Electrophysiological Characteristics and Radiofrequency Ablation of Focal Atrial Tachycardia Originating From the Superior Vena Cava

Transcription

1 Jpn Circ J 2001; 65: Electrophysiological Characteristics and Radiofrequency Ablation of Focal Atrial Tachycardia Originating From the Superior Vena Cava Kuan-Cheng Chang, MD; Yu-Chin Lin, MD; Jan-Yow Chen, MD; Hsiang-Tai Chou, MD, PhD; Jui-Sung Hung, MD The initiation of focal atrial tachycardia (AT) from the superior vena cava (SVC) remains unclear. In 3 patients (2 females, 1 male; aged 57, 66 and 50 years, respectively) with focal AT arising from different parts of the SVC, the AT occurred spontaneously, rather than being induced by electrical stimulation. The cycle length of the tachycardia was highly variable, ranging between 190 and 300ms in patient 1, 180 and 320ms in patient 2, and 200 and 300 ms in patient 3. The clinical or associated arrhythmias were atrial fibrillation (AF) (patients 1, 3) and atrial flutter (AFL) (patients 2, 3). A presumed SVC potential that was earlier than the activation of all the other mapping sites was recorded during AT at the lower anterior (15-mm above the atriocaval junction), the mid-anterior (25-mm above the atriocaval junction) and the lower posterior aspect of the SVC (17-mm above the atriocaval junction. Radiofrequency (RF) ablation targeting the SVC focus with the SVC potential promptly eliminated the focal AT in all 3 patients. The coexistent typical AFL was ablated, but the AF was not. The follow-up period was 6, 6, and 3 months, respectively, for each of the patients under no antiarrhythmic medication; there has not been a recurrence of symptomatic palpitation. In conclusion, focal electrical firing in the SVC can initiate AT and this type of focal AT is always associated with AFL or AF. RF ablation guided by the presumed SVC potential is safe and highly effective in eliminating the tachycardia. (Jpn Circ J 2001; 65: ) Key Words: Focal atrial tachycardia; Radiofrequency ablation; Superior vena cava Atrial tachycardia (AT) originating focally from diverse anatomical structures in both atria has been well described. 1 8 Focal AT is distinguished from macroreentrant AT by its electrophysiological characteristics and electropharmacological responses, and by the approaches to mapping and ablation of the tachycardia. 7,8 The underlying mechanism of focal AT is thought to be automaticity, triggered activity or atrial microreentry. 7,8 Recently, Tsai et al reported an unusual form of focal atrial fibrillation (AF) triggered by ectopic beats originating from the superior vena cava (SVC), and radiofrequency (RF) ablation of the triggering SVC focus was safe and highly effective in eliminating the focal AF. 9 Theoretically, it is possible that focal electrical firing in the SVC could initiate AT in addition to AF; however, the mechanism of this type of focal AT remains unclear. In this study, we describe the distinct electrocardiograms and electrophysiological characteristics in 3 patients with focal AT originating from various parts of the SVC. The location of the successful RF site in the SVC was proven by multi-plane SVC angiography. All 3 patients underwent uneventful ablations of their SVC foci within a few seconds of RF current application. (Received May 28, 2001; revised manuscript received August 17, 2001; accepted August 23, 2001) Division of Cardiology, Department of Medicine, China Medical College Hospital, Taiwan, ROC Mailing address: Kuan-Cheng Chang, MD, Division of Cardiology, Department of Medicine, China Medical College Hospital, 2 Yuh-Der Road, Taichung, Taiwan, ROC. kcc45@ms36.hinet.net Methods Patients The study group were 3 patients with drug-refractory atrial tachyarrhythmias who were admitted for electrophysiological study and RF ablation therapy. Each patient had a focal AT originating from the SVC that had been diagnosed and confirmed by the electrophysiological study, SVC angiography and RF ablation. Two (cases 1 and 2) of the 3 patients did not have significant organic heart diseases detectable by physical examination, chest roentgenograms, echocardiography, and coronary angiography. The other patient (case 3) had one-vessel coronary artery disease. The definition of focal AT was based on previously established criteria. 7,10 12 Electrophysiological Study and RF Ablation The electrophysiological study was performed in a postabsorptive state after each patient gave written informed consent. All antiarrhythmic drugs except amiodarone were discontinued for at least 5 half-lives before the study. Two 6F quadripolar electrode catheters with a 5-mm interelectrode spacing were positioned at the high right atrium and the right ventricular apex, respectively, for pacing and recording. Another 6F quadripolar electrode catheter with a 10-mm interelectrode spacing was positioned across the tricuspid annulus to record the His bundle potential. A 6F decapolar electrode catheter with a mm interelectrode spacing (Daig Corp) was positioned in the coronary sinus for recording and pacing. A 7F deflectable duodecapolar halo catheter (Cordis-Webster Co) was positioned along the tricuspid annulus for recording. A 7F

2 Focal Atrial Tachycardia From SVC 1035 Fig 1. Comparison of the P wave morphology between atrial tachycardia (AT) and sinus rhythm (SR). In all 3 cases, the P wave polarity and configuration of the AT beats in the 12-lead ECG were very similar to that of SR. More prominent positive deflection of the P wave over leads II, III and avf was observed in AT as opposed to that of SR in case 3. Fig 2. Mapping of the focal AT in case 1. Intracardiac recordings and simultaneous surface ECG (leads I, avf and V1) were displayed during AT. A low-amplitude spiky potential (arrowhead), followed by an atrial electrogram, was recorded by the distal bipoles of the ablation catheter (ABL DS), which was positioned at the SVC. Note the marked variation of the spike s cycle length and spontaneous Wenckebach block between the spiky potentials and the atrial electrograms. HRa, high right atrium; HIS, His bundle; CS, coronary sinus; DS, distal; MD, middle; PX, proximal; HAL 1-10, the first to the tenth bipoles of the duodecapolar halo catheter. The most distal pair of electrodes (HAL 1) was placed in the lower lateral right atrium and the most proximal pair (HAL 10) in the interatrial septum. Arrows indicate the direction of wavefront propagation. deflectable catheter with a 4-mm tip and a mm interelectrode spacing (Mansfield Division, Boston Scientific Corp or Cordis-Webster Co) was used as a roving catheter for pacing, mapping and ablation. Surface ECG leads I, avf and V1, as well as intracardiac electrograms, were simultaneously displayed on a multi-channel oscilloscopic recorder (Prucka Engineering Inc) and stored on a magnetic optical disc. The pacing stimuli were provided by a digital programmable stimulator (Bloom and Associates, DTU-215) at twice the diastolic threshold and 2 ms in duration. In all 3 patients, atrial stimulation was not performed for specifying the mechanism of the tachycardia during AT because of the risk of inducing AF and difficulty in interpreting the stimulation results. RF ablation was performed by a generator (Osypka 300 Smart) that delivered a continuous unmodulated current at 500 khz with temperature feedback. The target temperature during each RF current application was set at 60 C and the duration was 30 s per pulse. If the patient developed chest pain, cough or severe bradycardia during ablation, the RF current application was terminated immediately and the target temperature was decreased to C. At the presumed successful ablation site, further RF pulses lasting for a total of s were needed to ensure complete elimination of the arrhythmic focus. Post-ablation clinical follow-up at intervals of 1 week, 2 weeks, 4 weeks and then 3 months consisted of history taking, physical examination, a 12-lead ECG at each visit and a 24-h Holter ECG at 2 weeks. When there were symptoms suggestive of arrhythmia, another Holter recording or a follow-up electrophysiological study was performed. Telephone interviews were carried out for those patients who were lost to regular clinical follow-up. Results Patient 1 This patient, a 57-year old female, had experienced episodic palpitations for 10 years. A 12-lead ECG during palpitation showed an irregular narrow QRS complex tachycardia with an average ventricular rate of 166 beats/min, consistent with AF. A direct current shock of 50 J was required to restore sinus rhythm because of the drugrefractory tachycardia. The resting ECG displayed left axis deviation and non-specific ST T changes. During the baseline electrophysiological study, spontaneous initiation of sustained AF occurred before starting the programmed electrical stimulation. An attempt to restore sinus rhythm by intravenous administration of 300 mg of amiodarone

3 1036 CHANG K-C et al. Fig 3. Ablation of focal AT in case 1. Delivery of radiofrequency (RF) current at the SVC focus, which had the spiky potential, terminated the AT in 4.6 s. Asterisk, sinus beat; S, seconds; W, watts. Abbreviations as in Fig 2. Fig 4. Mapping of focal AT in case 2. Intracardiac recordings and simultaneous surface ECG (leads I, avf and V1) were shown during initiation of AT. At the anterior aspect of the SVC, the ablation catheter recorded a high-amplitude spiky potential. During sinus beat (asterisk), the spiky potential (arrowhead) was activated after an atrial electrogram, whereas it preceded the atrial electrogram at initiation of AT. Slight variation of the spike s cycle length was present in this recording. Abbreviations as in Fig 2. was ineffective, so an external direct current shock of 200 J was applied and the AF was terminated. However, after 2 sinus beats, a persistent AT was initiated by an ectopic beat originating from the high right atrial region. The cycle length of the tachycardia varied from approximately 190 to 300 ms and during tachycardia, the P wave deflection was upright over leads II, III, avf, downward over lead avr, and biphasic over lead V1, a P wave morphology that was very similar to that of the sinus beat (Fig1). At the anterior aspect of the SVC approximately 15 mm above the atriocaval junction, as evaluated from a left lateral SVC angiogram, the roving catheter recorded a low-amplitude ( mv) spiky potential, which was followed by an atrial electrogram during AT. Spontaneous Wenckebachtype exit block was observed between the spiky potential and the atrial electrogram (Fig2). The low-amplitude spiky potential preceded the onset of the surface P wave by 40 ms, which was measured from a beat with the shortest conduction time from the spike to the atrial electrogram. Delivery of RF current at the SVC focus with the spiky potential terminated the AT in 4.6 s (Fig 3). After ablation, neither AT nor AF occurred spontaneously. During isoproterenol infusion, programmed stimulation could not induce either atrial tachyarrhythmias and the patient has been free of symptomatic palpitations and syncope for 6 months under no antiarrhythmic medications. Patient 2 Patient 2, a 66-year old woman, had suffered from paroxysmal palpitation for 2 years. A 12-lead ECG during palpitation displayed atrial flutter (AFL) with 2:1 atrioventricular (AV) conduction and a ventricular rate of 150 beats/min. The resting ECG was normal. During the baseline electrophysiological study, rapid atrial pacing from the high right atrium at a cycle length of 250 ms induced the clinical AFL (flutter cycle length=250ms) with 2:1 AV conduction. Pacing from the coronary sinus ostium at cycle lengths shorter than the flutter cycle length revealed

4 Focal Atrial Tachycardia From SVC 1037 Fig 5. Continuous recordings from case 2. At a more upward SVC location, recordings from the distal bipoles of the ablation catheter showed alternate occurrence of a low-amplitude and a high-amplitude spiky potential. Note the lowamplitude spiky potential, which was very similar in morphology to the spiky potential recorded in case 1, but displayed exit block to the atrial activation. The activation of the high-amplitude spiky potential preceded the onset of the surface P wave by 75 ms. Delivery of RF current at this SVC focus terminated the tachycardia immediately. Double asterisk, sinus beat after ablation of AT. Abbreviations as in Fig 2. Fig 6. Recordings from case 3 showing mapping during AT. The ablation catheter, located at the SVC, recorded a high-amplitude spiky potential (arrowhead). This potential, followed by a low-amplitude atrial electrogram, preceded the onset of the surface P wave by 61 ms during AT. The AT suddenly terminated during mapping, when the ablation catheter was positioned at the SVC focus with the spiky potential. During sinus rhythm (asterisk), the spiky potential was activated after the low-amplitude atrial electrogram, which was in contrast to the recordings that occurred during AT. Abbreviations as in Fig2. concealed entrainment. The post-pacing interval measured from the coronary sinus ostium was approximately that of the flutter cycle length. Linear RF ablation targeting the septal isthmus terminated the AFL. However, spontaneous repetitive bursts of AT with a varying tachycardia cycle lengths ranging between 180 and 320 ms ensued. During tachycardia, the P wave was upright over leads II, III, avf, downward over lead avr, and biphasic over lead V1, a P wave morphology resembling that of the sinus beat (Fig 1). Intracardiac recordings disclosed local atrial activation at the high right atrium that was earlier than that of all the other recording sites. Detail mapping of the high right atrial region was accomplished by using the roving catheter and at the anterior aspect of the SVC, it recorded a high-amplitude ( mv) spiky potential, which appeared immediately after an atrial electrogram during sinus beats. However, during bursts of AT, the activation of the spiky potential preceded that of the atrial electrogram (Fig 4). At a more upward location of the anterior SVC, approximately 25 mm above the atriocaval junction, as evaluated from a left lateral SVC angiogram, the roving catheter recorded the alternate occurrence of a low-amplitude ( mv) and a high-amplitude ( mv) spiky potentials. The low-amplitude spiky potential, which was very similar in morphology to the spiky potential recorded in case 1, displayed exit block to the atrial activation (Fig5). The activation of the high-amplitude spiky potential was earlier than that of all the other atrial electrograms and preceded the onset of the surface P wave by 75ms. RF ablation at this SVC focus with both high and low-amplitude spiky potentials terminated the tachycardia in 1 s, and after ablation, there were no spontaneous bursts of AT with or without isoproterenol provocation. The patient has been free of symptomatic tachycardia for 6 months without antiarrhythmic medications.

5 1038 CHANG K-C et al. Fig 7. SVC angiograms showing the successful RF ablation sites (arrowhead) in the SVC. RAO, right anterior oblique projection; LAO, left anterior oblique projection. See text for full discussion. Patient 3 The third patient, a 50-year old male, presented with recurrent palpitations for 6 months after a previous ablation procedure at another institution 18 months ago. The 12- lead ECGs during palpitation episodes recorded 2 types of atrial tachyarrhythmia: one displaying an irregular narrow QRS complex tachycardia with an average ventricular rate of 190 beats/min consistent with AF, the other showing a repetitive form of non-sustained AT. The resting ECG was normal. During isoproterenol infusion at a rate of 1 g/min, rapid atrial pacing from the high right atrium at a cycle length of 170 ms reproducibly induced AF. The AF subsequently changed to AFL (flutter cycle length=200ms) with 2:1 AV conduction. Pacing from the coronary sinus ostium at cycle lengths of 180 and 170ms demonstrated concealed entrainment. The post-pacing interval measured from the coronary sinus ostium was approximately that of the flutter cycle length. However, this subeustachian isthmus-dependent typical AFL subsequently degenerated into AF and then terminated spontaneously. Following termination, an incessant, repetitive form of AT with almost beat-to-beat variations in tachycardia cycle length (ranging between 200 and 300 ms) occurred spontaneously. During tachycardia, the P wave was upright over leads II, III and avf, downward over lead avr, and biphasic over lead V1, a P wave morphology that very similar to that of the sinus beats (Fig1). We decided to ablate the AT first. As the local atrial activation at the high right atrium was earlier than that of all the other recording sites, we used the roving catheter to perform detail mapping of the high right atrial region. At the posterior aspect of the SVC, approximately 17 mm above the atriocaval junction, as evaluated from a left lateral SVC angiogram, the roving catheter recorded a high-amplitude ( mv) spiky potential during AT. This potential, followed by a relatively low-amplitude atrial electrogram, preceded the onset of the surface P wave by 61 ms during tachycardia. The AT terminated suddenly during mapping, when the roving catheter was positioned at the SVC focus with the spiky potential. During sinus rhythm, the spiky potential was activated after the lowamplitude atrial electrogram, which was in contrast to the recordings that occurred during AT. Applying RF current to this SVC focus resulted in complete elimination of the AT. Following ablation of the AT, linear ablation of the subeustachian isthmus to eradicate the typical AFL was then performed during sinus rhythm. Post-ablation rapid atrial pacing at the same cycle length of 170ms could not induce AF with or without isoproterenol provocation. The patient has been free of symptomatic palpitation for 3 months without antiarrhythmic medications. Discussion The principle finding of the current study is that a distinct type of focal AT can originate from various parts of the SVC. The clinical and electrophysiological characteristics of this focal AT include (1) P wave morphology during AT that is very similar to that of the sinus beat; (2) tachycardia that is initiated spontaneously; (3) an association with AFL or AF; (4) a presumed SVC potential with highly variable cycle lengths that can be recorded during AT; (5) the SVC potential possibly displaying 1:1 conduction or second-degree exit block to the atrial electrograms during AT; and (6) safe and highly effective elimination of the focal AT by application of RF current to the SVC focus with the SVC potential. Markowitz et al 7 defined focal AT by its electrophysiological characteristics of a centrifugal atrial activation pattern, almost dissociation of both atria from tachycardia with atrial extrastimuli, an identifiable early local atrial activation relative to the surface P wave, and inability to demonstrate manifest entrainment. 7 In the present study, delivery of atrial extrastimuli and using overdrive pacing to specify the tachycardia mechanism was not performed in the 3 patients because of the risk of inducing AF and difficulty in interpreting the stimulation results because of marked beat to beat variations in the tachycardia cycle length. However, the AT in each patient all displayed an eccentric atrial activation sequence and had a presumed SVC potential preceding the onset of the surface P wave during tachycardia. RF ablation targeting the SVC focus with the SVC potential promptly abolished the AT. These findings suggest a focal origin of the AT. Focal AT has been reported to arise from diverse anatomical structures in the right or the left atrium. 1 8 More recently, Ino et al described a case with a 2:1 exit block of repetitive focal activity in a SVC focus manifesting as a high right AT. 13 Furthermore, we found that AT can occur focally from different parts of the SVC, which was confirmed by the SVC angiogram showing the location of

6 Focal Atrial Tachycardia From SVC the successful ablation sites at various distances above the atriocaval junction and different positions within the SVC (Fig 7). These SVC foci were all located within 30 mm above from the atriocaval junction, which is in accordance with the histological and electrical connections of the musculature between the SVC and the right atrium Although it remains unclear why the atrial muscle extending into the SVC becomes arrhythmogenic, Ito et al have shown phase 4 depolarization-initiated automatic activity 14 and Yanaga demonstrated aconitine-induced abnormal automaticity and fibrillation in the musculature of the SVC. 18 It is likely that if the musculature in SVC contains pacemaker activity derived from its early embryological development, then under certain circumstances, it could become arrhythmogenic. We propose abnormal automaticity or triggered activity as the possible underlying mechanism of the tachycardia in the 3 patients because the tachycardia occurred spontaneously rather than being induced by programmed electrical stimulation. The difference in the P wave configuration in the 12-lead ECG during AT, when compared with one during sinus rhythm, is an important criterion to establish the diagnosis of AT. The P wave polarity during AT in the 3 patients was identical to that of the sinus beats, the only difference being a slightly larger positive deflection of the P wave over the inferior leads during AT in case 3. This indicates that the spread of intraatrial activation might be similar between the SVC-initiated focal AT and the sinus rhythm. Furthermore, the P wave configurations observed in the current study fulfilled the classification of cristal AT as suggested by Tada et al, 19 therefore, one should be aware that focal AT originating from the SVC could be erroneously classified as cristal AT if simply based on the ECG criteria for defining the site of origin of AT. Recently, Tsai et al demonstrated and characterized a spiky SVC potential in focal AF, which appeared as a large spiky deflection and preceded a low-amplitude far-field atrial electrogram during SVC ectopy or SVC ectopy-initiated AF. 9 In contrast to the recordings during AF, the SVC potential was activated after the far-field atrial electrogram during sinus rhythm. Ablation of the SVC ectopy was highly effective in terminating and/or preventing re-induction of AF. However, in the current study, we recorded a low-amplitude spiky SVC potential preceding an atrial electrogram during AT in case 1. The spiky potential exhibited an intermittent second-degree Wenckebach block to the atrial electrogram and there was an inverse relationship between the conduction time from the spiky potential to the atrial electrogram and the spike s cycle length. Both the spiky potential and the AT were eliminated within a few seconds of RF current application. These findings indicate that the AT was initiated and maintained by the focal SVC firing, which had an AV node-like decremental conduction property to the atria. From cases 2 and 3, we recorded a high-amplitude spiky SVC potential preceding an atrial electrogram during AT, as opposed to the recording that occurred during sinus beat. The reversed activation sequence between the atrial electrogram and the spike during AT was similar to that described by Tsai et al 9 and Haissaguerre et al 20 in focal AF. In case 3, the high-amplitude spike and the atrial electrogram exhibited a 1:1 relationship, whereas, at the successful ablation site of case 2, the recordings showed an alternate occurrence of a highamplitude and a low-amplitude spiky potential. The lowamplitude spiky potential, which was very similar in 1039 morphology to the spiky potential recorded in case 1, displayed exit block to the atrial activation. Application of RF current at the SVC foci with the spiky potentials promptly eliminated both the spikes and the AT in cases 2 and 3, indicating that the SVC foci also played a critical role in generating and maintaining the tachycardia in these patients. Finally, an association with other paroxysmal atrial tachyarrhythmias was present in all 3 patients; that is, AF in cases 1 and 3, and typical AFL in cases 2 and 3. The coexistent AFL was ablated, but the AF was not. However, during the follow-up period, we found that none of these atrial tachyarrhythmias recurred. Conclusions Focal electrical firing in the SVC can initiate AT in addition to AF. This type of focal AT is always associated with paroxysmal AFL or AF. RF ablation guided by the presumed SVC potential is safe and highly effective in eliminating the tachycardia. References 1. Narula O: Sinus node reentry: A mechanism for supraventricular tachycardia. Circulation 1974; 50: Wu D, Amat-Y-Leon F, Denes P, Dhingra RC, Pietras RJ, Rosen KM: Demonstration of sustained sinus and atrial reentry as a mechanism of paroxysmal supraventricular tachycardia. Circulation 1975; 51: Allessie MA, Bonke FIM: Direct demonstration of sinus node reentry in the rabbit heart. Circ Res 1979; 44: Gomes JA, Hariman RJ, Kang PS, Chowdry IH: Sustained symptomatic sinus node reentrant tachycardia: Incidence, clinical significance, electrophysiologic observations and the effects of antiarrhythmic agents. J Am Coll Cardiol 1985; 5: Iesaka Y, Takahashi A, Goya M, Soejima Y, Okamoto Y, Fujiwara H, et al: Adenosine-sensitive atrial reentrant tachycardia originating from the atrioventricular nodal area. J Cardiovasc Electrophysiol 1997; 8: Kalman JM, Olgin JE, Karch MR, Hamdan M, Lee RJ, Lesh M: Cristal tachycardias : Origin of right atrial tachycardia from the crista terminalis identified by intracardiac echocardiography. J Am Coll Cardiol 1998; 31: Markowitz SM, Stein KM, Mittal S, Slotwiner DJ, Lerman BB: Differential effects of adenosine on focal and macroreentrant atrial tachycardia. J Cardiovasc Electrophysiol 1999; 10: Chen SA, Chiang CE, Yang CJ, Cheng CC, Wu TJ, Wang SP, et al: Sustained atrial tachycardia in adult patients: Electrophysiological characteristics, pharmacological response, possible mechanism, and effects of radiofrequency ablation. Circulation 1994; 90: Tsai CF, Tai CT, Hsieh MH, Lin WS, Yu WC, Ueng KC, et al: Initiation of atrial fibrillation by ectopic beats originating from the superior vena cava: Electrophysiological characteristics and results of radiofrequency ablation. Circulation 2000; 102: Gillette PC, Crawford FC, Zeiger VL: Mechanism of atrial tachycardia. In: Zipes DP, Jalife J, editors. Cardiac electrophysiology: From cell to bedside. Philadelphia: Saunders Publishing Co, 1999: Scheinman MM, Basu D, Hollenberg M: Electrophysiologic studies in patients with persistent atrial tachycardia. Circulation 1974; 50: Wen MS, Yeh SJ, Wang CC, Lin FC, Wu D: Radiofrequency ablation therapy in three patients with paroxysmal atrial tachycardia. Pacing Clin Electrophysiol 1993; 16: Ino T, Miyamoto S, Ohno T, Tadera T: Exit block of focal repetitive activity in the superior vena vaca masquerading as a high right atrial tachycardia. J Cardiovasc Electrophysiol 2000; 11: Ito M, Arita M, Saeki K, Tanoue M, Fukushima I: Function properties of sinocaval conduction. Jpn J Physiol 1967; 17: Spach MS, Barr RC, Jewett PH: Spread of excitation from the atrium into thoracic veins in human beings and dogs. Am J Cardiol 1972; 30: Zipes DP, Knope RF: Electrical properties of the thoracic veins. Am J Cardiol 1972; 29:

7 1040 CHANG K-C et al. 17. Hashizume H, Ushiki T, Abe K: A histological study of the cardiac muscle of the human superior and inferior vena cavae. Arch Histol Cytol 1995; 58: Yanaga T: Studies on sinocaval conduction. Jpn Circ J 1966; 30: Tada H, Nogami A, Naito S, Suguta M, Nakatsugawa M, Horie Y, et al: Simple electrocardiographic criteria for identifying the site of origin of focal right atrial tachycardia. Pacing Clin Electrophysiol 1998; 21(Pt II): Haissaguerre M, Jais P, Shah DC, Takahashi A, Hocini M, Quiniou G, et al: Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998; 339: