm and AT. The EnGuide navigation signals were simultaneously recorded from each.


Figure 1. Radiographs showing a multielectrode array balloon catheter (Balloon) in the right atrium,a decapolar catheter in the coronary sinus (CS),and an ablation catheter (ABL) around crista termi-nalis.A and B, Right and left anterior oblique views, respectively.
site for geometric annotation of location and for generation of virtual electrograms that can be compared with the associated contact electrograms. Simultaneous recording of the bipolar and unipolar electrogram from the distal tip of the contact catheter was performed.Signals for both contact and noncontact electrograms were filtered with a bandwidth of 2 to 300 Hz. Electrogram morphologies, activation time difference, and electrogram voltage between contact and noncontact electrograms that were taken from the same endo-cardial sites were compared by use of a well-described template comparison algorithm.8,9,11,12

Catheter Ablation and Follow-Up
Catheter ablation (40 to 50 W, 50¡ãC to 60¡ãC, 40 seconds) was performed using a 4-mm electrode-tipped ablation catheter con-nected to an EPT-1000 generator (Boston Scientific Co). We first delivered RF energy on the origin or the proximal portion of preferential conduction from the origin. After catheter ablation, the same stimulation protocols used to induce AT before ablation were performed to make sure AT was noninducible. Successful catheter ablation was defined as inability to reinduce focal ATs. After hospital discharge, the patients were followed up closely (every 1 to 3 months) in the outpatient clinic. Long-term efficacy was assessed clinically on the basis of the resting surface ECG, 24-hour Holter monitoring, event recorder, and clinical symptoms.

Statistical Analysis
Continuous data were expressed as meanSD. For validation of the mapping accuracy, correlation between contact and noncontact electrogram was explored by calculating Pearson¡¯s correlation coef-ficients and using Bland-Altman technique for agreement. Differ-ences were considered significant at P0.05.

Results
Electrophysiological Characteristics
Fourteen focal ATs were demonstrated. Four ATs were incessant. Six ATs occurred spontaneously at the laboratory with or without isoproterenol, 2 ATs were initiated after right atrial stimuli, and 2 ATs were initiated after isoproterenol infusion plus atrial stimuli. Mean AT cycle length was 39481 ms, and the mean earliest activation time was 5122 ms before the onset of the P wave (Table 1). Furthermore, 11 of 14 ATs originated from inside the LVZ or border of LVZ.



Validation Data
The correlation between contact and noncontact unipolar electrograms showed the correlation coefficient of electro-gram morphology was 0.88 (P0.001), time difference was
2.252.79 ms, and correlation value of peak negative voltage (PNV) was 0.77 (P0.001). These correlations showed the similar results in the area outside LVZ (correlation coefficient of electrogram morphology, 0.86; P0.001; time difference,3.13.0 ms; correlation value of PNV, 0.77; P0.001) and inside LVZ (correlation coefficient of electrogram morphol-ogy, 0.89; P0.001; time difference, 1.72.5 ms; correlation
value of PNV, 0.93; P0.001). The agreement analysis of voltage for all validation points outside and inside LVZ showed only 3%, 5%, and 3% of validation points were outside 2 SD, respectively (Figure 2). For the timing differ-ence, 7% of validation points were outside 2 SD. The mean absolute value of the contact unipolar electro-gram PNV was 0.940.54 mV (sinus rhythm,1.030.62; AT, 0.930.54 mV) in the area outside LVZ and 0.350.19 mV (sinus rhythm, 0.480.17; AT,0.280.16 mV) inside LVZ. The mean absolute value of
the noncontact virtual unipolar electrogram PNV was1.020.41 mV (sinus rhythm, 1.610.69; AT,0.960.32 mV) in the area outside LVZ and 0.420.20 mV (sinus rhythm, 0.520.12; AT, 0.370.21 mV)inside LVZ.

Response to Adenosine
Adenosine could not terminate AT in patient No. 2 with incessant AT. However, adenosine (3 to 12 mg) terminated ATs in the other 5 patients (Nos. 3, 8, 11, 12, and 13).Noncontact mapping demonstrated 2 types of adenosine-induced termination of AT. The first type (n
1) showed shifting of AT origin before termination. The second type (n
4) showed no change of AT origin. All AT termination episodes showed disappearance of focal activation at origin,
Figure 2. Bland-Altman analysis showing the relationship between contact and noncontact electrogram voltages in all val-idation points (A), outside LVZ (B), and inside LVZ (C).



Figure 3. Schematic representations of anatomical relation between origin, pref-erential
conduction, and breakout site during AT (Nos. 1 through 13). Focal activation arises from origin with prefer-ential conduction. Schema in case No.1 indicated 2 AT origins. The other cases had a single AT origin. The asterisk rep-resents origin of AT, the curved line rep-resents the preferential conduction, and the bifurcation point of the curved line
represent the breakout site. IVC indicates inferior vena cava. not attributable to block at the area of preferential conduction or exit site.

Anatomical Relation Between Origin, Preferential Conduction, and Breakout Site
Noncontact mapping clearly showed the anatomic locations and activation wavefronts from origins with preferential conduction and activated the myocardium surrounding the
origin (Figures 3 through 5). The distance from the balloon center to AT origins was 21.66.2 mm (Table 2). Catheter Ablation and Follow-Up For case No. 2, RF energy was applied on the proximal portion of preferential conduction from the origin because of
continuous shifting of the origin within a small area (1.51.0 cm 2 ). For case No. 3, RF energy was applied on the proximal portion of preferential conduction, because this
patient had severe chest pain when we applied RF energy on the origin. For cases Nos. 1 and 7, RF energy was applied on the origin only. For the other 9 patients, RF energy was
applied on the origin and proximal portion of preferential conduction. A mean of 117 RF applications were required for eliminating AT, and 79 RF applications were used for
insurance ablation in the origins or preferential conduction. For case No. 3, AT cycle length increased to approximately 500 ms; because of the long procedure time, patient preferred discontinuation of the procedure, and had regular follow-up of the clinical symptoms. There was no complication during the ablation procedure. During a follow-up of 85 months, only case No. 1 had recurrent AT originating from crista terminalis (CT), and the second procedure had successful ablation of AT from the old focus. The other 11 patients did not have recurrence of focal AT.

Discussion
Major Findings
To the best of our knowledge, this is the first study demon-strating the anatomical relation and electrophysiologic char-acteristics of origin with preferential conduction of focal AT



Figure 4. Noncontact mapping of focal AT (No. 5) in the posterolateral view. sopotential map shows overlapping of the activation wavefront located at origin, preferential conduction, and breakout site. Isopotential maps showing the activation sequence of AT. Color scale for each isopotential map has been set so that white indicates most negative potential and purple indicates least negative poten-tial. The focal activation originates from the middle portion of CT, and the wavefront propagates up to the upper portion of CT through the preferential conduction area, then reaches the breakout site, and spreads to the whole atrium. The noncontact unipolar electrogram reveals a QS pattern at origin. The spread from the breakout site activates the myocardium surrounding the AT origin. Noncontact elec-trograms at the origin and proximal site of the preferential conduction reveal multicomponents of activation. Origin indicates origin of AT; BO, breakout site of AT; and RAA, right atrial appendage. using the noncontact mapping system. Catheter ablation ofthe origin or proximal portion of preferential conduction is effective in eliminating focal AT.

Electrophysiological Characteristics
In this study, 8 (57%) AT origins were located at CT. Kalman et al 13 have demonstrated the predominant origin of focal AT along the CT. CT showed an area of prominent anisotropy
with slow conduction property and might play an important role in development of microreentry.14 Furthermore, cardio-myocytes in CT may have pacemaking activity or abnormal automaticity,15 and thus focal AT may originate from CT. The preferential conduction of 9 ATs was located along or across the CT; this finding suggests CT can be the part of preferential conduction that AT activation wavefronts pass through.
Information about the relationship between substrate prop-erty and AT origins is limited. Previous studies on the atrial specimen resected from the area with atrial arrhythmias
showed a slow-response action potential with spontaneous depolarization.16¨C18 Josephson et al 19 also showed the slow response or depressed fast response action potential from the atrial specimen resected from human AT. These findings suggest focal AT may originate from diseased atria and explain the possible mechanism of 11 (78.6%) ATs originat-ing
from LVZ or border zone around the LVZ. The noncon-tact unipolar electrograms in the LVZ demonstrated wide, low-amplitude, and fractionated electrograms, suggesting a delayed and nonuniform anisotropic conduction through the diseased right atrium. This may be related to atrial fibrosis resulting from proliferation of smooth muscle cells and collagen fibers beneath the endocardial lining.20 In previous myocardial infarction models that were used
for validation of Carto system, the mean reduction of unipolar electrogram voltage in the infarcted area was 40.4% to 53.2%.21¨C23 When we selected the 30% of maximum peak negative voltage as the criteria for low voltage area, we found the mean reduction of contact unipolar electrogram voltage in the low voltage zone was 49% of the contact unipolar
electrogram in the region outside the low voltage area. This is compatible with the previous study of myocardial infarction model.21¨C23 Using relative ratio as a low voltage zone criteria



Figure 5. Noncontact mapping of focal AT (case No. 11) in the lateral caudal view. Color scale for each isochronal map has been set so that white indicates earliest activation and purple indicates latest activation time. The focal activation originates from the anterior
portion of RA-inferior vena cava junction, and the wavefront propagates up to the middle portion of CT and then spreads out. The con-tact and noncontact unipolar electrograms reveal a QS pattern at origin and an rS pattern at breakout site. The contact and noncontact unipolar electrograms at the origin and proximal site of the preferential conduction reveal multicomponents of electrogram deflections. Noncontact unipolar electrograms recorded at origin, preferential conduction, and breakout site were nearly identical with the contact electrograms obtained from these areas simultaneously. O indicates origin of AT; P, preferential conduction; BO, breakout site of AT; and IVC, inferior vena cava.account for functional variation in unipolar peak negative voltage and low voltage zone of both beat to beat and patient to patient would make voltage data more comparable.

New Insight of Focal AT From RF Catheter Ablation
Conventionally, intracardiac mapping of focal AT using 1 or 2 roving catheters was used to localize the origin of AT.1¨C5 A QS pattern in the unipolar recording is highly predictive of the successful ablation site. Recently, noncontact mapping systems have provided an accurate guide for mapping and ablation of focal tachycardia.6,7 Several investigators have demonstrated and validated the accuracy of noncontact mapping system in mapping of atrial
and ventricular arrhythmias if the distance from noncontact mapping system balloon center was 4 cm.8,9 This study showed that the distance from balloon center to origin, area of
preferential conduction, or breakout site was within 4 cm in 13 of the 14 ATs. Validation of the origin, area of preferential conduction, and breakout site showed the significant correla-tion between contact and noncontact electrogram morphol-ogy, time difference, and voltage. These correlations also were demonstrated inside and outside the LVZ.
The present study also showed that applications of RF energy on the origin or proximal portion of preferential conduction could eliminate focal AT. This finding raises the
issues of selecting appropriate sites for RF ablation. Previous studies showed impulses generated within the sinus node could not propagate to the atrium when conduction in the
zone of perinodal fibers becomes depressed (attributable to pathological conditions).24,25 In case No. 2, the noncontact mapping showed continuous shifting of focal activation site
within a small area, and we observed disappearance of AT after application of RF energy on the proximal portion of preferential conduction only. In case No. 3, who had severe
chest pain and vagal reflex during ablation of the AT origin, we changed the ablation target from origin to preferential conduction and could decrease the AT rate. Marchlinski et al 26 have reported focal AT with preferen-tial conduction using magnetic electroanatomical mapping.The concepts from previous studies may explain that ablation
of specific fiber connected from the origin can create the exit



B indicates border zone; CT-L, M, and U, lower, middle, and upper portions of crista terminalis; Isthmus-M, middle cavotricuspid isthmus; IVC, inferior vena cava;LVZ, low voltage zone; O-LVZ, outside the LVZ; RA-HAL, HL, LAL, LPL, MAL, and ML, high nterolateral, high lateral, low anterolateral, low posterolateral, middle anterolateral, and middle lateral right atrium; RA-A/IVC, anterior portion of RA-IVC junction; RA-P/SVC, posterior portion of RA-SVC junction; and TA-S, superior portion of tricuspid annulus.

block from origin.27,28 These findings suggest that preferen-tial conduction may play a critical role in ablation of focal AT, and we can choose the ablation target according to the concept of origin and preferential conduction of AT. Al-though the true mechanism of preferential conduction was unknown, this conduction may be preferential in 1 direction
because of anisotropic conduction, anatomic obstacles, or conduction through islands of scar or poor conduction. Such conduction may not be electrically protected either perma-nently or functionally.

Study Limitations
Although we have demonstrated the accuracy of voltage correlation inside and outside the LVZ, future research using the experimental model with pathologic study is still necessary.

Conclusions
Noncontact mapping successfully demonstrated precise loca-tions and electrophysiologic characteristics of origin and preferential conduction of focal AT. Focal AT originates
from a small area, conducts through a preferential area, and spreads out to the whole atrium. Application of RF energy on origin or proximal portion of preferential conduction was effective in eliminating focal AT.

Acknowledgments
This work was supported by grants NSC92-2314-B-010-052,NSC92-2314-B-038-050, VGH92-37, 238, RFCM92-01-009, and SKH-TMU-92-17.

References
1. Tracy CM, Swartz JF, Fletcher RD, et al. Radiofrequency catheter ablation of ectopic atrial tachycardia using paced activation sequence mapping. J Am Coll Cardiol. 1993;21:910 ¨C917.
2. Kay GN, Chong F, Epstein AE, et al. Radiofrequency ablation for treatment of primary atrial tachycardias. J Am Coll Cardiol. 1993;21:901¨C909.
3. Lesh MD, Van Hare GF, Epstein LM, et al. Radiofrequency catheterablation of atrial arrhythmias: results and mechanisms. Circulation. 1994;89:1074 ¨C1089.
4. Chen SA, Chiang CE, Yang CJ, et al. Sustained atrial tachycardia in adult patients: electrophysiological characteristics, pharmacological response,possible mechanisms, and effects of radiofrequency ablation. Circulation.1994;90:1262¨C1278.
5. Chen SA, Tai CT, Chiang CE, et al. Focal atrial tachycardia: reanalysis of the clinical and electrophysiologic characteristics and prediction of suc-cessful radiofrequency ablation. J Cardiovasc Electrophysiol. 1998;9:355¨C365.
6. Friedman PA, Asirvatham SJ, Grice S, et al. Noncontact mapping to guide ablation of right ventricular outflow tract tachycardia. J Am CollCardiol. 2002;39:1808 ¨C1812.
7. Hindricks G, Kottkamp H. Simultaneous noncontact mapping of left atrium in patients with paroxysmal atrial fibrillation. Circulation. 2001;104:297¨C303.
8. Schilling RJ, Peters NS, Davies DW. Simultaneous endocardial mappingin the human left ventricle using a non-contact mapping catheter: com-parison of contact and reconstructed electrograms during sinus rhythm.Circulation. 1998;98:887¨C898.
9. Kadish A, Hauck J, Pederson B, et al. Mapping of atrial activation with a non-contact, multielectrode catheter in dogs. Circulation. 1999;99:1906¨C1913.
10. Tai CT, Huang JL, Lin YK, et al. Non-contact three-dimensional mapping and ablation of upper loop re-entry originating in the right atrium. J Am Coll Cardiol. 2002;40:746 ¨C753.
11. Ropella KM, Sahakian AV, Baerman JM, et al. The coherence spectrum: a quantitative discriminator of fibrillatory and nonfibrillatory cardiac rhythms. Circulation. 989;80:112¨C119.
12. Abboud S, Sadeh D. The use of cross-correlation function for the alignment of ECG waveforms and rejection of extrasystoles. ComputBiomed Res. 1984;17:258 ¨C266.
13. Kalman JM, Olgin JE, Karch MR, et al. ¡°Cristal tachycardias¡±: origin of right atrial tachycardias from the crista terminalis identified by intra-cardiacechocardiography. J Am Coll Cardiol. 1998;31:451¨C459.
14. Spach MS. Discontinuous Cardiac Conduction: Its Origin in Cellular Connectivity With Long-Term Adaptive Changes That Cause Arrhyth-mias.Armonk, NY: Futura Publishing Company; 1997:5¨C58.
15. Boineau JP, Canavan TE, Schuessler RB, et al. Demonstration of a widely distributed atrial pacemaker complex in the human heart. Circulation.1988;77:1221¨C1237.
16. Ten Eick RE, Singer DH. Electrophysiological properties of diseased human atrium, I: low diastolic potential and altered cellular response to potassium. Circ Res. 1979;44:545¨C557.
17. Rabine LM, Albert A, Pham TD, et al. The relationship of human atrial cellular electrophysiology to clinical function and ultrastructure. Circ Res. 1983;52:188 ¨C199.
18. Hordof AJ, Edie R, Malm JR, et al. Electrophysiologic properties and response to pharmacologic agents of fibers from diseased human atria.Circulation. 1976;54:774 ¨C779.
19. Josephson ME, Spear JF, Harken AH, et al. Surgical excision of automatic atrial tachycardia: anatomic and electrophysiologic correlates.Am Heart J. 1982;104:1076 ¨C1085.
20. Bharati S, Lev M. Histology of the normal and disease atrium. In: FalkRH, Podrid PJ, eds. Atrial Fibrillation: Mechanisms and Management.New York: Raven Press; 1992:15¨C39.
21. Kornowski R, Hong MK, Gepstein L, et al. Preliminary animal and clinical experiences using an electromechanical endocardial mapping procedure to distinguish infracted from healthy myocardium. Circulation.
1998;98:1116 ¨C1124.
22. Callans DJ, Ren JF, Michele J, et al. Electroanatomic left ventricular mapping in the porcine model of healed anterior myocardial infarction: correlation with intracardiac echocardiography and pathological analysis.Circulation. 1999;100:1744 ¨C1750.
23. Gepstein L, Goldin A, Lessick J, et al. Electromechanical characterization of chronic myocardial infarction in the canine coronary occlusion model.Circulation. 1998;98:2055¨C2064.
24. Sano T, Iida Y. The sino-atrial connection and wandering pacemaker. J Electrocardiol. 1968;1:147¨C154.
25. Strauss HC, Bigger JT Jr, Bassett AL, et al. Actions of diphenylhydantoin on the electrical properties of isolated rabbit and canine atria. Circ Res.1968;23:463¨C477.
26. Marchlinski F, Callans D, Gottlieb C, et al. Magnetic electroanatomical mapping for ablation of focal atrial tachycardias. Pacing Clin Electro-physiol.1998;21:1621¨C1635.
27. Strauss HC, Bigger JT. Electrophysiological properties of the rabbit sinoatrial perinodal fibers. Circ Res. 1972;31:490 ¨C506.
28. Kirchhof CJ, Allessie M. Sinus node automaticity during atrial fibrillation in isolated rabbit hearts. Circulation. 1992;86:263¨C271.