Catheter Ablation Approach on the Right SideOnly for Paroxysmal Atrial Fihrillation Therapy:Long-Term Results

ANDREA NATALE,* FABIO LEONELLI,* SALWA BEHEIRY,* KEITH NEWBY.tENNIO PISANO,* DOMENICO POTENZA,* KATHLEEN RAJKOVICH,*BRANDON WIDES,* LISA CROMWELL,* and CERY TOMASSONI+

From the *University of Kentucky, Lexington, Kentuc:ky, tDuke University Medi[:a] Center,Durham, North Carolina, and tCasa Sollievo della Sofferenza San Ciovanni Rotondo, Italy

NATALE, A., ET Al..: Catheter Ablation Approach on the Right Side Only for Paroxysmal Atrial Fibrilla-tion Therapy: Long-Term Results. We report the long-term follow-up of a right side only catheter ablationapproach for paroxysmal AF. Eighteen patients with AE refractory to drugs entered the study. Ablationwas attempted in the right atrium only by creating linear lesions based on a specific design including fromtwo to four linear lesions. Induction of AF was attempted before ablation and after placement of the le-sions. A septal lesion was performed in nine patients. In ten patients atrial defibrillation thresholds(ADETs) before ablation and following creation of the linear lesions were compared. After a mean follow-up of 22 ±11 months, seven patients had recurrence of AE, and another nine patients experienced atria!flutter or atrial tachycardia. Eive patients remained in sinus rhythm without medications and four re-quired the use of drugs. Three patients had sporadic AF and six were in chronic AE. The recurrence ratewas similar in patients with and without the septal lesion. However, a cure with right side ablation ap-peared to be predicted by the presence of disorganized and earlier activity in the high right atrium andcrista terminalis. Linear lesions in the right atrium were associated with a lower ADET (pre 2.6 ± 04 J vspost 1.7 ± 0.6 J). In conclusion, in a small number of patients, control of AE can be obtained with a rightside only approach. Gertain activation patterns may identify patients suitable to this approach. No spe-cific lesion pattern appeared more effective. Right atrial linear lesions resulted in lower ADET. (PAGE2000; 23:224-233}

atrial fibrillation, oblation, atrial defibrillation

Introduction

In recent years, atrial fibrillation (AF) has be-come a subject of increased interest and investi-gation. This has resulted in great advances in ourunderstanding of some of the mechanisms respon-sible for this arrhythmia. While catheter ablationfor several supraventricular tachycardias has be-come an established curative approach, clinicalinvestigation of catheter ahlation for AF is in its

Preliminary results of this study were preseiiteti al th<: Ameri-can Heart Association Meeting, November 1996.

Address for reprints: Andrea Natale, M.D., Clevoland ClinicFoundation, Cardiology Desk Fl.'i. 9500 Euclid Ave., Cleve-land, OH 44195. Fax: (216) 444-4428.

Received February 22, 1999; revised April 29, 1999; acceptedMav28. 1999,

infancy. Although, more recently, a subtype of pa-tients with AF has been reported that may beamenable to the curative ablation procedure hyplacing focal lesions,^'^ the initial effort was di-rected toward replication of the surgical maze asdesigned by Cox et al.^ However, placement of lin-ear lesions in the right and the left atrium has sev-eral limitations, including the potential for a longprocedural time and its associated fluoroscopy ex-posure and a significant risk for thromboembolicevents. A more limited approach was reported byHaissaguerre et al.'* who created linear lesions inthe right atrium using a multipolar catheter placedin two strategic locations. Similarly to their ap-proach and based on some preliminary animalwork/'*^ we assessed the efficacy of a right sideonly ablation procedure for patients with frequentand symptomatic episodes of AF, refractory to an-

224 February 2000 PACE, Vol. 23

RIGHT SIDE ABLATION FOR ATRIAL FIBRILLATION

tiarrhythmic drugs. This article includes the long-term success rate and the clinical implication andobservation learned by performing this procedureusing the technical tools presently available.

Methods

Study Population

Eighteen patients highly symptomatic for AFrefractory to antiarrhythmic drugs, includingamiodarone were included in the study. Patientsvi'ere considered for this protocol if (1] they hadmore than two symptomatic episodes per monthof paroxysmal AF, (2) they had failed at least twoantiarrhythmic drugs, including amiodarone, and(3) they had normal left atrial size and an ejectionfraction of the left ventricle > 45%. Patients wereexcluded from this protocol if (1) they had evi-dence of other atrial arrhythmias (atrial tachycar-dia, atrial flutter, atrioventricular (AV) nodal reen-try, and AV reentry), and (2) they had a leftventricular ejection fraction < 45% and/or leftatrial enlargement.

Electrophysiological Study

Informed, written consent was obtained fromall patients hefore entering the study. Before theprocedure, each patient was monitored for severalmonths with a loop recorder and Holter monitor todocument the initiation of the episodes of AF andensure the correlation with symptoms and arrhyth-mias. All patients underwent standard electro-physiological study, including catheter placementin the high right atrium, right ventricle, and coro-nary sinus, and His-bundle recording. Programmedelectrical stimulation was performed hefore and af-ter the ahlation from the high right atrium and thedistal coronary sinus electrode pair. The pacingprotocol included extra stimulus testing using upto three extra stimuli, and burst pacing down to 200ms or loss of capture. If atrium fibrillation couldnot he induced before the ablation in a drug-freestate, isoproterenol infusion was given starting at 2mcg/min. In the first live patients, ablation wasperformed during AF. In the remaining patients, le-sions were placed during sinus rhythm. Once re-producible induction of AF was established, pa-tients were cardioverted back to sinus rhythm andthe ablation procedure was then initiated.

Ablation ProcedureRadiofrequency lesions were created using a

6- or 8-mm electrode deflectable catheter advancedin the right atrium using a long, rigid 8 Fr Daigsheath (SRO) [Daig Corp., Minnetonka, MN, USA)to gain better catheter control during the proce-dure. In two patients, the ablation procedure wasperformed using a 4-mm tip deflectable catheter inconjunction with a nonfluoroscopic mapping sys-tem (Carto, Biosense, Inc.). In these two patients,the lesion pattern was repeated using an 8-mm tipelectrode deflectable catheter. The lesion patternin the first seven patients (Fig. lA) consisted of twointercaval lesions from the superior vena cava tothe inferior vena cava in front and behind the cristaterminalis. A third lesion was performed acrossthe right atrial isthmus from the tricuspid annulusto the hiferior vena cava, and a fourth lesion wason the anterior wall of the right atrium from the ap-pendage to the anterior aspect of the tricuspid an-nulus. In the remaining 11 patients (Fig. IB), theanterior wall lesion was eliminated and replacedby a septal lesion and only one intercaval line wasperformed. Tn these 11 patients, after the intercavaland isthmus lesions were performed, programmedelectrical stimulation was repeated and if AF be-came noninducible, the septal lesion was not cre-ated. In each patient, the first lesion performed wasthe isthmus line. Radiofrequency energy was de-livered between the catheter electrode and a cuta-neous patch electrode placed over the left scapula.The catheter was dragged from one point to theother of the imaginary line. Ablation over the imag-inary line was repeated several times until at least60%-70% reduction of the electrogram amplitudewas observed. Ablation was initiated with a powersetting of about 25-30 W, which u'as progressivelyincreased up to a setting of 50 W.

Internal Defibrillation Protocol

In ten patients, atrial defibrillation threshold(ADFT) was determined before ablation and aftertbe lesion pattern was completed. Energy was de-livered using custom-made catheters designed forrecording, pacing, and defibrillation. ADFT was ob-tained using a step-up protocol. The starting energylevel was 0.2 J, and approximately 0.2 J incrementswere tested until AF was terminated. Shocks weredelivered after 30 seconds of stable AF. The record-

PACE, Vol. 23 Fehruary 2000 225

NATALE.ETAL.

RAOAon a

Tricuspidvalve

Figure 1. Schematic diagram of the lesion patterns performed in the right atrium that is shownopened in tho right anterior oblique (RAO) view. In seven patients (panel A) four lesions werecreated including: (a) behind the crista terminalis, (b) in front of the crista termlnalis, (c) in theanterior wall from the appendage to the anterior aspect of the tricuspid annulus. and (dj in thetricuspid valve/inferior vena cava isthmus. In the remaining 11 patients (pane! B), after creatingthe isthmus (c) and intercaval (b) lines, a third lesion along the septum (a) was performed if atrialfibrillation remained inducible.

B

ing and defibrillation system consisted of two sep-arate nonapolar electrode catheters placed in theright atrial appendage and coronary sinus, or a sin-gle catheter with eight distal electrodes advancedin the coronary sinus, or eight additional electrodesfloating in the high right atrium, separated by a 7-or 9-cm gap. Energy was delivered using an exter-nal defibrillator with a capacitance of 100 (xF.Biphasic shocks with 50% fixed tilt were used.

Postablation Management

After the ablation, patients were observed for24 honrs and subsequently discharged home offtheir antiarrhythmic. Anticoagulant therapy wasreinitiated in every patient 24 hours after the pro-cedure. Before discharge from the hospital, atransthoracic echocardiography was performed toexclude pericardial effusion. Patients were seen inthe outpatient clinic for follow-up 1, 3,6,8,12,16,20, and 24 months after the procedure. Based onthe patient's symptoms, either an event recorderor Holter monitoring recordings were obtained. Incase of arrhythmia recurrence, drug therapy was

initiated using the same antiarrhythmic that wasineffective before the procedure. All patients wereinstructed to assess their peripheral pulse andrecord frequency and regularity of their rhythm. Inasymptomatic patients without any evidence ofAF, an event recorder with two random transmis-sions every 24 hours for 4 weeks was ohtained 2,6, and 12 months after the procedure. In patientsexperiencing recurrence of AF, repeat electro-physiological study was proposed. In addition, anelectrophysiological study was performed inthose patients considered candidates for AV nodalablation at follow-up. All variables presented inthe study are expressed as mean value ± SD.

Results

Patient Population

Eighteen patients [16 men, 2 women) withparoxysmal self-terminating AF entered the study.The mean age was 61 ± 7 years. Sixteen patientshad no evidence or history of cardiovascular dis-ease. Two patients had history of coronary arterydisease with a structurally normal heart at the

226 February 2000 PACE, Vol. 23

RIGHT SIDE ABLATION FOR ATRIAL FIBRILLATION

time of ablation. The modality of AF induction be-fore ablation was as follows: incremental pacingin 8 patients, extra stimulus testing in 6 patients,hurst pacing in 3 patients, and 1 patient was non-inducible despite programmed stimulation per-formed in a drug-free state and during isopro-terenol infusion. In four patients, isoproterenolinfusion was required for induction of AF and intwo of them, AF became incessant.

Ablation Results

The procedure was performed during AF inthe first five patients. Subsequently, in view of therecurrence of severe sinus node dysfunction intwo of these patients, ablation was performed dur-ing sinus rhythm unless AF became difficult toterminate before ahlation as observed in two pa-tients after administration of isoproterenol. In thefirst seven patients, the lesion pattern shown inFigure lA was concluded before reattempting AFinduction. In four of these seven patients, place-ment of the intercaval line posterior to the cristaterminalis resulted in hemidiaphragm paralysis.This appeared reversible in two of these patients.In the remaining 11 patients, AF induction was re-assessed after performing the intercaval and isth-mus lesions. In case of persistence of inducibleAF, the septal lesion was performed in this othergroup of patients (Fig. IB). In two of these pa-tients, AF hecame noninducible after the first twolesions. Therefore, the septal lesion was not per-formed. In one of these two patients, AF hecameincessant on isoproterenol infusion and was ter-minated after completing the intercaval line.

In the remaining nine patients, the third septallesion was concluded. Following the ablation pro-cedure, only 2 of 18 patients had inducible AF withright-side stimulation. An example is shown inFigure 2. However, using left side atrial stimulationfVom the coronary sinus catheter, AF was initiatedin 12 of the 18 patients. Three patients had sinusnode dysfunction at the end of the procedure. In allthree patients, ablation was performed in AF. Intwo patients, sinus node dysfunction was persis-tent and required pacemaker implantation,whereas in the third patient, recovery of the sinusnode function was observed 2 days after the abla-tion. The mean fluoroscopy time was 155 ± 27minutes. Tbe mean procedural time was 7.9 ± 3.3

Figure 2. Example showing induction of atrialfibrillation during programmed stimulation with a singleextra stimulus before ablation (panel A): followingablation programmed stimulation with three extrastimuli was unable to induce any arrhythmia (panel B).Ill = electrocardiographic recording of lead III; RA =recording from tbe high right atrium; HBE = His-bandlerecording: CS = five coronary sinus recordings from theproximal (top) to the distal (bottom).

hours. After a mean follow-up of 22 ± 11 months,13 (72%) of the 18 patients experienced recurrenceof atrial arrhythmias. Eleven patients experiencedarrhythmia recurrence within days or no more than2 months after the procedure, whereas the two re-maining patients had recurrence of AF at 15 and 21months after the procedure. Atrial arrhythmia re-currence was observed in a similar number of pa-tients regardless of the presence or absence of theseptal linear lesion. In fact, 5 (71%) of 7 patientsundergoing an ablation protocol not including theseptal lesion experienced atrial arrhythmias at fol-low-up. Of the two patients receiving only the in-tercaval and the isthmus lines, one patient experi-enced recurrence of AF. Atrial arrhythmias wereseen in seven [77%) of the nine patients undergo-ing ahlation including the septal line. Based on theelectrocardiographic appearance, the arrhythmiasobserved with recurrence were atrial tachycardiain 3 patients, atrial flutter in 6 patients, primary AF

PACE, Vol. 23 February 2000 227

NATALE, ET AL.

in 2 patients, AF associated with severe mitral in-sufficiency secondary to ischemia in 1 patient, andAF associated with an episode of pneumonia in 1patient. The latter patient experienced a secondepisode of AF at about 25 months of follow-up inthe absence of comorbidity. Classification intoatrial tachycardia or flutter was based on the elec-trocardiographic appearance and the cycle lengthof the arrhythmia. Arrhythmias with a cycle length< 300 ms were called atrial flutter, whereas ar-rhythmias with a cycle length > 300 ms off medi-cation were called atrial tachycardia. Of interestwas the fact that AF was paroxysmal before the ah-lation in each patient, whereas all arrhythmias ob-served at the time of recurrence were persistent. Atthe final follow-up, 5 (27%) of 18 patients were insinus rhythm without the need of medication. Fouradditional patients were in sinus rhythm with theuse of medication previously ineffective. Six pa-tients were in chronic atrial flutter/fibrillation andthe remaining three patients had sporadic episodesof atrial flutter/fibrillation while being treated withmedication. Two of these three patients requestedAV nodal ablation and permanent pacing due tothe incomplete suppression achieved with medica-tions. The patients more likely to respond to medi-cation at the time of recurrence were those showingslower atrial tachycardialike rhythms (three ofthree patients) [Fig. 3). This patient had an atrialtachycardia with the rate of 125 beats/min at the

^ f I'L f^—L—L,

Figure 3. Recording from a patient experiencingrecurrent postablation. An atrial tachycardia (125beats/min) was documented. The top tracing is lead IIfrom the electrocardiogram and the bottom tracingshows an intracardiac recording from a catheter placedacross the tricuspid valve. VE - ventricular electrogram.

time of recurrence. This arrhythmia responded tosotalol (80 mg twice daily), which before the abla-tion was ineffective even at a higher dose. Theoverall results of the procedure are summarized inTable I.

Observations During the Ablation Procedureand FoIIow-Up

During ablation, mapping of different regionsof the right atrium showed that the areas with themost organized electrical activity were the coro-nary sinus, the high right atrium, and the cristaterminalis recordings. In 12 patients, the most dis-organized activity with the shortest cycle lengthwas observed in the septal and posterior region ofthe right atrium. However, after placement of oneintercaval line and the isthmus line, more disor-ganized fibrillatory activity, limited to the cristaterminalis region and high right atrium with theremaining of the atrial chambers showing an orga-nized flutterlike activity, was observed in two ofthe patients as shown in Fig. 4.

Patients showing more disorganized and frag-mented electrograms in the septum, and a left-to-right activation in the coronary sinus recordingswere more likely to develop a recurrence (Fig. 5).On the contrary, patients with disorganized activ-ity localized predominantly in the high rightatrium and crista terminalis regions, which wasassociated with a right-to-left activation in thecoronary sinus recording, were more likely to becured by the right side ablation (4 of the 5 patientsremaining in sinus rhythm without medication,Figs. 4 and 6).

At the time of recurrence, the majority of pa-tients showed electrocardiographic evidence of aflutterlike (6 patients) or atrial tachycardialike (3patients) cycle length (9 of 13 patients). In four ofthe six patients showing on the surface electrocar-diogram flutterlike waves, instrumentation andrecording from the coronary sinus documented aleft side fibrillatory activity (Fig. 7). The arrhyth-mias showing organized activity in both right andleft atrium (two patients) appeared to have the ear-liest right side electrograms in the septum, or highposteroseptal region, suggesting a left side rhythm(Fig. 5B). In both patients, atrial flutter was inces-sant and concealed entrainment could not bedemonstrated by pacing along the tricuspid annu-

228 February 2000 PACE, Vol. 23

RIGHT SIDE ABLATION FOR ATRIAL FIBRILLATION

Pt#

1.2.

3.4.5.

6.7.8.

9.

10.11.

12.

13.14.15.16.

17.18.

LesionSet

Fig. 1aFig. 1a

Fig. 1aFig. laFig. l a

Fig, l aFig. l aFig. 1b

Fig. 1b

Fig. 1bFig. 1b

Fig. 1b

Fig. 1bFig. 1bIC + IsthmusFig. 1b

Fig. 1bIC + Isthmus

RhythmDuring

Ablation

AFAF

AFAFAF

SRSRSR

SR

SRSR

SR

SRSRSRSR

SRAF

Table 1.

Abiation and Outcome Data

Complications

SNDATPNPSND/PNP

——

PNP

—TPNP

Trans PRProiong

Trans PRProiong

——

Trans PRProlong

————

—

TSND

Site ofShorter

F-F

CTPS

CTPSPS

PSPSCT

PS

PSPS

PS

PSPSCT

Undeter

PSCT

Arrhythmiaw/ Recurr

—

AF

—ATAF

ATA. Flutter

—

A. Fiutter

ATAF

A. Fiutter/AF

A. Flutter/AF—AF

A. Fiutter/AF

A. Fiutter/AF—

Outcomeat

Fol low-Up

SR + DDD-PPMRare AF w/drug +

DDD-PPMSRSR w/drugAVN Abi + DDD-PPM

Rare AF w/drugSR w/drugChronic A. FiutterSR

Chornic A. Fiutter

SR w/drugAVN Abl + DDD-PPM

Rare AF w/drugChronic AF AVN

Abi + VVi PPMChronic AFSRSR w/drugChronic AF AVN

Abl + VVi PPMChronic AFSR

AF ^ atrial fibrillation; SR = sinus rhythm; IC = intercavai iine; Isthmus - iiner lesion in the tricuspid valve/inferior vena cava isthmus;SND - sinus node dysfunction; TPNP/PNP = transient or persistent phrenic nerve paralysis; Trans PR Prolong = transient PR inten/aiprolongation; CT = crista terminalis; PS = posteroseptai region of the right atrium; AT = atrial tachycardia-like rhythm;A. Flutter = atriai flutter; A. Flutter/AF = atrial flutter on the surface ECG witfi fibrillatory activity in the left atrium;SR w/ drug = persistence of sinus rhythm with antiarrhythmic therapy; Undeter = more than one site with short F-F interval;AVN Abl = AV nodal ablation; DDD-PPM =• dual chamber pacemaker; Rare AF w/ drug = less or equal to three episodes of atrialfibrillation per year while treated with antiarrhythmic medications; Arrhythmia w/ recurr = type of arrhythmia observed at follow-up.

lus/inferior vena cava isthmus. However, atrialflutter secondary to the presence of incompletelines cannot be excluded.

ADFT Data

In ten patients, ADFT was measured beforeablation and following the lesion pattern wascompleted. The mean ADFT hefore the procedurewas 2.8 ± 0.4 I which was reduced after lesionswere performed in the right atrium to 1.6 ± 0.6 ] (P< 0.05). The average impedance before and afterthe ahlation was unchanged (53 ± 5 ohms preab-

lation and 55 ± 6 ohms postablation). However,most of the arrhythmias observed at follow-upwere persistent and associated with early recur-rence following cardioversion (median 8 days,from 2 to 120 days).

Discussion

AF has been described as a disorganized andrandom phenomenon. Later, experimental dataprovided firm evidence of the reentrant nature ofthis arrhythmia estimating that even four to six si-multaneous reentry wavelets were sufficient to

PACE, Vol. 23 February 2000 229

NATALE, ET AL.

Figure 4. Intracardiac recording in a patient showingdisorganized fibrillatory activity limited to the cristaterminalis (recordings from ABLp and ABLd) followingplacement of one intercaval tine and the isthmus line.In this patient placement of a linear lesion anterior tothe crista resulted in termination of atria! fibrillation andlong-term cure. A septal lesion was not performed in thispatient. RA = high right atrium; ABLp-d — recordingsfrom tbe proximal and distal electrode of the ablationcatheter placed in along the crista: HBE ^ proximalrecordings from the His bundle catheter: CS ^ coronarysinus recordings from proximal (top) to distal (bottom).

maintain AF.^ Based on the understanding thatmultiple wavelets need a critical size and mass oftissue to perpetuate. Cox et al.^ devised a surgicalprocedure that involved transmural incisionsstrategically placed to interrupt all potential reen-

try circuits without interfering grossly with thenormal conduction of the sinus impulse. Theproof of the effectiveness of atriotomies in alteringthe arrhythmia substrate have prompted andpaved the way to catheter-hased procedures. Thisapproach has the advantage of not requiring gen-eral anesthesia and thoracotomy to achieve thesame result. The initial pioneering observation bySwartz et al." was encouraging but raised new con-cerns and challenges that have to be resolved tomake this procedure technically practical. It be-came clear that extensive ablation in the leftatrium may be associated with a significant risk ofembolic stroke. In addition, with the technicaltools available at that time, the procedure was ex-tremely lengthy and required a long fluoroscopyexposure with the possibility of future detrimentto the patient and the operator. Recognizing theselimitations and encouraged by the results of pre-liminary experimental studies,'''* we emharked ona right side only ablation approach for treatmentof AF in patients with a paroxysmal nature of thisarrhythmia. In our experience, patients were se-lected after they failed a large number of antiar-rhythmics and were considered as possible candi-dates for AV nodal ahlation and pacemakerimplantation. As a requirement for the study, pa-tients were expected to he severely symptomatic

Ml

HPS

ABL

Figure 5. In panel A intracardiac recordings show disorganized activation in the high and lowseptum (HPS and LS). In addition, the coronary sinus recordings suggest a left-to-right activationsequence. In this patient despite creation of a septal line, a flutterlike arrhythmia was observedat follow-up (panel B). This appeared to have an earliest site in the septum with no evidence ofconcealed antrainment from any site in the right atrium. HPS = recordings from the high postero-septal region; ABL = recording from the ablation catheter placed in the low septum.

230 February 2000 PACE. Vol. 23

RIGHT SIDE ABLATION FOR ATRIAL FIBRILLATION

Figure 6. In this patient recordings during atrialfihrillation shows the most disorganized activity alongthe crista terminalis (RA). In addition the coronary sinusrecording seems to show a proximal to distal activationsequence. In tbis patient lesions based on tbe pattern inFigure la resulted in cure from atrial fibrillation. HA —recordings from a catheter placed in the high portion oftbe crista terminalis.

despite drug therapy. Our ablation approachchanged over time, which provided us with someinformation about the effectiveness of different le-sion patterns. Our study shows that in a smallnumber of patients, right side only ablation is ableto eliminate AF alone or with the use of antiar-rhythmic medications previously ineffective. Oursuccess is overall similar to the other two series'*'̂

Figure 7. Becordings from a patient experiencingarrhythmia recurrence at follow-up. The surface ECCshowed a flutterlike arrhythmia. However, instru-mentation of the coronary sinus showed fibriUatoryactivity originating from tbe left atrium. In this patient.AV nodal ablation was performed during tbe sameprocedure because tbe recurrent arrbytbmia remainedrefractory to drug tberapy. BA = high right atriumrecording; HBE = recording from tbe catbeter close to theHis-bundle position.

previously published, which included a differentlesion pattern than the one used in our protocol.

Based on the results observed in experimen-tal study,^" it is believed that a linear lesion inthe septum may be important in achieving suc-cessful ablation with the right side only ap-proach. In humans, Gaita et al.^ has reported suc-cess by placing a lesion in the septnm. In thisarea, the investigators observed a more disorga-nized and chaotic activity, which they believe isresponsible for the persistence of the arrhythmia.In our series, the presence or absence of a septalline did not change the long-term success rateand control of AF. As observed in one of our pa-tients, linear lesions limited to the intercaval re-gion and the tricuspid-annulus/inferior venacava isthmus may be sufficient in long-term elim-ination of paroxysmal AF. In addition, in few pa-tients, a very disorganized and chaotic fibrilla-tory activity was limited to the crista terminalisand high right atrium. As a matter of fact, in ourseries, patients showing such irregular activationin the high and mid-crista terminalis region werecured by the right side approach. However, pa-tients with irregular and shorter FF intervals inthe septum were more likely to experience recur-rence originating from the left atrium. It is possi-ble tbat the atrial activity recorded in the septumreflects wandering wavelets on both sides of thisstructure.

Animal studies, predominantly in dog mod-els, have supported the concept that linear le-sions in the intercaval region,'• '̂•^^ across the in-teratrial septum at the level of the foramenovalis,^" or in the pulmonary vein region^^ areable to terminate and/or prevent reinduction andmaintenance of AF. Our results seemed to suggesttbat in the majority of patients, right atrial lesionsin the septum or in other locations are not consis-tently effective in preventing recurrence of AF, al-though this was the case in a few patients. As itappeared from the cases experiencing recurrencethat underwent reinstrumentation and mapping,most of these arrhythmias originated in the leftatrium. It is possible that in the majority of pa-tients, ablation of the left atrium should be at-tempted first. This could be consistent with therecent observation of a significant number of focalAF originating from the left pulmonary vein re-gion. ̂ -̂ However, one could argue that based on

PACE, Vol. 23 February 2000 231

NATALE, ET AL.

the response observed in a few of our patients,specific groups have to he identified that couldrespond to specific lesion patterns. In addition, inour series transmurality and continuity of the le-sions were not verified. This alone could have asignificant impact on the outcome and could ex-plain the high incidence of recurrence.

It is clear that we are dealing with a heteroge-nous type of arrhythmia, which needs to be exten-sively evaluated and studied to properly identifygroups that could respond to specific and differentablation approaches.

Of interest, the majority of patients experi-encing recurrence of tachyarrhythmias had orga-nized rhythms defined as atrial flutter of atrialtachycardia. In addition, even those patients whoexperienced AF as documented hy endocardialrecording had a surface electrocardiographic ap-pearance of atrial flutter and showed a right-sidedatrial activation compatible to a flutterlike ar-rhythmia. This suggests that our lesions certainlychanged the electrical substrate of the arrhythmiaexperienced hy the patients. If one considered thatablation or atrial flutter is associated with about a10% recurrence rate despite demonstration of acontinuous bidirectional line of block after theprocedure,^'' it is conceivable to expect that eachlinear lesion may result in an exponentially higherrecurrence rate. This certainly represents a prob-lem that has to be considered and could partiallybe responsible for the results observed in ourstudy. Finally, similarly to what is reported byother studies where linear lesions were used totreat AF, our experience showed an excessive flu-

oroscopy time, reflecting an inadequate catheterdesign to perform such procedures.

Whether linear lesions in the right atriumcould render other device-based therapy more ef-fective for termination of AF is also unclear. Aninteresting observation was the reduction of theenergy requirement for termination of AF afterplacement of linear lesions in the right atrium.This was consistent with a preliminary ohserva-tion in an animal model.̂ "^ This approach could beconsidered in patient candidates for the im-plantable atrial defibrillator with ADFT ahove theoutput of present devices. However, our studyshows tbat some of the arrhythmias observed atthe time of recurrence were nearly incessant orpermanent and tended to reoccur. This could cer-tainly be considered a proarrhythmic effect and alimitation of our intervention.

In conclusion, right atrium only linear lesionsin patients with frequent episodes of paroxysmalAF refractory to antiarrhythmic drugs result in amodification of the underlying arrhythmogenicsuhstrate with cure in a small number of patientsand more persistent and organized arrhythmias inothers. Activation patterns may he helpful to iden-tify subjects amenable to this approach. The pre-sent technology represents a major limitation forthis procedure. Although reduction of the energyrequirements for termination of AF after rightatrial lesions could render feasible the implanta-tion of atrial defibrillators in patients with highADFT, this approach may he curbed by the nearlyincessant nature of the arrhythmias observed atfollow-up.

References

1. Jais P, Haissaguerre M, Shah DC, et al. A focalsource of atrial fihrillation treated by discreteradiofrnquency ablation. Circulation 1997; 95:572-576.

2. Haissaguerre M, |ais P. Shar DC, et al. Spontaneousinitiation of atrial fibrillation hy ectopic beats orig-inating in the pulmonary veins. N Engl J Med 1998;339:659-666.

3. Cox JL, Buineau IP, Scuessler RB, et al. Five yearexperience with the Maze procedure for atrial fih-rillation. Ann Thorac Surg 1993; 56:814-824.

4. Haissaguerre M, lais P, Shah DC. et al. Right andleft atrial radiofrequency catheter therapy ofparoxysmal atrial fibrillation. I Cardiovasc Electro-physiol 1996; 7:1132-1143.

5. Avitall B. Hare I. Mughal K, et al. Ahlation of atrialfibrillation in a dog model, (abstract) | Am CollCardio! 1994; 276A.

6. Avitall B. Hare J. Mughal K, et al. Right-sideddriven atrial fihrillalion in a storilii pericarditisdog model, [abstract) PACE 1994; 17(Part II):774.

7. Allessie MA, Lammers WEJEP, Bonkc FIM, et al.Experimental evaluation of Moe's multiplewavelet hypothesis of atrial fibrillation. In DPZipes. ) lalife (eds.): Cardiac Electrophysiologyand Arrhythmias. Orlando, Grune & Stratton,1985, pp. 265-275.

8. Swartz JF, Pellersels C, Silvers J, et al. A catheter-hased curative approach to atrial fihrillation in hu-mans. Circulation 1994; 90(Part 2):I-335.

232 February 2000 PACE, Vol. 23

RIGHT SIDE ABLATION FOR ATRIAL FIBRILLATION

9. Gaita F, Riccardi R, Calo L, et al. Atrial mapping 12.and radiofreqnency catheter ablation in patientswith idiopathic atrial fihrillation. Circulation1998; 97:2136-2145.

10. Tondo G, Scherlag BJ, Otomo K, et al. Critical atrialsite for ablation of pacing-induced atrial fibrille- 13.tion in tho normal dog heart. J Gardiovasc Electro-physiol 1997:8:1255-1265.

11. Li H, Hare J, Mugbal K. et a l Distribution of atrial 14.electrogram types during atrial fibrillation: Effectof rapid atrial pacing and intercaval junction abla-tion. } Am Coll Cardiol 1996; 27:1713-1721.

Morillo GA. Klein G|, Jones DL, ot al. Chronicrapid atrial pacing: Siructural. functional, andelectrophysiologic characteristics of a new modelof sustained atrial fibrillation. Circulation 1995;91:1588-1595.Schumacher B, Lowalter T. Wolpert G, et al. Ra-diofrequency ablation of atrial flutter. ] GardiovascElectrophysiol 1998; 9:S139-S145.Kalman |, Olgin |, Karch M, et al. Effect of rightatriai linear lesions on atria] defibrillation thresh-old. Implications for "hybrid therapy", (abstract)PAGE 1996; 19 (Part 1I):625.

PACE, Vol. 23 February 2000 233