Physiology of "Atypical" Atrioventricular Junctional Reentrant Tachycardia Occurring Following Radiofrequency Catheter Modification of the Atrioventricular Node JEFFREY GOLDBERGER, RODNEY BROOKS, and ALAN KADISH From the Department of Medicine, Division of Cardiology, Northwestern University Medical School, Chicago, Illinois

GOLDBERGER, J., ET AL.: Physiology of "Atypical" Atrioventricular Junctional Reentrant Tachycardia Occurring Following Radiofrequency Catheter Modification of the Atrioventricular Node. The physiology 0/ atypical atriovenlricuiar junclional reentrant (achycardia fAV/RT) occurring following catheter modi/ication of the AV node is poorly defined. Six patients undergoing radio/requency current catheter modi/icotion of the AV node had inducibJe atypical AVJRT before or after AV nodal modification. Typical AVJHT was differentiated from atypical AVjHT by a ventriculoatrial (VA) time < 60 msec in the Hisbundle eiectrogram recording. Five of six patients had typical AVfRT and two had atypical AVfRT prior to AV nodal modification. FoJIowing anterior approach AV nodal modi/ication, previously undetected atypical AVJET was induced in four patients. Earliest retrograde atrial activation in the posterior septum was documented in aJJ patients with atypical AVJRT prior to modification and in three o//our patients with atypical AV/RT following modification. The AH intervals during tachycardia were 320 ± 52 msec in typical AV/HT, 88 ± 33 msec in the premodification atypical AVfHTs, and 172 ± 12 msec in the postmodification atypical AV/RTs (P - 0.0001). The AH/HA ratios were 4.1 ± 0.9 in typical AVJHT, 0.5 ± 0.2 in the premodification atypical AV/HTs, and 0.9 ± 0.2 in the postmodification atypical AVJRTs {P = 0.0001). Two pafients with postmodificafion atypical AV/HT underwent further posterior approach AV node modification that resulted in VA block. One pafient with postmodification atypical AVJRT had further anterior approach AV nodal modification that resulted in heart block. The retrograde limb of the atypical AVJRT seen following anterior approach AV nodal modification is a posterior, slow pathway. (PACE, Vol. 15, December 1992) atrioventricular nodal modification, radiofrequency ablation, atrioventricular node reentry

Introduction Since the advent of radiofrequency current catheter modification of the atrioventricular (AV] node, there have been reports of "atypical" AV junctional reentrant tachycardia (AVJRT) occurring following AV nodal modification.^-^ While several theories have been advanced to explain the

Address for reprints: Jeffrey Goidberger, M.D., Northwestern Memoriai Hospital, 250 East Superior Street, Suite 524. Chicago, IL 60611. Fax: (312) 908-6003. Received May 15. 1992; revision July 1, 1992; accepted July 16, 1992.

2270

appearance of these new tachycardias, there has been no validation of these potential mechanisms. We, therefore, report here a series of six patients who had various forms of atypical AVJRT, occurring either prior to or following radiofrequency current catheter modification of the AV node, in order to elucidate the physiology of the new tachycardias induced following AV nodal modification.

Methods Inclusion criteria for this study were: (1) a baseline electrophysiology test performed in the drug-free state at Northwestern Memorial Hospital

December 1992

PACE, Vol. 15

ATYPICAL AVJRT FOLLOWING ABLATION

between July 1990 and August 1991, and (2) the presence of inducible paroxysmal supraventricular tachycardia with the cbaracteristics of AVJRT and a ventriculoatrial (VA) time > 60 msec in the His bundle recording ("atypical" AVJRT) at eitber tbe baseline electrophysiology study or after anterior approach AV nodal modification. Six consecutive patients who met tbe inclusion criteria form tbe study population; during the study period 22 patients witb AVJRT were evaluated. No patient bad any evidence of preexcitation or a concealed accessory patbway. All patients underwent radiofrequency current catbeter modification of tbe AV node. Five patients were male and one was female, Their mean age was 54 ± 24 years (range 23-79 years). All patients bad symptomatic episodes of supraventricular tacbycardia. Medical tberapy bad failed in five of tbe six patients. Tbe sixtb patient presented witb syncope and opted for ablative tberapy. Radiofrequency current catbeter modification of tbe AV node was approved by tbe Nortbwestern University Committee on Human Researcb. All patients provided written informed consent prior to tbe procedure. Electrophysiology Study All patients underwent standard baseline electropbysiology studies in tbe nonsedated, fasting state after cardiac medications bad been discontinued for at least 5 half-lives. Tbree qnadripolar catbeters (6 Frencb [USCI, Division of CR Bard. Billerica, MA, USA] or 7 Frencb [Mansfield-Webster, Boston Scientific Corp., Watertown, MA, USA]) were positioned in tbe bigb rigbt atrium. His bundle region, and rigbt ventricle through sheatbs placed in tbe femoral vein. A multipolar catbeter was positioned in tbe coronary sinus in eight of tbe nine electropbysiology studies during wbicb supraventricular tachycardia was inducible. Intracardiac electrograms were recorded with bandpass filters of 40 to 400 Hz. Electrocardiographic leads I, II, III, VI, V6, and the intracardiac electrograms were displayed on an oscilloscope and recorded at a paper speed of 100 to 200 mm/ sec (Astromed MT-96000, West Warwick, RI, USA). Pacing was performed witb a programmable stimulator (Bloom Associates, Reading, PA, USA) using rectangular stimuli at twice the diastolic tbresbold witb a stimulus duration of 2 msec. AV

PACE, Vol. 15

nodal refractory periods were determined using the extrastimulus technique. Incremental atrial and ventricular pacing were performed to determine tbe longest cycle lengtb at which one to one AV and VA conduction, respectively, was no longer preserved (Wenckebacb cycle lengtb). Standard tecbniques for the induction of supraventricular tacbycardia were used."" If supraventricular tacbycardia could not be induced in the baseline state, intravenous isoproterenol at a dose of 1 to 2 fjLg/min was begun and tacbycardia was induced with programmed stimulation. During supraventricular tachycardia, diastole was scanned with single ventricular extrastimuli from the rigbt ventricular apex or septum at 10msec decrements until ventricular refractoriness was reacbed, in order to examine tbe ability to affect tbe atrial electrogram. Two ventricular extrastimuli were applied if single ventricular extrastimuli failed to advance tbe atrial electrogram. Definition of Induced Arrhythmias Typical AVJRT was diagnosed wben atrial activation occurred simultaneously witb ventricular activation;^ tbis was considered to occur wben tbe VA time in tbe His bundle recording was < 60 msec during tacbycardia,** In addition, these tachycardias were characterized by earliest retrograde atrial activation in the His bundle recording and tbe requirement of a critically prolonged AH interval for tacbycardia induction. Atypical AVJRT was suspected wben tbe VA time during tacbycardia was > 60 msec in tbe His bundle recording and earliest retrograde atrial activation was noted along tbe interatrial septum. Tbese tacbycardias were differentiated from atrial tacbycardia by: 1. spontaneous termination of tbe tacbycardia witb an atrial electrogram; 2. termination of tbe tachycardia with single or double premature ventricular extrastimuli witbout retrograde conduction to the atrium. These tacbycardias were differentiated from orthodromic AV reentrant tacbycardia utilizing a septal bypass tract by: 1. the presence of intermittent 2:1 AV block; 2. inability of premature ventricular extrastimuli to advance tbe succeeding atrial electrogram/-^ Tbough this criterion does not com-

December 1992

2271

GOLDBERGER, ET AL.

pletely exclude orthodromic AV reentrant tachycardia in combination with the other findings, the tacbycardias induced were not likely ortbodromic AV reentrant tachycardia. Radiofrequency Current Catheter Ablation Procedure All ablation procedures were performed using a large tip (4 mmj steerable 7 French quadripolar catbeter with eitber 2 or 5 mm interelectrode distance (Mansfield-Webster). Tbe endpoint of radiofrequency ablation was the inability to induce sustained supraventricular tachycardia. Radiofrequency energy was delivered as a continuous unmodulated sine wave output from an electrosurgical unit at 350 kHz (Radionics RFCSB Lesion Generator, Burlington, MA, USA), Two techniques for ablation were used for the patients in this study: 1. anterior or "fast pathway" approach.' In tbis approacb tbe catbeter was initially positioned in tbe His bundle position to record a large His bundle electrogram. The catheter was then withdrawn slowly witb clockwise torque until the largest possible atrial to ventricular electrogram ratio in tbe presence of a small His bundle electrogram was obtained. Using tbis approacb, radiofrequency energy was applied between tbe distal electrode of tbe ablation catbeter and a large skin electrode using graded doses of energy. Initially, 20 volts (V) were applied (approximately 4 watts [WJ). If no evidence of beart block was noted, the voltage was increased to 30 V and, if tolerated, to 40 V. If no evidence of beart block was noted, energy was delivered for a total of 60 seconds. Tbe occasional higb number of reported energy deliveries are related to counting eacb energy level as a separate application. Energy delivery was stopped immediately wben an impedance rise was observed and the catbeter was removed and any adberent coagulant wiped off. Four patients were initially treated witb tbis approach. 2. Posterior or "sJovv pathway" approach.^ In this approach tbe catbeter was positioned near the coronary sinus os or sligbtly anterior to it. Wben possible, tbe site of earliest retrograde atrial activation during retrograde conduction over tbe "slow" AV nodal patbway (during tacbycardia or ventricular pacing) was identified for ablation. All abla-

2272

tions using this approacb were done well away from tbe His bundle (below the mid-point separating tbe catheter recording tbe His bundle potential and tbe catheter in tbe coronary sinus). Using this approacb, radiofrequency energy was applied between tbe distal electrode and a large skin electrode using 40 to 60 V for 30 to 60 seconds or until an impedance rise was observed. If an impedance rise occurred, tbe catbeter was removed and any adherent coagulant wiped off. Data Analysis Measurements made during tacbycardia included tbe tacbycardia cycle length, the AH interval, the HA interval, and the VA times at various sites along tbe interatrial septum, tbe coronary sinus, and tricuspid annulus. All measurements were made by a single observer (JC) to tbe nearest 5 msec. Data are presented as mean ± standard deviation. Tbe characteristics of tbe typical AVJRTs, tbe atypical AVJRTs noted prior to ablation, and tbe atypical AVJRTs induced following anterior approacb AV nodal modification were compared using an analysis of variance. Scbeffe's procedure was used to compare tbe differences among tbe three groups wben a significant result was obtained. A P value < 0.05 was considered statistically significant.

Results Four patients had dual anterograde AV nodal physiology. Four patients bad typical AVJRT, one patient had atypical AVJRT, and tbe final patient bad typical and atypical AVJRT induced prior to AV node modification. Following anterior AV node modification, previously undetected atypical AVJRT was induced in four patients. All tacbycardias were sustained and reproducibly initiated (at least twice). Table I demonstrates tbe cbaracteristics of tbe induced tacbycardias. Cumulative data by tacbycardia type are demonstrated in Table II. Table III demonstrates the baseline AH interval, AV node Wenckebacb cycle lengtb, and VA Wenckebach cycle lengtb prior to AV node modification and following tbe final AV node modification.

December 1992

PACE, Vol, 15

ATYPICAL AVJRT FOLLOWING ABLATION

Table I. Characteristics of the Typical and Atypicai AV Junctional Reentrant Tachycardias Induced Prior To and Following AV Node Modification in Six Patients Patient

Study #

1

1 2 1 2 1 1 1 1 1 1 1 1 2

2 3 4 5

6

Induced by A V A V Cath V Spont A V A A V V

ISO

+ -1—

-1-

+ + -1-

Mechanism TYP ATYP" TYP ATYP* TYP ATYPATYP TYP ATYP ATYP ATYP TYP ATYP*

AH/HA

CL (msec)

AH (msec)

HA (msec)

VA/HBE (msec)

4.6 0.6 3.3 1.1 3.4 0.8 0.4 4.0 0.2 0.5 0.6 5.4 10

450 450 390 310 370 360 320 330 310 250 290 460 370

370 170 300 165 285 165 90 265 50 80 130 380 190

80 280 90 145 85 195 230 65 260 170 160 80 180

20 210 30 100 40 150 200 25 210 110 110 20 110

VA/post (msec) 35 40 85 135 180 195 90 95 40 80

' Tachycardias that were induced following AV node modification. A = atrial pacing; AH = AH interval; HA = HA interval; V = ventricular pacing; Cath = catheter induced; Spont = spontaneous; Iso = isoproterenoi; AV = atrioventricular; TYP = typical AV junctional reentrant tachycardia (AVJRT); ATYP = atypical AVJRT: CL = tachycardia cycle length; VA/HBE = VA time in His bundle eiectrogram recording; VA = ventriculoatrial; VA/post = VA time recorded posteriorly near coronary sinus.

Typical AVJRT Five patients had inducible typical AVJRT at their initial electrophysiology study. During typical AVJRT the mean VA time in the His hundle recording was 27 ± 8 msec (range 20-40 msec). The earliest atrial activation was documented to

be in the His bundle eiectrogram in thee out of the five tachycardias [a coronary sinus catheter was not in place during the other two tachycardias). These tachycardias were induced by atrial pacing that resulted in a prolonged AH interval (250 to 320 msec) prior to initiation of typical AVJRT in four out of five patients. Aside from meeting the

Table M.

Characteristics of the Typical and Atypical AVJRT Induced Prior To and Following AV Node Modification in Six Patients Tachycardia Typical AVJRT (N - 5) Atypical AVJRT; Pre AV node mod (N = 4) Post AV node mod (N - 4)

CL (msec)

AH/HA

AH (msec)

400 ± 55

4.1 ± 0.9

320 ± 52

292 ± 3 1 ' 372 ± 58

0.5 ± 0.2" 0.9 ± 0.2"

88 ± 3 3 " ,t 172 ± 1 2 "

Earliest A ANT POST POST (3/4)

In one patient with atypical AVJRT following AV node modification, recordings were not made posteriorly along the atrial septum. The site of earliest retrograde atrial activation during tachycardia is, therefore, not known. * P < 0,05, compared to typical AVJRT; " P < 0.001, compared to typical AVJRT; ^ P < 0.05, compared to post AV node mod. ANT = anterior in the HBE recording; AVJRT = atrioventricular junctional reentrant tachycardia; CL = tachycardia cycle length; Earliest A = earliest retrograde atrial activation during tachycardia; POST = posterior near the coronary sinus; Post AV node mod = AVJRT induced folloviring AV node modification; Pre AV node mod = AVJRT induced prior to AV node modification.

PACE, Vol. 15

December 1992

2273

GOLDBERGER, ET AL.

Figure 1. Simultaneous recordings of surface leads I, II. Ill, Vl, and V6. with intracardiac recordings from the high right atrium (HRA), His-hundle electrograms (HBE), proximal coronary sinus (CSp), and distal coronary sinus (CSdJ. and right ventricular apex (RVA). All tracings are at a (continued next pagej

2274

December 1992

PACE, Vol. 15

ATYPICAL AVJRT FOLLOWING ABLATION

Table III. Electrophysiological Characteristics During Baseline Electrophysioiogical Testing Prior to AV Node Modification and Following the Final AV Node Modification Procedure Final

Baseline Patient

AH (msec)

AVNWCL (msec)

VAWCL (msec)

AH (msec)

AVNWCL (msec)

VAWCL (msec)

1 2 3 4 5 6

130

400 350 360 240 380 400

380 340 400 220 280' 330

HB 370 160 70 80 170

HB 520 400 350 390 400

HB No VA cond 420 510 500* 700

100 90 80 90 60

At the time of the final measurements, supraventricular tachycardia was not inducible except in patients 3 and 6 who had inducible atypical AV junctional reentrant tachycardia, ' All measurements except these were made without isoproterenol. AH = AH interval; AV = atrioventricular; AVNWCL = AV node Wenckebach cycle length; HB = heart block; No VA cond = no VA conduction was detected; VA = ventriculoatrial; VAWCL = VA Wenckebach cycle length.

criteria for typical AVJRT described in the methods section, spontaneous tachycardia termination with an atrial electrogram was observed in patient 5 and termination of tachycardia with two ventricular extrastimuli without retrograde conduction to the atrium was observed in patient 1. Atypical AVJRT Pre-AV Node Modification

Patient 4 initially had only atypical AVJRT during his baseline electrophysiology study. This tachycardia was differentiated from orthodromic AV reentrant tachycardia hy the presence of intermittent 2:1 AV block. Although this tachycardia was not conclusively differentiated from an atrial tachycardia, the AH interval during sinus rhythm was 10 msec greater than the AH interval during tachycardia, which suggests that an atrial tachycardia is unlikely. At the time of his electrophysiology study, the tachycardia could be terminated only briefly before spontaneous reinitiation. Following posterior AV nodal modification, the tachycardia was not inducible.

Patient 5 had inducible typical AVJRT and three different atypical AVJRTs (Fig. 1, Table I). Earliest retrograde atrial activation occurred near the coronary sinus os in all episodes of atypical AVJRT. The first two atypical AVJRTs differed predominantly in the retrograde limb. The first atypical AVJRT (Fig. IB) was slower (cycle length 310 msec vs 250 msec) and had a longer VA time (195 msec vs 90 msec). The second atypical AVJRT (Fig. lC) had a retrograde activation sequence and VA time (90 msec) that was identical to the third atypical AVJRT (Fig. ID). They differed in that the AH interval (the anterograde limb) was shorter (80 msec vs 130 msec) in the second atypical AVJRT, despite the fact that the tachycardia cycle length was shorter, suggesting two different anterograde conduction pathways in the AV node. This finding also makes atrial tachycardia an unlikely mechanism for these two tachycardias because an atrial tachycardia with the same activation sequence should not have a shorter AH when the cycle length is shorter. Premature extrastimuli did not

paper speed of 100 mm/sec. There were/our di/^erent types of inducible AV junctional reentrant tachycardias (AVJRT) in patient 5. (A) Typical AV junctionai reentrant tachycardia (AVJRT). (Bj Atypical AVJRT. Note that the earliest retrograde afriaJ activation occurs in the CSp (arrow). (C) A second atypical AV/RT with a different retrograde pathway manifested by a shorter VA time. The CS catheter during this recording was not near the os. (D) A third atypicai AV/RT. See text for details. TCL ^ tachycardia cycle length.

PACE. Vol. 15

December 1992

2275

GOLDBERGER. ET AL.

Figure 2. SimuJfaneous recordings of surface leads I, II, III, Vl and V6, with intracardiac recordings from the high right atrium (HRA), His-bundJe electrograms fHBEJ, proximal coronary sinus (CSp), distal coronary sinus (CSd), and right ventricular apex (RVA). All tracings are at a paper speed of 100 mm/sec. These tachycardias were induced in patient 2. (A) Typical AV junctional reentrant tachycardia (AVJRT}. (B) Induction of atypical AVJRT by ventricular overdrive pacing. Note the long VA time required to initiate tachycardia. fC) Atypical AVJRT. The earliest retrograde atrial activation occurs in the CSp recording farrow). See text for details.

Figure 3. Simultaneous recordings of surface leads I, II. Ill, Vl, and V6, with intracardiac recordings from the high right atrium fHRA), His-bundle electrograms (HBEj. proximal coronary sinus (CSp), distal coronary sinus (CSd), and right ventricular apex (RVAj. Recordings A and B were made at a paper speed of 100 mm/sec and recordings C and D were made at a paper speed of 50 mm/sec. These recordings were made during ventricular pacing in patient 2. (A) VA conduction at a ventricular paced cycle length of 400 ms prior to any radiofrequency current catheter modification of the AV node. Note the short VA time. fB) VA conduction at a ventricular paced cycle length of 400 ms during isoproterenoi infusion following anterior approach radiofrequency current catheter modification of the AV node. Note the longer VA times compared to the tracing in A. (C) Intermittent VA conduction at a ventricular paced cycle length of 750 ms during infusion of isoproterenoi following anterior and posterior AV nodal modification. The first paced ventricular beat is conducted retrograde with earliest atrial activation in the proximal coronary sinus recording (arrow). There is no retrograde conduction from the second paced ventricular beat. The last atrial beat (asterisk) is a sinus beat. (D) There is no VA conduction at a ventricular paced cycle length of 500 ms during isoproterenoi infusion following anterior and posterior AV nodal modification. All atrial beats are sinus beats (*}.

2276

December 1992

PACE, Vol. 15

ATYPICAL AVJRT FOLLOWING ABLATION

A

i;i|iiiiiiinMiiiilili|(iiiiiill|liMiiiii|iilliriii|iiiiiiiii|Mi

lOOms :|iiiiiiiii|ilTiiiTi'i|iin

|iiii

ij

iiii|iiiiiiiii|

B

100 ms

Figure 3.

PACE, VoL 15

December 1992

2277

GOLDBERGER, ET AL.

advance the following atrial eiectrogram during the third atypical AVJRT. This tachycardia was initiated with atrial pacing and required a long AH interval (210 msec] for initiation. Because the retrograde activation sequence and the VA times of the second atypical AVJRT were identical to those of the third atypical AVJRT and since retrograde conduction during the third atypical AVJRT was not likely due to an accessory pathway, it is unlikely that retrograde conduction during the second atypical AVJRT was due to an accessory pathway. The first atypical AVJRT was differentiated from a tachycardia utilizing an accessory AV pathway by the presence of intermittent 2:1 AV hlock. This tachycardia was differentiated from an atrial tachycardia by termination of the tachycardia with a premature ventricular extrastimulus that did not conduct retrograde to the atrium. This tachycardia was only initiated with ventricular pacing and required a long VA conduction time for initiation (stimulus to A interval of 245 msec). In summary, two patients were documented to have a total of four atypical AVJRTs prior to AV node modification. Earliest retrograde atrial activation occurred posteriorly, near the coronary sinus OS, in all four of these tachycardias (Tables I and II, Fig. IB). In one patient, tachycardia was eliminated by posterior approach AV nodal modification. The other patient with typical AVJRT and three types of atypical AVJRT initially underwent posterior AV modification that was not completely successful in abolishing all of his tachycardias. He, therefore, underwent anterior approach AV node modification as well and has had no inducible or spontaneous supraventricular tachycardia. Atypical AVJRT Following AV Node Modification Patient 1 initially had inducible typical AVJRT and underwent anterior AV node modification. At the end of the study, supraventricular tachycardia could not be induced. On follow-up electrophysiology testing 4 days later, atypical AVJRT was induced. Premature extrastimuli did not advance the following atrial eiectrogram. This tachycardia was not conclusively differentiated from an atrial tachycardia. However, it was only initiated with ventricular pacing and required a

2278

long VA conduction time for initiation (stimulus to A interval of 410 msec). This patient underwent further anterior AV node modification, which resulted in second-degree AV block. This ultimately progressed to complete heart block and the patient required a permanent pacemaker for symptomatic bradycardia. Patient 2 initially had inducible typical AVJRT (Fig. 2A) and underwent anterior AV node modification. At the end of the study, supraventricular tachycardia could not be induced. On follow-up electrophysiology testing 2 days later, atypical AVJRT was induced [Figs. 2B and 2C). This tachycardia was differentiated from a tachycardia utilizing an accessory bypass tract by the presence of intermittent 2:1 AV block. This tachycardia was differentiated from an atrial tachycardia by spontaneous termination with an atrial eiectrogram. This tachycardia was only initiated with ventricular pacing (Fig. 2B] and required a long VA conduction time for initiation (stimulus to A interval of 300 msec). This patient underwent posterior approach AV node modification near the coronary sinus os. After application of 11 radiofrequency current lesions, there was marked prolongation of the AV and VA conduction times. At follow-up electrophysiology study 3 days later, no tachycardia was inducible and there was no VA conduction without isoproterenoi. Figure 3A demonstrates the baseline VA conduction presumably secondary to fast pathway conduction prior to AV node modification. Following anterior approach AV node modification, VA conduction persisted (Fig. 3B) but with a much longer VA time. Although no VA conduction was noted at baseline following anterior and posterior AV nodal modification, when isoproterenoi was administered at 2 (xg/min, intermittent VA conduction was noted (Fig. 3C). Figure 3D demonstrates that there is no VA conduction at a paced cycle length of 500 msec. Figure 4 demonstrates the AV node function curves in this patient. Dual AV node physiology is not demonstrated on the initial study. Following anterior AV node modification, the A2H2 curve is shifted upward and to the right. Following anterior and posterior AV node modification, the AV node function curve is shifted further upward and to the right.

December 1992

Patient 3 initially had inducible typical

PACE, Vol. 15

ATYPICAL AVjRT FOLLOWING ABLATION

o INITUL

o INITIAL

350-1

a ANT

a ANT 300-

A ANT+POST

A2H2 (ms)

3iro-

400-

A

a D

250200-

•

A2H2

150-

• 300-

100200

oc

200-

QifcQD

100-

350 450 550 650 750

A1A2 (ms) Figure 4. AV nodal function curve (A2H2 versus Al A2) from patieni 2 obtained prior to AV node modi/icalion fINITIALJ, following anterior approach AV nodal modification {ANT}, and foiloiving anlerior and posterior AV nodal modification (ANT+POSTJ. The first two AV nodal function curves were obtained at a drive cycle length of 550 ms. The final AV nodal function curve was obtained at a drive cycle length of 700 ms.

AVJRT and underwent anterior AV node modification. Following seven applications of radiofrequency energy, typical AVJRT was no longer inducible. However, atypical AVJRT was induced. Premature extrastimuli did not advance the following atrial electrogram. This tachycardia was differentiated from an atrial tachycardia by termination of the tachycardia with a premature ventricular extrastimulus that did not conduct retrograde to the atrium. This tachycardia was only initiated with ventricular pacing and required a long VA conduction time for initiation (stimulus to A interval of 300 msec). Posterior approach AV node modification was then performed slightly anterior to the coronary sinus os. After the application of six radiofrequency current lesions, there was no VA conduction. On follow-up electrophysiology testing 1 month later, VA conduction had recovered and atypical AVJRT was induced with the same cycle length and VA time as during the initial study. As the patient had not had any symptomatic recurrences, no further AV node modification was

PACE. Vol. 15

400

500

600

A1A2 (ms)

On

250

300

Figure 5. AVnodai function curve (A2H2 versus A3A2) from patient 3 obtained prior to AV node modification (INITIAL), and foilowing anterior and posterior approach AV nodal modification fANTj. The latter AV nodal function curve was obtained during the patient's foiiow-up eiectrophysiology study when atypical AVJRT was induced (identical to atypical AV/RT induced foliowing anterior approach AV nodal modification). Both AV nodal function curves were obtained at

a drive cycle length of 550 ms. Note the presence of duaJ AV nodal physioiogy at baseline and following anterior approach AV nodal modification.

performed. Figure 5 demonstrates the AV node function curves in this patient. Dual AV node physiology is demonstrated on the initial study. During the follow-up study, the A2H2 curve is shifted upward and to the right and dual AV node physiology persists. Patient 6 initially had inducible typical AVJRT and underwent anterior AV node modification. Following one application of radiofrequency energy, which resulted in an increase in the PR interval from 140 to 220 msec, typical AVJRT was no longer inducible. Nonsustained episodes of atypical AVJRT were inducible. On follow-up electrophysiological testing 2 days later, sustained atypical AVJRT was induced. Premature extrastimuli did not advance the following atrial electrogram. This tachycardia was not conclusively differentiated from an atrial tachycardia. This tachycardia was initiated only with ventricular pacing (stimulus to A interval of 325 msec). Posterior approach AV nodal modification was ahorted secondary to chest pain. In summary, atypical AVJRT was initiated in

December 1992

2279

GOLDBERGER, ET AL.

four patients following AV node modification. Earliest retrograde atrial activation was documented to occur posteriorly, near the coronary sinus OS, in three of four of these tachycardias. Further anterior AV node modification in one patient resulted in complete heart block. Further posterior AV nodal modification in two patients guided by the earliest retrograde atrial activation during atypical AVJRT resulted in loss of VA conduction. Comparison of Tachycardia The anterograde conduction times among the pre- and postmodification atypical AVJRTs and the typical AVJRTs were significantly different (P = 0.0001; Table 11). The AH interval during the premodification tachycardias (88 ± 33 msec] was significantly shorter than the AH interval during the postmodification tachycardias (172 ± 12 msec; P < 0.05). The HA interval during typical AVJRT (78 ± 10 msec) was significantly shorter than the HA interval during the premodification AVJRTs (199 ± 46 msec; P < 0.005) and the postmodification AVJRTs (208 ± 7 msec; P < 0.005). There was no significant difference between the HA intervals in the pre- and postmodification atypical AVJRTs. Because the atypical AVJRTs demonstrated earliest retrograde atrial activation in the posterior atrial septum (near the coronary sinus os) associated with long HA conduction times, the retrograde limb for these tachycardias is a slow AV nodal pathway.'°'^^

Discussion Six patients with atypical varieties of AVJRT were presented. The retrograde limh of these tachycardias was identified as a slow AV nodal pathway because the earliest retrograde atrial activation occurred posteriorly near the coronary sinus OS. These tachycardias were noted in some patients prior to and in other patients following radiofrequency current catheter modification of the AV node. The characteristics of the retrograde limb of the atypical AVJRTs that occur following anterior approach AV node modification are similar to those of the premodification atypical AVJRTs. Furthermore, posterior approach AV node modification of two of the postmodification

2280

tachycardias resulted in VA block. These findings confirm that the retrograde limb of these tachycardias is a posterior, "slow" AV nodal pathway and not a damaged fast AV nodal pathway. There are several reports,^^"^^ describing atypical forms of AVJRT unrelated to AV nodal modification. In 1977, Wu et al.,^'* described three patients with retrograde dual AV nodal physiology. One of these patients had inducible "fast—slow" AVJRT with an AH/HA ratio of 0.21. Sung et a!.^^ also described five cases of atypical AVJRT. These patients had tachycardias with an AH/HA ratio < 1. However, the characteristics and activation sequence of these tachycardias were not descrihed. McGuire et al.^'' descrihed three patients with both typical and atypical AVJRT. During atypical AVJRT, earliest retrograde atrial activation occurred near the coronary sinus os. The premodification atypical AVJRTs noted in this study have similar characteristics to the atypical AVJRTs in these previous reports. Since the advent of radiofrequency AV node modification, there have been reports of atypical varieties of AVJRT occurring following AV node modification.^"^ Several hypotheses have been suggested to explain the appearance of these new arrhythmias postablation. Langberg et al.^ reported nine patients with atypical AVJRT following AV node modification. The AH/HA ratios for the atypical AVJRTs was not presented. However, the mean tachycardia cycle length was 357 msec and the mean AH interval was 140 msec, similar to the characteristics of the postmodification tachycardias in this study. They postulated several mechanisms for these tachycardias: (1) persistent atypical AVJRT with retrograde upper common pathway delay; (2) persistent typical AVJRT with longer ventriculoatrial conduction times due to a damaged fast pathway; (3) atypical AVJRT with retrograde conduction via a slow AV nodal pathway and anterograde conduction via a fast AV nodal pathway; and (4) atypical AVJRT with anterograde and retrograde conduction via slow AV nodal pathways. Similar hypotheses were considered in the report by Lee et al.^ Based on the findings of the present study, it seems unlikely that the retrograde limb of the atypical AVJRTs occurring following anterior approach AV node modification could be a damaged or modified fast AV nodal pathway or upper common pathway. It is unlikely

December 1992

PACE, Vol. 15

ATYPICAL AVJRT FOLLOWING ABLATION

for anterior AV node modification to alter the sequence of activation of the fast pathway from anterior to posterior. Furthermore, posterior AV node modification in the patients with atypical AVJRT following anterior AV node modification resulted in VA block, confirming that there was no retrograde fast pathway conduction (retrograde fast pathway conduction should not have been abolished by posterior ablation). Thus, the retrograde limb of these atypical AVJRTs is a slow AV nodal pathway. This rules out mechanisms 1 and 2. The anterograde pathway in the postmodification AVJRTs is uncertain. The AH intervals were intermediate between those of typical AVJRT (when the anterograde limb is presumably a slow pathway) and the premodification atypical AVJRTs [when the anterograde limb is presumably a fast pathway). We have shown that the anterograde conduction properties of these tachycardias are significantly different than the anterograde conduction properties of the preablation atypical AVJRTs. Thus, although the anterograde limb may be either a slow pathway or a "damaged" fast pathway, its conduction characteristics are different than those typical of the fast AV nodal pathway. The anterograde conduction times for the atypical AVJRTs occurring following AV nodal modification were also significantly shorter than those for typical AVJRT. This may be explained by the use of a slow AV nodal pathway with different properties than the slow AV nodal pathway utilized in typical AVJRT. Alternatively, the conduction times during typical AVJRT were measured without isoproterenoi in four out of five cases, whereas isoproterenoi was required to induce the postmodification atypical AVJRTs in all cases. Isoproterenoi likely affects the conduction time in the slow AV nodal pathway significantly. Thus, the anterograde limb of these tachycardias may still be a slow AV nodal pathway. These tachycardias may then represent slow—slow AVJRT.^^ If so, at least four of our patients had more than one slow pathway. During the study period. 22 patients with AVJRT were evaluated. This gives an incidence of muhiple slow pathways of at least 18%. This is consistent with the findings of Beckman^^ who reported the incidence of two to four slow pathways to be at least 31%. There are other reports^^'^^"^^ that also indicate the presence of more than two AV nodal pathways. It is, therefore, not surprising

PACE, Vol. 15

that occasionally atypical AVJRT using two slow AV nodal pathways can be induced, particularly after fast pathway ablation. Slow—slow AVJRT may not be noted premodification because retrograde conduction occurs preferentially over the fast pathway. Limitations The atypical AVJRTs occurring following AV node modification in patients 1 and 6 were not conclusively differentiated from an atrial tachycardia. However, these tachycardias were induced only with ventricular pacing, which suggests that atrial tachycardia is unlikely. A tachycardia utilizing a concealed posteroseptal accessory pathway with AV nodallike properties could only be definitively excluded in patients 2, 4. and 5 in whom intermittent 2:1 AV block was noted during tachycardia. The clinical manifestations of the atypical AVJRTs in the other patients were not characteristic of this type of tachycardia.^^ Implications for Catheter Modification of the AV Node There are reports of successful AV node modification using the anterior^ and posterior approaches.^ Following AV node modification, atypical AVJRT may be induced in a minority of patients. The clinical implications of the induction of atypical AVJRT following anterior approach AV node modification are unclear. In the report by Langberg et al.^ only one patient out of nine had a clinical occurrence of slow—slow AVJRT and on follow-up electrophysioiogy study, this tachycardia was not inducible. We have seen one patient with recurrent symptomatic atypical AVJRT following anterior approach AV node modification. If atypical AVJRT does require further AV node modification, the approach taken should be carefully considered. While the retrograde limh of these tachycardias has been shown to be a slow AV nodal pathway, the anterograde limb may be either a slow AV nodal pathway or a damaged fast AV nodal pathway. Because the anterograde limb of these tachycardias may be a slow pathway, anterior approach AV node modification may not cure the tachycardia while maintaining AV conduction. Posterior approach AV node modification may be used with the ablation site being deter-

December 1992

2281

GOLDBERGER, ET AL.

mined by the earliest retrograde atrial activation during tachycardia. While posterior approach AV node modification was successfully used in our patients who had already undergone anterior ap-

proach AV node modification without producing AV block, the risks of performing AV node modification at both sites must be carefully considered and discussed with the patient.

References 1. Lee M, Morady F, Kadish A, et al. Catheter modification of the atrioventricular junction with radiofrequency energy for control of atrioventricular nodal reentry tachycardia. Circulation 1991; 83(3): 827-835. 2. Langberg J, Kou W, Calkins H, et al. Conversion from typical to "atypical" AV nodal reentry tachycardia (AVNRT) after radiofrequency catheter modification of the AV junction, (ahstract) J Am Coll Cardiol 1991; 17(2):337A. 3. Langberg J, Kim Y, Goyal R, et ai. Conversion of typical to "atypical" atrioventricular nodal reentrant tachycardia after radiofrequency catheter modification of the atrioventricular junction. Am J Cardiol 1992; 65:503-508. 4. Leitch}, Klein C. Yee R, et al. Invasive electrophysiologic evaluation of patients with supraventricular tachycardia. Cardiol Clinics 1990; 8{3]: 465-477. 5. Epstein L, Scheinman M, Langberg I, et al. Percutaneous catheter modification of the atrioventricular node. Circulation 1989; 80:757-768. 6. Benditt D, Iritchett E, Smith W, et al. Ventriculoatrial intervals: Diagnostic use in paroxysmal supraventricular tachycardia. Ann Intern Med 1976; 91(2):161-166. 7. Miles W, Yee R, Klein C, et al. The preexcitation index: An aid in determining the mechanism of supraventricuiar tachycardia and localizing accessory pathways. Circulation 1986; 74(3):493-5G0. 8. Coldberger |, Wang Y, Scheinman M. Stimulation of the summit of the right ventricular aspect of the ventricular septum during orthodromic atrioventricular reentrant tachycardia. Am J Cerdiol 1992; 70:78-85. 9. Roman C, Wang X. Friday K, et al. Catheter technique for selective ablation of slow Pathway in AV nodal reentrant tachycardia, (abstract) PACE 1990; 13:498. 10. Sung R, Waxman H, Saksena S, et al. Sequence of retrograde atrial activation in patients with dual atrioventricular nodal pathways. Circulation 1981; 64:1059-1067. 11, Dhala A, Sra J, Blanck Z, et al. Markedly different sequence of atrial activation during retrograde conduction over "fast" and "slow" pathways in patients with co-existent common and uncommon atrioventricular nodal reentry: Isn't the atrium a

2282

12. 13. 14.

15.

16.

17.

18.

19.

20.

21.

22.

December 1992

necessary part of reentrant circuit? (abstract) Circulation 1991; 84(4):II-235. Kuhlkamp V, Haasis R, Seipel L. AV nodal reentrant tachycardia using three different AV nodal pathways. Eur Heart J 1990; ll(9);857-862. Strasberg B. Swiryn S, Bauernfeind R, et al. Retrograde dual etrioventricular nodal pathways. J Am Coll Cardiol 1981; 48:639-646. Wu D, Denes P. Amat-y-Leon F, et al. An unusual variety of atrioventricular nodal reentry due to retrograde dual atrioventricular nodal pathways. Circulation 1977: 56:50-59. Sung R, Styperek ], Myerburg R, et al. Initiation of two distinct forms of atrioventricular nodal reentrant tachycardia during programmed ventricular stimulation in man. J Am Coll Cardiol 1978: 42: 404-415. McGuire M, Lau K, Johnson D, et al. Patients with two types af atrioventricutarjunctional (AV nodal) reentrant tachycardia Circulation 1991; 83: 1232-1246. Baerman }, Wang X, Jackman W. Atrioventricular nodal reentry with an antegrade slow pathway and retrograde slow pathway: Clinical and electrophysiologic properties, (abstract) J Am Coll Cardiol 1991; 17(2):338A. Beckman K, Jackman W, McClelland J, et al. Evidence for multiple slow pathways in patients with atrioventricular nodal reentrant tachycardia, (abstract) Circulation 1991; 84(4):II-235. Ruder M, Mead R, Smith N, et al. Comparison of pre- and postoperative conduction patterns in patients surgically cured of atrioventricular node reentrant tachycardia. J Am Coll Cardiol 1991; 17(2):397-402. Sebag C, Chevalier P, Davy J, et al. Triple antegrade nodal pathway in a patient with supraventricular paroxysmal tachycardia } Electrocardiol 1986; 19(l):85-90. Swiryn S, Bauernfeind R, Palileo E, et al. Electrophysiologic study demonstrating triple antegrade AV nodal pathways in patients with spontaneous and/or induced supraventricular tachycardia. Am Heart I 1982; 103:168-176. Critelli G, Gallagher J, Monda V, et al. Anatomic and electrophysiologic substrate of the permanent form of junctional reciprocating tachycardia. J Am Coll Cardiol 1984; 4(3):601-610.

PACE, Vol. 15