br perinodal ablationof nodal reentrant efficacy and success

Br HeartJ 1995;74:268-276

Perinodal slow potential as a local guide fortranscatheter radiofrequency ablation ofatrioventricular nodal reentrant tachycardia:therapeutic efficacy and electrophysiologicalmechanisms of success

Jiunn-Lee Lin, Fang-Yue Lin, Huey-Ming Lo, Chuen-Den Tseng, Tin-Fu Cheng,Jin-Jer Chen, Yung-Zu Tseng, Wen-Pin Lien

Division ofCardiology,Department ofInternal Medicine,National TaiwanUniversity Hospital,Taipei, Taiwan,Republic ofChinaJ-L LinC-D TsengJ-J ChenY-Z TsengW-P LienDivision ofCardiovascularSurgery, Departmentof Surgery, NationalTaiwan UniversityHospital, Taipei,Taiwan, Republic ofChinaF-Y LinDepartment ofMedicine, ProvincialTaipei Hospital,Taipei, Taiwan,Republic ofChinaT-F ChengDepartment ofMedicine, ProvincialTao-Yuan Hospital,Tao-Yuan, Republic ofChinaH-M Lo

Correspondence to:Dr W-P Lien, Division ofCardiology, Department ofInternal Medicine, NationalTaiwan University Hospital,No 7 Chun-Shan SouthRoad, Taipei, Taiwan 100,Republic of China.

Accepted for publication13 March 1995

AbstractBackground-A specific local indicatorin the Koch's triangle could be critical tothe complication-free treatment of atrio-ventricular nodal reentrant tachycardiaby transcatheter radiofrequency abla-tion. Recording of perinodal slow poten-tial reflects a slow conduction area, andprobably indicates the location of theslow pathway component of the circuit.Specific ablation of the slow pathwaywould carry the least risk ofatrioventricu-lar block.Methods and results-Guided by themapped perinodal slow potential, atri-oventricular nodal reentrant tachycardiawas successfully eliminated in all of 55consecutive patients in one session. Fiftytwo patients (94.5%) had confirmed slowpotential at the final success sites.Despite the good result, the underlyingelectrophysiological mechanisms of earlysuccess from slow-potential-guidingcatheter ablation were heterogeneous:selective slow pathway eradication in 31patients (56-4%, group A), selective slowpathway modification in 18 patients(32.7%, group B), inadvertent fast path-way damage in six patients (1090/o, groupC). Group B patients had the preserva-tion of dual atrioventricular nodal path-ways, adequate atrio-Hisian delay, fastpathway facilitation, and a higher fre-quency of inducible, single non-con-ducted nodal echo (15118, 83'3% v 6/31,19*4% in group A, P (( 0001). The uppercommunicating path of the circuit wasimplicated as another site of radiofre-quency destruction. Three recurrenceswere documented in follow up study.However, reablation by the sameapproach caused complete atrioventricu-lar block in one patient (1-7%, 1/58 proce-dures). None of the local characteristicsof ablation sites was an independent pre-dictor ofprocedure outcome.Conclusions-Perinodal slow potential isnot a specific slow pathway indicator intranscatheter radiofrequency ablation ofatrioventricular nodal reentrant tachy-cardia. Multiple strategic sites of thereentry circuit may be damaged throughsimilar local signals.

Keywords: perinodal slow potential; reentrant tachy-cardia; transcatheter radiofrequency ablation

(Br Heart_1995;74:268-276)

The recent advance in the definitive treatmentof the atrioventricular nodal reentrant tachy-cardia is astonishing, given the persistent con-troversy in the exact tissue components of thereentry circuit. With the present knowledge oflocal anatomy and traditional electrophysio-logical study, both surgery' 2 and catheterablation'34 have been shown to be successful ineliminating reentrant tachycardia by selectivedamage of either the anterograde slow path-way or the retrograde fast pathway. However,the lack of a specific electrical marker for adestructive procedure in the critical atrioven-tricular junctional area has always been adrawback in the achievement of a successfulresult without major complication, that is,high grade atrioventricular block.

Perinodal slow potential is a local signalrecordable from Koch's triangle of the atrio-ventricular junctional area. The small low fre-quency local potential with decrementalconduction property was first studied for thepurpose of therapeutic catheter ablation byHaissaguerre et al in 1992.4 He showed theslow potential to be a specific electricalmarker for complication-free slow pathwayablation of atrioventricular nodal reentranttachycardia. Later, direct contact mappings57by high density, high resolution plaque elec-trodes have also confirmed the existence ofsimilar slow potentials located between theatrioventricular node and the coronary sinusostium in canine, porcine, and human hearts.However, the histological, functional, andclinical correlations of the slow potential tothe atrioventricular nodal reentrant tachycardiaare still speculative, and by no means conclusive.With the recognition of the importance

of perinodal atrial tissue89 and the relevantconduction slowing in atrioventricular nodalreentrant tachycardia,'-5 we hypothesised therecording of the slow potential in theKoch's triangle as a reasonable indicator forthe slow pathway component of the reentrycircuit. The present prospective study wasthus conducted to evaluate the therapeuticeffect, and the underlying electrophysiologicalmechanisms of the functional outcome, of

268 on O

ctober 24, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.74.3.268 on 1 S

eptember 1995. D

ownloaded from

http://heart.bmj.com/

Perinodal slow potential as a local guidefor transcatheter radiofrequency ablation of atrioventricular nodal reentrant tachycardia

transcatheter radiofrequency ablation ofatrioventricular nodal reentrant tachycardiaguided by the perinodal slow potential, so asto elucidate the clinical implications of thelocal electrical landmark in the atrioventricularjunction.

MethodsPATIENTSBetween May 1992 and October 1993, 55consecutive patients with paroxysmalsupraventricular tachycardia, which thenproved to be slow-fast atrioventricular nodalreentrant tachycardia, were referred for trans-catheter radiofrequency ablation treatment.They were 17 men and 38 women. Their ageranged from 16 to 70 years, mean 53 (SD 8)years. Mean duration of clinical recurrenttachycardia was 8-0 (3-0) years. Coexistentcardiovascular diseases were rheumatic valvu-lar disease in two patients (mitral stenosisafter percutaneous mitral valvuloplasty in one;mitral regurgitation in one), congenital heartdisease in two patients (secundum atrial sep-tal defect in one; perimembranous ventricularseptal defect in one), coronary artery diseasein one patient, and combined atrioventricularreentrant tachycardia incorporating concealedleft posterior accessory atrioventricular path-way in one patient. Two patients had had pre-vious fast pathway modification, and therecurrence of slow-fast atrioventricular nodalreentrant tachycardia. One of the latter twopatients had first degree atrioventricularblock.

BASELINE ELECTROPHYSIOLOGY STUDYBefore transcatheter radiofrequency ablation,every patient underwent a complete electro-physiological study, which was performed in apostabsorptive, non-sedated state. Informedwritten consent was obtained before thestudy. Our investigational protocol wasapproved by the hospital research committee,and was in compliance with local ethics stan-dards. During the study, three 6F quadripolarelectrode catheters with 1 cm interelectrodedistance (USCI, or Mansfield-Webster) wereintroduced through femoral veins, advancedinto the right atrium, then positioned at highright atrium, His bundle region, and rightventricular apex. A fourth 7F decapolar elec-trode catheter (USCI, interelectrode distanceof 0-5 cm) or orthogonal electrode catheterwith three groups of circumferentiallyarranged electrode pieces (Mansfield/Webster, interelectrode distance of 1 mmbetween orthogonal pieces) was insertedthrough the jugular vein and positioned in thecoronary sinus. The proximal pair or group ofthis catheter was carefully positioned in closeproximity to the coronary sinus ostium.Surface electrocardiogram leads (I, aVF, V1),as well as intracardiac local electrograms frommultiple sites, were simultaneously displayedand recorded on a multichannel oscilloscoperecorder (EVR-1 10, PPG Biomedical) at apaper speed of 100 or 150 mm/s. The electricalstimulation was performed with a digital pro-

grammable stimulator (Bloom Associates).Pacing stimuli were twice the diastolic thresh-old in strength and 2 ms in duration. Thestimulation protocol encompassed incremen-tal atrial and ventricular pacing, andextrastimulation coupling on two basicdriving cycle lengths (usually 600 ms or 500ms, and 400 ms). The induction of atrioven-tricular nodal reentrant tachycardia wasattempted repeatedly to find out the most reli-able and reproducible protocol of tachycardiainitiation. If the atrioventricular nodal reen-trant tachycardia could not be induced or sus-tained in the baseline state, isoprenaline wasinfused and titrated to an increase of at least20% of the heart rate. The same pacing proto-col was then repeated. With such study proto-col, all 55 patients were shown to have dualatrioventricular nodal pathways with thedemonstration of discontinuous atrioventricu-lar nodal conduction curve through atrialextra-stimulation (50 patients), incrementalatrial pacing (39 patients), ventricularextrastimulation (15 patients), and incrementalventricular pacing (seven patients). All 55patients in the present study had reproduciblesustained atrioventricular nodal reentranttachycardia by the stimulation protocol.Twenty two of the 55 patients required iso-prenaline infusion to facilitate the tachycardiainduction.

PERINODAL SLOW POTENTIAL RECORDINGOur working hypothesis for effective slowpathway ablation assumed the perinodal slowpotential to be an electrical landmark of thecritical slow conduction area, that is, the atri-oventricular nodal slow pathway, of the reentrycircuit. Systematic mapping was conductedby a deflectible, large tip (4 mm) quadripolarcatheter with 2-5-2 mm interelectrode dis-tance (Mansfield/Webster) along the anterior,median, and posterior septal zones of the tri-cuspid annular border of Koch's triangle.Bipolar local electrogram recorded from thedistal electrode pair was magnified at0-05-0-2 mV/cm with a filter range of 30 and500 Hz. Any focus- manifesting multiphasicatrial electrogram or distinct interpolated lowfrequency potential was selected and verifiedby low rate atrial overdriving, usually 20-50%above the sinus rate. If a rounded or biphasicsmall potential was dissociable from the localatrial electrogram and diminished in ampli-tude with the gradual increase of atrial pacerate, the selected site was confirmed to havethe perinodal slow potential (fig 1). If pacingverification was impossible due to catheterinstability or too low signal amplitude, the sitewith a suspicious slow potential was stillpicked up as an ablation target, when therewas no better choice. Since the slow potentialswere frequently recorded at two or even threecontiguous zones between the His bundleposition and the coronary sinus ostium, thetranscatheter radiofrequency ablation wasusually targeted first to the most postero-medial zone. Local His potential, atrialrepolarisation and local artefacts wereexcluded by the following properties of the

269 on O



/B


eptember 1995. D

ownloaded from


Lin, Lin, Lo, Tseng, Cheng, Chen, et al

I A A-A -A

VI

H1 11- 700- -700

-1 - tV r--AP---------IS- 4-.11! ~ ~-I F- 1MOPT VI

VI

,

Figure 1 Recording and pacing verification ofperinodal slow potential. The lowfrequency biphasic potential (arrow) is interposed between the local atrial and ventricularelectrograms. The slow potential could be dissociated graduallyfrom the atrial componentand merged towards the ventricular component with a decrease in right atrial pacing cyclelength from 700 ms to 600 ms and 500 ms, as shown in the four panels. For each panel,the tracings include surface electrocardiograms, I, V,, intracardiac electrograms at highright atrium (HRA), His bundle (HBE), and the mapping site (Map)

slow potential: (1) rate dependent gradualmigration and convergence of slow potentialinto local ventriculogram; (2) fixed His-ven-tricular interval over a range of atrial pacingcycle lengths; (3) relative constancy of thelocal activation time of the slow potential; (4)specific anatomical distribution of the slowpotential.

CATHETER ABLATIONAfter the selection of an appropriate target siteby the above local electrographic criteria,radiofrequency energy of 500 kHz was trans-mitted between the 4 mm bulbous tip (27rrn2 surface area, Mansfield/Webster) of theablation catheter and a dispersive patch elec-trode on the left posterior chest wall. Theenergy was supplied by a commercially avail-able generator (Radionics 3C) in a continuousunmodulated sine wave mode. Power, volt-age, current, local impedance, and timingwere displayed on the monitor screen of thegenerator machine but the data were notstored. For each ablation attempt, the powersetting was always started from 10-15 W andgradually increased to 20-30 W, each for 30 sduration. The surface electrocardiogram was

closely monitored. In case of appearance ofjunctional tachycardia during energy delivery,atrial overdrive pacing was applied to guaran-tee a better observation of atrioventricularconduction. After each ablation, tachycardiainducibility was immediately retested by simi-lar stimulation protocol as in the baselinestudy. The therapeutic end point was the non-inducibility of atrioventricular nodal reentranttachycardia and no more than single non-con-ducted echo beat using the programmed atrialand ventricular stimulation protocol with iso-prenaline facilitation.

FOLLOW UPAfter the laboratory success, all 55 patientswere followed regularly in the outpatient

clinic or by the local referring doctor for atleast six months (mean 18 (5) months, range9 to 27 months). On monthly follow up visits,detailed history taking of any symptomaticrecurrence of tachycardia, physical examina-tion, 12-lead surface electrocardiogram, andoccasional 24 hour Holter electrocardiogramstudy were carefully evaluated. After explana-tion, 40 (72-7%) of the 55 patients acceptedlaboratory electrophysiological restudy 3-4months after the initial procedure.

ELECTROPHYSIOLOGIC DATA ANALYSISBefore the catheter ablation, the diagnosis ofatrioventricular nodal reentrant tachycardiawas established according to the criteria sug-gested by many investigators'01: (1) demon-stration of discontinuous atrioventricularnodal conduction curve, and the induction oftachycardia when the critical atrio-Hisianinterval is achieved; (2) atrial activation simul-taneous with, or earlier than, ventricularactivation during tachycardia; (3) no advance-ment of atrial activation by ventricularextrastimulation introduced when His bundlewas refractory; (4) identical septal sequence ofretrograde atrial activation during tachycardiaand ventricular pacing; (5) constant His-atrialinterval of the return cycle after an introducedatrial extrastimulation over a wide range ofcoupling intervals during tachycardia. Criteria2 and 3 excluded the incorporation of anaccessory atrioventricular pathway. Criteria 4and 5 excluded the possibility of atrial tachy-cardia.

In order to evaluate the underlying electro-physiological mechanism, the atrioventricularnodal function was carefully comparedbetween the electrophysiological studies per-formed before, early after (30 minutes to onehour), and late (three to four months) afterthe transcatheter radiofrequency ablation.The electrophysiological data included thebasic sinus cycle length, the atrio-Hisian inter-val, the atrioventricular nodal effective refrac-tory period, the atrioventricular nodalblocking cycle length, 1:1 conducting cyclelength of the fast and slow pathways, theeffective refractory periods of the fast andslow pathways, the atrial effective refractoryperiod, and the inducibility of atrioventricularnodal reentrant echoes or tachycardia. Theeffective refractory periods were compared atthe same, or the closest possible driving cyclelength in the baseline state or the same auto-nomic stress state, or both, by single atrialextrastimulation technique. Atrioventricularnodal conduction curve was depicted, classi-fied, and compared among the three studyphases using incremental atrial pacing oratrial extrastimulation technique.

For the recorded perinodal slow potential,the following characteristics were analysed:local atrio-slow potential interval, normalisedslow potential activation time, slow potentialamplitude, and the anatomical location of theslow potential in Koch's triangle. Classically,the His bundle recording area (bipolar signal,1 cm interpolar distance) represented theanterior atrionodal input zone of the nonnal

270

on October 24, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.74.3.268 on 1 Septem

ber 1995. Dow

nloaded from


Perinodal slow potential as a local guide for transcatheter radiofrequency ablation of atrioventricular nodal reentrant tachycardia

Figure 2 Normalisedslow potential activationtime. The slow potentialactivation time relative tothe normal atrioventricularconduction is estimated bythe ratio ofcorrected localatrio-slow potentialinterval versus atrio-Hisian interval ofHisbundle electrogram(HBE), that is,c(A-SP)IA-H. H, Hisbundle. Otherabbreviations as in fig 1.Time scale is shown at thebottom.

I -

aVFVi

0 1 2 3 4 5 6 7 8 9 *10omS

11lllIll111II IIIIllillll 1 IIll11lIIII 1llll

atrioventricular conduction in human. Therelative timing of the slow potential activationas normalised to the atrio-Hisian interval (fig2) was to highlight the relative activationsequence of the slow potential in normal atri-oventricular conduction.

STATISTICSBasic comparative statistics was calculated bythe paired Student t test for dependent con-

tinuous data, and the x2 test with Yates cor-

rection for categoric data. All continuous dataare expressed as mean (SD). A probabilityvalue of less than 0 05 was considered signifi-cant.

In order to understand the contribution ofperinodal slow potential to the successfulcatheter ablation of atrioventricular nodalreentrant tachycardia, univariate followed bymultivariate logistic regression analysis was

applied to extract the independent factors pre-

dicting procedure outcome, by means of theStatistical Analysis System (SAS) program.

Sensitivity, specificity, positive and negativepredictive values, and predictive accuracy

were calculated for the selected predictors, ifany.

ResultsPERINODAL SLOW POTENTIAL AND CATHETERABLATIONAlong the tricuspid border of Koch's triangle,meticulous mapping followed by pacing verifi-cation confirmed the existence of definiteperinodal slow potential at the successfulablation sites in 52 of the 55 patients (94 5%).However, all 55 patients had had successfulelimination of the slow-fast atrioventricularnodal reentrant tachycardia in one single pro-

cedure. The lack of a definite slow potentialdid not prohibit the successful catheter abla-tion in the remained three patients, who hadonly a suspicious local signal. Mean radio-frequency pulse number required for the suc-

cessful outcome was 2-4 (2 9) (range 1 to 20).Mean total procedure time was 2-8 (1N1)hours. Mean fluoroscopic time allocated tomapping and ablation was 23-4 (14-9) min-utes, ranging from 50-60 minutes in thelearning period to 10-15 minutes in the latterhalf of the series. The confirmed slow poten-tials could be recorded from one, two, or even

all three zones in the Koch's triangle: the pos-teroseptal zone alone in three patients, themedioseptal zone alone in 32 patients, theanteroseptal zone alone in one patient, themedioseptal and posteroseptal zones in 13patients, and all three zones in three patients.The most posteromedial site was selected forthe first ablation.The mere recording of perinodal slow

potential in the critical triangle of Koch didnot guarantee the success of transcatheterradiofrequency ablation of atrioventricularnodal reentrant tachycardia. Analysis of localelectrograms of the selected targets of the first

Table 1 Local electrogram characteristics of the successful andfailed target sites of thefirst and overall ablation attempts.Values are means (SD)

First ablation site All ablation sites

Success Failure P < Success Failure P <

PNSP (+) 31/32 (96 9%) 17/23 (73 9%) 0-012 52/55 (94 5%) 37/71 (52-1%) 0.00000A-SP (ms) 50-8 (10-3) 54-4 (11-6) NS 50 6 (10-2) 53-0 (9-3) NSc(A-SP)/A-H (%) 73-5 (19-2) 87-5 (22 0) 0-024 75-8 (18-9) 86-5 (18-7) 0-0095PNSP amplitude (mV) 0-041 (0 022) 0-036 (0-020) NS 0-040 (0 025) 0-031 (0-016) 0 054Koch's triangle zone: 0-012 0-001

posteroseptal 6 13 9 34medioseptal 25 10 43 34anteroseptal 1 0 3 3

HeatJT (+) 26/32 (81-3%) 12/23 (52-2%) 0-021 42/55 (76 4%) 23/71 (32-4%) 0 00000

PNSP (+), with perinodal slow potential; A-SP, local atrio-slow potential interval; c(A-SP), corrected local atrio-slow potentialinterval; A-H, atrio-Hisian interval at His bundle electrogram; Koch's triangle zone, specific location of the ablation site at thetricuspid border of the Koch's triangle; Heat JT (+), with junctional tachycardia during radiofrequency ablation.

271 on O



/B


eptember 1995. D

ownloaded from



and overall ablation attempts revealed only64-6% (31/48) and 58-4% (52/89) of the slowpotential recording sites had subsequentlyresulted in the successful elimination of thetachycardia. Other than the existence ofperinodal slow potential, the success sites bysingle radiofrequency pulse usually had sev-eral other characteristics (table 1): (1) cor-rected activation time of the slow potentialcorresponded to 73T5 (19^2)% (range 33-3%to 113%) of the atrio-Hisian interval on theHis bundle electrogram (versus 87-5 (22.0)%,range 50%-142%, at failure sites, P < 0024);(2) more frequent localisation at the medio-septal zone of Koch's triangle (78' 1 %, versus43 5% at failure sites, P < 0 012 byX2); (3)more frequent emergency of junctional tachy-cardia during radiofrequency energy delivery(81-3%, versus 52-2% at failure sites, P <0-021). Accumulated analysis of the localelectrograms of overall ablation attemptsshowed the same trend (table 1).

However, further multivariate analysis wasunable to isolate any local characteristic inde-pendently predicting the procedure outcomeof the first ablation attempt in the 55 patients.Only when neglecting the possible effect ofaccumulated tissue damage and pooling thedata from all ablation attempts did the record-ing of perinodal slow potential and heatingjunctional tachycardia emerge as independentpredictors of success. However, the positivepredictive value and the predictive accuracyremained inadequate: 58-4% and 68&3% forperinodal slow potential alone, 64-2% and71-4% for junctional tachycardia alone,69-4% and 75-4% for the two together.

ELECTROPHYSIOLOGICAL MECHANISMS OFEARLY SUCCESSAfter the procedure success, three mainmechanisms of atrioventricular nodal modifi-cation were found by repeated electrophysio-logical study, and recognised as responsiblefor the non-inducibility of the clinical tachy-cardia. The three electrophysiological mecha-nisms elucidated by the atrioventricular nodalconduction patterns (fig 3) were the selectiveeradication of the slow pathway, the selectivemodification of the slow pathway, and theselective modification of the fast pathway.

Selective slow pathway eradicationIn 31 of the 55 patients (56'4%) (group A)the discontinuous atrioventricular nodal con-duction curve disappeared (Al, A2 in fig 3)early after the success of transcatheterradiofrequency ablation, suggesting the eradi-cation of the slow pathway. Overall, the 31patients had shortening of the sinus cyclelength, the 1:1 conducting cycle length of thefast pathway, and the effective refractoryperiod of the fast pathway after the ablation(table 2). None of the 31 patients had sus-tained 1:1 conduction of the slow pathway.Maximum atrio-Hisian interval obtainableafter ablation was less than the minimumrequirement of the atrio-Hisian delay for theinduction of the reentrant tachycardia.Among the 31 patients, 13 were shown tohave premature interruption of the atrioven-tricular nodal conduction (subgroup Al) (Alin fig 3), that is, prolongation of the atrioven-tricular nodal effective refractory period. Theother 18 patients had only the "smoothing" of

Al

0

*0

-0 00

00 0 0 0.0

0 0

50 300 350 400 450AlA2

B

0

00

o

l

500 550 600

450400350300

I 250<g 200

150100500

0 Before0 After

I(N4

300

250

200

150

_ A20

_ * 0 .

- 000

!00 250 300 350 400 450 500 550 600A1A2

C

0

0

00*o o o000

0

100

50

0250 300 350 400 450 500

AlA2

0

00*0 0

250 300 350 400A1A2

450 500

Figure 3 Comparison of atnoventricular nodal conduction curves before and after the success of transcatheterradiofrequency ablation of atrioventnicular nodal reentrant tachycardia. Each panel shows the paired data of atrio-Hisianintervals (A2H2) and atrial coupling intervals (AJA2) derivedfrom single atrial extrastimulus study. Thefour panelsrepresent examples ofselective slow pathway eradication (panels Al, A2), selective slow pathway modification and uppercommunicating path damage (panel B), andfast pathway damage (panel C). Panels A andA2 show the difference ofatrioventricular nodal refractoriness between the two corresponded subgroups ofpatients, despite similar disappearance ofthe slow pathway behaviour after ablation. Panel B shows the postablation preservation of the dual atrioventricular nodalpathways. The effective refractory period of the residual slow pathway is prolonged after ablation in this particular group Bpatient, whereas it is unchanged or shortened in many othersfrom the same group. Panel C shows the all slow pathwaybehaviour with the inadvertent damage of the antegrade fast pathway. Intervals ofA2H2 andAlA2 in ms.

450400350300

N 250ze.j 200< 150

10050

2

350300250

c'4 200II(N 150

10050A I

272

on October 24, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.74.3.268 on 1 Septem

ber 1995. Dow

nloaded from



Table 2 Electrophysiological characteristics of the 31 groupA patients before and soon after the catheter ablation. Valuesare means (SD), ms

Subgroup Al (n = 13) SubgroupA2 (n = 18)

Before After P 6 Before After P6

Sinus CL 800-8 (111-2) 727-2 (119-2) 0-070 725-0 (113-4) 661-1 (103-1) 0 001A-H 73-4 (15-2) 73 9 (16-1) NS 78-9 (38-3) 76-1 (34 2) NSMax (A-H) 370-0 (73-8) 176-2 (37 3) 0-00002 360-0 (89-3) 201-6 (62 5) 0 00001Min (A-H)sv-r 261-5 (72 3) - - 288-3 (91-4) - -

FPERP 387-7 (101-2) 336-2 (42 5) 0-025 296-1 (47 2) 224-4 (47 6) 0 00009SPERP 261-5 (36 3) - - 238-8 (35 6) - -AERP 126-2 (89 6) - - 58-1 (37 6) - -1:1 FP 476-2 (112-1) 396-2 (69 6) 0-002 375-6 (76 4) 329-4 (60 6) 0-0021:1 SP 367-8 (53 8) - - 356-7 (75-7) - -1:1 VA 413-1 (80 4) 353-9 (69 5) 0-043 352-2 (84-1) 332-2 (49 3) NS

Sinus CL, sinus cycle length; Max (A-H), maximum atrio-Hisian interval obtainable; Miin (A-H)sv, minimum atrio-Hisianinterval required for the induction of reentrant tachycardia; FPERP, SPERP, effective refractory periods of the fast pathway and theslow pathway, AERP, difference ofthe effective refractory periods ofthe fast and the slow pathways; 1:1 FP, 1:1 SP, minimum 1:1conducting cycle lengths of the fast pathway and the slow pathway; 1:1 VA, miniimum 1: 1 conducting cycle length of retrogradeventriculo-atrial conduction. P value, comparison of data before and early after catheter ablation.

the atrioventricular nodal function curve(subgroup A2) (A2 in fig 3), that is, the sameor shorter atrioventricular nodal effectiverefractory period. The frequency of inductionof single, non-conducted slow-fast atrioven-tricular nodal reentrant echo was rathergreater in patients of subgroup A2 (5/18,27 8%) than those of subgroup Al (1/13,7.7%, P > 0-05). The subgroup Al (13patients, 23-6% of all 55 patients) mightrepresent those who had more definite eradi-cation of the slow pathway of the atrioventric-ular nodal reentrant circuit early after thesuccess.

Selective slow pathway modificationEighteen (32 7%) (group B) of the 55 patientshad the persistence of the atrioventricularnodal dual pathways after the successfulcatheter ablation (B in fig 3). Electro-physiologically, they had also had the shorten-ing of the sinus cycle length, the 1:1 conduct-ing cycle length of the fast pathway, and theeffective refractory period of the fast pathway(table 3). None of the 18 patients had sus-tained 1:1 conduction of the slow pathwayeither. However, the duality of the atrioven-tricular nodal conduction persisted. Theeffective refractory period of the slow pathwaydetermined by extrastimulus study was littlechanged after the catheter ablation. The max-imum atrio-Hisian interval obtainableremained more than that needed for theinduction of the tachycardia. The frequencyof the induction of single, non-conducted

Table 3 Electrophysiological characteristics of the 18 group B patients before, and soonafter the catheter ablation. Values are means (SD), ms

Before After P <

Sinus CL 711-7 (77-9) 647-2 (88 7) 0-002A-H 70 0 (14-1) 68-6 (15-7) NSMax (A-H) 406-1 (109-5) 338-3 (85-1) 0-016Mim (A-H)sv-r 256-7 (55 7) - -FPERP 326-1 (68-3) 287-8 (29 8) 0-008SPERP 236-9 (31-6) 232-5 (37 7) NSAERP 101-3 (53 3) 61-9 (41-2) 0-0531:1 FP 417-1 (92 7) 354-1 (37 9) 0 0071:1 SP 351-3 (78 6) - -1:1 VA 360-0 (80 6) 306-7 (58-5) 0-031

Abbreviations as in table 2.

slow-fast atrioventricular nodal echo wasmuch more than in patients of group A(15/18, 83-3% in group B, versus 6/31, 19-4%in group A, P 4( 0001). The implicated site ofablation to the reentry circuit was probablytwofold: the modification of the slow pathwaycomponent, and damage of the upper com-municating path between the fast and the slowpathways. Particularly, the preferential dam-age of the latter alone might well result in thepostablation electrophysiological profile of thegroup B patients: facilitated fast pathway, per-sistent slow pathway, inducible non-con-ducted atrioventricular nodal reentrant echo,but no sustained reentrant tachycardia.

Selective fast pathway modificationWith the guidance of perinodal slow potential,inadvertent fast pathway ablation occurred insix of the 55 patients (10-9%) (group C; C infig 3): four with antegrade and retrograde fastpathway damage, two with selective retro-grade fast pathway elimination. One of thelatter two patients had had simultaneouseradication of the slow pathway, as shown bythe loss of the discontinuity of the antegradeatrioventricular nodal conduction. Nonedeveloped complete atrioventricular blockfrom the first ablation procedure. The corre-sponding ablation sites for the inadvertent fastpathway damage were located at theanteroseptal zone in one patient, in themedioseptal zone in three patients, and in theposteroseptal zone in one patient.

FOLLOW UP ELECTROPHYSIOLOGICAL STUDYThree months after the success of catheterablation, 40 patients (72 7%) agreed to have afollow up electrophysiological study (table 4):23 from group A (10 from subgroup Al, 13from subgroup A2), 13 from group B, andfour from group C. Under the same studyprotocol, including isoprenaline facilitation,three patients (3/40, 7-5%) were shown tohave reinducible sustained slow-fast atrioven-tricular nodal reentrant tachycardia. Amongthe three patients, two had clinical recurrenceof the tachycardia. All three patients had beenshown to have the eradication of atrioventricu-lar nodal conduction duality early after the

273 on O



/B


eptember 1995. D

ownloaded from



Table 4 Late phase electrophysiological data in 40follow up patients. Values are means(SD), ms

GroupAl (n=10) A2(n=13) B(n=13) C(n=4)

Sinus CL 866-0 (156-0) 848-5 (107-6) 839-2 (100-5) 890-0 (40 8)A-H 80-0 (13-3) 85.4 (43-9) 75 0 (12-2) 172-5 (49 9)AVNERP 443-0 (201-9) 346-2 (99-5) 374-6 (64-8) 345-0 (183-4)FPERP 457 0 (189-6) 365-4 (92 6) 406-9 (71-2) -AVBCL 497-0 (180-9) 420-8 (101-3) 464-6 (90 1) 447-5 (128-2)1:1 VA 472-2 (135-8) 382-3 (84-5) 396-2 (82-6) 580, 460

AVNERP, effective refractory period of atrioventricular node; AVBCL, blocking cycle length of1:1 atrioventricular nodal conduction; other abbreviations as in table 2.

previous catheter ablation: two from subgroupAl, one from subgroup A2. Analysis of thefollow up electrophysiological data showedthat return of the physiology of dual atrioven-tricular nodal pathways was more commonthan its disappearance: 13 of the 23 patientsinitially belonging to group A were reclassifiedas group B, while only three of 13 patientsfrom group B converted to group A. Onepatient from the initial subgroup Al had latedisappearance of retrograde fast pathway con-duction. Two of the three patients with initialbidirectional fast pathway modificationremained the same, one with weak return ofretrograde ventriculoatrial conduction. Onepatient with initially pure retrograde fast path-way damage had had late antegrade slowpathway modification on follow up. A singlenon-conducted or conducted atrioventricularnodal echo was inducible in 14 of 21 patientswith antegrade atrioventricular nodal dualityand no sustained tachycardia. None of 16patients with eradication of the antegradeslow pathway or retrograde fast pathway hadan inducible echo. The effective refractoryperiod and the 1:1 conducting cycle length ofthe atrioventricular node were generally pro-longed, supporting the progressive modifica-tion (but not eradication) of the slow pathwayconduction as the main electrophysiologicalmechanism for the long term effectiveness.The three patients with recurrent tachycar-

dia received a second ablation procedure, alsoguided by the perinodal slow potential. Twowere successfully reablated of the slow path-way and the tachycardia. The third patientunfortunately developed complete atrio-ventricular block, needing permanent pace-maker implantation. The ablation energy(30 W for only 10 s) was focused at amedioseptal site with clear recording of theslow potential. The use of the perinodal slowpotential as a local landmark resulted overallin 13.8% inadvertent fast pathway damage(8/58), including 1-7% complete atrioventric-ular block (1/58).

DiscussionThe present study confirmed the high thera-peutic efficacy of transcatheter radiofrequencyablation in the eradication of slow4ast atrio-ventricular nodal reentrant tagycardia..However, «the Xnereording of peinodalaow potersd did am appea to. be a mof the success of the procedure. Only 64-6%and 58-4% of the slow potential recordingsites from the Koch's triangle had subse-

quently resulted in the successful eliminationof the reentrant tachycardia on the first andoverall ablation attempts. Furthermore, theelectrophysiological mechanisms of the resul-tant clinical success were actually very com-plex. Only 56-4% of the patients studied hadthe expected selective slow pathway eradica-tion under the guidance of the perinodal slowpotential. For the others, the combined dam-age to the upper communicating path and theslow pathway component of the reentry cir-cuit (32-7%), or even inadvertent damage ofthe fast pathway (10.9%), were thought to bethe responsible mechanisms. A single localelectrical landmark for specific slow pathwayablation does not seem to exist. Other thanthe recording of appropriately timed perinodalslow potential, the present study has actuallyrecommended two concomitant characteris-tics common to a success site of catheter abla-tion, that is, the medioseptal location and theheating junctional tachycardia.

PERINODAL SLOW POTENTIAL AND THE SLOWPATHWAYBy contact mapping of high density plaqueelectrodes in human hearts, McGuire et allhave recently documented the existence oflow frequency potentials between the atrio-ventricular node and the coronary sinus inKoch's triangle. Further correlative study bysimultaneous extracellular and intracellularpotential recordings from the same investiga-tors suggests that the nodal type cells distrib-uting subendocardially near the tricuspidannular border are the responsible tissue com-ponent of the slow potential with decrementalconduction property. These observations con-firm the single catheter finding of the samepotential by Damato et al2 in 1969 andHaissaguerre et aP in 1992. However, thefunctional correlation of the slow potentialrecording tissue to the slow pathway compo-nent of atrioventricular nodal reentry circuitremains largely speculative. The exact histo-logical differentiation of the origin of the slowpotential between atrionodal transitionalcell,"3 dead end fibres,'4 and even the atrioven-tricular node'3 15 itself is also lacking.As expected, success in the utilisation of

the perinodal slow potential as a local land-mark for transcatheter radiofrequency abla-tion of atrioventricular nodal reentranttachycardia in the present study was also theresult of complex mechanisms. The heteroge-neous results rflect the heterogeneoussources of the slow potentials, and the corre-sponding tissue components in the regionwhere the atrioventricular nodal reentry islocated. The slow potential recorded in theKoch's triangle is apparently not a specificindicator of selective slow pathway ablation byradiofrequency current. Up to the present, theachievement of complication-free slow path-way by catheter ablation stilleqtures.- cd 'kodwledge of the localanaaory,* ee slow onductkon zone evalua-tion, for example, the slow potential area,4'the junctional tachycardia during energydelivery,19 and so on.

274 on O



/B


eptember 1995. D

ownloaded from



FAST PATHWAY FACILITATION ANDNON-CONDUCTED NODAL ECHOIn our study, the facilitation of fast pathwayconduction was manifested in most patientsafter the successful catheter'ablation of theatrioventricular nodal reentrant tachycardia.In contrast to other studies,320 we demon-strated that the shortening of effective refrac-tory period and 1:1 conducting cycle length ofthe fast pathway would occur whether or notthe slow pathway was totally eliminated.

Partly as a result of a concomitant decreasein the sinus cycle length,'42' the mechanismof the fast pathway facilitation was usuallyattributed to the postablation heightening ofsympathetic tone or to the withdrawal ofparasympathetic balance.22 However, recentstudies20 23 with chemical autonomic blockadedisputed the role of autonomic tone, and sug-gested that electrotonus interaction betweenthe fast pathway and the slow pathway mightrather be the determining factor. If the latteris correct, the ablation site in patients withpostablation atrioventricular nodal duality(group B) in the present study must haveinvolved the intervening atrial myocardium orthe communicating fibres between the twopathways, particularly at the proximal por-tion. This is also compatible with the observa-tion of more frequent induction of single,non-conducted slow-fast atrioventricularnodal reentrant echo in patients with residualslow pathway than those without it, early andlate after the catheter ablation. The existenceof non-conducted atrioventricular nodal reen-trant echo theoretically implicates a missinglink between the retrograde fast pathway andthe antegrade slow pathway in the circuit ofreentry, that is, the upper communicatingpath. A combination of the findings of fastpathway facilitation and non-conducted nodalreentrant echo in these patients strongly sug-gests strategic damage to the proximal com-municating structure between the fastpathway and the slow pathway components ofthe reentry circuit.

FOLLOW UPLong term electrophysiological follow up inthe present study showed increasing reappear-ance of the slow atrioventricular nodal path-way. Non-conducted atrioventricular nodalreentrant echo, but never sustained tachycar-dia, remained easily inducible in those withresidual dual atrioventricular nodal pathways.Persistent block of the upper communicatingpath is apparently at least as important as pro-gressive modification of the slow pathway inpreventing the recurrence of the previousreentrant tachycardia.

LIMITATIONSFirst, discriminating the postablation electro-physiological mechanisms contributing tosuccess of the procedure relied heavily onanalysis of the atrioventricular nodal conduc-tion curve, derived by classical extrstimula-tion or incremental pacing. Studies haveprovided firm evidence that dual atrioventricu-lar nodal pathways could exist with various

conduction patterns, with or without a jumpof atrio-Hisian intervals.24 25 However, anaccepted and standardised electrophysio-logical study with better evaluation of theatrioventricular nodal physiology is still pend-ing in clinical practice. Second, the targetmapping performed strictly along the tricus-pid border of the Koch's triangle might havemissed better sites locating away from theannulus.5 Technically, a more thoroughsearch of the slow potential in the Koch's tri-angle is definitely feasible. However, the cost-benefit balance of ablation over the atrial sideof the tricuspid annulus and the risk ofcomplete atrioventricular block has to becarefully evaluated. Third, the lack of temper-ature monitoring in radiofrequency energydelivery26 might lead to misjudgment on theeffect of an ablation attempt. The guidingaccuracy of the perinodal slow potential couldhave been underestimated. A prospectivestudy incorporating a thermometer tippedablation catheter would be needed to answerthis question.

CONCLUSIONThis prospective study has shown the inade-quacy of the perinodal slow potential in guid-ing selective slow pathway ablation ofslow-fast atrioventricular nodal reentranttachycardia by transcatheter radiofrequencycurrent. The underlying electrophysiologicalmechanisms of the high clinical efficacy of thisprocedure involve selective eradication of theslow pathway, the intervening upper commu-nicating path, or even the fast pathway.Careful evaluation of the local anatomy, heat-ing junctional tachycardia, and the atrioven-tricular conduction status, in addition to themapping of perinodal slow potential, is impor-tant for complication-free curative treatmentof the atrioventricular nodal reentrant tachy-cardia. A specific and reliable electrical land-mark for the slow pathway location in theKoch's triangle remains unknown.

The authors thank Miss Ying-Hwa Lin for the preparation ofthe manuscript.

1 Ross DL, Hohnson DC, Denniss AR, Cooper MJ,Richards DA, Uther JB. Curative surgery for atrioven-tricular junctional ("AV nodal") reentrant tachycardia.J7Am Coll Cardiol 1985;6:1383-92.

2 Cox JL, Holman WL, Cain ME. Cryosurgical treatment ofatrioventricular node reentrant tachycardia. Circulation1987;76: 1329-36.

3 Jackman WM, Beckman KJ, McClelland JH, Wang X-Z,Friday KJ, Roman CA, et al. Treatment of supraventric-ular tachycardia due to atrioventricular nodal reentry byradiofrequency catheter ablation of slow-pathwayconduction. N EnglJ Med 1992;327:313-8.

4 Haissaguerre M, Gaita F, Fischer B, Commenges D,Montserrat P, d'Ivemois C, et al. Elimination of atrio-ventricular nodal reentrant tachycardia using discreteslow potentials to guide application of radiofrequencyenergy. Circulation 1992;85:2162-75.

5 McGuire MA, Bourke JP, Robotin MC, Johnson DC,Meldrum-Hanna W, Nunn GR, et al. High resolutionmapping of Koch's triangle using sixty electrodes inhumans with atrioventricular junctional (AV nodal)reentrant tachycardia. Circulatinm 1993;88[part 1]:2315-28.

6 McGuam MA, de efks JMT, Veraltens JT, Opthof T,Becker AC, Jsm Mi. Origin and agaskance of doublepotentil near te euioventricular node: cortrelation ofextraceMular potentials, intracellular potentials, andhistology. Circulation 1994;89:2351-60.

7 de Bakker JMT, Coronel R, McGuire MA, VermeulenJT, Opthof T, Tasseron S, et al. Slow potentials in

275 on O



/B


eptember 1995. D

ownloaded from



atrioventricular junctional area of patients operated onfor atrio-ventricular node tachycardia and in isolatedporcine hearts. JAm Coll Cardiol 1994;23:709-15.

8 Mazgalev T, Dreifus LS, Bianchi J, Michelson EL. Themechanism of AV junctional reentry: role of the atrio-nodal junction. Anat Rec 1981;201:179-88.

9 Iinuma H, Dreifus ]LS, Mazgalev T, Price R, MichelsonEL. Role of the perinodal region in atrioventricularnodal reentry: evidence in an isolated rabbit heart pre-paration. JAm Coil Cardiol 1983;2:465-73.

10 Wu D, Yeh S-J, Wang C-C, Wen M-S, Chang H-J, LinF C. Nature of dual atrioventricular node path-ways andthe tachycardia circuit as defined by radiofrequencyablation technique. JAm Coll Cardiol 1992; 20:884-95.

11 Akhtar M, Jazayeri MR, Sra J, Blanck Z, Deshpande S,Dhala A. Atrioventricular nodal reentry: clinical, electro-physiological, and therapeutic considerations. Circulation1993;88:282-95.

12 Damato AN, Lau SH, Berkowitz WD, Rosen KM, LisiKR. Recordings of specialized conducting fibers (AVnodal, His bundle, and right bundle branch) in manusing an electrode catheter technic. Circulation 1969;39:435-47.

13 Anderson RH, Janse MJ, van Capelle FJI, Billette J,Becker AE, Durrer D. A combined morphological andelectrophysiological study of the atrioventricular node ofthe rabbit heart. Circ Res 1974;35:909-22.

14 van Capelle FJL, Janse MJ, Varghese PJ, Freud GE, MaterC, Durrer D. Spread of excitation in the atrioventricularnode of isolated rabbit hearts studied by multiple micro-electrode recordings. Circ Res 1972;31:602-16.

15 Paes de Carvalho A, de Almeida DF. Spread of activitythrough the atrioventricular node. Circ Res 1960;8:801-9.

16 James TN. Morphology of the human atrioventricularnode, with remarks pertinent to its electrophysiology.Am HeartJ 1961;62:756-71.

17 Janse MJ, Anderson RH, McGuire MA, Ho SY. "AVnodal" reentry: Part I: "AV nodal" reentry revisited.J Cardiovasc Electrophysiol 1993;4:561-72.

18 Lauer MR, Young C, Munsif A, Yu J, Kuo C-T, Liem LB,et al. Identification of patients at high risk of developingcomplete atrioventricular block during ablation of slowpathway conduction [abstr]. Circulation 1993;88(part 2):I-202.

19 Thakur RK, Klein GJ, Yee R, Stites HW. Junctionaltachycardia: a useful marker during radiofrequency abla-tion for atrioventricular node reentrant tachycardia.J Am Coil Cardiol 1993;22:1706-10.

20 Natale A, Klein G, Yee R, Thakur R. Shortening of thefast pathway refractoriness after slow pathway ablation:effects of autonomic blockade. Circulation 1994;89:1103-8.

21 Ehlert FA, Goldberger IJ, Brooks R, Miller S, Kadish AH.Persistent inappropriate sinus tachycardia after radio-frequency current catheter modification of the atrioven-tricular node. Am ICardiol 1992;69:1092-5.

22 Kay GN, Epstein AE, Dailey SM, Plumb VJ. Selectiveradiofrequency ablation of the slow pathway for thetreatment of atrioventricular nodal reentrant tachycar-dia: evidence for involvement of perinodal myocardiumwithin the reentrant circuit. Circulation 1992;85:1675-88.

23 Lesh MD, Gibb WH, Epstein LM. Electrotonic inter-action between dual AV nodal pathways: evidence fromRF ablation and a computer model [abstr]. Circulation1992;86(suppl ):I-130.

24 Tchou P, Mazgalev T, Keim S, Mowrey K, Cheng YN,Rist K, et al. Towards a new definition of dual atrioven-tricular nodal pathways [abstr]. Circulation 1993;88[part2] :I-202.

25 Tchou P, Mazgalev T, Mowrey K, Cheng YN, Rist K.Interaction of dual pathways producing complex atrio-ventricular nodal conduction curves [abstr]. Circulation1993;88[part 2]:I-203.

26 Hindricks G, Haverkamp W, Gulker H, Rissel U, BuddeT, Richter KD, et al. Radiofrequency coagulation ofventricular myocardium: improved prediction of lesionsize by monitoring catheter tip temperature. Eur Heart J1989;10:972-84.

276 on O



/B


eptember 1995. D

ownloaded from


br perinodal ablationof nodal reentrant efficacy and success

Documents