Localization of the origin of the atrioventricular junctional rhythm induced during selective ablation of slow-pathway conduction in patients with atrioventricular node reentrant tachycardia

Localization of the origin of the atrioventricular junctional rhythm induced during selective ablation of slow-pathway conduction in patients with atrioventricular node reentrant tachycardia

J o h n C.L. Yu, MD, Michael R. Lauer , MD, PhD, Char l ie Young, MD, L. Bing Liem, DO,

Char l e s Hou, MD, and Ruey J. Sung, MD Stanford and San Jose, Calif.

During radiofrequency catheter ablation of slow atrioventricular node pathway conduction in patients with atrioventricular node reentrant tachycardia, an atrioventricular junctional rhythm is frequently observed. The origin and relation to ablation success of this junctional rhythm was examined in this study. By using standard intracardiac electrophysiology techniques, we studied the radiofrequency energy- induced atrioventricular junctional rhythm in 43 consecutive patients with atrioventricular node reentrant tachycardia undergoing selective ablation of slow-pathway conduction. The frequency of atrioventricular junctional activity was correlated with successful and unsuccessful attempts at ablation of slow-pathway conduction. Also, we compared the sequence of retrograde atrial activation of radiofrequency energy-induced atrioventricular junctional beats in a subgroup of 22 patients with the retrograde activation sequence observed during pacing from the right ventricular apex and the site of successful ablation of slow-pathway conduction. A total of 201 radiofrequency-energy applications was delivered in 43 patients with ->5 atrioventricular junctional beat(s) induced during 110 (55%) of 201 ablation attempts. Atrioventricular junctional activity was noted during 98% of successful ablations but only 43% of the unsuccessful attempts (sensitivity, 98%; specificity, 57%; negative predictive value, 99%). The mean time to appearance of atrioventricular junctional beats was 8.8 ± 4.1 sec (mean ± SD) after the onset of radiofrequency-energy application, In 22 (100%) of 22 patients in whom detailed atrial mapping was performed, the retrograde atrial activation sequence of the radiofrequency-induced atrioventricular junctional beats was earliest in the anterior atrial septum, identical to that seen during pacing from the right ventricular apex. Earliest retrograde atrial activation was at the posterior septum in all

From the Cardiac Electrophysiology Laboratories, Stanford University Medical Center, and Kaiser Foundation Hospital-Santa Teresa. Received for publication Aug. 24, 1995; accepted Oct. 1, 1995. Dr. Hou is a research fellow supported by the MacKay Memorial Hospital, Taipei, Taiwan. Reprint requests: Ruey J. Sung, MD, Cardiac Electrophysiology Service, Room H2146, Stanford University Medical Center, 300 Pasteur Drive, Stanford, CA 94305. Copyright © 1996 by Mosby-Year Book, Inc. 0002-8703/96/$5.00 + 0 4/1/70748

patients during pacing from the successful ablation site, a markedly different activation pattern compared with that seen during either radiofrequency ablation or ventricular pacing. Whereas the occurrence of atrioventricular junctional activity during radiofrequency ablation does not necessarily herald a successful ablation of slow atrioventricular node pathway conduction, its absence strongly suggests that the energy is being applied in an unsuccessful fashion. Furthermore, it appears that radiofrequency energy-induced atrioventricular junctional beats originate not from the endocardium in contact with the ablating catheter tip but instead appear to exit remotely from the anterior atrial septal region. This finding supports the existence of specialized tissues in the atrioventricular junction that preferentially transmit the effects of radiofrequency energy to an anterior exit site, possibly identical to the atrial exit site of the retrograde fast atrioventricular node conduction pathway. (AM HEART J 1996;131:937-46.)

Selective rad iof requency ca the te r ab la t ion of slow a t r ioven t r i cu la r (AV) node p a t h w a y conduct ion is a wel l -accepted t r e a t m e n t for pa t i en t s wi th AV node r e e n t r a n t tachycard ia . 1,2 N u m e r o u s inves t iga tors have noted the deve lopmen t of an AV junc t iona l r h y t h m dur ing the appl ica t ion of rad iof requency en- e rgy in the AV junct ion. 3-7 The absence of a n y rad iof requency e n e r g y - i n d u c e d AVjunc t iona l activ- i ty has, in fact, been considered a m a r k e r for an unsuccessful ab la t ion a t t emp t . However , the site of origin and m e c h a n i s m of in i t ia t ion of these abla t ion- re la ted AV junc t iona l bea t s are still in dispute. The objective of th is s tudy was prospect ive ly to eva lua te the usefu lness o f rad io f requency e n e r g y - i n d u c e d AV junc t iona l r h y t h m as a m a r k e r for successful abla- t ion of s low-pa thway conduct ion and localize its origin.

METHODS Patients and electrophysiologic studies. Foz~y-three

consecutive patients with AV node reentrant tachycardia

937

May 1996 938 Yu et al. American Heart Journal

Fig, 1. Radiographic recordings of intracardiac positions of multipolar electrode catheters for selective ablation of slow AV node pathway conduction. Right anterior oblique, anterior-posterior, and left anterior oblique views are shown. Catheters are placed in high right atrium (HRA), anterior atrial septa, or His bundle region (HBE), coronary sinus (CS), and right ventricular apex (RVA). Mapping and ablation catheter (ME) is positioned in posteroseptal region close to ostium of coronary sinus.

undergoing electrophysiology study and radiofrequency catheter ablation were studied after obtaining informed consent according to local institutional review board guide- lines. Included were 15 men and 28 women with an aver- age age of 49.6 _+ 18.4 years. All antiarrhythmic drugs were withheld at least five half-lives before the study.

Standard intracardiac electrophysiology techniques were used. Multipolar electrode catheters were positioned by using percutaneous technique in the coronary sinus/ great cardiac vein, the high right atrium, anterior atrial septal (His bundle) region, and right ventricular apex (Fig. 1). The decapolar electrode catheter in the coronary sinus had its most proximal electrode pair positioned just inside the ostium. Continuous intra-arterial blood pressure monitoring was performed. Programmed extrastimulation (Model DTU-125, Bloom Associates, Ltd., Reading, Pa.) was performed to assess AV node conduction properties and induce AV node reentrant tachycardia. Intracardiac electrograms from the high right atrium, His bundle, and coronary sinus electrode catheters, along with surface electrocardiogram leads I, aVF, and V1 were continuously recorded (EP Lab, Biomedical Instrumentation, Inc., Markham, Ontario, Canada, and Cath 2000 Crossover System, Gould, Inc., Valley View, Ohio). The electrophysiologic diagnosis of AV node reentrant tachycardia was confirmed by using standard criteria, s-l°

Endocardial mapping and catheter ablation. Intracar- diac endocardial mapping of the triangle of Koch along the atrial side of the tricuspid annulus was performed by using a deflectable quadripolar 7F catheter (Mansfield, Boston, Mass., or EP Technologies, Sunnyvale, Calif.) introduced through the right femoral vein. During fixed-cycle-length (CL) electrical pacing (CL = 600 msec) from the right ventricular apex with and without single ventricular extrastimuli, the atria] activation sequence during ventriculoatrial conduction was recorded from (1) the anteriorly located His bundle electrode catheter, (2) the posteriorly positioned coronary sinus electrode catheter, and (3) the mapping electrode catheter within the triangle of Koch. In all patients, ventriculoatrial conduction was present (without or with isoproterenol infusion) 11 with a short ventric-

uloatrial conduction time and earliest retrograde atrial activation mapped to an anterior location near the His bundle recording site (fast pathway; Fig. 2). 12 In some patients, incremental or extrastimulus pacing techniques uncovered an additional conduction pathway exhibiting a long ventriculoatrial conduction time and earliest retrograde atrial activation recorded posteriorly by the proximal electrode pair at the ostium of the coronary sinus (slow pathway). 12 Before each ablation attempt, fixed-CL electrical pacing was performed with and without single extrastimuli from both the right ventricular apex and the tip of the ablation catheter at the putative site of slow- pathway conduction. During ablation attempts, unmodulated radiofrequency current (Radionics, Model RFG-3C, Burlington, Mass.)at a power of 30 W was delivered for 30 sec. The ablation attempt was immediately terminated ff an impedance increase occurred. Any coagulum present was cleaned from the catheter tip. After every ablation attempt, regardless of the presence or absence of AV junctional activity during the energy application, atrial and ventricular pacing with and without extrastimuli was performed to assess the effectiveness of the ablation attempt. During the period of this study, an ablation was defined as successful if slow-pathway conduction could no longer be demonstrated in both the antegrade and retrograde direction and AV node reentrant tachycardia was no longer inducible in the absence or presence of intravenous isoproterenol infusion) 1 (Subsequent to the period of this study, we redefined a successful ablation for AV node reentrant tachycardia as complete elimination of inducible tachycardia [without or withisoproterenol infusion], even though residual antegrade slow-pathway conduction [with or without single AV node echo beats] may be present). An AV junctional rhythm during radiofrequency-energy application was defined as the occurrence of five or more AVjunc- tional premature depolarizations in which atrial activation proceeded initially from the AV junctional region to the high right atrium (low to high sequence) and the QRS complex structure and axis were identical to that seen with sinus beats.

In 22 of 43 patients, the atrial activation sequence ofra-

Volume 131, Number 5 American Heart Journal Y u et al. 9 3 9

diofrequency energy-induced junctional beats was examined and compared with the retrograde atrial activation sequence seen during pacing from both the right ventricular apex and the tip of the ablation catheter whi]eit was still positioned at the site of the ablation. The difference in local atrial activation times (SAAT) was calculated between the appearance of the atrial electrograms recorded at the anteriorly located distal His bundle (HiSd) recording site and the posteriorly located proximal coronary sinus (CSp) electrode pair positioned at the ostium of the coronary sinus during (1) radiofrequency energy-induced AV junctional rhythm, (2) pacing from the right ventricular apex, or (3) pacing from the successful ablation site of slow-pathway conduction.

Stat is t ics . Means were expressed --1 standard devia- tion. Statistical significance at the ~<5% level was calculated by using an unpaired t test (Statview, Brain Power, Inc., Calabasas, Calif.).

R E S U L T S

All 43 patients underwent successful selective ablation of slow AV node conduction, receiving a total of 201 radiofrequency applications or 5.4 _+ 4.2 applications per patient (Table I). The mean impedance during radiofrequency application was 104 _+ 12 ohms. There were no complications from the procedures, and no patient developed complete heart block or had ablation of fast-pathway conduction. In patients in whom AV junctional rhythm occurred during the successful radiofrequency application (42 [98%] of 43), AV junctional activity was recorded 8.8 _+ 4.1 sec after the onset of the radiofrequency delivery during t h e successful attempt. The radiofrequency energy-induced AVjunctional rhythm was irregular and exhibited ventricular rates between 90 and 130 beats/min (109 _+ 11 beats/min). During the radiofrequency-induced AV junctional rhythm, retrograde atrial activation occurred before or simulta- neous with the inscription of the QRS complex, and it was never associated with retrograde block in the AV node (H-A block). The frequency of occurrence of AV junctional activity during successful and unsuccessful radiofrequency-ablation attempts is shown in Table I. The overall incidence of AVjunctional activity during 201 radiofrequency applications was 110 (55%) of 201, with AVjunctional activity absent during 91 (45%) of 201 radiofrequency deliveries. The presence ofradiofrequency energy-induced AVjunc- tional rhythm was a very sensitive indicator (98%) of a successful ablation at tempt but was relatively nonspecific (57%). Consequently the absence of radiofrequency-induced AV junctional activity correlated highly with an unsuccessful ablation at tempt (negative predictive value of 99%), whereas the presence of an AV junctional rhythm during radiof-

Table I. Outcome of attempted ablation of AV node slow- pathway conduction and development of AV junctional rhythm in 43 patients

A V junctional rhythm Outcome of

attempted ablation Present Absent Total

Successful 42 1 43 Unsuccessful 68 90 158

Total 110 91 201

requency ablation did not necessarily indicate a successful outcome (positive predictive value of 38%).

The retrograde atrial activation sequence during the radiofrequency energy-induced AV junctional rhythm was closely examined to clarify its origin and potential electrophysiologic mechanism in 22 of the 43 patients. Of these 22 patients, 19 (86%) patients had the typical (slow-fast) form of AV node reentrant tachycardia induced, whereas 3 (14%) patients had the atypical (fast-slow) form induced. The successful slow AV node pathway ablation site was in the atrial septum along the tricuspid annulus midway between the His bundle recording site and the coronary sinus ostium in 12 (55%) of 22 patients, and posteriorly near the coronary sinus ostium in 10 (45%) of 22 patients.

Fig. 2 displays the characteristic sequence of retrograde atrial activation in a patient with retrograde dual AV node conduction pathways. Note that during retrograde fast-pathway conduction, the earliest atrial activation occurred anteriorly, as recorded in the His bundle catheter, whereas the proximal coronary sinus electrodes near the ostium recorded the earliest retrograde atrial activation during retrograde slow-pathway conduction. Both these activation patterns differ significantly from that seen during antegrade AV conduction. When the atrial activation sequence of radiofrequency energy-induced AV junctional activity was examined in patients with either typical or atypical AV node reentry (Figs. 3 and 4), it was found to match closely the retrograde atrial activation sequence seen during pacing from the right ventricular apex, whereas it was significantly different from the sequence seen during pacing from the site of successful ablation of slow-pathway conduction. In the case of both radiofrequency-induced AV junctional beats and right ventricular pacing, the initial atrial activation was consistently recorded by the His bundle catheter in the anterior atrial septum (HiSd). In the case of pacing at the successful site of ablation, initial activation was consistently recorded posteriorly near the coronary sinus ostium (CSp; Figs. 3 and 4). Fig. 5

May 1996 940 Yu #t al. American Heart Journal

S S

i l IRA

CSI ' _ .

,cs3 A ~ - ....

cs+

CSS ~A

200 ms

h . . . . , i . . , . i t l

Fig. 2. Intracardiac recordings of atrial activation sequence during shift from retrograde fast AV node pathway conduction to retrograde slow AV node pathway conduction. Displayed from the top to bottom are surface electrocardiographic leads I, aVF, and V1 intracardiac recordings from the high right atrium (HRA), proximal to distal coronary sinus (CS1 to CS5, respectively), proximal (HBEz) and distal His bundle (HBE2) region, and right ventricular apex (RVA). CS1 electrode pair was located at ostium of coronary sinus. RVA pacing at fLxed CL of 400 msec is performed, followed by premature ventricular stimulus delivered at CL of 360 msec. Shown here are electrograms recorded from last stimulus of drive train followed by premature stimulus. Last drive stimulus results in ventriculoatrial conduction with retrograde atrial activation through fast AV node conduction pathway with the earliest atrial activation seen anteriorly near the His bundle region (HBE1 or HBE2). Premature ventricular stimulus results in retrograde fast-pathway conduction block, with atrial activation occurring instead through posterior slow-conduction pathway near ostium of coronary sinus (CSz). After retrograde conduction, there is ventricular echo beat with conduction through antegrade fast AV node conduction pathway. A, Atrial electrogram; H, His bundle electrogram resulting from antegrade activation; H-, His bundle electrogram resulting from retrograde activation; S, stimulus artifact.

summarizes the results from all 22 patients and shows the individual and mean differences in local atrial activation times (SAAT) between the anterior His bundle region (Hisd) and posteriorly near the os- t ium of the coronary sinus (CSp).

DISCUSSION Clinical significance of radiofrequency energy-in-

duced AV junctional rhythm. Our results, along with those of others, 3-7 show that although the development of an AV junctional rhythm during attempted radiofrequency ablation of slow AV node pathway conduction is not necessarily indicative of a successful ablation, the absence of such activity is a sensi- rive marker of an unsuccessful ablation attempt. During attempted ablation of slow-pathway conduction, failure to note an AV junctional rhythm within 15 sec after initiating radiofrequency-energy delivery should prompt termination of the radiofrequency application and catheter repositioning, Early termination of inevitably unsuccessful radiofrequency applications may result in reduced procedure time, less radiation exposure to patients, physicians, nurses,

and laboratory staff, 13 and may limit unnecessary and unintended radiofrequency-induced damage to the AV junction region, such as complete AV block. 14, 15

The finding of AV junctional activity without a successful ablation of slow-pathway conduction would appear to indicate that adequate tissue heating is occurring, but the tissue being damaged does not include the slow-conduction pathway. On the other hand, the failure of radiofrequency energy to induce AV junctional activity does not necessarily imply that the catheter tip is incorrectly positioned far from the slow-pathway location. It may simply indicate that heating of tissue is inadequate, either because of poor tissue contact or insufficient current flow to the tissue as a result of high impedance between the catheter tip and the tissue (coagulum for- mation) or the body-surface grounding patch. A high impedance, from any cause, would be readily de- tected by noting the impedance measurement from the radiofrequency generator. In this study, lack of development of AV junctional rhythm during radiofrequency ablation was not associated with a high

Volume 131, Number 5 American Heart Journal Yu et al. 941

1

aVF

V1 H.RA

i S1 A

- V A

A CS1

CS2 _ . ~ ' A

CS3 _,~

CS4 ~ ~ ~ CS5

HBE2

R V A

ABLp

ABLd -

ABLATION CATli~T.R PACING

B

V A

JUNCTIONAL IIF_,AT

C Sl

A

J

V ~ l E t ~ A l l FACING 200 ms

I . . . . . . . . . I . . . . . . . . . I

Fig. 3. Intracardiac recordings of atrial-activation sequence in patient with typical (slow-fast) form of AV node reentrant tachycardia during (A) pacing from successful site of ablation, (B) radiofrequency energy- induced AVjunctional rhythm, and (C) pacing from right ventricular apex. Recordings displayed from top to bottom are identical to those shown in Fig. 2, with the addition of recordings from proximal (ABLp) and distal ablation catheter (ABLd), which is positioned in region of slow-pathway conduction in posterior triangle of Koch. CS1 electrode pair was located at ostium of coronary sinus. Note similarity in atrial-activation sequence during radiofrequency energy-induced AVjunctional rhythm and right ventricular pacing (anterior atrial activation earliest in distal His bundle region [HBE2]), both of which are distinctly different from that seen during pacing from successful site of ablation in right atria (posterior atrial activation earliest near ostium of coronary sinus [CSd). A, Atrial electrogram; H, His bundle e]ectrogram resulting from antegrade activation; H-, His bundle electrogram resulting from retrograde activation; $1, stimulus artifact.

impedance. Poor tissue contact would be more diffi- cult to detect. Mapping of the retrograde slow-pathway atrial exit site (if possible), the amplitude and appearance of the recorded local electrogram, and the pacing threshold from the tip of the ablation catheter are among a number of preablation vari- ables that may be useful in evaluating correct catheter positioning and tissue contact. Monitoring of catheter-tip temperature 16, 17 which was not performed in these patients--would provide a more re- liable indication of actual tissue heating during the ablation attempt.

Origin and mechanism of initiation of radiofrequency energy-induced AV junctional rhythm. Our study showed that the radiofrequency energy-induced AV

junctional rhythm during ablation of slow-pathway conduction results in earliest atrial activation near the retrograde atrial exit of the AV node fast- conduction pathway, similar to that seen either with pacing from the right ventricle or during the slow- fast form of AV node reentrant tachycardia. Con- trary to what might be intuitively assumed, the retrograde atrial exit of the radiofrequency-induced AV junctional rhythm (anterior location) appears distinctly different from that seen during electrical pacing from the site of successful slow-pathway ablation (posterior location). Our observations are con- sistent with recently reported findings by Goldreyer et al., is although these investigators used different reference points for the anterior and posterior atrial


1

aVF

VI

HRA

CSI

CS2

CS3

CS4

CS5

HBE1

HBE2

RVA

ABLp

ABLd

A Sl

A

_ • A "

A

- /

L- (

B

~ h

H h

C

A

A

ABLATION CATHETER PACING JUNCTIONAL BEAT

VENTRICULAR PACING

200 ms [,,,,,,,,,I,,,,,,,,,I

Fig. 4. Intra•ardia•re••rdings•fatrial•a•ti•ati•nsequen•einpatientwithatypical(fast•s••w)f•rm•fAV node reentrant tachycardia during (A) pacing from successful site of ablation, (B) radiofrequency energy- induced AV junctional rhythm, and (C) pacing from right ventricular apex. Recorded signals from top to bottom are same as in Fig. 3. CS1 electrode pair was located at ostium of coronary sinus. As with patient with slow-fast form of AV node reentry (Fig. 2), there is marked similarity in atrial-activation sequence during radiofrequency energy-induced AV junctional rhythm and ventricular pacing (anterior atrial activation earliest in distal His bundle region [HBE2]), both of which are distinctly different from that seen during pacing from successful site of ablation in right atria (posterior atrial activation earliest near ostium of coronary sinus [CS1]). A, Atrial eleetrogram; H, His bundle eleetrogram resulting from antegrade activation; H-, His bundle eleetrogram resulting from retrograde activation; 81, stimulus artifact.

septal regions (His bundle electrogram and slow- pathway potentials, respectively).

The origin and mechanisms underlying the generation of the radiofrequency energy-induced AVjunc- tional rhythm remain to be explained. Of possible significance, Nath et al. 19 showed that superfusion of guinea pig papillary muscles with a hyperthermic Tyrode's solution results in membrane depolarization and spontaneous rhythmic activity at low levels of membrane potentials. Clinically, however, radiofrequency energy-induced automatic activity is a very uncommon finding during radiofrequency ablation in ventricular tissues, suggesting that tissue heating, per se, is not sufficient to induce an accelerated automatic rhythm. In addition, it is not clear how radiofrequency-energy application at the site of

slow-pathway conduction (posterior triangle of Koch) results in the development of an AV junctional rhythm that exits distant from the retrograde atrial exit site of the fast-conduction pathway (anterior triangle of Koch). The effect of radiofrequency application in the posterior AV junction is clearly different from electrical pacing in the same region. In pacing, the atria in the posterior AV junction is the site of earliest activation, whereas in the former case, the atria in the anterior AV junction is the site of earliest activation. Bipolar electrical pacing between two widely separated electrodes (5 mm) by using a high DC current square wave brings cardiac cells to their action-potential threshold and is a distinctly different biophysical stimulus from the unmodulated high- frequency (500 to 750 kHz) radiofrequency current

Volume 131, Number 5 American Heart Journal Yu et al. 943

5O

"~' 40

| |

-30

.4O

Patient No, 9 10 11 12 13 14 15 16 17 18 19 20 21 22

• ~ . ~ ' ~ . ~ : ~

AV Junctional Beats [--] L.._J

RVA Pacing

A AAT (ms)

+ 1 7 + 9

+ 1 9 + 1 1

P value

NS

Ablation-Site Pacing ~ -11 + 12 <0.05

Fig, 5. Differences in atrial activation times (~AAT) as recorded at distal His bundle region and proximal coronary sinus in 22 patients during pacing from successful site of ablation or right ventI~cular apex and with radiofrequency energy-induced AV junctional rhythm. Measurements of 3AAT are made relative to atrial electrogram recorded from distal His bundle region, Negative values mean that atrial electrogram recorded in proximal coronary sinus (CSp) was recorded earlier than atrial electrogram recorded in distal His bundle region (Hiss), whereas positive values indicate that CSp atrial electrogram was recorded later than HiSd atrial electrogram.

that does not stimulate action-potential generation in cardiac cells. ~° Consequently, unlike electrical pacing, it is not surprising that the production ofradiofrequency lesions in the posterior septum does not cause action-potential generation emanating from this region.

The application of radiofrequency energy to cardiac tissues results in resistive heating of tissue and the development of coagulation necrosis. 2~'23 The spread of the heating appears equal in all directions, resulting in a concentric spherical or hemispherical lesion radiating from the radiofrequency source. 20"23

The effect on directly adjacent cardiac cells exposed to this type of heating is immediate depolarization and death. More important, however, the focal depolarization in the border zone results in an injury current that exerts distant effects by spreading electro- tonically to adjacent, undamaged regions of tissue. It is critical to recognize, however, that electrotonic spread of the injury current, unlike the radiofrequency lesion itself, is not equal in all directions. Ac- cording to cable theory, 242~ the decremental injury current will conduct farthest, fastest, and decayleast along the pathway of lowest intracellular or extra-

May 1996 944 Yu 8t al. American Heart Journal

cellular resistivity or both. It is expected, therefore, that the injury current induced by radiofrequency- energy application will travel farthest from its site of application if a low-resistance pathway exists for conduction of the injury current.

An appreciation of the anatomy of the AV junction 27 provides clues to explain how an injury current generated in the posterior triangle of Koch may induce an AVjunctional rhythm that exits anteriorly. Lying at the apex of the triangle of Koch, the compact node penetrates the central fibrous body to become the His bundle. Tawara 2s described cells transitional in structure extending from the posterior aspect of the compact node to the coronary sinus ostium. From these initial observations, along with subsequent contributions by other investigators, 29, 30 Becker and Anderson 31 grouped these transitional cells into three zones: superficial, deep, and posterior. The superficial zone was continuous with the anterior and superior aspect of the compact node; the posterior zone joined the inferior and posterior part of the compact node; and the deep zone connected the left atrial septum to the deep part of the compact node. Hecht et al . 32 used the term "nodal approaches" to describe these transitional cell zones. In the canine heart, Racker 33 similarly observed discrete atrionodal bundles--superior, medial, and lateral--to form a prommal AV bundle, which in turn was contiguous with the compact node. The proximal AV bundle cor- responded to the zone of transitional cells described by Tawara. 2s Racker 34 further demonstrated that the superior atrionodal bundles and the proximal AV bundle possessed functional properties of specialized conducting tissues, distinctly different from those of the working atrial myocardium. 35 In the rabbit heart, there is a dual input to the compact node during antegrade conduction36-39--an anterior input entering the node as a broad wavefront anterior to the coronary sinus ostium and a posterior input entering the node beneath the coronary sinus ostium via the crista terminalis. During retrograde conduction, the earliest exit to the atrium is in the interatrial septum, anterior to the coronary sinus ostium, at the same location as the anterior input during antegrade conduction; the crista terminalis is activated much later than the interatrial septum. 36"4°

These anatomic studies provide evidence of a direct low-resistance connection between the posterior and anterior portions of the triangle 0fKoch. The injury current may conduct decrementally from its site of origin in the slow-conduction pathway near the ostium of the coronary sinus to the anteriorly located compact AV node. In the AV junction, cells re-

siding in the compact node or proximal His bundle region exhibit spontaneous pacemaker activityY These, or similar cells in the AVjunction, may be the source of the AV junctional activity induced by radiofrequency-energy application. The injury current may enhance diastolic depolarization by pro- viding a net depolarizing current. The increased rate of diastolic depolarization would allow these cells to surface as lead pacemakers, discharging at a rate faster than that of the sinus node pacemaker, resulting in active potential conduction retrograde to the atria in the anterior retrograde fast-conduction pathway, and antegrade in the His bundle to the ventricles. Also, ifradiofrequency-current delivery to the AV node region enhances postganglionic release of norepinephrine from sympathetic nerve endings in excess of acetylcholine release from vagal nerve endings, this may also increase AV junctional auto- maticity 42-44 by increasing the rate of diastolic depolarization. 45

Study limitations, Localization of the origin of the AV junctional activity may have been aided by mea- suring of the H-A interval during the radiofrequency energy-induced AV junctional rhythm and compar- ing it with the H-A interval recorded during AV node reentrant tachycardia or pacing from the right ventricular apex. However, the CL of the AV junctional activity induced during radiofrequency ablation is irregular and highly variable and rarely equal to the CL of the inducible AV node reentrant tachycardia or the CLs used during pacing from the right ventricular apex. Because retrograde conduction from the His bundle through the AV node to the atria is decremental and highly dependent on the input CL, comparisons between H-A intervals recorded during radiofrequency energy-induced AV junctional activity and AV node reentrant tachycardia or right ventricular pacing would not be helpful. Furthermore, moment-to-moment variations in autonomic tone, which may also dramatically affect retrograde tran- snodal conduction, would preclude useful comparisons of H-A intervals recorded at different times. Fi- nally, because radiofrequency application in the AV junction may directly stimulate postganglionic para- sympathetic and sympathetic fibers, comparison of H-A interval data during radiofrequency ablation with that recorded before or after radiofrequency application would be of little value.

The failure to induce AVjunctional activity during radiofrequency applications in the posterior AV junction region does not necessarily imply that the ablation catheter is mispositioned away from the slow-conduction pathway. As mentioned, poor tissue

Volume 131, Number 5 American Heart Journal Yu et al. 9 4 5

contact and inadequate tissue heating cannot be ex- cluded as the cause for the lack of AV junctional activity. Monitoring of tip temperature of the ablating catheter during radiofrequency delivery would have been a useful addition to this study, but this technol- ogy was not available during the period of this project.

Conclusions. Although the development of an AV junctional rhythm during selective radiofrequency ablation of slow-pathway conduction does not necessarily herald a successful ablation, the absence of induced AVjunctional activity within 15 sec of onset of radiofrequency application appears to be a sensitive indicator of an unsuccessful ablation attempt, prob- ably because of insufficient tissue heating or subop- timal catheter positioning. The AV junctional activity induced appears to exit to the atria from the anterior triangle of Koch--possibly in the retrograde fast-conduction pathway even though the radiofrequency energy is applied in the posterior triangle of Koch near the ostium of the coronary sinus. These findings support the existence of specialized tissues in the AVjunction region that allow the transmission of the effects of radiofrequency ablation from the posterior to the anterior regions of Koch's triangle.

REFERENCES

1. Goy J J, Fromer M, Sch]aapter J, Kappenberger L. Clinical efficacy of radiofrequency current in the treatment of patients with atrioventricular node reentrant tachycardia. J Am Col] Cardiol 1990;16:418-23.

2. Jackman WM, Beckman KJ, McClelland JH, Wang X, Friday K, Roman C, Moulton KP, Twidale N, Hazlitt A, Prior MZ, Oren J, Overholt ED, Lazzara R. Treatment of supraventricular tachycardia due to atrioventricular nodal reentry by radiofrequency ablation of slow-pathway conduction. N Engl J Med 1992;327:313-8.

3. Wang X, McClelland JH, Beckman KJ, et al. Accelerated junctional rhythm during slow pathway ablation [Abstract]. Circulation 1991;84 (suppl II):582.

4. Thakur RK, Klein GJ, Yee R, Stites HW. Junctional tachycardia: a useful marker during radiofrequency ablation for atrioventricular node reentrant tachycardia. J Am Coll Cardiol 1993;22:1706-10.

5. Kelly PA, Mann DE, Adler SW, Fuenzalida CE, Baitey WM, Reiter MJ. Predictors of successful radiofrequency ablation of extranodal slow pathways. PACE 1994;17:1143-8.

6. Jentzer JH, Goyal R, Williamson BD, et al. Analysis of junctional ec- tepy during radiofrequency ablation of the slow pathway in patients with atrioventricular nodal reentrant tachycardia. Circulation 1994;90: 2820-6.

7. Alison JF, Yeung-Lai-Wah JA, Schulzer M, Kerr CR. Characterization of junctional rhythm after atrioventricular node ablation. Circulation 1995;91:84-90.

8. Denes P, Wu D, Dhingra RC. Demonstration of dual AV nodal pathways in patients with paroxysmal supraventricular tachycardia. Circulation 1973;48:549-55.

9. Denes P, Wu D, Dhingra RC. Dual atrioventricular nodal pathways. A common electrophysiologic response. Br Heart J 1975;37:1069-76.

10. Sung RJ, Styperek JL, Myerburg RJ, Castellanous A. Initiation of two distinct forms of atrioventricular nodal reentrant tachycardia during programmed ventricular stimulation in man. Am J Cardiol 1978;42:404- 15.

11. Huycke EC, Lai WT, Nugyen NX, Keung EC, Sung RJ. Role ofintra-

venous isoproterenol on the electrophysiologic induction of atrioventricular node reentrant tachycardia in patients with dual atrioventricular node pathways. Am J Cardiol 1989;64:1131-7.

12. Sung RJ, Waxmaa HL, Saksena S, Juma Z. Sequence of retrograde atrial activation in patients with dual atrioventricular nodal pathways. Circulation 1981;64:1053-67.

13. Calkins H, Niklason L, Sousa J, El-Atassi R, Langberg J, Morady F. Radiation exposure during radiofrequency catheter ablation of accessory atrioventricular connections. Circulation 1991;84:2376-82.

14. Dimarco JP, Garan H, Ruskin JN. Complications in patients undergoing cardiac electrophysiologic procedures. Ann Intern Med 1982;97: 490-3.

15. Hendricks G. The Multicentre European Radiofrequency Survey (MERFS): complications of radiofrequency catheter ablation of arrhythmias. Eur Heart J 1993;14:1644-53.

16. Langberg JJ, Calkins H, El-Atassi R, et al. Temperature monitoring during radiofrequency catheter ablation of accessory pathways. Circu- lation 1992;86:1469-74.

17. Calkins H, Prystowsky E, Carlson M, Klein LS, Saul JP, Gillette P, and the Atakr Multicenter Investigators Group. Temperature monitoring during radiofrequency catheter ablation procedures using closed loop control. Circulation 1994;90:1279-86.

18. Goldreyer BN, Stephens If, Tobias SA. The origin of "junctional" rhythms seen during radiofrequency ablation of AV nodal slow pathway conduction [Abstract]. PACE 1994;17:748.

19. Nath S, Lynch C, Whayne JG, Haines D. Cellular electrophysiologi- cal effects of hyperthermia on isolated guinea pig papillary muscle: implications for catheter ablation. Circulation 1993;88(Part I): 1826-31.

20. Borggrefe M, Hindricks G, Haverkamp W, Budde T, Breithardt G. Radiofrequency ablation. In: Zipes D, Jalife J, eds. Cardiac electrophysiology: from cell to bedside. Philadelphia: Saunders, 1990: 997-1004.

21. Haines DE. Determinants of lesion size during radiofrequency catheter ablation: the role of electrode-catheter contact pressure and dura- tion of energy delivery. J Cardiovasc Electrophysiol 1991;2:509-15.

22. Haines DE, Watson DD. Tissue heating during radiofrequency catheter ablation: a thermodynamic model and observations in isolated per- fused and superfused canine right ventricular free wall. PACE 1989; 12:962-76.

23. Huang SKS. Radio-frequency catheter ablation of cardiac arrhythmias: appraisal of an evolving therapeutic modality. AM HEART J 1989;118: 1317-23.

24. Fozzard HA. Conduction of the action potential. In: Berne RM, ed. Handbook of physiology. Section 2: the cardiovascular system. Vol I: the heart. Bethesda: American Physiological Society, 1979:335-56.

25. Hodgkin AL, Rushton WAH. The electrical constants of a crustacean nerve fibre. Proc R Soc Lond 1946;133:444-79.

26. Jack JJB, Noble D, Tsien RW. Electric current flow in excitable cells. Oxford: Clarendon Press, 1975.

27. Sung RJ, Lauer MR, Chun H. Atrioventricular node reentry: current concepts and new perspectives. PACE 1994;17:1413-30.

28. Tawara S. Das Reizleitungssystem des Saugetierherzens: Jena: Gustav Fischer, 1906.

29. Truex RC, Smythe MO. Reconstruction of the human atrioventricular node. Anat Rec 1967;158:11-20.

30. James TN. Morphology of the human atrioventricular node, with remarks pertinent to its electrophysiology. AM HEART J 1961;62:756- 71.

31. Becker AE, Anderson RH. Morphology of the human atrioventricular junctional area. In: Wellens H J J, Lie KI, Janse M J, eds. The conduction system of the heart: structure, function and clinical implications. Leiden: Sterfest Kroese BV, 1976:263-86.

32. Hecht HH, Kossmann CE, Childers RW, Langendorf R, Lev M, Rosen KM, Pruitt RD, Truex RC, Uhley HN, Watt TB. Atrioventricular and intraventricular conduction--revised nomenclature and concepts. Am J Cardiol 1973;31:232-43.

33. Racker DK. Atrioventricular node and input pathways: a correlated gross anatomical and histological study of the canine atrioventricular junctional region. Anat Rec 1989;224:336-54.

34. Racker DK. Sinoventricular transmission in 10 M K + by canine atrio-


ventricular nodal inputs: superior atrionodal bundle and proximal atrioventricular bundle. Circulation 1991;83:1738-53.

35. DeMello WC, Hoffman BF. Potassium ions and electrical activity of specialized cardiac myofibers. Am J Physio] 1960;199:1125-30.

36. Janse MJ. Influence of the direction of the atrial wavefront on A-V nodal transmission in isolated hearts of rabbits. Circ Res !969;25:439- 49.

37. Janse MJ, van Capelle FJL, Anderson RH, Touboul P, Billette J. Elec- trophysielogy and structure of the atrioventricular node of the isolated rabbit heart. In: Wellens HJJ, Lie KI, Janse MJ, eds. The conduction system of the heart: structure, function and clinical implications. Leiden: Sterfert Kroese BV, 1976:296-315.

38. Meijler FL, Janse MJ. Morphology and electrophysiology of the mam- malian atrioventricular node. Physiol Rev 1988;68:608-47.

39. Zipes DP, Mendez C, Moe GK. Evidence for summation and voltage dependence in rabbit atrioventricular nodal fibers. Circ Res 1973; 32:170-7.

40. Janse MJ, Anderson RH, McGuire MA, Ho SY. "AV nodal" reentry re- visited: part I. J Cardiovasc Etectrophysiol 1993;4:561-72.

41. Hoffman BF, Cranefield PF. Electrophysiology of the heart. Mt. Kisco, NY: Futura, 1960:123.

42. Randall WC, Arde]] JL. Nervous control of the heart: anatomy and pathophysiology. In: Zipes D, Jalife J, eds. Cardiac electrophysiology: from cell to bedside. Philadelphia: Saunders, 1990:291-9.

43. Urthaler F, Neely BH, Hageman GR. Differential sympathetic-para- sympathetic interactions in sinus node and AV junction. Am J Physiol 1986;250:H43-H51.

44. Wallick DW, Felder D, Levy MN. Autonomic control of pacemaker activity in the atrioventricular junction of the dog. Am J Physiol 1978; 235:H308-13.

45. Irisawa H, Giles WR. Sinus and atrioventricular node cells: cellular electrophysiology. In: Zipes DP, Jalife J, eds. Cardiac electrophysiology: from cell to bedside. Philadelphia: Saunders, 1990:95-102.

Documents

Localization of the origin of the atrioventricular junctional rhythm induced during selective ablation of slow-pathway conduction in patients with atrioventricular node reentrant tachycardia