the role of purkinje and pre-purkinje potentials in the reentrant circuit of verapamil-sensitive...

IntroductionVerapamil-sensitive idiopathic left ventricu-

lar tachycardia (ILVT) with a right bundle branchblock (RBBB) configuration and left-axis deviation(LAD) has been suggested to originate from thePurkinje network of the left posterior fascicle,1,2

and the mechanism of ILVT is likely to be reen-try.1–6 However, the location of the actual reen-trant circuit and its slow conduction area are notclearly understood. Catheter ablation of ILVT hasbeen performed at the site of the earliest Purkinjepotential (PP) during ILVT because this site wassuggested to be the exit site of the reentrant cir-cuit.7 Recent studies in patients with ILVT re-

ported successful catheter ablations at sites thatwere far away from the earliest PP recordingsite.8–10 In these cases, because the appropriatesite of catheter ablation was often determined bythe effect of ventricular stimulation from the abla-tion catheter and the “bump” phenomenon, themapping catheter had to be moved to attempt pac-ing stimulation for determination of the appropri-ate ablation site away from the earliest PP.

Some studies have reported a comparativelydull potential preceding the PP (pre-PP) recordedduring tachycardia close to the site of successfulcatheter ablation,9,11 and the entrainment phe-nomenon of the preceding potential was exam-ined by pacing from outside of the reentrant cir-cuit of ILVT.12 However, the clinical significanceof pre-PP and its role in the reentrant circuit areunknown. This study examined the relationshipbetween the pre-PP and PP during ILVT to eluci-date the role of these potentials in the reentrantcircuit of ILVT.

The Role of Purkinje and Pre-Purkinje Potentialsin the Reentrant Circuit of Verapamil-SensitiveIdiopathic LV TachycardiaTAKESHI AIBA, KAZUHIRO SUYAMA, NAOHIKO AIHARA, ATSUSHITAGUCHI, WATARU SHIMIZU, TAKASHI KURITA, and SHIRO KAMAKURAFrom the Division Of Cardiology, National Cardiovascular Center, Osaka, Japan

AIBA, T., ET AL.: The Role of Purkinje and Pre-Purkinje Potentials in the Reentrant Circuit of Verapamil-Sensitive Idiopathic LV Tachycardia. Although the mechanism of verapamil-sensitive idiopathic left ven-tricular tachycardia (ILVT) is usually reentry, the actual reentrant circuit is not clearly understood. Thisstudy examined the relationship between the Purkinje potential (PP) and a dull potential preceding PP(pre-PP) during ILVT to elucidate the roles of these potentials in the reentrant circuit of ILVT. Electro-physiological studies and radiofrequency catheter ablation were performed in ten patients (7 men, 3women, mean age 29 years) who had an ILVT with a right bundle branch block configuration and left-axisdeviation. Left ventricular endocardial mapping using an octapolar catheter and entrainment and reset-ting studies during VT was performed by pacing from the right ventricular outflow tract (RVOT) and eachsite of the left ventricular mapping catheter. PP and pre-PP were recorded simultaneously during VT inall patients. The earliest PP during VT was recorded at the inferoposterior septum, and PP was activatedbidirectionally toward the proximal (basal) and distal (apical) sites along the left posterior fascicle. In con-trast, pre-PP was recorded at sites slightly proximal to the earliest PP recording site, and was activated to-ward the earliest PP site. Pacing from RVOT confirmed manifest entrainment, and the stimulus to pre-PPinterval was prolonged with a shorter pacing cycle length. Concealed entrainment was demonstrated bycapture of the PPs of the left ventricular mapping catheter in six patients, and the postpacing interval ateach PP site was equal to the tachycardia cycle length. The pre-PP was orthodromically activated from theproximal to the distal site during pacing. More rapid pacing also produced delay in activation from PP topre-PP, indicating slow conduction in ILVT. Catheter ablation was performed at the pre-PP recording siteduring VT, and was successful in all patients. The reentrant circuit of ILVT could be constructed basedon the pre-PP, PP, and slow conduction between the PP and pre-PP. Catheter ablation of ILVT was suc-cessful at the pre-PP recording site. (PACE 2001; 24:333–344)

pre-Purkinje potential, Purkinje potential, ventricular tachycardia, reentry, entrainment, catheterablation

PACE, Vol. 24 March 2001 333

Address for reprints: Kazuhiro Suyama, M.D., Division of Car-diology, National Cardiovascular Center, 5-7-1 Fujishirodai,Suita, Osaka, 565-8565 Japan. Fax: 81-6-6872-7486.

Received April 18, 2000; revised July 20, 2000; accepted Au-gust 3, 2000.

Reprinted with permission fromJOURNAL OF PACING AND CLINICAL ELECTROPHYSIOLOGY , Volume 24, No. 3, March 2001Copyright © 2001 by Futura Publishing Company, Inc., Armonk, NY 10504-0418.

MethodsPatients

This study consecutively included ten pa-tients (7 men, 3 women) with a mean age of 29years (range 13–41 years). They were referred forcatheter ablation of medically refractorymonomorphic sustained ventricular tachycardia(VT) that was inducible by electrophysiologicaltesting (Table I). Electrocardiograms (ECGs)recorded during tachycardia exhibited RBBB andLAD in all patients. No patient had apparent struc-tural heart disease as judged by resting ECG, max-imal treadmill exercise testing, chest X ray, andtwo-dimensional and Doppler echocardiography.VT was slowed and terminated by intravenous ve-rapamil in each patient.

Electrophysiological Study

Electrophysiological testing was performed ina fasting, drug-free, nonsedated state, after writ-ten, informed consent had been obtained. All an-tiarrhythmic drugs had been discontinued at least5 half-lives before electrophysiological testing.Then 6 Fr quadripolar catheters (EP Technologies,Mountain View, CA, USA) were introduced viasheaths inserted into the femoral vein and placedunder fluoroscopic guidance in the high rightatrium, His-bundle position, and right ventricularapex (RVA) or outflow tract (RVOT). Stimulationwas performed with rectangular impulses of 2-msduration at twice the diastolic threshold by a pro-grammed stimulator (EP-3, EP Med System, Inc.,Mt. Arlington, NJ, USA). The stimulation protocolconsisted of programmed ventricular stimulationfrom the RVA and the RVOT using eight drivestimuli with up to three premature extrastimuli

and programmed atrial stimulation with up to twopremature extrastimuli. Additionally, incremen-tal atrial and ventricular stimulation with con-stant cycle length were performed. The stimula-tion protocol was terminated when sustained VTwas induced. If sustained VT was not induced,isoproterenol was administrated by continuousinfusion at 1 mg/min, and incremental pro-grammed ventricular stimulation was repeated.Surface 12-lead ECG and endocardial electro-grams were displayed on a computer monitor us-ing an EP-Lab system (Quinton ElectrophysiologyCorp., Seattle, WA, USA) or Bard Electrophysiol-ogy system (C.R. Bard. Inc. Billerica, MA, USA).All bipolar endocardial electrograms were filteredbetween a band-pass of 30–500 Hz and recordedsimultaneously with a 12-lead ECG. Data werestored on a computer disk for further evaluation.

Endocardial Mapping and Entrainment StudyAfter ILVT was induced, endocardial mapping

was performed during VT using a 6 Fr 2–mm-spaced octapolar catheter with a 2-mm tip elec-trode (EP Technologies) in all patients. These map-ping catheters were introduced into the rightfemoral artery and advanced retrogradely into theposteroapical septum of the left ventricle. The lo-cation of mapping electrodes was identified usingbiplane fluoroscopy. The authors searched for theactivation sequence of pre-PP and PP during VT,and the mapping catheter was manipulated and ad-justed to record the pre-PP and PP simultaneously.

The entrainment and resetting studies wereperformed during ILVT while recording the pre-PPand PP. Pacing from the RVOT and the mappingcatheter in the left ventricle was attempted at somecycle lengths of 10–20 ms shorter than the VT cy-

AIBA, ET AL.

334 March 2001 PACE, Vol. 24

Table I.

Patient Characteristics

Pt. No Sex Age VT Onset (yr.) Symptom VT Morph Induction Mode VTCL (ms) Follow-Up (months)

1 M 28 18 palpitation RBBB/LAD RVA LV pacing 347 322 M 25 18 palpitation RBBB/LAD HRA RVA pacing 400 303 M 41 11 palpitation RBBB/LAD RVA LV pacing 500 224 M 41 40 palpitation RBBB/LAD RVOT pacing 333 225 F 26 24 palpitation RBBB/LAD RVA LV pacing 324 186 F 13 12 palpitation RBBB/LAD RVA LV pacing 330 157 M 31 18 palpitation RBBB/LAD RVA pacing 400 128 M 14 10 palpitation RBBB/LAD RVA pacing 273 49 F 39 37 palpitation RBBB/LAD RVA pacing 300 2

10 M 33 27 palpitation RBBB/LAD HRA LV pacing 372 2

mean 6 SD 29 6 10 21 6 11 358 6 64 16 6 11

RBBB 5 right bundle branch block; LAD 5 left-axis deviation; RV A 5 right ventricular apex; RVOT 5 right ventricular outflow tract;HRA 5 high right atrium; LV 5 left ventricle; VTCL 5 ventricular tachycardia cycle length.

cle length. These stimuli from the mappingcatheter were performed at low output (0.5–1.0mA, 1.0–2.0 ms) to capture only the Purkinje net-work. The responses of these pacing stimuli to pre-PP and PP were examined to determine the roles ofthese potentials in the reentrant circuit of ILVT.

Radiofrequency Catheter AblationThe VT was first mapped using the octapolar

catheter. Afterward, we removed this catheter, re-placed it with a 7 Fr radiofrequency (RF)quadripolar electrode catheter (Marinr, MedtronicInc., Minneapolis, MN, USA) and mapped VTwith the ablation catheter. RF catheter ablationwas performed at the pre-PP recording site duringtachycardia, and RF current was applied for 60seconds when the tachycardia was terminatedwithin 10 seconds. However, if tachycardia wasnot terminated within 10 seconds, the energy ap-plication was discontinued. Energy delivery wasterminated immediately in the event of an in-crease in impedance or displacement of thecatheter electrode. After each application of RFcurrent, programmed ventricular stimulation wasrepeated with and without administration of iso-proterenol. If the VT was inducible, another ap-plication of RF current was performed after addi-tional mapping. These procedures wereterminated when ILVT could not be induced for aperiod of 30 minutes after the last application ofRF energy with isoproterenol administration at adose exceeding that required for induction of VTprior to ablation. Intravenous heparin in a bolus of50 U/kg and as an infusion of 1,000 U/hour wereadministered during the procedure.

Postablation ManagementPatients were monitored for 4 to 7 days after

catheter ablation, and a maximal treadmill exer-cise test was performed prior to discharge. All pa-tients were followed without any antiarrhythmicdrugs, and 24-hour Holter ECGs were obtained atapproximately yearly intervals.

Statistical Analysis

The continuous variables were expressed asmean 6 SD and compared using an unpaired andpaired t-test when appropriate. A P value of ,0.05 was considered statistically significant.

ResultsElectrophysiological Study

In all patients, monomorphic sustained VTwith RBBB configuration and LAD was inducedby programmed stimulation from the RVA, RVOT,or left ventricular septum. The mean cycle lengthof the induced VT in the absence of isoproterenol

was 358 6 64 ms. Nonclinical VT was not inducedin any patient.

Left Ventricular Endocardial Mapping with anOctapolar Catheter

Left ventricular endocardial mapping duringILVT was recorded pre-PP and PP simultaneouslyin all patients. The earliest PP was recorded at theinferoposterior septum, and the PP activationspread to the basal and apical sites of the left ven-tricle along the posterior fascicle of the left bundlebranch. In contrast, the pre-PP was recordedwithin a small area slightly proximal to the earli-est PP recording site, and activated from the basalto the apical site of the left ventricle toward theearliest PP site (Fig.1B). During sinus rhythm im-mediately after termination of VT, this mappingcatheter recorded only the PP that was activatedfrom the proximal (basal) to the distal (apical) siteof the left ventricle along the posterior fascicle ofthe left bundle branch, and the pre-PP was notrecorded (Fig.1A). The positions of the octapolarmapping catheter recording pre-PP and PP areshown in Figure 2.

The earliest pre-PP and the earliest PP pre-ceded the onset of QRS complex by 53 6 22 msand 17 6 6 ms, respectively. The distance betweenthe earliest pre-PP and the earliest PP on the fluo-roscopy image was 22 6 4 mm (Table II). Thus, thepre-PP was recorded within a few centimetersproximal of the earliest PP recording site.

Entrainment Study

Pacing from RVOT confirmed manifest en-trainment in all patients, and the pre-PP was cap-tured orthodromically by these pacing stimuli(Fig. 3A). More rapid pacing at a shorter cyclelength (10 ms) also prolonged the stimulus to pre-PP interval but not the pre-PP to PP interval (Fig.3B).

Low output (0.5–1.0 mA, 1.0–2.0 ms) pacingwas attempted from the distal and proximal PPrecording sites of the left ventricular mappingcatheter, where pre-PP could be recorded simulta-neously. In six of ten patients, these pacing stim-uli could capture only the PP of each site and con-firmed concealed entrainment as follows.

During pacing from the distal PP recordingsite, the postpacing interval of this site was equalto the cycle length of VT, and the stimulus to QRSinterval was equal to the PP to QRS interval dur-ing VT. The interval from the PP to pre-PP at theproximal site during pacing was long and thesame as that during VT (Fig. 4A). However, morerapid pacing produced a delay in activation fromthe PP to pre-PP at that site, and the postpacing in-terval was longer than the VT cycle length. This

REENTRANT CIRCUIT OF VERAPAMIL-SENSITIVE VT



Figure 2. Radiographs from the right and left anterior oblique projections showing the locationof the left ventricular mapping catheter (LVmap) in patient 3. Four other catheters were locatedon the high right atrium (HRA), His-bundle region (His), right ventricular apex (RVA), and rightventricular outflow tract (RVOT). The mapping catheter was located on the middle of the leftventricular septum.

Figure 1. Electroendocardial recordings during ventricular tachycardia (VT) (B) and sinus rhythm(A) in patient 2. Surface electrocardiograms (ECGs) (II,V1) and His-bundle electrogram (HBE) areshown at the top, along with recording from the proximal to distal sites of the octapolar mappingcatheter. This mapping catheter was located on the middle of the left ventricular septum. DuringVT, the earliest Purkinje potential (PP) was recorded at LV4–5 and PP was activated toward thebasal and apical sites of the left ventricle. Moreover, the pre-Purkinje potential (pre-PP) wasrecorded at LV7–8 to LV4–5, which was activated toward the earliest PP. During sinus rhythm,the PP was activated from the proximal to distal site of the left ventricle and the pre-PP was notrecorded. A 5 local atrial electrogram, H 5 His potential; V 5 local ventricular electrogram; H’5retrograde His potential.

AIBA, ET AL.



A

B

Figure 3. Electroendocardial recordings during ventricular tachycardia (VT) with pacing fromthe right ventricular outflow tract in patient 5. Two surface electrocardiograms (ECGs) (II ,V1) areshown at the top, along with recording from the proximal to distal site of the mapping catheter(LV prox, mid, and dist). Pacing from this site confirmed manifest entrainment, and the intervalbetween pre-Purkinje potentials (pre-PP) before and after the last pacing stimulus was equal tothe pacing cycle length. Thus, pre-PP was orthodromically activated by these stimuli. Moreover,as the pacing cycle length was changed from 285 ms (A) to 275 ms (B), the stimulus to pre-PPinterval was prolonged from 278 ms to 292 ms, but the activation time from the earliest pre-PP toPurkinje potential (PP) was not changed. S 5 stimulation.

A

B

Figure 4. Electroendocardial recordings during ventricular tachycardia (VT) in patient 6. Four surfaceelectrocardiograms (ECGs) (I, II, V2,V4) are shown at the top, along with recordings from the proximal to the distal siteof the octapolar mapping catheter (LV prox, mid, and dist). (A) During VT, at a cycle length of 320 ms, pacing (cyclelength 310 ms) from the distal site could capture only the Purkinje potential (PP) and caused concealed entrainment.The postpacing interval at this site was equal to the VT cycle length. Thus, this pacing site was on the reentrant circuitof VT. The stimulus to QRS interval (St-QRS 20 ms) was equal to the PP to QRS interval during VT. The PP wasactivated from the distal to proximal site by pacing stimuli. However, the pre-Purkinje potential (pre-PP) was activatedfrom the proximal to the distal site by pacing stimuli, and the interval between the pre-PPs before and after the lastpacing stimulus was equal to the pacing cycle length. Thus, the pre-PP was orthodromically activated by these pacingstimuli. The activation time from the PP to pre-PP at the proximal site during pacing was long (235 ms) and the sameas that during VT. (B) The pacing cycle length was changed from 310 to 300 ms during VT. This pacing also causedconcealed entrainment, and the postpacing interval was longer (10 ms) than the VT cycle length. The activation timefrom the PP to pre-PP at the proximal site during pacing (245 ms) was longer than that during VT (235 ms). Thisprolongation was equal to the prolongation of the postpacing interval. Thus, this prolongation can be explained bythe decremental conduction of slow conduction during pacing. The St-QRS interval was not changed at this distalpacing site. S 5 stimulation; I 5 last paced QRS complex.

prolongation of the postpacing interval was equalto the prolongation of the PP to pre-PP interval atthe proximal site (Fig. 4B).

During pacing from the proximal PP recordingsite, the postpacing interval of this site was equalto the cycle length of VT, and the stimulus to QRSinterval was equal to the PP to QRS interval duringVT. The interval from stimulus to pre-PP was longand equal to the interval from the PP to pre-PP dur-ing VT (Fig. 5A). However, more rapid pacing pro-duced a delay in activation from stimulus to pre-PP, and the postpacing interval was longer than theVT cycle length. This prolongation of the postpac-ing interval was equal to the prolongation of thestimulus to pre-PP interval (Fig. 5B).

RF Catheter AblationRF catheter ablation was delivered during VT

at the pre-PP recording site by the ablationcatheter in all patients. This ablation site was notalways the earliest pre-PP recording site by the oc-tapolar catheter (Fig. 6). The pre-PP to QRS inter-val of ablation site was significantly shorter thanthat found with the octapolar catheter (42 6 13 vs53 6 22 ms, P , 0.05) (Tables II and III). Patient 4needed three applications, patient 1 needed twoapplications, and the other patients needed onlyone application for termination of the VT (TableIII). No complications occurred in any patient.

Follow-Up

Patients have been followed for 2 ; 32months (median 18 months) without antiarrhyth-

mic drugs. None of the patients has had a recur-rence of VT.

DiscussionThe main findings of this study were that the

pre-PP could be recorded within a small areaslightly proximal to the earliest PP recording siteduring VT, and that it was activated from the prox-imal (basal) to the distal (apical) sites of the leftventricle toward the earliest PP site. The directionof activation of pre-PP was different from those ofPP. Concealed entrainment was demonstrated bycapture of the PPs at the distal and proximal sitesof the mapping catheter, and the postpacing inter-val of these PPs was equal to the cycle length ofVT. The stimulus to QRS interval was equal to thePP to QRS interval during VT at each pacing site.More rapid pacing during VT also suggested adecremental property existed in the activationfrom PP to pre-PP, which might explain the slowconduction in ILVT. Thus, the reentrant circuit ofILVT could be demonstrated by these three com-ponents: PP, pre-PP, and slow conduction be-tween the PP and pre-PP. RF catheter ablation wassuccessful at the pre-PP recording site in all pa-tients.

Relationship Between the Pre-P Potential and PPotential

It has been suggested that catheter ablation ofverapamil-sensitive ILVT is successful at the siteof recording the earliest PP.7 Previous reports sug-gested that the PP might be on the exit site of the



Table II.

The Pre-PP and PP in Octapolar Mapping Catheter

Activation Time (msc)Distance (mm)

earliest earliest earliest pre-PP-Entrainment Study

Pt. No pre-PP-QRS PP-QRS earliest PP RVOT LV Map

1 86 28 26 1 22 42 11 22 1 1

3 74 18 30 1 2

4 23 16 22 1 2

5 40 15 18 1 1

6 86 8 26 1 1

7 50 14 18 1 18 42 18 18 1 1

9 43 21 18 1 2

10 39 20 18 1 1

mean 6 SD 53 6 22 17 6 6 22 6 4

pre-PP 5 pre-Purkinje potential; PP 5 Purkinje potential; RVOT 5 right ventricular outflow tract; LV map 5 leftventricular mapping catheter.

A

B

Figure 5. Electroendocardial recordings during ventricular tachycardia (VT) in patient 6. Four surface electrocardiograms(ECGs) (I, II, V2, V4) are shown at the top, along with recordings from the proximal to distal sites of the octapolar mappingcatheter (left ventricular [LV] prox, mid, and dist). (A) During VT, at a cycle length of 315 ms, pacing (cycle length 300 ms)from the proximal site could capture only the Purkinje potential (PP) and caused concealed entrainment. The postpacinginterval at this site was equal to the VT cycle length. Thus, this pacing site was on the reentrant circuit of VT. The thirdQRS complex (I) was the last paced QRS complex, because the R-R interval (300 ms) was equal to the pacing cycle length.The stimulus to QRS interval (St-QRS 320 ms) was long and equal to the PP-QRS interval during VT. The pre-Purkinjepotential (pre-PP) was activated from the proximal to the distal site by pacing stimuli, and the interval between the pre-PPs before and after the last pacing stimulus were equal to the pacing cycle length. Thus, the pre-PP was orthodromicallycaptured by pacing stimuli. The interval from the stimulus to pre-PP at the proximal site was long (230 ms) and equal tothe interval from the PP to pre-PP during VT. (B) The pacing cycle length was changed from 300 to 290 ms during VT. Thispacing also caused concealed entrainment. The postpacing interval at this pacing cycle length was longer than the VT cyclelength. The second QRS complex (I) was the last paced QRS complex, because the R-R interval (290 ms) was equal to thepacing cycle length. The St-QRS interval (35 ms) at this pacing cycle length was much shorter than that at a cycle lengthof 300 ms. The pre-PP was orthodromically captured by pacing stimuli, and the activation time from the stimulus to pre-PP (255 ms) was longer than that from the PP to pre-PP (235 ms) during VT. This prolongation of the activation time (20ms) was equal to the prolongation of the postpacing interval. S 5 stimulation; I 5last paced QRS complex



Figure 6. Electroendocardial recordings during ventricular tachycardia (VT) from octapolarcatheter (A) and ablation catheter at the successful ablation site (B) in patient 3. Surfaceelectrocardiograms (ECGs) (II,V1) are shown at the top, along with recording from the proximalto distal sites of the octapolar catheter (left ventricular [LV] prox, mid, and dist) in A, and alongwith recording from the distal site of ablation catheter (Abl) in B. The pre-Purkinje potential (pre-PP) and Purkinje potential (PP) were recorded in both catheters. The pre-PP to QRS interval ofablation site was 17-ms shorter than that found with the octapolar catheter. V 5 local ventricularelectrogram.

Table III.

Catheter Ablation

VT TerminationPt. No Abl (n) W/Sec/°C pre-PP-QRS(ms) PP-QRS(ms) pre-PP-PP(ms) Effect (seconds)

1 1 48/60/47 50 25 25 1 3.62 47/41/47 50 15 35 1 9

2 1 40/60/47 45 16 29 1 4.53 1 47/60/46 57 11 46 1 VT acceleration4 1 28/60/53 2 13 2 2

2 5/30/59 2 11 2 23 45/60/48 24 6 18 1 bump

5 1 38/60/49 32 10 22 1 4.66 1 47/60/45 65 25 70 1 47 1 33/60/52 38 14 24 1 18 1 36/60/51 28 10 18 1 19 1 43/60/49 34 20 14 1 5

10 1 33/60/52 44 18 26 1 1.8

mean 6 SD 1.3 6 0.6 43 6 13 13 6 7 30 6 16

pre-PP 5 pre-Purkinje potential; PP 5 Purkinje potential; Ab 5 number of radiofrequency catheter ablation applications.

reentrant circuit in ILVT.8,9,13 Recent studies re-ported that a comparatively dull potential (pre-PP)was recorded during ILVT preceding the PP. Theauthors previously reported a similar case of ILVTin which the mid-diastolic potential was on thereentrant circuit.11 Tada et al.9 reported two casesof ILVT with two distinct presystolic potentialsduring tachycardia in the mid-septal or inferoapi-cal area. Their two distinct presystolic potentialswere similar to the pre-PP and PP. Moreover, theyreported the decremental property of the pre-PPduring atrial or ventricular pacing, and suggestedthat the Purkinje tissue with decremental proper-ties was responsible for the tachycardia mecha-nism. However, the relationship between thesetwo presystolic potentials was unclear because themapping site area was small and limited.

In this study, using an octapolar mappingcatheter, the sequences of the PP and pre-PP wereclearly determined. The earliest PP was recordedat the inferoposterior septum in the left ventricleduring VT, and the PP activated toward the basaland apical sites of the left ventricle along the pos-terior fascicle. However, the pre-PP was recordedwithin a small area slightly proximal to the earli-est PP recording site, and it was activated from thebasal to apical site toward the earliest PP site. Thedirection of activation of the pre-PP was differentfrom that of the PP. Thus, the pre-PP clearly dif-fered from the PP.

Entrainment and Resetting StudiesSome studies have reported the response of

pacing stimuli during tachycardia delivered fromoutside the reentrant circuit of ILVT.6,12 Tsuchiyaet al.12 reported a late-diastolic potential (LDP) pre-ceding the PP in ILVT, and suggested that LDP wasthe potential at the entrance site of a slow conduc-tion zone because a delay in conduction existed be-tween the LDP and PPs in response to the increas-ing rate of RVOT pacing. They, therefore, suggestedthat slow conduction of ILVT existed between theLDP and PP. However, they did not demonstrateelectroendocardial recordings obtained during pac-ing from the LDP site or other sites on the reentrantcircuit of ILVT, and the relationship between theactual reentrant circuit and the LDP and/or PPscould not be determined because of the small num-ber of mapping electrodes. Their LDP and our pre-PP are not the same potential, because the re-sponses to these potentials on pacing from RVOTdiffer completely: more rapid pacing from RVOT inthis study prolonged the activation time from stim-ulus to pre-PP, but not that from the pre-PP to PP.

While recording pre-PP and PP during VTwith the mapping catheter, low output pacing

from the distal or proximal PP sites could captureonly the PP and caused concealed entrainment.The postpacing interval of these sites was equal tothe VT cycle length. These findings suggested thatthese distal and proximal pacing sites were on thereentrant circuit of ILVT. The reentrant circuit ofILVT is shown schematically in Figure 7. The pre-PP was activated from the proximal to the distalsite by pacing stimuli, and the interval betweenthe pre-PPs before and after the last pacing stimu-lus was equal to the pacing cycle length. Thus, thepre-PP was orthodromically activated by thesepacing stimuli. The activation time from the PP topre-PP during pacing was long, and the same asthat during VT. Moreover, more rapid pacing pro-duced a delay in activation from the PP to pre-PPbut not in the pre-PP itself. This prolongation ofactivation was equal to that of the postpacing in-terval. These findings suggested that a characteris-tic of decremental property existed between thePP and pre-PP at the proximal site of the mappingcatheter that caused the slow conduction in ILVT.

The stimulus to QRS interval was equal to thePP to QRS interval at the distal and proximal pac-ing sites. However, more rapid pacing from theproximal site, but not from the distal site, distinctly

AIBA, ET AL.


Figure 7. Suggested schema for the reentrant circuit ofidiopathic left ventricular tachycardia (ILVT) in the rightanterior oblique projection. During ventriculartachycardia (VT), the pre-Purkinje potential (pre-PP) wasslowly activated from the basal to the apical site of the leftventricular septum toward the earliest Purkinje potential(PP), and the PP was quickly activated toward the basaland apical sites of the left ventricle along the posteriorfascicle. Slow conduction of this reentrant circuit existsbetween the PP and pre-PP at the proximal site.

shortened the stimulus to QRS interval. The reen-try model in the present study (Fig. 8) can explainthis phenomenon: the site of collision between theprevious (n-1) and current (n) wave was movedfrom the PP to the pre-PP by a shorter pacing cyclelength. Thus, the change in stimulus to QRS inter-val resulted from this antidromic activation.

Reentrant Circuit of ILVT

While many previous studies have suggestedthat the mechanism of ILVT is reentry and thattachycardia might originate from the Purkinjefiber network of the posterior fascicle of the leftbundle branch,1–8 the size of this reentrant circuithas not been clearly determined.8,10,14,15 Wen etal.8 reported that the slow conduction zone of thereentry circuit was of considerable size, extendingfrom the mid-septum to the inferior apical septumof the left ventricle. Lai et al.10 reported that thedistance from the entrance to the exit site of thistachycardia was . 2 cm. In the present study, en-

trainment and resetting studies demonstrated thatthe earliest PP site (distal pacing site) and the siteof PP along with the earliest pre-PP (proximal pac-ing site) were on the reentrant circuit of ILVT.Slow conduction of ILVT might exist nearby at thesite of the PP along with the earliest pre-PP. Thesefindings suggest that the reentrant circuit of ILVTis at least of considerable size.

The reentrant circuit of ILVT could be con-structed from the PP, pre-PP, and the slow conduc-tion between PP and pre-PP. During VT, the pre-PPwas slowly activated from the basal to the apicalsite of the left ventricular septum toward the earli-est PP, and the PP was quickly activated bidirec-tionally toward the basal and apical sites of the leftventricle along the posterior fascicle. Moreover, thedecremental property existed between the PP andthe earliest pre-PP at the proximal site, indicatingslow conduction of ILVT. However, during sinusrhythm, the pre-PP could not be recorded. Thus,the pre-PP appears to be functional.



Figure 8. Suggested schema for the reentrant circuit and the entrainment from the proximal anddistal sites of the Purkinje potential (PP) in idiopathic left ventricular tachycardia (ILVT). Thereentrant circuit of ILVT was constructed from the pre-Purkinje potential (pre-PP), PP, and theslow conduction between the PP and pre-PP. Pacing from these sites during VT could capture onlythe Purkinje network and caused concealed entrainment. The postpacing interval of these PPswas equal to the VT cycle length. Thus, these pacing sites were on the reentrant circuit. The pre-PP was orthodromically captured during pacing. The PP was antidromically captured during theproximal pacing, but orthodromically captured during the distal pacing. When the pacing cyclelength was shortened, a conduction delay occurred between the PP and pre-PP at the proximalsite, and the site of collision between the previous (n-1) and current (n) waves moved from the PPto pre-PP site. Thus, the stimulus to QRS interval was changed from long to short by this antidromicactivation during proximal pacing. However, during distal pacing, the stimulus to QRS intervalwas short and unchanged. () 5 pacing cycle length.

Clinical Implications

In a previous study, catheter ablation of ILVTwas performed at the site of the earliest PP duringtachycardia.7 This method was based on the ideathat the earliest PP was the potential of the exit siteof this tachycardia. However, it was unclear whichsite was the earliest PP site, and endocardial map-ping to search for the earliest PP might consumemuch time. In this study, we performed left ven-tricular endocardial mapping during sinus rhythmand VT using an octapolar catheter, and the se-quences of the PP and pre-PP could be recorded si-multaneously in all patients. Catheter ablation wasperformed at the pre-PP recording site and wassuccessful in all patients. This ablation site wasnot always the same as the earliest pre-PP record-ing site by octapolar mapping catheter. The pre-PPto QRS interval of successful ablation sites was sig-nificantly shorter than the earliest pre-PP to QRSinterval of the mapping catheter. These findingssuggested that catheter ablation might be success-ful anywhere pre-PP can be recorded, even if it isnot the earliest one. Thus, pre-PP is a new markerthat can easily determine a site for successful abla-tion away from the exit site of ILVT.

Study LimitationsIn four of ten patients, concealed entrainment

pacing from the PP of the distal and proximal pairof the mapping catheter could not be demon-strated. This is because the amplitude of the PPwas so small that pacing from the catheter elec-trodes captured not only the PP of reentrant cir-

cuit but also the local ventricular myocardium,which was out of the reentrant circuit. Thus, itwas difficult to demonstrate the concealed en-trainment by pacing from these sites in ILVT.Moreover, capture of the slow conduction or thepre-PP itself could not be demonstrated, becausethe amplitude of these potentials was muchsmaller. Thus, other interpretations could not becompletely rule out, like pre-PP was not a part ofthe reentrant circuit but a bystander. However, theactivation time from the PP to pre-PP at the proxi-mal site was related to the VT cycle length, and theprolongation of the postpacing interval was equalto the prolongation of activation from PP to pre-PP. These findings could not be explained if thepre-PP were a bystander.

The role of pre-PP and PP in the reentrant cir-cuit of ILVT could be determined. However, the ac-tual slow conduction pathway connecting the PPand pre-PP was not determined because it was dif-ficult to record the actual slow conduction poten-tial by the mapping and recording methods weused.

ConclusionThe authors’ endocardial mapping, entrain-

ment, and resetting studies using an octapolarcatheter during VT can explain the role of the pre-PP and PP in ILVT. It appears that the reentrantcircuit of ILVT can be constructed from the pre-PP, PP, and the slow conduction between the PPand pre-PP, and the circuit is of considerable size.Catheter ablation of ILVT was successful at thepre-PP recording site.

AIBA, ET AL.


References1. Ohe T, Shimomura K, Aihara N, et al. Idiopathic sustained left ven-

tricular tachycardia: Clinical and electrophysiologic characteris-tics. Circulation 1988; 77:560–568.

2. German LD, Packer DL, Bardy GH, et al. Ventricular tachycardia in-duced by atrial stimulation in patients without symptomatic car-diac disease. Am J Cardiol 1983; 52:1202–1207.

3. Belhassen B, Rotmensch HH, Laniado S. Response of recurrentsustained ventricular tachycardia to verapamil. Br Heart J 1981;46:679–682.

4. Okumura K, Matsuyama K, Miyagi H, et al. Entrainment of idio-pathic ventricular tachycardia of left ventricular origin with evi-dence for reentry with an area of slow conduction and effect of ve-rapamil. Am J Cardiol 1988; 62:727–732.

5. Kottkamp H, Hindricks G, Willems S, et al. Idiopathic left ventric-ular tachycardia: New insights into electrophysiological character-istics and radiofrequency catheter ablation. PACE 1995;18:1285–1297.

6. Okumura K, Yamabe H, Tsuchiya T, et al. Characteristics of slowconduction zone demonstrated during entrainment of idiopathicventricular tachycardia of left ventricular origin. Am J Cardiol1996; 77:379–383.

7. Nakagawa H, Beckman KJ, McClelland JH, et al. Radiofrequencycatheter ablation of idiopathic left ventricular tachycardia guidedby a Purkinje potential. Circulation 1993; 88:2607–2617.

8. Wen MS, Yeh SJ, Wang CC, et al. Successful radiofrequency abla-tion of idiopathic left ventricular tachycardia at a site away fromthe tachycardia exit. J Am Coll Cardiol 1997; 30:1025–1031.

9. Tada H, Nogami A, Naito S, et al. Retrograde Purkinje potential ac-tivation during sinus rhythm following catheter ablation of idio-pathic left ventricular tachycardia. J Cardiovasc Electrophysiol1998; 9:1218–1224.

10. Lai LP, Lin JL, Hwang JJ, et al. Entrance site of slow conductionzone of verapamil-sensitive idiopathic left ventricular tachycardia:Evidence supporting macroreentry in the Purkinje system. J Car-diovasc Electrophysiol 1998; 9:184–190.

11. Aiba T, Suyama K, Matsuo K, et al. Mid-diastolic potential is re-lated to the reentrant circuit in a patient with verapamil-sensitiveidiopathic left ventricular tachycardia. J Cardiovasc Electrophysiol1998; 9:1004–1007.

12. Tsuchiya T, Okumura K, Honda T, et al. Significance of late dias-tolic potential preceding Purkinje potential in verapamil-sensitiveidiopathic left ventricular tachycardia. Circulation 1999;99:2408–2413.

13. Chen YJ, Chen SA, Tai CT, et al. Radiofrequency ablation of idio-pathic left ventricular tachycardia with change ECG morphology.PACE 1998; 21:1668–1671.

14. Kottkamp H, Chen X, Hindricks G, et al. Radiofrequency catheterablation of idiopathic left ventricular tachycardia: Further evi-dence for microreentry as the underlying mechanism. J CardiovascElectrophysiol 1994; 5:268–273.

15. Thakur RK, Klein GJ, Sivaram CA, et al. Anatomic substrate for idio-pathic left ventricular tachycardia. Circulation 1996; 93:497–501.

the role of purkinje and pre-purkinje potentials in the reentrant circuit of verapamil-sensitive...

Documents