Available online at www.sciencedirect.com

y 41 (2008) 160–164www.jecgonline.com

Journal of Electrocardiolog

Multiple macroreentrant ventricular tachycardias exhibitingcentrifugal endocardial activations from the scar

border zone after myocardial infarctionTakumi Yamada, MD,a,* Yoshimasa Murakami, MD,b Masahiro Muto, MD,b

Naoki Yoshida, MD,b G. Neal Kay, MDa

aDivision of Cardiovascular Diseases, University of Alabama at Birmingham, Birmingham, AL, USAbDivision of Cardiology, Aichi Prefectural Cardiovascular and Respiratory Center, Ichinomiya, Japan

Received 8 August 2007; accepted 29 October 2007

Abstract A 53-year-old man with a ventricular tachycardia (VT) electrical storm during the chronic phase of

⁎ CorrespondingRhythm ManagementBirmingham, AL 35294720.

E-mail address: ta

0022-0736/$ – see frodoi:10.1016/j.jelectroc

an extensive anteroseptal myocardial infarction underwent electrophysiologic testing and catheterablation. An electroanatomical map during 7 induced macroreentrant VTs demonstrated multiplecentrifugal endocardial activations from sites that were located at the circumferential border zone of alarge scar area. Interestingly, during the radiofrequency catheter ablation of 4 of the VTs, theelimination of the substrate of the previous VTs converted one VT to another probably because thoseVTs might have shared a central common pathway of the macroreentrant circuit with different exits.© 2008 Elsevier Inc. All rights reserved.

Keywords: Ventricular tachycardia; Centrifugal endocardial activation; Scar border zone; Myocardial infarction;

Radiofrequency catheter ablation

Introduction

Reentry is the major mechanism of ventricular tachycar-dia (VT) associated with myocardial infarction (MI) scars.1

The location and size of the reentrant circuits of the post-MIVTs are variable. Endocardial electroanatomical mappingmay be useful for identifying the entire reentrant circuits ofsome post-MI VTs that are recognized to have a head-meets-tail activation pattern.2 However, in all post-MI VTs, theentire reentrant circuit cannot always be identified becauseintramural or epicardial myocardial tissue may be involvedin the reentrant circuit. It has been reported that the scarborder zone often creates a critical isthmus of the reentrantcircuit of the post-MI VTs because small islets of survivingmyocardial tissue that interspersed with the scar tissue in thatzone exhibit slow conduction that creates an opportunity forelectrical reentry.3-6

author. Division of Cardiovascular Diseases, CardiacLaboratory, University of Alabama at Birmingham,4-0019, USA. Tel.: +1 205 975 4724; fax: +1 205 975

kumi-y@fb4.so-net.ne.jp

nt matter © 2008 Elsevier Inc. All rights reserved.ard.2007.10.003

Case report

A 53-year-old man with a history of an extensiveanteroseptal myocardial infarction 21 years before theonset of VT underwent an implantation of an implantablecardioverter defibrillator (ICD) 3 years previously. He wasadmitted for multiple ICD shock deliveries because of a VTelectrical storm. After written, informed consent wasobtained, an electrophysiologic study was performed. Leftventriculography demonstrated an aneurysmal change of theanteroseptal wall to apex and a severely reduced leftventricular systolic function (ejection fraction, 29%). Thebaseline heart rhythm was sinus rhythm. First, electroana-tomical mapping of the left ventricle was performed using a7F, 4-mm tip ablation catheter (Navi-Star, Biosense Webster,Diamond Bar, CA) introduced through a long sheath via theright femoral artery during right ventricular pacing with theICD lead to identify the substrate of the VTs on the voltagemap. The voltage map revealed a wide low voltage area(b1.5 mV) and localized scar area (b0.1 mV) in theanteroseptal wall to apex (Fig. 1). After that, a clinical sus-tained VT (VT1, right bundle branch block, and a superioraxis QRS morphology, cycle length = 320 milliseconds)was induced by burst pacing from the right ventricular

Fig. 1. A voltage map of the left ventricle and activation maps during multiple VTs. The purple indicates the area with a voltage of the local bipolar electrogrammore than 1.5 mV; red, with a voltage of less than 0.1 mV; and gray, with a voltage of less than 0.05 mV in the voltage map. Note that all the origins of thecentrifugal endocardial activations were located in the circumference of the scar border zone. AP indicates anteroposterior. (Available in color online at www.sciencedirect.com and www.jecgonline.com.)

161T. Yamada et al. / Journal of Electrocardiology 41 (2008) 160–164

apex (Figs. 2 and 3). Although the hemodynamics degen-erated during VT1 (systolic blood pressure, 70-80 mm Hg),quick pacing and electroanatomical mapping were success-fully performed. An activation mapping of VT1 wasperformed in a limited area of interest on the remap that

Fig. 2. Twelve-lead electrocardiograms of the VTs. The number

was made by extracting the anatomical frame of the baselinemap during right ventricular pacing. The activation maprevealed centrifugal endocardial activation from the site inthe septal wall near the apex where a presystolic ventricularpotential preceding the QRS onset by 50 milliseconds was

s in the parentheses indicate the cycle length of the VTs.

Fig. 3. Cardiac tracings recorded during entrainment pacing and the successful ablation site of VT1. The red arrow indicates the successful ablation site of VT1.At the successful ablation site, a presystolic potential was recorded, and the PPI was almost identical to the cycle length of VT1. ABL d(p) indicates the distal(proximal) electrode pair of the ablation catheter; MA, mitral annulus; RVA, right ventricular apex. The other abbreviations are as in Fig. 1. (Available in coloronline at www.sciencedirect.com and www.jecgonline.com).

162 T. Yamada et al. / Journal of Electrocardiology 41 (2008) 160–164

recorded (Fig. 3). At that site, the postpacing interval (PPI)was almost identical to the cycle length of VT1, though noentrainment with concealed fusion was obtained (Fig. 3). Afew radiofrequency (RF) applications with a target tempera-ture of 60°C and maximum power output of 50 W weredelivered at that site and resulted in the conversion to anotherVT (VT2, left bundle branch block, and a superior axis QRSmorphology, cycle length = 330 milliseconds) (Figs. 2and 4). The same procedure as for VT1 was repeated duringVT2. The origin of the centrifugal endocardial activationshifted to the midseptal wall, and the same electrophysio-logic findings as for VT1 were obtained. A few RFapplications with the same setting as in VT1 were deliveredat that site and resulted in the conversion to another VT(VT3, left bundle branch block, and an inferior axis QRSmorphology, cycle length = 350 milliseconds) (Fig. 4). Afterthe conversion, the local ventricular activation time relativeto the QRS onset at the ablation site became significant laterthan that before the conversion (Fig. 4). The same proceduresas for the previous VTs were repeated during VT3. Theorigin of the centrifugal endocardial activation shifted to theanteroseptal wall on the basal side, and the same electro-physiologic findings as in the previous VTs were obtained. Afew RF applications with the same setting as in the previousVTs were delivered at that site and resulted in the conversionto a VTsimilar to VT1 (VT1′, right bundle branch block, anda superior axis QRS morphology, cycle length = 340milliseconds) (Fig. 5) and thereafter another VT (VT4,right bundle branch block, and a superior axis QRS

morphology, cycle length = 360 milliseconds) (Fig. 2).During the first beat of VT4 during the conversion fromVT1′ to VT4, the local ventricular activation with almost thesame coupling interval as the previous beats, precedingthe QRS onset as well as the local ventricular activation withthe activation time relative to the QRS onset as late as thefollowing beats, was recorded (Fig. 5). The same proceduresas in the previous VTs were repeated during VT4. The originof the centrifugal endocardial activation shifted to the lateralwall near the apex (Fig. 1), and the same electrophysiologicfindings as in the previous VTs were obtained. A few RFapplications with the same setting as in the previous VTswere delivered at that site with the interruption of the VT.Thereafter, 3 different VTs (VT5 to VT7) (Fig. 2) thatexhibited a centrifugal endocardial activation from adifferent origin were induced one after another by burstpacing from the right ventricular apex after the eliminationof the other VT (Fig. 1). At all of the origins, the PPI wasidentical to the VT cycle length that tended to prolong afterthe elimination of the other VT. In all of the 7 VTs, theorigins of the centrifugal endocardial activations werelocated in the circumferential scar border zone of the largescar area. During all of the 7 VTs, middiastolic potentialscould be recorded in the scar area apart from the origin ofthe centrifugal endocardial activations; however, they werenot annotated in the activation maps. After the eliminationof VT7 by a few RF applications at that site, no VT couldbe induced despite programed electrical stimulation as wellas an isoproterenol infusion. The total procedure and

Fig. 4. Conversion from VT2 to VT3 during the RF ablation and activation maps during VT2 and VT3. CL indicates cycle length. The other abbreviations are asin Fig. 3. (Available in color online at www.sciencedirect.com and www.jecgonline.com).

163T. Yamada et al. / Journal of Electrocardiology 41 (2008) 160–164

fluoroscopy times were 273 and 67 minutes, respectively,and the total number of radiofrequency applications was36. No complications occurred. During more than 3 yearsof follow-up, the patient has been free of any ICD shockdeliveries with the same drugs (mexiletine and carvedilol)as before the ablation.

Fig. 5. Cardiac tracings showing the conversion from VT1′ to VT4 during the RF ablation. The other abbreviations are as in Fig. 4.

Discussion

De Bakker et al3 reported that in 13% of the patients withpost-MI VTs undergoing antiarrhythmic surgery, intraopera-tive endocardial mapping during the VTs demonstrated acentrifugal endocardial activation from the scar border zone.

164 T. Yamada et al. / Journal of Electrocardiology 41 (2008) 160–164

However, to the best of our knowledge, this is the first reportdemonstrating multiple post-MI VTs with centrifugalendocardial activations from the scar border zone usingelectroanatomical mapping. The mechanism of all of the VTsexhibiting centrifugal endocardial activations in this casewas considered to be reentry from the results of theentrainment pacing. However, which reentry was themechanism of those VTs, micro or macro? De Bakker etal3 reported that the VTs exhibiting centrifugal endocardialactivations in their study, analyzing the activation sequenceof the middiastolic potentials during the VTs and entrain-ment pacing, favored a macroreentry as the mechanism. Inthis case, microreentry was considered to be unlikelybecause entrainment with concealed fusion could not beobtained at the successful ablation site. Furthermore,interestingly the elimination of the substrate of the previousVTs by RF ablation converted VT1 to VT2, VT2 to VT3, andVT3 to VT4. If the mechanism of those VTs had beenmicroreentry, the previous VTs should have entrained thefollowing VTs. However, it would have been unlikely. Wespeculated that those VTs might have shared a centralcommon pathway of the macroreentrant circuit and haddifferent exits. Actually, in this case, not much time could bespent in mapping the middiastolic potentials, resulting in theactivation map limited to the border zone of interest for eachVT because the VTs were symptomatic. Therefore, we couldnot reveal the entire circuit during the various VTs byelectroanatomical mapping to delineate the common centralpathway. Epicardial mapping might have provided somehelpful information for revealing the VT mechanismalthough it was not attempted in this case. However, webelieve that our speculation was the most likely one becauseit could explain the results of the entrainment pacing at thesuccessful ablation site and the prolongation of the VT cyclelength after any elimination of the VTs. Furthermore, aconversion from VT1′ to VT4 was suddenly caused by an

exit block in VT1′ without any preceding slowing of VT1′.That evidence might favor our speculation.

Marchlinski et al5 reported that RF linear endocardiallesions across the scar border zone revealed by substratevoltage mapping were effective in controlling unmappablepost-MI VTs. Verma et al6 reported that approximately 70%of the hemodynamically stable, monomorphic post-MI VTswere successfully ablated at the sites in the endocardial scarborder zone as defined by substrate voltage mapping. Thefindings in this case were consistent with the previousreports. However, this case might have demonstratedmultiple exits of the macroreentrant circuits in the circum-ferential border zone of the large scar area. Therefore, thesefindings suggest that circumferential RF ablation targetingthe scar border zone might be an effective technique foreliminating post-MI VTs such as ours.

References

1. Soejima K, Stevenson WG, Maisel WH, Sapp JL, Epstein LM.Electrically unexcitable scar mapping based on pacing threshold foridentification of the reentry circuit isthmus: feasibility for guidingventricular tachycardia ablation. Circulation 2002;106:1678.

2. de Chillou C, Lacroix D, Klug D, et al. Isthmus characteristics ofreentrant ventricular tachycardia after myocardial infarction. Circulation2002;105:726.

3. de Bakker JM, van Capelle FJ, Janse MJ, et al. Macroreentry in theinfarcted human heart: the mechanism of ventricular tachycardias with a“focal” activation pattern. J Am Coll Cardiol 1991;18:1005.

4. de Bakker JM, van Capelle FJ, Janse MJ, et al. Slow conduction in theinfarcted human heart. “Zigzag” course of activation. Circulation 1993;88:915.

5. Marchlinski FE, Callans DJ, Gottlieb CD, Zado E. Linear ablation lesionsfor control of unmappable ventricular tachycardia in patients withischemic and nonischemic cardiomyopathy. Circulation 2000;101:1288.

6. Verma A, Marrouche NF, Schweikert RA, et al. Relationship betweensuccessful ablation sites and the scar border zone defined by substratemapping for ventricular tachycardia post-myocardial infarction.J Cardiovasc Electrophysiol 2005;16:465.