Focal Atrial Fibrillation Associated with MultipleBreakout Sites at the Crista TerminalisTAKUMI YAMADA,* YOSHIMASA MURAKAMI,* TARO OKADA,*and TOYOAKI MUROHARA†From the *Division of Cardiology, Aichi Prefectural Cardiovascular and Respiratory Center, Ichinomiya, and†Department of Cardiology, Nagoya University Graduate School of Medicine, Nagoya, Japan

YAMADA, T., ET AL.: Focal Atrial Fibrillation Associated with Multiple Breakout Sites at the Crista Ter-minalis. A 34-year-old man with permanent atrial fibrillation (AF) underwent electrophysiologic testing.Spontaneous AF was observed even after successful pulmonary vein (PV) isolation of all four PVs. Intracar-diac electrograms recorded from a basket catheter deployed around the crista terminalis during triggeredatrial premature beats exhibited low-amplitude potentials which were suggested to reflect focal ectopicactivity, preceding high-frequency atrial potentials. The firing from those focal activity sites induced ashift in the breakout sites and conduction block to the right atrium, which suggested the participationof preferential conduction. Radiofrequency catheter ablation targeting the focal origin and preferentialconduction sites eliminated the AF. (PACE 2006; 29:207–210)

atrial fibrillation, crista terminalis, preferential conduction, radiofrequency catheter ablation

IntroductionRadiofrequency (RF) catheter ablation has

proven to be very effective in eliminating right fo-cal atrial tachycardias (ATs).1 However, it has beenreported to be difficult to locate the origins of someof those ATs because of the participation of prefer-ential conduction.2 Though focal atrial fibrillation(AF) originating from the right atrium3 has beenreported, the participation of preferential conduc-tion in that AF has not yet been reported. We reporta case of focal AF associated with preferential con-duction and multiple breakout sites at the cristaterminalis (CT).

Case ReportA 34-year-old man with permanent AF refrac-

tory to class I (disopyramide and flecainide) andclass III (bepridil) antiarrhythmic drugs was re-ferred for an electrophysiologic study (EPS) andRF catheter ablation. This patient had no historyof concomitant disease. There was no echocardio-graphic evidence of structural heart disease. Theleft atrial dimension was 35 mm and left ventric-ular ejection fraction 69%. Informed consent wasobtained, and the EPS was performed after all an-tiarrhythmic drugs had been discontinued for atleast five half-lives prior to the study. Anticoag-ulation with warfarin was begun starting from 1month before the study.

In the first session, pulmonary vein (PV) isola-tion with a 31-mm multielectrode basket catheter

Address for reprints: Takumi Yamada, M.D., Division of Cardi-ology, Aichi Prefectural Cardiovascular and Respiratory Cen-ter, 2135 Kariyasuka, Yamato-cho, Ichinomiya 491-0934, Japan.Fax: +81-586-45-6800; e-mail: takumi-y@fb4.so-net.ne.jp

Received August 12, 2005; revised October 11, 2005; acceptedOctober 19, 2005.

(MBC) (ConstellationTM, EP Technologies BostonScientific, San Jose, CA, USA) was performed bythe same technique as we reported previously.4Anticoagulation with warfarin was also contin-ued after the first session. Though sinus rhythmwas restored after successful isolation of all fourPVs, paroxysmal AF recurred 2 months later. Be-cause the paroxysmal AF was uncontrolled evenafter the administration of disopyramide and fle-cainide, a second session was performed 5 monthsafter the first. At the beginning of the secondsession, the PVs were remapped and isolationof all four PVs was confirmed. Thereafter, spon-taneous AF triggered by atrial premature beats(APBs) with a short coupling interval was in-duced by burst pacing from the distal CS undera continuous infusion of isoproterenol. The P-wave morphology of the APB triggering the AFwas positive in leads I, II, III, and aVF and bipha-sic in V1, suggesting that the AF might originatefrom the mid to high right atrium. Low-amplitudefractionated potentials with the earliest activationwere recorded from several lower electrode pairsof the MBC splines deployed at the mid poste-rior wall of the right atrium during the triggeredAPBs (Fig. 1). However, the earliest activation ofthe high-frequency potentials following those low-amplitude fractionated potentials was recordedfrom the lower electrode pairs of the medial spline.The activation sequence of those high-frequencypotentials recorded from several electrode pairs ofthe posterior spline was from superior to inferiorand the earliest low-amplitude fractionated poten-tials recorded from the lower electrode pairs of thesame spline did not connect directly to those high-frequency potentials.

Double potentials were recorded from all theelectrode pairs of the posteromedial spline during

C©2006, The Authors. Journal compilation C©2006, Blackwell Publishing, Inc.

PACE, Vol. 29 February 2006 207

YAMADA, ET AL.

Figure 1. Cardiac tracings showing the initiation of atrial fibrillation (AF) (left panel) and successful ablation site(right panels).The first and second atrial beats are sinus beats, and the third and fourth beats are atrial prematurebeats (APBs) triggering the AF. During the triggering APBs, low-amplitude fractionated potentials with the earliestactivation (arrowheads) were recorded from electrode pairs F7–8 and G7–8 deployed at the posterior wall of the rightatrium before the earliest high-frequency potentials recorded from electrode pair H5–6 (star). Those high-frequencypotentials exhibited a double potential configuration, suggesting the crista terminalis (CT). The activation sequenceof those high-frequency potentials recorded from electrode pairs 3–4 to 7–8 of splines F and G was from superior toinferior and the earliest low-amplitude fractionated potentials recorded from electrode pairs 7–8 of splines F and Gdid not connect directly to those high-frequency potentials. Splines A and B of the multielectrode basket catheter(MBC) were deployed on the side of the tricuspid annulus. The alphabetical letters A, B, F, and G indicate the splinesof the MBC. ABL = ablation catheter; CS = coronary sinus; LAO = left anterior oblique view; RAO = right anterioroblique view.

both sinus rhythm and triggered APBs (Fig. 2). Thelow-amplitude fractionated potentials with theearliest activation were always recorded from sev-eral lower electrode pairs of the posterior splinesduring the triggered APBs. The earliest activationsite and activation sequence of the high-frequencypotentials following those low-amplitude fraction-ated potentials changed from beat to beat duringthe triggered APBs. When the high-frequency po-tentials did not follow some beats in a run oflow-amplitude fractionated potentials, APBs werenever observed.

RF applications were delivered with the guid-ance of the MBC with a target temperature of 60◦Cand maximum power output of 50 W for 60 sec-onds, using an 8-mm tip catheter during sustainedAF. Though a few RF applications targeting the ear-liest low-amplitude fractionated potentials did notterminate the AF (Fig. 1), internal cardioversionrestored sinus rhythm. Thereafter, neither AF norAPBs could be induced with the same protocol as

was used before the ablation. No complications oc-curred.

Follow-up was performed at the outpatientclinic of our hospital at 2 weeks, 1 month, andevery month thereafter following the second pro-cedure, using 24-Holter and cardiac recordings.When the patient reported symptoms, an eventmonitor was given to document the cause of thesymptoms. During more than 1 year of follow-up,this patient was free of symptomatic AF withoutany antiarrhythmic drugs. Anticoagulation withwarfarin was stopped 6 months after the secondsession because of the absence of any AF recur-rence and no emboli had occurred throughout thefollow-up period.

DiscussionNoncontact balloon mapping has revealed

that some right atrial focal ATs are associated withpreferential conduction.3 However, to our knowl-edge there have been no reports about focal AF

208 February 2006 PACE, Vol. 29

AF WITH MULTIPLE CRISTA BREAKOUTS

Figure 2. Concealed ectopy.The first and third atrial beats are sinus beats, and the second, fourth,and fifth beats are triggered APBs. Double potentials, suggesting the CT were recorded from sev-eral electrode pairs of spline H during both sinus rhythm and triggered APBs. During the triggeredAPBs, the low-amplitude fractionated potentials with the earliest activation were always recordedfrom electrode pairs F7–8, G5–6, and G7–8 deployed at the posterior wall of the right atrium.However, the earliest activation of the high-frequency potentials following those low-amplitudefractionated potentials was recorded from the electrode pair H3–4 during the first APB, electrodepairs F7–8 and H3–4 during the second APB, and electrode pair F7–8 during the third APB, re-spectively. The activation sequence of the high-frequency potentials recorded along splines F andG changed according to the shift in the earliest activation site of those high-frequency potentials.The high-frequency potentials did not follow the low-amplitude fractionated potentials recordedimmediately after the first and second APBs (concealed ectopy). The conduction between thelow-amplitude fractionated potentials and high-frequency potentials exhibited Wenckebach typeblock. The arrowheads and stars indicate the earliest activation of the low-amplitude fractionatedpotentials and that of the high-frequency potentials, respectively.

associated with preferential conduction. In thiscase, the findings from the contact mapping withan MBC and the results of the RF catheter ab-lation suggested the participation of preferentialconduction in the focal AF originating from theright atrium.

PV isolation alone cannot cure AF completelybecause of non-PV AF foci.5 About 4% of thosenon-PV foci have been located at the CT.5 Anotherreport has shown that about 60% of right atrialfocal ATs were located on the CT.3 Some reportshave suggested that there are cardiomyocytes withpacemaker activity or abnormal automaticity6 andcatecholamine-sensitive ectopic foci5 in the CT.In this case, the AF originating from the poste-rior wall of the right atrium was catecholamine-sensitive and double potentials were recordedaround the AF origin. These findings suggestedthat the AF in this case might have originated fromthe CT.

In this case, the low-amplitude fractionatedpotentials were suggested to reflect the ectopicactivity of the AF origin and the high-frequencypotentials following those low-amplitude fraction-ated potentials were suggested to reflect the activ-ity of the atrium surrounding the AF origin. Theparticipation of preferential conduction was sug-gested in this AF because the earliest activation ofthe low-amplitude fractionated potentials and thatof the high-frequency potentials were observed atdifferent sites. Because the CT exhibits an area ofprominent anisotropy with slow conduction prop-erties,7 the CT could be the source of the preferen-tial conduction that the activation wavefronts passthrough in the ATs originating from the CT.3 TheCT could also be the source of the preferential con-duction in the AF in this case.

In this case, firing of ectopic focal activity in-duced a Wenckebach type of conduction block tothe atrial myocardium (concealed ectopy). This

PACE, Vol. 29 February 2006 209

YAMADA, ET AL.

Figure 3. Schema showing the AF origin, preferential conduction and multiple breakout sitesduring the three APBs in Figure 2.The activation from the single AF origin between electrode pairsF6–7 and G6–7 propagated through multiple preferential conduction areas toward the multiplebreakout sites around electrode pairs H3–4 and F7–8. Thereafter, the activation spread to thewhole atrium from one or two breakout sites which were present during each APB. The stars,solid lines, dotted lines, and gray marks at the end of the solid lines indicate the AF origins,preferential conduction sites, atrial activation sequence, and breakout sites, respectively. Thealphabetical letters F, G, and H and numbers indicate the splines of the MBC and electrodes onthe splines, respectively. IVC = inferior vena cava. The other abbreviation is as in Figure 1.

finding suggested that the preferential conductionassociated with the AF in this case might have hada decremental property.

Though the earliest activation site of the low-amplitude fractionated potentials never shiftedduring triggered APBs, that of the high-frequencypotentials altered between multiple sites beat tobeat. This finding could be explained by the ex-istence of one AF origin with multiple preferen-tial conduction and breakout sites (Fig. 3). Theanalysis of right atrial focal ATs by noncontact bal-loon mapping has not reported multiple preferen-tial conduction sites and their decremental proper-

ties.3 This is probably because in AF, ectopic focalactivity with a shorter coupling interval than theAT, can change the preferential conduction. Multi-ple preferential conduction and breakout sites mayplay a role in the initiation and perpetuation ofAF.

ConclusionsA preferential conduction can be associated

with focal AF originating from the CT as well asfocal AT from the CT. In the RF catheter ablationof focal AF originating from the CT, the singularityof that AF should be considered.

References1. Kay GN, Chong F, Epstein AE, Dailey SM, Plumb VJ. Radiofre-

quency ablation for treatment of primary atrial tachycardias. J AmColl Cardiol 1993; 21:901–909.

2. Higa S, Tai CT, Lin YJ, et al. Focal atrial tachycardia: New insightfrom noncontact mapping and catheter ablation. Circulation 2004;109:84–91.

3. Chen SA, Tai CT, Yu WC, et al. Right atrial focal atrial fib-rillation: Electrophysiologic characteristics and radiofrequencycatheter ablation. J Cardiovasc Electrophysiol 1999; 10:328–335.

4. Yamada T, Murakami Y, Muto M, et al. Computerized three-dimensional potential mapping with a multielectrode basket

catheter can be useful for pulmonary vein electrical disconnection.J Interv Card Electrophysiol 2005; 12:23–33.

5. Lin WS, Tai CT, Hsieh MH, et al. Catheter ablation of paroxysmalatrial fibrillation initiated by non-pulmonary vein ectopy. Circula-tion 2003; 107:3176–3183.

6. Boineau JP, Canavan TE, Schuessler RB, Cain ME, Corr PB, Cox JL.Demonstration of a widely distributed atrial pacemaker complexin the human heart. Circulation 1988; 77:1221–1237.

7. Spach MS. Discontinuous Cardiac Conduction: Its Origin in Cel-lular Connectivity with Long-term Adaptive Changes That CauseArrhythmias. Armonk, NY, Futura Publishing Company, 1997,5–58.

210 February 2006 PACE, Vol. 29