reinitiation of ventricular macroreentry within the his-purkinje system by back-up ventricular...

Reinitiation of Ventricular Macroreentry within theHis-Purkinje System by Back-Up Ventricular Pacing—AMechanism of Ventricular Tachycardia StormCHRISTOPHER REITHMANN, M.D., ANTON HAHNEFELD, M.D., NICO OVERSOHL, M.D.,MICHAEL ULBRICH, M.D., THOMAS REMP, M.D., and GERHARD STEINBECK, M.D.From the Medizinische Klinik I, Klinikum Grosshadern, Universitat Munchen, Germany

Background: We describe immediate reinitiation of macroreentry ventricular tachycardia (VT) involvingthe His-Purkinje system by ventricular pacing from the electrode of an implantable cardioverter defibril-lator (ICD) as a mechanism of VT storm refractory to ICD therapy.

Methods and Results: Repetitive reinitiation of bundle branch reentry tachycardia (BBRT), interfascic-ular tachycardia, or both VTs by ventricular pacing was identified in four ICD patients presenting with VTstorm or incessant VT. All patients had a pre-existing prolonged HV interval (75 ± 9 ms) and left bundlebranch block (LBBB) or bifascicular block during sinus rhythm. The VTs included BBRT with LBBB inthree patients and interfascicular tachycardia with right bundle branch block (RBBB) and left anterior orleft posterior fascicular block in two patients. The paced beats from the ICD electrode exhibited a LBBBpattern of depolarization in two patients and a RBBB contour in V1 and V2 with left axis deviation in twopatients. The QRS complex during pacing from the ICD electrode closely resembled that of the recurrentVT in all four patients suggesting that the pacing site of the ICD electrode was in proximity to the myocar-dial exit site of the bundle fascicle used for antegrade conduction during the reinitiated VT. Ventricularpacing from the ICD electrode after termination of the VT apparently encountered the retrograde refrac-toriness of this bundle fascicle and allowed immediate re-propagation of the wavefront orthodromicallyalong the VT circuit. BBRT was eliminated by ablation of the right bundle branch. Successful ablationof the interfascicular tachycardias was achieved by targeting (1) an abnormal potential of the distal leftposterior Purkinje network or (2) a diastolic potential during VT in the midinferior left ventricular (LV)septum.

Conclusions: Repetitive reinitiation of BBRT and interfascicular tachycardia by ventricular pacing fromthe ICD electrode should be considered as a mechanism of VT storm refractory to ICD therapy in patientswith a pre-existing conduction delay within the His-Purkinje system. (PACE 2007; 30:225–235)

ventricular tachycardia, bundle branch reentry tachycardia, interfascicular tachycardia,implantable cardioverter defibrillator, catheter ablation

IntroductionIn patients with a conduction delay within the

His-Purkinje system, macroreentry involving theright and left bundles can result in sustained ven-tricular tachycardia (VT). Bundle branch reentrytachycardia (BBRT) usually depends on a retro-grade block in the right bundle with slow retro-grade conduction through the left bundle, whichallows sufficient time for recovery of subsequentanterograde conduction through the right bun-dle.1–3 Ventricular depolarization begins at the

None of the authors have any financial associations that mightpose a conflict of interest in connection with the submittedarticle. Furthermore, none of the authors have any other kindsof associations that need to be disclosed.

Address for reprints: Christopher Reithmann, M.D., Medizinis-che Klinik I, Klinikum Grosshadern, Universitat Munchen, Ger-many. Fax: +49-8970958830; e-mail: [email protected]

Received July 24, 2006; revised September 8, 2006; October 25,2006.

distal end of the right bundle, and BBRT usu-ally presents with a typical left bundle branchblock (LBBB) pattern. Interfascicular tachycardiais another form of macroreentry within the His-Purkinje system, which usually proceeds in an an-terograde direction over the left anterior fascicleand retrograde through the posterior fascicle.4–6

In this case, the initial site of ventricular depo-larization is at the distal end of the left anteriorfascicle and the QRS morphology has a right bun-dle branch block (RBBB) and left posterior fasci-cular block (LPFB) pattern. Typically, conductionmay occur in the opposite direction leading to aVT with RBBB and left anterior fascicular block(LAFB) pattern.

The BBRT and interfascicular tachycardia canbe eliminated by catheter ablation. However, themajority of these patients have significant left ven-tricular dysfunction, and implantable cardioverterdefibrillator (ICD) therapy may be necessary totreat intramyocardial VT and to prevent sud-den cardiac death.7–9 Frequent ICD therapies are

C©2007, The Authors. Journal compilation C©2007, Blackwell Publishing, Inc.

PACE, Vol. 30 February 2007 225

REITHMANN, ET AL.

an important clinical problem for ICD patientswith previously undetected or newly developingmacroreentry VT within the His-Purkinje systembecause they have a high rate of recurrence and re-spond poorly to pharmacologic therapy. We reporton four ICD patients in whom repetitive reiniti-ation of BBRT and interfascicular tachycardia byback-up ventricular stimulation was identified asthe mechanism of VT storm refractory to ICD ther-apy.

Patients and MethodsPatients

Forty ICD patients with VT storm (≥5 appro-priate ICD therapies per day or ≥3 shocks perday) or incessant VT underwent electrophysiolog-ical (EP) study and VT ablation. Recurrent bundlebranch reentry tachycardia (BBRT) or interfascic-ular tachycardia was found in nine patients. ICDinterrogation and EP study revealed that repetitivereinitiation of BBRT or interfascicular tachycardiaby antibradycardia pacing from the ICD electrodesubsequent to antitachycardia pacing or defibrilla-tion was the mechanism of VT storm in four pa-tients (10%).

Among them, two patients had an ischemiccardiomyopathy, one patient had a dilated car-diomyopathy and a 79-year-old female patient had

Table I.

Patient Characteristics, Type of ICD, Antibradycardia, and Post-Shock Pacing Mode

Structural AntibradycardiaAge Initial Heart LV-EF Pacing Mode

Patient (Years) Clinical VT Disease (%) Type of ICD (Programmable) Pacing Mode

1 71 170/min ECGpattern: LSB

ICM 24 Medtronic intrinsicdual chamber(Minneapolis, MN,USA)

MVP, AAI/R 60option DDD/R 60

Blanking interval 520ms, followed byDDD, MVPde-activated for 60seconds

2 79 180/min ECGpattern: LSB

No 70 Medtronic maximodual chamber

DDI 50–120 Blanking interval 520ms, escapeinterval 1,200 msfollowed by VVI for25 events

3 78 Nodocumentationavailable

ICM 33 ELA alto 2 DR dualchamber (SorinGroup, Milan, Italy)

DDD/R 60 Blanking interval1,000 ms followedby DDD

4 42 160/min ECGpattern: RBBB+ LPFB

DCM 32 Medtronic intrinsicdual chamber

MVP, AAI/R 60option DDD/R 60

Blanking interval 520ms, followed byDDD, MVPde-activated for 60seconds

LV-EF = left ventricular ejection fraction; DCM = dilative cardiomyopathy; ICM = ischemic cardiomyopathy; MVP = managed ventricularpacing.

no detectable structural heart disease (Table I).During sinus rhythm, a complete LBBB was foundin three patients and a RBBB with LAFB in one pa-tient (Table II). First-degree atrioventricular block(PR ≥ 200 ms) was documented in all patients (PR265 ± 41 ms). All patients had been refractory toa mean of 3 ± 2 antiarrhythmic drugs includingamiodarone.

ICD Interrogation

Antitachycardia pacing, shock, antibradycar-dia pacing, and post-shock pacing were availablein all ICDs. The type of devices and mode of an-tibradycardia and post-shock pacing is given inTable I. Antitachycardia pacing, antibradycardiapacing, and post-shock pacing were delivered froma bipolar electrode system included in the ICD leadin three patients and from an additional bipolarpace-sense electrode in one patient (patient 2). Inone case, the ICD interrogation (Medtronic Max-imoDR, Minneapolis, MN, USA) at arrival of thepatient revealed an unprogrammable reset modefollowing multiple ineffective ICD interventionsfor incessant VT (patient 2).

Electrophysiologic Study

After giving informed consent, all patients un-derwent EP study. In one patient presenting with

226 February 2007 PACE, Vol. 30

VENTRICULAR TACHYCARDIA STORM IN ICD PATIENTS

Table II.

Electrocardiographic and Electrophysiologic Determinants

PR Bundle Branch ECG Pattern ECG PatternDuring Block During Ventricular During

Patient SR (ms) During SR HV (ms) Pacing from ICD Repetitive VTs EP Study

1 240 LBBB 70 LBBB + leftward axis LBBB + leftward axis BBRT2 300 LBBB 70 LBBB + normal axis LBBB + normal axis BBRT3 300 LBBB 88 RBBB + left axis

deviationVT1, LBBB

VT2, RBBB + LAFBVT1, BBRT

VT2, interfasc.tachy4 220 RBBB + LAFB 73 RBBB + left axis

deviationVT1, RBBB + LAFB

VT2, RBBB + LPFBVT1, interfasc.tachy

VT2, interfasc.tachy

PR = PR interval; SR = sinus rhythm; HV = His ventricular interval; VT = ventricular tachycardia; LBBB = left bundle branch block;LAFB = left anterior fascicular block; LPFB = left posterior fascicular block; RBBB = right bundle branch block; BBRT = bundle branchreentry tachycardia; interfasc.tachy = interfascicular tachycardia.

incessant VT, EP study and catheter ablation hadto be performed under general anesthesia in intu-bation narcosis (patient 2). Antitachycardia pac-ing and shock were deactivated during the proce-dure in all patients. Catheters were introduced tothe right ventricular (RV) apex and to the His bun-dle region or right bundle branch. Left ventricular(LV) mapping for interfascicular tachycardia wasperformed retrogradely with a steerable catheterwith a 4 mm electrode tip (Cordis-Webster Dia-mond Bar, CA, USA). Activation mapping duringinterfascicular tachycardia was followed by a de-tailed mapping of the Purkinje network of the LVseptum during sinus rhythm. The stimulation pro-tocol consisted of a programmed ventricular stim-ulation from the RV apex and the RV outflow tractat four cycle lengths with up to three extrastimuli.If ventricular tachycardia was not inducible by RVendocardial stimulation, stimulation was repeatedfrom the endocardial LV apex and from the ICDelectrode.

Catheter Ablation

Radiofrequency (RF) current was deliveredfrom the distal electrode of a 4-mm tipped standardablation catheter (Cordis-Webster). A 500 kHz RFablation unit (Cordis-Webster) was used for abla-tion. The current was initially applied at a powerof 10–20 W and was increased every 10–15 sec-onds to a maximum output of 30–40 W underimpedance control.10 All patients underwent anEP re-evaluation 30 minutes after the ablation pro-cedure.

Follow-Up

Follow-up information was obtained in ouroutpatient clinic and by ICD interrogation.

Definitions

Diagnostic criteria for BBRT included (1) theQRS morphology of the tachycardia exhibits atypical LBBB pattern; (2) the onset of ventricu-lar depolarization is preceded by His (H) bundleor right bundle (RB) potentials, an appropriate se-quence of H-RB activation and stable His ventricu-lar (HV)- and right bundle ventricular (RB-V) inter-vals; (3) spontaneous variations in interventricular(VV) intervals are preceded by similar changes inRB-RB intervals; (4) the induction of tachycardiaduring programmed stimulation is consistently de-pendent on achieving a critical conduction delayin the His-Purkinje system; and (5) BBRT is non-inducible after successful RB ablation.7

Diagnostic criteria for interfascicular tachy-cardia included (1) the QRS morphology of thetachycardia exhibits a typical RBBB and LPFB pat-tern, a typical RBBB and LAFB pattern or both QRSmorphologies alternately; (2) the onset of ventricu-lar depolarization is preceded by a Purkinje poten-tial (PP) and stable PP ventricular (PP-V) intervals;(3) spontaneous variations in VV intervals are pre-ceded by similar changes in PP-PP intervals.

ResultsReinitiation of Ventricular Tachycardia

Incessant VT and VT storm due to repeti-tive reinitiations of BBRT or interfascicular tachy-cardia by ventricular back-up pacing were docu-mented in four ICD patients. They included reini-tiation of VT following antibradycardia pacingsubsequent to antitachycardia pacing (Fig. 1A)and subsequent to internal or external defibrilla-tion (Fig. 1B). In two patients with incessant VTand repetitive ineffective ICD therapies, immedi-ate reinitiation of the VT following ventricular


REITHMANN, ET AL.

Figure 1. Immediate reinitiation of interfascicular tachycardia by antibradycardia ventricularpacing (DDD) from the ICD electrode after antitachycardia pacing (A) and shock (B) (patient 4).Programming of the antibradycardia pacing mode to atrial pacing (AAI) interrupted the incessantVT (C). VP, antibradycardia ventricular pacing.

back-up pacing was avoided by programming an-tibradycardia pacing mode to atrial pacing (AAI)(Fig. 1C). Additionally, spontaneous initiations ofVT were documented in two patients (patients 1and 3).

ICD Therapy

Dual chamber ICD implantation had been per-formed in our (n = 2) or another institution (n = 2).The clinical arrhythmia leading to implantation ofthe device was hemodynamically poorly toleratedVT in three patients and VT with syncope in onepatient (patient 2). Twelve-lead ECGs of the clini-cal VT leading to ICD implantation were availablefrom three patients (Table I). In these patients, theECG morphology of the clinical VTs documentedbefore ICD implantation was similar to at least onemorphology of the repetitive VTs documented af-ter ICD implantation. All patients had undergoneEP study in our (n = 2) or another institution (n =2) before ICD implantation but BBRT or interfas-cicular tachycardia was not inducible or had notbeen diagnosed in any case. The following ventric-

ular tachyarrhythmias were inducible during theinitial EP study: nonclinical high frequent VT inpatients 1 and 4; hemodynamically nontoleratedVT (cycle length 310 ms) with LBBB pattern simi-lar to the clinical VT in patient 2; no EP data wereavailable from patient 3. Following ICD implanta-tion, VT storm was recorded after an interval of ≤2months (mean 17 days) in 3 patients and after aninterval of 12 months in one patient (patient 3).

Repetitive ICD Therapies

Two patients (patients 2 and 4) were referredto our intensive care unit with incessant VT andmultiple ineffective ICD therapies. In two patientswith ischemic cardiomyopathy (patients 1 and 3),ICD interrogation revealed more than 100 effectiveor ineffective ICD therapies for VT within the pre-ceding 2 weeks. At arrival in our institution, threepatients (among them the two patients with inces-sant VT) had only paced ventricular beats (DDDRor VVIR mode) during the VT-free intervals andone patient had intrinsic atrioventricular conduc-tion during sinus rhythm.



Figure 2. Incessant VT in a patient (patient 2) with repetitive reinitiations of BBRT by ventricularpacing from the ICD electrode. Twelve-lead ECGs during sinus rhythm (SR), ventricular pacingfrom the ICD electrode (ICD-DDD) and during the incessant VT. The RV angiogram in right anterioroblique (RAO) projection shows the position of the additional ICD pace-sense electrode in themedial RV septum (arrow).

VT Morphologies

Mean VT cycle length was 413 ± 45 ms. Twopatients presented with a VT with LBBB morphol-ogy and normal or leftward axis (Fig. 2). One pa-tient with dilated cardiomyopathy presented witha VT with RBBB and LPFB pattern alternating witha VT with RBBB and LAFB pattern (Fig. 3). In one

Figure 3. Incessant VT in a patient (patient 4) with repetitive reinitiations of interfasciculartachycardias by ventricular pacing from the ICD electrode. Twelve-lead ECGs during atrial pacing(ICD-AAI), ventricular pacing (ICD-DDD), and during the alternating incessant VTs with RBBBand LAFB (VT1) and RBBB and LPFB (VT2). The LV angiogram in RAO projection shows theposition of the ICD electrode.

patient with ischemic cardiomyopathy associatedwith marked regional wall motion abnormalitieswith akinesis in the inferior-basal and apical LV(patient 3), a VT with LBBB pattern and normalaxis (cycle length 470 ms) alternating with a VTwith RBBB and LAFB pattern (cycle length 470ms) was documented.


REITHMANN, ET AL.

Twelve-Lead ECG During Ventricular Pacingfrom the ICD Electrode

The ICD lead was implanted in the apex of theRV septum in all four patients and an additionalpace-sense electrode was implanted in the medialRV septum in one patient (chest x-ray). In the twopatients with repetitive reinitiations of BBRT, ven-tricular pacing from the ICD electrode produceda LBBB pattern with normal or leftward axis re-sembling the QRS morphology during the BBRT(Fig. 2, Table II). In the patient with repetitivereinitiations of alternating forms of interfascicu-lar tachycardia, the paced beats from the ICD elec-trode exhibited a RBBB contour in V1 and V2 withleft axis deviation similar to the QRS morphologyduring the interfascicular tachycardia with RBBBand LAFB (Fig. 3, Table II). In the patient pre-senting with both, BBRT and interfascicular tachy-cardia, pacing from the ICD electrode produced aRBBB pattern of depolarization with left axis devi-ation resembling that of the interfascicular tachy-cardia with RBBB and LAFB.

EP Study

At baseline, the HV interval was prolonged inall patients (75 ± 9 ms). During the EP study, theclinical VTs could be induced by RV apex stimula-tion in three patients. In a patient with dilated car-diomyopathy (patient 4), the interfascicular tachy-cardias could not be induced by programmed RVstimulation but they were reproducibly inducedby pacing from the LV apex and from the ICD elec-trode.

During the VTs with LBBB pattern, a potentialfrom the His bundle and the right bundle branchpreceded the QRS complexes (Fig. 4A), and thecriteria of a BBRT were fulfilled. During the VTwith RBBB and LPFB pattern and the VTs withRBBB and LAFB pattern, a short, sharp poten-tial preceding ventricular activation was recordedover a relatively large area, representing the ac-tivation of the left bundle and Purkinje system(Fig. 5A). This PP was recorded progressively laterfrom the distal to the proximal electrode along theposterior fascicle during the VT with RBBB andLPFB pattern suggesting that the posterior fasci-cle was activated retrogradely during this tachy-cardia. In the two cases of interfascicular tachy-cardia, a large area of diastolic potentials (DP) wasfound in the medial and inferior left ventricularseptum during the VTs with a mean DP to PP in-terval of 140 and 180 ms and a PP to DP intervalof 320 and 300 ms, respectively (Fig. 5A). TheseDPs apparently represent activation of the dis-eased Purkinje network during the interfasciculartachycardias.

Figure 4. Elimination of BBRT in a patient (patient 2)with repetitive reinitiations of VT by ventricular pacingtargeting the right bundle branch (RBB) (A). Ablationterminated the tachycardia after few seconds resultingin complete RBBB and was not re-initiated by the sub-sequent antibradycardia ventricular stimulus from theICD electrode (B).

Mapping of the LV Septum During Sinus Rhythm

In the patients with interfascicular tachycar-dias, a low-amplitude diastolic abnormal potentialwas identified during sinus rhythm in a large areaof the midinferior and inferior LV septum (Fig. 5B).These abnormal potentials apparently represent aretrograde Purkinje activation (retroPP) with slowconduction within the diseased Purkinje network.The earliest retroPPs were recorded 400 and 450ms and the latest retro PPs were recorded 560 and580 ms after the PP and ventricular activation dur-ing sinus rhythm.

Reinitiation of VT by Ventricular Pacing fromthe ICD Electrode

Figure 6 demonstrates (mechanical) ter-mination and immediate reinitiation of the



Figure 5. Catheter ablation of interfascicular tachycardia in a patient (patient 4) with repetitivereinitiations of VT by ventricular pacing from the ICD electrode. (A) During the VT, a Purkinjepotential (PP) (arrow) with a short PP-ventriculo interval and a large area of diastolic potentials(DP) (PP-DP interval: 300 ms) was found in the medial and inferior left ventricular septum. (B)During sinus rhythm, a diastolic abnormal low potential (retro PP) (asterisks) was identified ina large area of the midinferior and inferior LV septum (B, left). The potentials of the His bundleand left bundle fascicle are indicated by arrow. Mapping of the earliest retro PP during sinusrhythm was used to guide ablation of the interfascicular tachycardia. The PP-retro PP intervalsare given. Fluoroscopic view of the position of the ablation catheter in left anterior oblique (LAO)projection showing where the VT could be successfully ablated (B mid). Energy delivery at thesuccessful ablation position induced ventricular ectopic beats with a QRS morphology similar tothe incessant VT with RBBB and LAFB (VT1) and RBBB and LPFB (VT2) (B, right).

interfascicular tachycardia with RBBB and LPFBby ventricular pacing from the ICD electrode. Af-ter de-activation of ventricular pacing from theICD, termination of the VT is followed by sinusbeats that do not reinitiate the VT (Fig. 6A). Af-ter re-programming the antibradycardia ventricu-lar pacing mode (DDDR), a subsequent ventricularstimulus from the ICD electrode activates the Purk-inje network with a similar morphology as the DPduring the VT and a slightly longer QRS-DP in-terval (Fig. 6B). This pacing condition apparentlyallows re-propagation of the wavefront ortho-dromically along the VT circuit and immediatereinitiation of the tachycardia.

Ablation of BBRT

The clinical VTs were successfully ablatedin four patients. Ablation of the right bun-

dle branch eliminated BBRT in three patients(Fig. 4B). Complete RBBB was achieved in all caseswithout occurrence of complete atrioventricularblock.

Ablation of Interfascicular Tachycardia

In the two patients with reinitiation of in-terfascicular tachycardias, the VTs could not besuccessfully ablated targeting the proximal com-ponents of the left specific conduction system.This suggested that the reentry circuit involvedmore distal components of the left fascicles anddiseased Purkinje network. In the patient with di-lated cardiomyopathy and alternating interfascic-ular tachycardias (patient 4), mapping the earli-est retro PP during sinus rhythm in the midinfe-rior septum was used to guide successful ablationof the interfascicular tachycardia (Fig. 5 B left).


REITHMANN, ET AL.

Figure 6. Intracardiac recordings during terminationand immediate reinitiation of the interfascicular tachy-cardia. Termination of the VT during mapping of themidinferior LV septum (MAP). (A) Deactivation of an-tibradycardia ventricular pacing (programmed to AAI).The VT terminates after the DP (asterisks) with loss ofthe subsequent PP. In the following sinus beats, a late po-tential (retro PP) (asterisks) is recorded after ventricularactivation (QRS—retro PP 400 ms). The following sinusbeats do not reinitiate the interfascicular tachycardia.CS, coronary sinus. (B) Antibradycardia ventricular pac-ing activated (programmed to DDD). After terminationof the VT and a subsequent ventricular stimulus (s), apotential (asterisks) is recorded with a similar morphol-ogy as DP during the VT and with a slightly prolongedQRS-DP interval (380 ms). The paced ventricular stim-ulus from the ICD electrode immediately reinitiates theVT.

Ablation during sinus rhythm induced ventricu-lar ectopic beats with a QRS morphology similarto the clinical VT with RBBB and LAFB pattern(VT 1) and with RBBB and LPFB (VT 2) pattern,

Figure 7. Schematic illustration of reinitiation ofmacroreentry VT within the His-Purkinje system by ven-tricular pacing from the ICD electrode. (A) After termina-tion of BBRT, a paced ventricular stimulus from the ICDelectrode in proximity to the myocardial exit of the rightbundle (RB) encounters the retrograde refractoriness ofthe RB (dotted line) and allows proceeding of the BBRTretrogradely via the left bundle (LB). (B) After termina-tion of interfascicular tachycardia, a paced ventricularstimulus from the ICD electrode in proximity to the exitof the left posterior fascicle (LPF) encounters the retro-grade refractoriness of the left anterior fascicle (LAF)and/or a unidirectional block within the distal Purkinjenetwork and, thus, reinitiates the VT retrogradely via theLPF. Typically conduction may occur in the opposite di-rection. The myocardial exit during the VT and duringpacing from the ICD electrode are indicated by arrow.

(Fig. 5B right) suggesting that the successful ab-lation site was located within the distal Purkinjenetwork connected with the distal components ofboth left fascicles.

In a patient with ischemic cardiomyopathy(patient 3), interfascicular tachycardia with RBBBand LAFB was inducible after elimination of BBRTand could be successfully ablated targeting DPsin the midinferior LV septum during the VT.Complete or higher-degree atrioventricular blockdid not occur after ablation of the interfasciculartachycardias.



Nonclinical VTs

In addition to BBRT and interfascicular VT, ahigh-frequent polymorphic VT was inducible dur-ing the EP/ablation study in one patient (patient 4)but was not targeted by ablation.

Follow-Up

ICD interrogation revealed one successful an-titachycardia pacing episode for VT in a patientwith ischemic cardiomyopathy (patient 1) and oneasymptomatic successful antitachycardia pacingepisode for VT in a patient with dilated cardiomy-opathy (patient 4) during the follow-up of 12 and9 months, respectively. ICD interrogation and re-programming was not possible due to an unpro-grammable reset mode in patient 2, but the patientrefused implantation of a new ICD system and iswell without arrhythmias after a follow-up of 9months. One patient with ischemic cardiomyopa-thy (patient 3) died of severe pneumonia unrelatedto the intervention 6 months after the procedurewithout postablation ICD therapies.

DiscussionRepetitive reinitiation of macroreentry VT

involving the His-Purkinje system by antibrady-cardia or post-shock ventricular pacing was themechanism of incessant VT or VT storm in fourICD patients. In these patients, the bundle branchblock pattern and QRS axis produced by ventric-ular pacing from the ICD electrode closely resem-bled that of the reinitiated VT suggesting that theICD electrode was implanted in proximity to thedistal end of the fascicle used for antegrade con-duction during the VT.

Repetitive Ineffective ICD Therapies

Frequent or repetitive shocks constitutean electrophysiological emergency.11 The maincauses of repetitive ICD therapies are (1) the fail-ure of ICD antitachycardia pacing or shock therapyto terminate VT or ventricular fibrillation due topatient-related or ICD system-related reasons, (2)the spontaneous reinitiation of ventricular tachy-arrhythmias immediately after effective ICD ther-apy, (3) delivery of unnecessary repetitive shocksfor recurring self-terminating episodes of VT, (4)the misclassification of supraventricular tachycar-dia (SVT) or oversensing leading to inadequateICD therapies, and (5) the misclassification of post-shock nonsustained VT or post-shock SVT as inef-fective ICD therapy. Repetitive ICD shocks are as-sociated with a markedly reduced quality of life.Life threatening VT storm or incessant VT refrac-tory to ICD therapy can also be caused by acuteischemia, exacerbation of heart failure, metabolicabnormalities, and drug effects. In this report, wedescribe immediate reinitiation of BBRT and in-

terfascicular tachycardia by back-up ventricularpacing as an underrecognized mechanism of VTstorm refractory to ICD therapy, which was foundin 10% of consecutive patients with VT storm orincessant VT undergoing catheter ablation in ourinstitution.

Ventricular Macroreentry Involving theHis-Purkinje System in ICD Patients

It is important to recognize BBRT or inter-fascicular tachycardia as the mechanism of fre-quent ICD therapies because they can be readilyand permanently eliminated by the use of catheterablation. In our study, BBRT or interfasciculartachycardia was not inducible or has not beencorrectly diagnosed during the initial EP studybefore ICD implantation in all patients. In threepatients, VT storm or incessant VT due to BBRT orinterfascicular tachycardia occurred after an inter-val of less than 2 months after ICD implantation.Comparison of the 12-lead ECGs of the “clinical”VTs before and after ICD implantation showed asimilar VT morphology suggesting a pre-existingmacroreentry VT involving the His-Purkinje sys-tem in these patients. In one patient the VT stormoccurred more than 1 year after ICD implantationand no 12-lead ECG of the clinical VT prior to ICDimplantation was available. Development of ven-tricular macroreentry involving the right or leftbundle over time after ICD implantation due toprogression of the cardiac disease may be a pos-sible explanation for the delayed occurrence ofBBRT and interfascicular tachycardia in this pa-tient. Antiarrhythmic drug-induced slowing ofconduction velocity can favor the occurrence ofmacroreentry VT within the His-Purkinje sys-tem.12 It is possible that amiodarone-inducedslowing of conduction velocity within the HisPurkinje system may have favored the occurrenceof recurrent BBRT or interfascicular tachycardiasin our patients. VT storm is associated with in-creased adrenergic tone13,14 which may have facili-tated the repetitive reinitiations of sustained BBRTand interfascicular tachycardia.

Position of the ICD Electrode

The ICD lead including the distal coil andthe tip-ring electrode are usually implanted in theapex of the RV septum. Therefore, it is not un-expected that the tip of the ICD electrode or ofan additional pace-sense electrode may be locatedin vicinity to the distal end of the right bundlebranch. As the initial site of ventricular muscledepolarization during typical BBRT begins at theend of the right bundle branch in the midinferioror apical RV septum, the QRS morphology duringbipolar pacing from the ICD electrode may resem-ble the QRS morphology during the BBRT. In thetwo cases of interfascicular tachycardias, the QRS


REITHMANN, ET AL.

morphology during pacing from the ICD lead ex-hibited a RBBB contour in V1 and V2 with left axisdeviation. This suggested that the ICD electrodewas deeply implanted in the inferior interventricu-lar septum or has migrated to the left ventricle. TheRBBB pattern and QRS axis during pacing resem-bled that of the interfascicular tachycardia withRBBB and LAFB suggesting the tip of the electrodemay be located in proximity of the myocardial exitof the left posterior fascicle.

Antibradycardia and Post-Shock Pacing Mode

Due to the pre-existing conduction delaywithin the His-Purkinje system, termination ofBBRT and interfascicular tachycardia can befollowed by short periods of temporary atrioven-tricular block. In two ICD patients, the antibrady-cardia pacing mode of the device was an atrial-based managed ventricular pacing mode (MVP).This new pacing mode, which was designed tocombine the advantages of AAI with the safety ofDDD pacing, aims to provide AAI/R pacing withventricular monitoring during intrinsic conduc-tion and DDD/R during atrioventricular block. Inboth patients, the device automatically switched toDDD/R due to blocked atrial events after the tachy-cardia which then immediately reinitiated the VT.Similarly, the automatic mode switch of the de-vice to DDD (for 60 seconds) in the post-shock pac-ing mode was followed by repetitive reinitiationsof the VT after internal defibrillation. In the othertwo patients, the antibradycardia and post-shockpacing mode of the ICD was a VVI and DDD mode,respectively, and ventricular pacing in response toatrioventricular block immediately reinitiated theVT.

Unidirectional Block in the Distal PurkinjeNetwork

Depending on the bundle branch or fasciclethat is used for antegrade propagation, a retro-grade block in this bundle branch or fascicle isassumed to be the prerequisite of BBRT or inter-fascicular tachycardia. The results of this studysuggest that unidirectional block within the distalPurkinje network of the LV septum may play a rolein the reentry circuit of interfascicular tachycar-dia. The electrophysiological findings during in-terfascicular tachycardias show some striking sim-ilarities to recent findings of patients with idio-pathic left ventricular tachycardias with RBBB andleft axis deviation. Wen et al. have recently de-scribed a slow conduction zone of considerablesize extending from the midseptum to the infe-rior apical septum of the LV as an essential partof the reentry circuit in these patients with fas-cicular tachycardia.15 Ouyang et al. have iden-tified abnormal potentials in the posterior Purk-inje fiber network during sinus rhythm (SR) co-

inciding with diastolic potentials during the VTin patients with idiopathic left ventricular tachy-cardias.16 The authors suggested that these abnor-mal potentials represent a retrograde Purkinje ac-tivation (retro PP) with slow conduction over aPurkinje-Purkinje or Purkinje-ventricular Purkinjeconnection attributable to unidirectional block.Similarly, abnormal retro PP was identified dur-ing SR in a large area of the midinferior left ven-tricular septum in the two patients with interfas-cicular tachycardias in this study. The data sug-gest that the circuit of interfascicular tachycardiainvolves both left bundle fascicles and abnormalPurkinje tissue with slow conduction and uni-directional block. The coexistence of a VT withRBBB and LAFB pattern and a VT with RBBBand LPFB pattern is a criterion for interfascicu-lar tachycardia. In one patient with a BBRT anda LV tachycardia involving the His-Purkinje sys-tem (patient 3), only the LV tachycardia with RBBBand LAFB pattern was documented. It is possi-ble that in this patient the LV tachycardia onlyinvolved one left bundle fascicle and diseasedPurkinje network instead of involving both leftbundle fascicles. Three-dimensional mappingstudies will probably be helpful to delineate moreclearly the reentry circuit of fascicular and inter-fascicular tachycardia in patients with structuralheart disease.

Reinitiation of VT by Ventricular Pacing

Immediate reinitiation of BBRT with LBBBby ventricular pacing can occur when the pac-ing stimulus encounters the retrograde refractoryperiod of the right bundle branch (Fig. 7A). Thiscan mainly be accomplished from a pacing sitein close proximity to the end of the right bun-dle branch when retrograde RBBB is easier to setup. The impulse propagation then proceeds ret-rogradely along the left bundle branch because ofits shorter refractoriness and the delay associatedwith transseptal conduction. Initiation and main-tenance of BBRT are critically dependent on de-layed retrograde left bundle branch conduction asthe prerequisite for the recovery of subsequent an-tegrade conduction through the right bundle.

Similarly, after termination of interfasciculartachycardia, a pacing stimulus from the inferior LVseptum may encounter a retrograde block in oneof the left bundle fascicles and/or a unidirectionalblock in the distal Purkinje network (Fig. 7B). Af-ter retrograde conduction via the other left bundlefascicle, antegrade propagation may occur by wayof the right bundle branch and the recovered re-maining left bundle fascicle.

In the cases of repetitive reinitiation of BBRTor interfascicular tachycardias presented in thisstudy, the QRS morphology during pacing from theICD exhibited a similar bundle branch pattern and



QRS axis as compared to the reinitiated VT. Thissuggests that repetitive reinitiation of macroreen-try VT involving the His-Purkinje system may par-ticularly occur if the pacing electrode is located inproximity to the myocardial exit site of the bundlefascicle antegradely conducting during the VT.

Catheter Ablation

BBRT was eliminated by ablation of the rightbundle branch in all patients without completeatrioventricular block. In the two patients withinterfascicular tachycardia, VT ablation from thebase of the left posterior fascicle was not success-ful, suggesting that the reentrant circuit involvedmore distal parts of the left bundle fascicles andthe Purkinje network in the inferior LV septum.In one patient (patient 4), mapping of the earli-est retro PP during SR as described by Ouyang etal.16 was used for guiding successful ablation. Inthe other patient (patient 3) presenting with BBRTand interfascicular tachycardia, the interfasciculartachycardia could be eliminated by targeting DPsin the midinferior LV septum during the VT withRBBB and LAFB.

Limitations

The BBRT and interfascicular tachycardiawere probably the clinical arrhythmia leading to

ICD implantation in three patients and shouldhave been eliminated by catheter ablation beforeinitiation of the device therapy or even insteadof the device therapy (in one patient). The BBRTand interfascicular tachycardia had not been diag-nosed during the initial EP study before ICD im-plantation. It is important to note that careful EPevaluation including a short long protocol and pro-grammed LV stimulation may be necessary to initi-ate BBRT and interfascicular tachycardia, respec-tively.

ConclusionsImmediate reinitiation of BBRT and interfas-

cicular tachycardias by antibradycardia ventricu-lar pacing can lead to incessant VT refractory toICD therapy. If the ICD electrode is implanted inproximity to the myocardial exit site of the bun-dle fascicle antegradely conducting during the VT,ventricular pacing from the ICD electrode can ap-parently easily encounter the retrograde refractori-ness of the antegradely conducting bundle fascicleand allow re-propagation of the wavefront ortho-dromically along the VT circuit. The BBRT andinterfascicular tachycardia should be definitivelyeliminated by the use of catheter ablation before,or in appropriate cases, even instead of ICD im-plantation.

References1. Akhtar M, Damato AN, Batsford WP, Ruskin JN, Ogunkelu JB, Var-

gas G. Demonstration of reentry within the His-Purkinje system inman. Circulation 1974; 50:1150–1162.

2. Caceres J, Jazayeri M, McKinnie J, Avitall B, Denker ST, TchouP, Akhtar M. Sustained bundle branch reentry as a mechanism ofclinical tachycardia. Circulation 1989; 79:256–270.

3. Blanck Z, Dhala A, Desphande S, Deshpande S, Sra J, Jazayeri M,Akhtar M. Bundle branch reentrant ventricular tachycardia: Cumu-lative experience in 48 patients. J Cardiovasc Electrophysiol 1993;4:253–262.

4. Simons GR, Sorrentino RA, Zimerman LI, Wharton JM, Natale A.Bundle branch reentry tachycardia and possible sustained interfas-cicular reentry tachycardia with a shared unusual induction pat-tern. J Cardiovasc Electrophys 1996; 7:44–50.

5. Berger RD, Orias D, Kasper EK, Calkins H. Catheter ablation ofcoexistent bundle branch and interfascicular reentrant ventriculartachycardias. J Cardiovasc Electrophysiol 1996; 7:341–347.

6. Crijns HJGM, Smeets JLRM, Rodriquez LM, Meijer A, Wellens HJ.Cure of interfascicular reentrant ventricular tachycardia by abla-tion of the anterior fascicle of the left bundle branch. J CardiovascElectrophysiol 1995; 6:486–492.

7. Narasimhan C, Jazayeri M, Sra J, Dhala A, Deshpande S, Biehl M,Akhtar M, et al. Ventricular tachycardia in valvular heart disease:Facilitation of sustained bundle-branch reentry by valve surgery.Circulation 1997; 96:4307–4313.

8. Mehdirad AA, Keim S, Rist K, Tchou P. Long-term clinical outcomeof right bundle branch radiofrequency catheter ablation for treat-ment of bundle branch reentrant ventricular tachycardia. PacingClin Electrophysiol 1995; 18:2135–2143.

9. Blanck Z, Jazayeri M, Dhala A, Deshpande S, Sra J, Akhtar M. Bun-dle branch reentry: A mechanism of ventricular tachycardia in theabsence of myocardial or valvular dysfunction. J Am Coll Cardiol1993; 22:1718–1722.

10. Reithmann C, Remp T, Hoffmann E, Matis T, Wakili R, SteinbeckG. Different patterns of the fall of impedance as the result of heat-ing during ostial pulmonary vein ablation: Implications for powertitration. Pacing Clin Electrophysiol 2005; 28:1282–1291.

11. Swerdlow CD, Shivkumar K. Implantable cardioverter-defibrillator: Clinical aspects. In: Zipes DP, Jalife J (eds.): CardiacElectrophysiology. From Cell to Bedside, 4th Ed. Philadelphia:Saunders, 2004, pp. 980–994.

12. Chalvidan T, Cellarier G, Deharo JC, Colin R, Savon N, Barra N,Peyre JP, et al. His-Purkinje system reentry as a proarrhythmic ef-fect of flecainide. Pacing Clin Electrophysiol 2000; 14:1639–1647.

13. Issa ZF, Zipes DP. Antiarrhythmic effects of targeted cardiac neuro-modulation: Is it time for clinical application? Heart Rhythm 2005;2:1363–1364.

14. Nademanee K, Taylor R, Bailey W, Rieders DE, Kosar EM. Treatingelectrical storm: Sympathetic blockade versus advanced cardiaclife support – guided therapy. Circulation 2000; 102:742–747.

15. Wen MS, Yeh SJ, Wang CC, Lin FC, Wu D. Successful radiofre-quency ablation of idiopathic left ventricular tachycardia at a siteaway from the tachycardia exit. J Am Coll Cardiol 1997; 30:1024–1031.

16. Ouyang F, Cappato R, Ernst S, Goya M, Volkmer M, Hebe J, Antz M,et al. Electroanatomic substrate of idiopathic left ventricular tachy-cardia. Unidirectional block and macroreentry within the purkinjenetwork. Circulation 2002; 105:462–469.


reinitiation of ventricular macroreentry within the his-purkinje system by back-up ventricular...

Documents