Remote Magnetic Navigation to Guide Endocardial and Epicardial Catheter Mapping of Scar-Related Ventricular Tachycardia

Remote Magnetic Navigation to Guide Endocardial andEpicardial Catheter Mapping of Scar-Related

Ventricular Tachycardia

Arash Aryana, MD; Andre d’Avila, MD; E. Kevin Heist, MD, PhD; Theofanie Mela, MD;Jagmeet P. Singh, MD, PhD; Jeremy N. Ruskin, MD; Vivek Y. Reddy, MD

Background—The present study examines the safety and feasibility of using a remote magnetic navigation system toperform endocardial and epicardial substrate-based mapping and radiofrequency ablation in patients with scar-relatedventricular tachycardia (VT).

Methods and Results—Using the magnetic navigation system, we performed 27 procedures on 24 consecutive patientswith a history of VT related to myocardial infarction, dilated cardiomyopathy, arrhythmogenic right ventricularcardiomyopathy, hypertrophic cardiomyopathy, or sarcoidosis. Electroanatomic mapping of the left ventricular, rightventricular, and ventricular epicardial surfaces was constructed in 24, 10, and 12 patients, respectively. Complete-chamber VT activation maps were created in 4 patients. A total of 77 VTs were inducible, of which 21 were targetedduring VT with the remotely navigated radiofrequency ablation catheter alone. With a combination of entrainment andactivation mapping, 17 of 21 VTs (81%) were successfully terminated in a mean of 8.4�8.2 seconds; for the remainder,irrigated radiofrequency ablation was necessary. The mean fluoroscopy times for endocardial and epicardial mappingwere 27�23 seconds (range, 0 to 105 seconds) and 18�18 seconds (range, 0 to 49 seconds), respectively. In concertwith a manually navigated irrigated ablation catheter, 75 of 77 VTs (97%) were ultimately ablated. Four patientsunderwent a second procedure for recurrent VT, 3 with the magnetic navigation system. After 1.2 procedures per patient,VT did not recur during a mean follow-up of 7�3 months (range, 2 to 12 months).

Conclusions—The present study demonstrates the safety and feasibility of remote catheter navigation to perform substratemapping of scar-related VT in a wide range of disease states with a minimal amount of fluoroscopy exposure.(Circulation. 2007;115:1191-1200.)

Key Words: ablation � catheter ablation � electrophysiology � magnetic resonance imaging � mapping� tachycardia � tomography

Significant advances have been made in catheter ablationof scar-related ventricular tachycardia (VT). These ad-

vances are due in part to an improved understanding of thepathophysiology governing these tachyarrhythmias and totechnological advances such as electroanatomic mapping(EAM) systems. This advancement has led to a paradigmshift in the strategy by which VT is mapped and ablated:substrate-based catheter ablation.1–4 Instead of mapping dur-ing VT, this approach involves identifying and ablating thearrhythmogenic myocardium predominantly during sinusrhythm. Although effective, this approach is limited by theneed for a degree of ventricular mapping accuracy and detailthat requires advanced operator skill with cathetermanipulation.

Clinical Perspective p 1200

When used in concert with a compatible EAM system, remotenavigation technology may facilitate cardiac mapping and abla-tion independently of operator dexterity. The magnetic naviga-tion system (MNS) uses highly flexible catheters equipped withsmall magnets embedded in the tip for catheter orientation withan external magnetic field. To date, this platform system hasbeen used in mapping and ablation of accessory pathways inpatients with AV nodal or AV reentrant tachycardia and in thetreatment of atrial fibrillation.5–7 The present study examines thehypothesis that substrate-based endocardial and epicardial re-mote magnetic mapping and ablation can be safely and effec-tively performed in patients with scar-related VT.

Received October 26, 2006; accepted December 22, 2006.From the Cardiac Arrhythmia Service, Massachusetts General Hospital, and Harvard Medical School, Boston, Mass.The online-only Data Supplement, consisting of movies, is available with this article at http://circ.ahajournals.org/cgi/content/full/

CIRCULATIONAHA.106.672162/DC1.Correspondence to Vivek Y. Reddy, MD, Cardiac Arrhythmia Service, Massachusetts General Hospital, 55 Fruit St, GRB–109, Boston, MA 02114.

E-mail [email protected]© 2007 American Heart Association, Inc.

Circulation is available at http://www.circulationaha.org DOI: 10.1161/CIRCULATIONAHA.106.672162

1191 by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

MethodsThe present study was approved by the Massachusetts GeneralHospital Institutional Review Board Committee and performedaccording to institutional guidelines.

Patient PopulationBetween November 2005 and October 2006, a total of 27 procedureswere performed on 24 consecutive patients with a history ofscar-related VT. The cause of VT substrate was diverse: post–myocardial infarction (MI), dilated cardiomyopathy, arrhythmogenicright ventricular (RV) cardiomyopathy, hypertrophic cardiomyopa-thy, and sarcoidosis. Twenty of 27 procedures (74%) were performedunder general anesthesia; the remainder were done under onlymoderate sedation. In 18 procedures (67%), intra-aortic balloonpump counterpulsation was used for either prophylaxis againstworsening heart failure or hemodynamic stabilization during arrhyth-mia induction and mapping. All intra-aortic balloon pumps wereremoved immediately at the completion of the procedure.

Remotely Guided MNSThe remotely controlled catheter navigation system used in thepresent study consists of 2 independent but communicating compo-nents: the Niobe Stereotaxis MNS (Stereotaxis, Inc, St Louis, Mo)and a prototype EAM system (CARTO-RMT, Biosense Webster,Inc, Diamond Bar, Calif). The MNS uses 2 large magnets positionedon either side of the procedure table to generate a compositemagnetic field for directional catheter orientation, as describedpreviously.5–7 The CARTO-RMT EAM platform is a prototypemagnetic localization system similar to the standard CARTO sys-tem8; the major important difference is its ability to localize theablation catheter without interference from the MNS magnetic field.This EAM system can localize both standard CARTO and thespecialized CARTO-RMT catheters. Accordingly, EAM can beperformed either in the conventional fashion with manual cathetermanipulation or remotely with the MNS.

The EAM system and the MNS communicate in a unidirectionalfashion. Three-dimensional locations can be prescribed on the EAMsfor transfer to the MNS. The MNS then calculates the vector requiredof the magnetic field to orient the catheter in this direction. If thecatheter fails to move to the desired location because of obstacles inits path (eg, trabeculae, chordae tendineae, papillary muscles), theoperator can simply withdraw the catheter remotely to free it of theobstacle(s) and then readvance to the desired location. Alternatively,the magnetic field can be remotely manipulated to incrementallymanipulate the catheter. The communication between the CARTO-RMT EAM and the MNS provides a synchronized view of the heartso that free hand manipulations of this magnetic vector can beperformed to iteratively move the catheter along the cardiac surface(Figure 1).

Cardiac Chamber Mapping andRadiofrequency AblationA transseptal puncture was performed under the guidance of fluo-roscopy and, if available, intracardiac echocardiography, and an 8.5FMullins sheath was placed near the mitral valve plane. Intravenousheparin was administered just before transseptal puncture. Leftventricular (LV) substrate mapping was performed with a combina-tion of transseptal and retrograde aortic approaches in all patients;endocardial RV mapping was performed with femoral venousaccess. In selected patients, epicardial mapping was performed withthe percutaneous subxiphoid needle puncture technique.9–11

Two types of catheters were used for mapping and ablation: aremotely guided, 4-mm-tip, quadripolar (RMT) catheter with 3embedded magnets to align with the MNS-generated magnetic field(Navistar-RMT, Biosense-Webster, Inc) and a manually directed,externally irrigated, 3.5-mm-tip catheter (Thermocool, Biosense-Webster, Inc). For the last 15 procedures, the RMT catheter had athermocouple embedded in its tip for temperature monitoring.

Programmed stimulation included up to 3 extrastimuli and rapidpacing from 2 ventricular sites (right or left, depending on thelocation of the scar) to document cycle lengths and 12-lead ECGmorphologies of all inducible VTs. In all patients, baseline ventric-ular substrate-based mapping was performed remotely with the MNSand the RMT mapping catheter. Mapping consisted of constructing3-dimensional electroanatomic voltage maps of the chamber(s) ofinterest (LV, RV, and/or ventricular epicardial surface) during sinusrhythm or RV pacing, displaying peak-to-peak bipolar electrogramamplitude with a fill threshold of at least 15 and 20 mm forendocardial and epicardial maps, respectively. As previously de-scribed, a bipolar electrogram voltage amplitude �1.5 mV wasdefined as normal myocardium in post-MI patients.11–14 For otherpathologies, the amplitude scales were adjusted to best identify thediseased myocardial tissue.11–14 Fluoroscopy exposure times duringventricular chamber mapping were recorded in most patients. Thechamber mapping time was defined as the time elapsed starting justafter the RMT catheter was placed into the relevant chamber andending after the chamber map was completed just before radiofre-quency ablation was begun. After detailed mapping to fully definethe scar borders, ventricular activation and entrainment mappingduring sustained VT was performed. In a few selected patients withsustained hemodynamically stable VT, full-chamber activation mapswere generated; most underwent partial activation mapping only.

In all patients, a combination of entrainment, late potential, andpace mapping was used during substrate mapping. Briefly, if ahemodynamically stable VT was inducible (stable for even a fewseconds), standard entrainment maneuvers were used to identify andablate the critical pathway of the circuit. Typically, the RMT catheterwas used to deliver these radiofrequency applications remotely alongthe endocardial and/or epicardial surfaces. Power titration was basedon impedance monitoring or, when available, temperature monitor-ing to achieve 55°C to 65°C and 65°C to 85°C for endocardial and

Figure 1. View-synchronization of theMNS and EAM systems. Shown are theMNS navigation screen with integratedfluoroscopic images (A) and the EAM ven-tricular surface on the CARTO-RMT screen(B). The top-right panel on the MNSscreen (A) is view-synchronized to theright panel on the EAM screen (B). Tomove the catheter in the direction shownon the EAM (yellow arrow), the operatorsimply moves the icon on the MNS screenin a similar fashion (orange arrow). If thecatheter does not reach the desired loca-tion because of an obstacle such as apapillary muscle or chordae tendineae, thecatheter is simply remotely withdrawn tofree if of the obstacle(s) and re-advancedto the desired location.

1192 Circulation March 13, 2007

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

epicardial lesions, respectively. If the arrhythmia(s) failed to termi-nate or remained inducible after ablation with the RMT catheter, amanual irrigated catheter was used to eliminate the VT. In addition,the manual catheter was used in most patients to ablate additionalremaining putative target sites such as late potentials, pace map siteswith a good QRS match to an inducible VT and a long stimulus toQRS time, and VT exit sites identified by pace mapping along scarborders. With the manual catheter, power was titrated to achieve animpedance fall of �10%. The postablation stimulation protocolincluded at least as aggressive a stimulation protocol as that used toinitially induce the VTs at the beginning of the procedure.

Follow-UpVT recurrence was identified by history and clinical symptoms andthrough device interrogation. All patients received anticoagulationwith warfarin or aspirin after the procedure. Antiarrhythmic drugtherapies that patients had been prescribed long term were eithercontinued at the same or reduced dosage or, in selected cases,discontinued. Antiarrhythmic agents that had been initiated recentlyto control multiple/incessant VT were discontinued. Patients wereseen in the implantable cardioverter-defibrillator (ICD) clinic fordevice interrogation at 1 to 2 months and every 3 months thereafter.All data are presented as mean�SD.

The authors had full access to and take full responsibility for theintegrity of the data. All authors have read and agree to themanuscript as written.

ResultsPatient CharacteristicsA total of 27 procedures were performed on 24 consecutivepatients with a history of scar-related VT. Patient characteristics

are shown in Table 1. The mean age of the study cohort was61�15 years (range, 21 to 83 years). Twenty-one patients (87%)were men. The mean LV ejection fraction as assessed byechocardiography was 37�18% (range, 15% to 83%), with amean LV dimension of 57�11 mm (range, 40 to 77 mm) at enddiastole and 46�13 mm (range, 22 to 69 mm) at end systole.Eleven patients (46%) had advanced heart failure as defined bya New York Heart Association functional class III or IV. Theorigin of VT substrate was related to MI in 15 patients (62%),dilated cardiomyopathy in 3 (13%), arrhythmogenic RV cardio-myopathy in 3 (13%), hypertrophic cardiomyopathy in 2 (8%),and sarcoidosis in 1 (4%). Nineteen patients (79%) had an ICD.In addition, 19 patients (79%) were receiving antiarrhythmicdrug therapy. All patients had symptomatic or recurrent mono-morphic VT; ICD therapy or cardioversion was required in 19cases (79%). Fifteen of 24 patients (62%) presented with eitherincessant VT or ICD storm, defined as �3 appropriate ICDtherapies in a 24-hour period. Four patients (17%) had a historyof previously failed VT ablation using a conventional approach(ie, not using remote navigation). All patients had evidence of atleast a single VT morphology on a 12-lead ECG or stored ICDelectrograms.

Mapping ApproachIn 24 of 27 procedures, LV endocardial mapping was per-formed via both transseptal and retrograde aortic approaches.

TABLE 1. Clinical Characteristics of Patients Undergoing Remote Mapping and Ablation

Patient Age, y LVEF, % NYHA HF Class VT Substrate ICD AAD History

1 68 27 IV MI Yes A, BB Syncope, slow incessant MMVT, DCCV

2 68 34 III DCM Yes BB Near syncope, sustained MMVT, ICD storm

3 77 15 III MI Yes A, BB Near syncope, slow incessant MMVT, ICD storm, prior failed RFA

4 49 45 IB MI No BB Symptomatic sustained MMVT

5 75 42 II MI Yes BB MMVT degenerated into PMVT, ICD storm

6 49 45 II MI No BB Symptomatic MMVT

7 67 18 III–IV MI Yes BB, M Sustained MMVT, ICD storm, prior failed RFA

8 54 53 IB ARVC Yes A, F, S Recurrent MMVT, ICD storm, prior failed RFA

9 76 33 III MI Yes BB, M Recurrent MMVT, ICD therapy

10 47 30 IB Sarcoidosis Yes M, S Symptomatic sustained MMVT, ICD storm

11 62 30 III MI Yes A, BB Near syncope, sustained MMVT, ICD therapy

12 83 27 II MI Yes A, BB Syncope, recurrent MMVT, ICD storm

13 50 83 IB HCM Yes S Recurrent MMVT, ICD storm

14 31 52 IB ARVC Yes BB, M Symptomatic sustained MMVT, ICD storm

15 65 20 III–IV DCM Yes A, BB, M Symptomatic sustained MMVT, ICD storm

16 55 23 III–IV MI Yes A, BB, M MMVT, ICD therapy

17 69 32 II MI Yes BB Recurrent MMVT, ICD storm

18 64 55 II MI No A, BB Symptomatic MMVT

19 72 20 III–IV MI Yes BB, S Recurrent MMVT, ICD storm

20 53 42 III DCM No A, BB, M Symptomatic recurrent MMVT

21 21 70 IB HCM Yes BB, M Recurrent PMVT, ICD therapy

22 64 61 IB ARVC No M, S Symptomatic MMVT (below ICD detection zone), DCCV, prior failed RFA

23 62 20 IB MI Yes BB, M Incessant MMVT, ICD storm

24 81 15 III–IV MI Yes A, BB Symptomatic slow incessant MMVT (below ICD detection zone)

LVEF indicates LV ejection fraction; NYHA HF, New York Heart Association heart failure; AAD, antiarrhythmic drug; A, amiodarone; BB, �-blocker; F, flecainide; M,mexelitine; S, sotalol; MMVT, monomorphic VT; DCCV, external direct current cardioversion; DCM, dilated cardiomyopathy; RFA, radiofrequency ablation; PMVT,polymorphic VT; ARVC, arrhythmogenic RV cardiomyopathy; and HCM, hypertrophic cardiomyopathy.

Aryana et al Remote Ventricular Substrate Mapping 1193

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

LV mapping was not performed in 2 arrhythmogenic RVcardiomyopathy patients (patients 8 and 14) and during theinitial procedure on the patient with sarcoidosis (patient10–1). RV endocardial mapping was performed using a longsheath advanced to the approximate plane of the tricuspidvalve for stabilization. Pericardial access was successful in 13of 16 patients (81%), including 2 patients with prior cardiacsurgery. Pericardial access could not be achieved in theremaining patients because of previous cardiac surgery (1patient) or the presence of pericardial adhesions resultingfrom prior MI (1 patient) or sarcoidosis (1 patient).

Remote Substrate-Based Ventricular MappingMaps of LV and RV endocardium and ventricular epicardiumwere constructed in 24, 10, and 12 procedures, respectively.The mean LV and RV chamber and pericardial space vol-umes were 220�104, 264�111, and 651�145 cm3, respec-tively. The mean total points collected during mapping of LVand RV endocardium and ventricular epicardium were

142�75, 76�73, and 123�37, respectively. Of these, themean points collected with the RMT catheter were 111�67(78%), 74�63 (97%), and 122�34 (99%), respectively.Remote mapping of these chambers required a mean durationof 84�44 minutes (48�18 seconds per point), 66�48 min-utes (42�18 seconds per point), and 75�34 minutes (36�12seconds per point), respectively. As shown in Table 2, thefluoroscopy times required to perform endocardial and epi-cardial remote mapping were 27�23 seconds (range, 0 to 105seconds) and 18�18 seconds (range, 0 to 49 seconds),respectively; the total fluoroscopy time to complete remotemapping was 34�32 seconds (range, 1 to 136 seconds). Thetotal procedural fluoroscopy time was 21�7 minutes (range,9 to 33 minutes). Bipolar ventricular voltage amplitude mapswere generated in all patients. Abnormal ventricular scartissue, defined by low-voltage electrogram amplitude, wasseen in all but 3 patients. Examples of ventricular EAMsconstructed in patients with post-MI, arrhythmogenic RVcardiomyopathy, and hypertrophic cardiomyopathy–relatedVTs are shown in Figures 2 and 3.

TABLE 2. Results of Arrhythmia Induction and Fluoroscopy Usage Time During Mapping and Ablation

Induced VTs Fluoroscopy Time Using RMT

Patient SVT, n CLs of Induced VTs, ms Total Fluoroscopy Time, min Endocardial, s Epicardial, s Total, s

1 1 585 22 N/A � � � N/A

2 0 · · · 13 N/A N/A N/A

3–1* 2 524, 405 24 N/A � � � N/A

3–2† 1 520 16 3 � � � 3

4 2 280, 220 16 N/A � � � N/A

5 1 236 30 N/A N/A N/A

6 0 295 (NSVT)‡ 12 13 � � � 13

7 5 417, 420, 435, 473, 562 25 12 � � � 12

8 2 550, 232 15 105 31 136

9 1 330 24 41 � � � 41

10–1* 3 330, 326, 421 26 38 � � � 38

10–2† 1 540 16 20 � � � 20

11 5 482, 345, 406, 438, 470 26 44 � � � 44

12 7 470, 520, 412, 386, 540, 480, 440 27 18 � � � 18

13 0 · · · 14 52 33 85

14 3 380, 320, 262 15 40 0 40

15 8 304, 349, 358, 384, 389, 421, 426, 433 23 38 � � � 38

16 6 731, 482, 459, 445, 473, 132 31 19 � � � 19

17 6 360, 335, 520, 339, 328, 344 25 32 � � � 32

18–1* 3 384, 398, 417 33 17 � � � 17

18–2† 3 365, 283, 208 29 32 49 81

19 1 380 27 29 � � � 29

20 4 471, 316, 297, 365 13 15 18 33

21 0 · · · 9 3 11 14

22 4 351, 379, 330, 201 19 N/A N/A N/A

23 5 426, 304, 271, 337, 276 22 5 0 5

24 3 562, 712, 325 21 0 1 1

SVT indicates sustained VT; CL, cycle length; NSVT, nonsustained VT; and N/A, data not available.*First mapping and ablation procedure for this patient.†Second mapping and ablation procedure for this patient.‡Induction of nonsustained VT only.

1194 Circulation March 13, 2007

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

VT Induction and Catheter AblationA total of 77 VTs were inducible during 23 of 27 procedures(85%), whereas in 4 patients (patients 2, 6, 13, and 21),sustained monomorphic VT could not be induced (Table 2).In 1 of these 5 patients (patient 6), nonsustained monomor-phic VT was induced repeatedly and thus targeted forablation. Although no VTs were inducible in the other 3, in 1patient (patient 2), typical atrial flutter was induced repeat-edly and eliminated by cavotricuspid isthmus ablation. Dur-ing sustained hemodynamically stable VT, full-chamber ac-tivation mapping was performed in 4 patients during 5procedures, and partial activation mapping was performed inthe remaining. Entrainment mapping was performed in 20procedures. Pace mapping and targeting of late and fraction-ated potentials were used in 20 and 23 procedures, respec-tively. Figures 4 and 5 illustrate examples of substratemapping and ablation in patients with hemodynamicallystable and unstable VTs.

Of the 77 inducible VTs, 21 were targeted for ablationduring VT with the remote catheter. Of these, a total of 17VTs (81%) were successfully terminated during 15 proce-dures at a mean duration of 8.4�8.2 seconds (Table 3). In 2procedures, the entire ablation was successfully completedwith the remote catheter alone, whereas in 22 procedures, the

manual, irrigated catheter also was used to enhance proce-dural safety and efficacy. In concert with the latter catheter,75 of 77 VTs (97%) were eliminated altogether. The meantotal duration of radiofrequency delivery per procedure was26�11 minutes. The mean total radiofrequency lesions de-livered with the remote and manual catheters were 8�8 and24�12 (total, 31�12), respectively. During 4 procedures(patients 3–1, 4, 9, and 17), arrhythmia termination did notoccur with remote ablation but was achieved with the manualcatheter. In 3 additional cases (patients 5, 10–1, 12, and 15),termination of VT was performed solely with the manualcatheter. In 2 cases, a “pop” occurred while radiofrequencyenergy was delivered with the standard remote catheter, butneither was associated with thromboembolism.

Ventricular arrhythmias were inducible in 5 patients at theend of the procedure (Table 3). In 3 cases (patients 3–2, 4,and 22), this induced rhythm was ventricular flutter. Anotherpatient (patient 15) was found to have 8 inducible VTs, withsuccessful elimination of all but 1. In the final patient (patient23), programmed stimulation resulted in only nonsustainedpolymorphic VT.

ComplicationsThe 30-day mortality from the procedure was zero. Onepatient (patient 7) with prior cardiac surgery who underwent

Figure 2. Full chamber remote activation mapping during VT. Electroanatomical activation (A) and voltage (B) maps were constructedduring VT (cycle length of 524 ms), in a post-MI patient (#3) with an apical LV scar. The large area of normal voltage in panel B repre-sents the abutting septal portion of the RV. The entire cycle length of the tachycardia was mapped. The insets in panel A represent thesequence of intracardiac ECGs as they traverse through electrical diastole (a facile visual representation of the VT is shown in a propa-gation map, Movie I, online-only Data Supplement ). The arrows point to entrainment sites (dark green). At each of these site (darkgreen points), except the last site (far left), entrainment with concealed fusion was noted with post-pacing intervals equal to thetachycardia cycle length; entrainment at the last site (far left) revealed again the tachycardia cycle length equaling the post-pacinginterval, but manifest fusion was noted. Ablation at the constrained portion of the channel (red point) with the remote catheter did notterminate the rhythm, but manual irrigated ablation did terminate the rhythm, albeit a late termination (15 sec). While VT was not induc-ible at the end of the procedure, the patient presented with another VT that was also remotely mapped as traversing this channel, butin the opposite direction. During the second procedure, this channel was completely ablated with irrigated radiofrequency energy toeliminate VT; no VT recurred after this second procedure. Also shown are electroanatomical activation (C & D) and voltage (E) mapsduring VT (cycle length of 550 ms) in a patient with arrhythmogenic right ventricular cardiomyopathy (#8). Half of the tachycardia cyclelength was mapped in this patient. A diastolic potential is seen in the inset; at this location, entrainment with concealed fusion and apost-pacing interval equal to 550 ms was observed. Endocardial ablation terminated the VT, but it was re-inducible; epicardial ablationat the opposite site completely eliminated the VT.

Aryana et al Remote Ventricular Substrate Mapping 1195

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

pericardial mapping with the manual catheter developed aloculated pericardial effusion in the posterior aspect of theleft atrium with partial compression requiring surgical peri-cardial decompression. One patient (patient 22) who under-went RV epicardial mapping only (no LV endocardial map-ping) developed transient right ulnar nerve palsy after theprocedure. It was unclear whether the palsy represented anembolic stroke or was the result of prolonged immobilizationunder general anesthesia. Another patient (patient 23) devel-oped uncomplicated bilateral lower-extremity deep venousthrombosis that was successfully treated with anticoagulation.Finally, 1 patient (patient 1) died more than a month after theprocedure as a result of advanced, medication-refractory heartfailure.

Follow-UpDuring follow-up, repeat VT ablation was required in 4patients. One patient (patient 17) with prior cardiac surgerycontinued to have recurrent VT associated with ICD dis-charges and underwent successful epicardial ablation with amanual approach after minimally invasive surgical subxi-phoid access to the pericardial space.15 The other 3 patients(patients 3, 10, and 18) presented with slow VT without ICDdischarges and underwent successful repeat VT ablation withthe MNS. During follow-up, inappropriate ICD dischargesoccurred in 1 patient (patient 9) as a result of atrial fibrilla-tion. In toto, there were no VT events after a mean follow-upof 7�3 months (range, 2 to 12 months). On the other hand, ifassessed after a single procedure only, procedural successwas achieved in 20 of 24 patients (83%).

DiscussionA number of advances have been made in recent years incatheter ablation of complex arrhythmias. Nevertheless, ma-nipulation of ablation catheters during substrate-based map-ping of scar-related VT requires adequate experience andmanual dexterity and can be limited by the technical skillrequired for detailed mapping of ventricular myocardium. Byobviating this skill requirement, remotely controlled magnet-ic/robotic navigation systems may enhance catheter-directedarrhythmia mapping and ablation. Magnetic navigation ofcardiac catheters was first reported over a decade ago in aneonate with complex congenital heart disease in whom itwas shown to enhance catheter guidance and manipulation.16

Since then, similar systems have been used safely in a varietyof invasive cardiovascular procedures, including ablation ofsupraventricular cardiac arrhythmias.5–7

The present study demonstrates the safety and efficacy ofendocardial and epicardial substrate mapping of VT with theMNS in the setting of a variety of cardiac pathologies. Inaddition to mapping, a subset of these patients also underwentablation in the LV and RV chambers and the pericardialspace. In toto, this remote navigation system proved capableof each of the 3 major components of substrate-basedmapping and VT ablation: (1) delineating and identifying thediseased myocardium, (2) performing the necessary electro-physiological maneuvers required to identify the arrhythmo-genic zones within the scar critical for VT maintenance, and(3) in a subset of induced VTs, delivering radiofrequencyenergy to terminate VT.

Figure 3. Correlation of electroanatomical mapping with 3-dimensional imaging. Anterior-posterior (A) and posterior-anterior (B) viewsfrom an electroanatomical voltage map, and a delayed-enhancement cardiac magnetic resonance imaging image of the LV inferior scar(C) are shown from a patient with magnetic resonance imaging–related VT (#6). A late potential is shown in the inset. The hyperen-hanced infracted tissue is visualized as white along the LV inferior wall, which is the same location as the infarct seen on the electro-anatomical maps.18 Panel (D) shows an epicardial and panel (E) a mesh view of an epicardial map with an embedded endocardial map.On the other hand, in a patient with hypertrophic cardiomyopathy (#13), there was no significant amount of scarred tissue visualizedduring either epicardial (D) or endocardial (E) mapping; the lower electrogram amplitude region overlying the base and acute margins ofthe RV simply represent epicardial fat (there was a significant lack of late/double/fractionated potentials in this region).19 Consistentwith this image, delayed-enhancement computer tomography imaging (F) revealed scattered intramural hyperenhanced regions consis-tent with no significant scar20 (magnetic resonance imaging could not be performed because of the patient’s ICD).

1196 Circulation March 13, 2007

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

Delineation and Identification ofDiseased MyocardiumAccurate EAMs of the LV, RV, and ventricular epicardiumcould be constructed remotely in patients with a wide varietyof disease states. The MNS-compatible RMT catheter offersseveral potential advantages over the conventional manualcatheter during chamber mapping. Because its orientation isguided entirely by magnetic field and no deflection wires arerequired, the RMT catheter is softer than traditional deflect-able catheters along its distal segment. This feature couldresult in several clinically significant benefits. First, it ispossible that less endocardial trauma (a common occurrenceduring standard mapping, albeit of unclear clinical signifi-cance) would result from the use of an RMT catheter. Inparticular, the risk of remote cardiac perforation should below. Second, the softer touch of the RMT catheter is likely tocause less deformation of cardiac chambers than manualmapping, potentially resulting in a more accurate rendering ofcardiac chambers. Although the software for the CARTO-RMT system used in the present study did not supportintegration with 3-dimensional computer tomography/mag-netic resonance imaging, it is possible that the registrationprocess to perform image-guided therapy may be facilitatedby a more precise rendering of the chamber volume. How-ever, the absence of a comparative manual mapping groupprevents us from making definitive conclusions on any thesepoints.

Third, there was a minimal amount of fluoroscopy useduring remote MNS mapping, �1 minute in most cases,regardless of the mapping approach, endocardial or epicar-dial. This was related in a large part to use of the RMTcatheter because it can be manipulated inside cardiac cham-bers with minimal concern for trauma. In addition, becausethe catheter tip can always be visualized in a real-time fashionby EAM, confirmation of position by fluoroscopy is rarelynecessary. It is important to note that manual ablation wasperformed in most cases. Therefore, it is likely that additionalreductions in total fluoroscopy will be realized once MNS-compatible, irrigated radiofrequency ablation catheters be-come available for clinical use. This could result in a markedreduction in radiation exposure for both patients and opera-tors during these complex procedures.

Identification of Arrhythmogenic Zones Requiredfor VTTo identify the arrhythmogenic zones within a scar, 4mapping strategies commonly are used in clinical practice:activation mapping, entrainment mapping, late potential map-ping, and pace mapping. Partial activation mapping wasperformed in all patients with sustained VT. Although notalways required, full activation maps were generated during 5procedures. Although the lack of comparative manual map-ping precludes a definitive conclusion, our qualitative assess-

Figure 4. Remote mapping and ablation of hemody-namically stable VT. Shown are the clinical slow VTat 585 ms (A), inferior views of the electroanatomicalactivation (B) and voltage (C) maps during VT, and acardiac computed tomography scan showing a cal-cified LV inferobasal scar (D) from a patient withpost-MI VT (#1). E, At the start of an attempt atentrainment from an inferior wall site deep within thescar (denoted by the black arrow in panel B), thefirst paced beat terminated the VT without manifestglobal ventricular capture. F, Just apical to this site(denoted by the red arrow in panel B), stable diastol-ic potentials are seen during VT; entrainment withconcealed fusion and a post-pacing interval equal to585 ms were observed at this location. G, Duringremote radiofrequency ablation at this site, the VTwas eliminated in � 4 s of commencing energydelivery.

Aryana et al Remote Ventricular Substrate Mapping 1197

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

ment was that catheter-induced premature ventricular beatswere less common than typically observed during manualmapping. This clinical observation is consistent with our priorexperience of a marked reduction in premature ventricularbeats during remote mapping compared with manual map-ping in an experimental porcine model of healed MI.17

Electrogram stability is important during ventricular map-ping to perform electrophysiological pacing maneuvers suchas entrainment and pace mapping. In the present study,remote mapping demonstrated the requisite catheter stabilityto provide stable beat-to-beat electrogram morphology andconsistent endocardial or epicardial ventricular capture dur-ing pacing. This was true whether performed during sinusrhythm or VT for entrainment or pace mapping, respectively.Detailed chamber mapping also was feasible during sinusrhythm to identify late potentials.

Delivery of Radiofrequency Energy toTerminate VTIn a subset of the patient cohort, radiofrequency ablation withthe MNS and RMT catheter could be performed safely andfeasibly in the LV, RV, and epicardial space. The initialprocedure proved successful in 20 of 24 patients (83%). Ofthe 4 patients with recurrences, 3 had a successful repeatablation procedure with the MNS (2 requiring a combinedendocardial/epicardial approach), whereas the fourth patientwas ablated manually with a pericardial approach aftersurgical pericardial access.15 Seventeen VTs were success-fully terminated in 15 procedures with radiofrequency appli-cation with the RMT catheter at a mean duration of 8.4seconds. Of note, most terminations achieved by endocardialdelivery of radiofrequency energy occurred in �10 seconds(in 9 of 13), whereas most epicardial ablations took longer to

terminate. This is consistent with our previous observationand may in fact be related to the presence of epicardial fatserving as an insulating barrier to rapid and effective radio-frequency energy transmission to the target site.

Although in 2 cases successful ablation was performedentirely with the RMT catheter alone, it should be empha-sized that the manual irrigated catheter also was used in mostthe cases. The reason was the safety and efficacy limitationsof standard 4-mm-tip ablation in left-sided cardiac chambersas a result of a higher thromboembolic potential, not areflection of its maneuverability or stability. It remains to bedetermined whether the ablation process could have beencompleted entirely with an irrigated RMT catheter, a hypoth-esis that can be tested once this irrigated catheter becomesavailable for clinical use.

Study LimitationsFirst, the fluoroscopic visual field was partially compromisedduring the procedures. With the magnets in place, it is usuallynot possible to fluoroscopically visualize the entire ventricu-lar cavity. This was also evident during pericardial mapping,which generally requires a larger field of visualization.Nonetheless, this did not prove to be a major limitationbecause the need for fluoroscopy is greatly minimized withEAM. Second, the present study was not a randomizedcomparison of the safety and efficacy of remote and manualmapping and ablation. It was designed predominantly toaddress the feasibility of remote ventricular mapping. There-fore, definitive conclusions comparing these approaches can-not be reached without a formal comparative study. Thisincludes assessments of catheter stability, premature ventric-ular beat frequency, and endocardial trauma. Third, althoughcertain technical limitations of manual mapping are overcome

Figure 5. Mapping and ablation of hemodynam-ically-unstable VTs. Four hemodynamically-unstableVTs were induced in a patient with dilated cardiomy-opathy–related VT (#20). Remote epicardial voltagemapping (top) revealed a large low-voltage areawithout the presence of late or fractionated poten-tials, most likely representing epicardial fat.20 Theremote endocardial voltage map (bottom) revealedan anterior-septal scar with associated fractionatedand late potentials (inset). Pace mapping (rightpanel) from this area resulted in a QRS morphologysimilar to the first VT and with a delay between thestimulus to QRS. Irrigated radiofrequency ablation oflate and fractionated potentials and good pace mapsites resulted in successful elimination of all induc-ible VTs. The relationship of the endocardial and epi-cardial EAMs is better appreciated in Movie II in theonline-only Data Supplement.

1198 Circulation March 13, 2007

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

with remote mapping, one must still master the other skillsrequired to perform remote ventricular mapping, includingmaneuvering the somewhat complex software architecture ofremote navigation. Fourth, although both retrograde andtransseptal approaches were used during the present study,the transseptal approach was preferable because of the en-hanced response of the catheter tip to remote advancement/retraction. That is, during movement of the catheter duringretrograde aortic mapping, the “slack” of the catheter alongthe arch of the aorta results in a relatively slow response time formovement of the catheter tip, a phenomenon less pronouncedwith transseptal mapping. This requires, however, that theoperator be familiar with the transseptal puncture technique.

ConclusionsThe present study presents clinical evidence for the feasibilityof remote catheter navigation to perform ventricular

substrate-based mapping in humans in a wide range ofdisease pathologies. The enhanced maneuverability of theRMT catheter permitted accurate mapping of difficult-to-reach areas. The remote approach was safe and efficacious,and it was possible with a minimal amount of fluoroscopytime and radiation exposure to both patients and operators. Byobviating the need for the advanced operator skill oftenrequired for detailed ventricular mapping, substrate-based VTmapping with this approach may become much more wide-spread and effective.

Sources of FundingThis work was supported in part by a National Institutes of HealthK23 award (HL68064) to Dr Reddy.

TABLE 3. Results of VT Ablation

Time Duration and Location ofRemote VT Termination

RFA Lesions(by Catheter), n

Arrhythmia RecurrenceDuring Follow-Up

Patient

VTsTerminatedRemotely by

RFA, msTime,

s Location RMT

ManuallyIrrigatedCatheter

TotalRFA

Time,min

ArrhythmiasInducible at

End ofProcedure VT

ICDEvent

1 585 3.9 LV 4 25 26 None No No

2 � � � � � � � � � 3 28 � � � � � � No No

3–1* Not terminated � � � � � � � � � � � � 21 None MMVT No

3–2† 520 7.5 LV 3 20 18 VFL No No

4 Not terminated � � � � � � 4 22 14 VFL No � � �

5 � � � � � � � � � 2 20 15 None No No

6 � � � � � � � � � 0 28 18 None No � � �

7 435, 562 2.7, 11.9 LV 1 18 32 None No No

8 550 2 EPI (RV) 2 31 10 None No No

9 Not terminated � � � � � � 12 0 20 None No Yes

10–1* � � � � � � � � � 18 23 23 None MMVT No

10–2† 540 1.4 RV 3 27 14 None No No

11 470 0.9 LV 5 7 22 None No No

12 � � � � � � � � � 4 24 46 None No No

13 � � � � � � � � � 30 21 � � � None No No

14 380 11.9 EPI (RV) � � � � � � 19 None No No

15 � � � � � � � � � 18 0 44 MMVT No No

16 482 18.4 LV 6 44 36 None MMVT No

17 Not terminated � � � � � � 8 33 50 None No No

18–1* 417 31.6 LV 3 56 26 None MMVT � � �

18–2† 365 9.3 LV 1 27 27 None No � � �

19 380 16.4 LV 9 30 22 None No No

20 365 2.7 LV 4 23 23 None No � � �

21 � � � � � � � � � 6 15 � � � None No No

22 351 13.8 EPI (RV) � � � � � � 28 VFL No � � �

23 426 1.8 LV 24 14 36 PMVT No No

24 562, 712 3.6, 2.7 LV 10 36 46 None No No

Mean � � � 8.4 � � � 8 24 26 � � � � � � � � �

SD � � � 8.2 � � � 8 12 11 � � � � � � � � �

RFA indicates radiofrequency ablation; MMVT, monomorphic VT; VFL, ventricular flutter; EPI, epicardial; and PMVT, polymorphic VT.*First mapping and ablation procedure for this patient.†Second mapping and ablation procedure for this patient.

Aryana et al Remote Ventricular Substrate Mapping 1199

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

DisclosuresDr Reddy has served as a consultant to Biosense-Webster, Inc. DrRuskin has served on the Medical Advisory Board of Stereotaxis,Inc. The remaining authors report no conflicts.

References1. Marchlinski FE, Callans DJ, Gottlieb CD, Zado E. Linear ablation lesions for

control of unmappable ventricular tachycardia in patients with ischemic andnonischemic cardiomyopathy. Circulation. 2000;101:1288–1296.

2. Soejima K, Suzuki M, Maisel WH, Brunckhorst CB, Delacretaz E, BlierL, Tung S, Khan H, Stevenson WG. Catheter ablation in patients withmultiple and unstable ventricular tachycardias after myocardialinfarction: short ablation lines guided by reentry circuit isthmuses andsinus rhythm mapping. Circulation. 2001;104:664–669.

3. Reddy VY, Neuzil P, Taborsky M, Ruskin JN. Short-term results ofsubstrate mapping and radiofrequency ablation of ischemic ventriculartachycardia using a saline-irrigated catheter. J Am Coll Cardiol. 2003;41:2228–2236.

4. Arenal A, Glez-Torrecilla E, Ortiz M, Villacastin J, Fdez-Portales F,Sousa E, del Castillo S, de Isla LP, Jimenez J, Almendral J. Ablation ofelectrograms with an isolated, delayed component as treatment ofunmappable monomorphic ventricular tachycardias in patients withstructural heart disease. J Am Coll Cardiol. 2003;41:81–92.

5. Faddis MN, Chen J, Osborn J, Talcott M, Cain ME, Lindsay BD.Magnetic guidance system for cardiac electrophysiology: a prospectivetrial of safety and efficacy in humans. J Am Coll Cardiol. 2003;42:1952–1958.

6. Ernst S, Ouyang F, Linder C, Hertting K, Stahl F, Chun J, Hachiya H,Bansch D, Antz M, Kuck KH. Initial experience with remote catheterablation using a novel magnetic navigation system: magnetic remotecatheter ablation. Circulation. 2004;109:1472–1475.

7. Pappone C, Vicedomini G, Manguso F, Gugliotta F, Mazzone P, GullettaS, Sora N, Sala S, Marzi A, Augello G, Livolsi L, Santagostino A,Santinelli V. Robotic magnetic navigation for atrial fibrillation ablation.J Am Coll Cardiol. 2006;47:1390–1400.

8. Eldar M, Ohad D, Bor A, Varda-Bloom N, Swanson DK, Battler A. Aclosed-chest pig model of sustained ventricular tachycardia. Pacing ClinElectrophysiol. 1994;17:1603–1609.

9. Sosa E, Scanavacca M, D’Avila A, Oliveira F, Ramires JAF. Nonsurgicaltransthoracic epicardial catheter ablation to treat recurrent ventriculartachycardia occurring late after myocardial infarction. J Am Coll Cardiol.2000;35:1442–1449.

10. D’Avila A, Scanavacca M, Sosa E, Ruskin JN, Reddy VY. Pericardialanatomy for the interventional electrophysiologist. J Cardiovasc Electro-physiol. 2003;14:422–430.

11. Reddy VY, Wrobleski D, Houghtaling C, Josephson ME, Ruskin JN.Combined epicardial and endocardial electroanatomic-mapping in aporcine model of healed myocardial infarction. Circulation. 2003;107:3236–3242.

12. Callans DJ, Ren JF, Michele J, Marchlinski FE, Dillon SM. Electro-anatomic left ventricular mapping in the porcine model of healed anteriormyocardial infarction: correlation with intracardiac echocardiography andpathological analysis. Circulation. 1999;100:1744–1750.

13. Wrobleski D, Houghtaling C, Josephson ME, Ruskin JN, Reddy VY. Useof electrogram characteristics during sinus rhythm to delineate the endo-cardial scar in a porcine model of healed myocardial infarction. J Car-diovasc Electrophysiol. 2003;14:524–529.

14. Marchlinski FE, Garcia F, Siadatan A, Sauer W, Beldner S, Zado E, HsiaH, Lin D, Cooper J, Verdino R, Gerstenfeld E, Dixit S, Russo A, CallansD. Ventricular tachycardia/ventricular fibrillation ablation in the settingof ischemic heart disease. J Cardiovasc Electrophysiol. 2005;16:S59–S70.

15. Soejima K, Couper G, Cooper JM, Sapp JL, Epstein LM, Stevenson WG.Subxiphoid surgical approach for epicardial catheter-based mapping andablation in patients with prior cardiac surgery or difficult pericardialaccess. Circulation. 2004;110:1197–1201.

16. Ram W, Meyer H. Heart catheterization in a neonate by interactingmagnetic fields: a new and simple method of catheter guidance. CatheterCardiovasc Diagn. 1991;22:317–319.

17. Basu Ray I, Houghtaling C, McPherson C, Kastelein N, Ruskin JN,Reddy VY. Use of a remote magnetically-guided catheter for ventricularelectroanatomical substrate mapping and ablation in a porcine model ofhealed myocardial infarction. Heart Rhythm. 2005;2:S158. Abstract.

18. Kim RJ, Fieno DS, Parrish TB, Harris K, Chen EL, Simonetti O, BundyJ, Finn JP, Klocke FJ, Judd RM. Relationship of MRI delayed contrastenhancement to irreversible injury, infarct age, and contractile function.Circulation. 1999;100:1992–2002.

19. Lardo AC, Cordeiro MA, Silva C, Amado LC, George RT, Saliaris AP,Schuleri KH, Fernandes VR, Zviman M, Nazarian S, Halperin HR, WuKC, Hare JM, Lima JA. Contrast-enhanced multidetector computedtomography viability imaging after myocardial infarction: character-ization of myocyte death, microvascular obstruction, and chronic scar.Circulation. 2006;113:394–404.

20. Abbara S, Desai JC, Butler J, Nieman K, Reddy VY. Mapping epicardialfat with multidetector computed tomography to facilitate percutaneoustransepicardial arrhythmia ablation. Eur J Radiol. 2006;57:417–422.

CLINICAL PERSPECTIVESignificant advances have been made in recent years in catheter ablation of scar-related ventricular tachycardia as a resultof both (1) an improved understanding of its pathophysiology and (2) technological advances that aid in performing theprocedure. In particular, substrate mapping, in which the ventricle is mapped predominantly during sinus rhythm, allowssuccessful catheter ablation of virtually all ventricular tachycardias, regardless of their hemodynamic effect. Althoughhighly effective, ventricular tachycardia ablation remains an uncommon procedure, in part because of the advancedoperator skill required to perform detailed ventricular mapping. The present study provides clinical evidence for the safetyand feasibility of remote catheter navigation in performing ventricular substrate mapping in a wide range of diseasepathologies. Used in concert with a compatible electroanatomic mapping system, remote magnetic navigation technologyproved capable of performing each of the 3 major components of substrate-based ventricular tachycardia ablation:(1) delineating and identifying endocardial and epicardial scarred tissue, (2) performing the necessary electrophysiologicalmaneuvers required to identify those arrhythmogenic zones critical for maintaining tachycardia, and (3) deliveringradiofrequency energy to terminate both endocardial and epicardial ventricular tachycardias. The enhanced maneuverabil-ity of the remotely navigated catheter using this system allowed accurate mapping of otherwise difficult-to-reach areas.Finally, the “soft touch” of the remotely guided catheter permitted this detailed ventricular mapping with minimalfluoroscopy use. Thus, by obviating the need for the advanced operator skill required for substrate mapping, remotenavigation technology may result in more widespread and effective catheter ablation of ventricular tachycardia.

1200 Circulation March 13, 2007

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from

Ruskin and Vivek Y. ReddyArash Aryana, Andre d'Avila, E. Kevin Heist, Theofanie Mela, Jagmeet P. Singh, Jeremy N.

Scar-Related Ventricular TachycardiaRemote Magnetic Navigation to Guide Endocardial and Epicardial Catheter Mapping of

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 2007 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/CIRCULATIONAHA.106.672162

2007;115:1191-1200; originally published online February 12, 2007;Circulation. 

http://circ.ahajournals.org/content/115/10/1191World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circ.ahajournals.org/content/suppl/2007/02/19/CIRCULATIONAHA.106.672162.DC1.htmlData Supplement (unedited) at:

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialCirculationin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on July 10, 2015http://circ.ahajournals.org/Downloaded from