MRI-Guided ventricular tachycardia ablation: integration of late gadolinium-enhanced 3D scar in...

1941-3084American Heart Association. All rights reserved. Print ISSN: 1941-3149. Online ISSN: 2011 Copyright ©

Avenue, Dallas, TX 72514Circulation: Arrhythmia and Electrophysiology is published by the American Heart Association. 7272 Greenville

DOI: 10.1161/CIRCEP.110.958744 published online January 26, 2011;Circ Arrhythm Electrophysiol

Matthew Peters, Anastasios Saliaris, Magdi Saba, Stephen Shorofsky and Jean JeudyTimm Dickfeld, Jing Tian, Ghada Ahmad, Alejandro Jimenez, Aharon Turgeman, Richard Kuk,

Enhanced 3D Scar in Patients with ICDMRI-Guided Ventricular Tachycardia Ablation: Integration of Late Gadolinium

World Wide Web at:

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

http://circep.ahajournals.org/content/suppl/2011/01/26/CIRCEP.110.958744.DC1.html Data Supplement (unedited) at:

initial publication. Advance online articles must include the digital object identifier (DOIs) and date ofpublication priority; they are indexed by PubMed from initial publication. Citations to available prior to final publication). Advance online articles are citable and establishnot yet appeared in the paper journal (edited, typeset versions may be posted when Advance online articles have been peer reviewed and accepted for publication but have

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

[email protected] West Camden Street, Baltimore, MD 21201-2436. Phone: 410-528-4050. Fax: 410-528-8550. E-mail:Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters Kluwer Health,

http://circep.ahajournals.org/site/subscriptions/Subscriptions: Information about subscribing to Circulation: Arrhythmia and Electrophysiology is online at

by guest on February 21, 2013circep.ahajournals.orgDownloaded from

http://circep.ahajournals.org/site/subscriptions/

mailto:[email protected]

http://www.lww.com/reprints

http://circep.ahajournals.org/

1

MRI-Guided Ventricular Tachycardia Ablation: Integration of Late

Gadolinium Enhanced 3D Scar in Patients with ICD

Running title: Dickfeld et al.; MRI-Guided VT Ablation in ICD Patients

Timm Dickfeld, MD, PhD*, Jing Tian, MD, PhD*, Ghada Ahmad, MD*,

Alejandro Jimenez, MD*, Aharon Turgeman, M.Sc. M.B.A^, Richard Kuk, MD*,

Matthew Peters, MS*, Anastasios Saliaris, MD*, Magdi Saba, MD*,

Stephen Shorofsky, MD, PhD*, Jean Jeudy, MD#

Division of Cardiology* and Radiology#, University of Maryland, Baltimore Biosense Webster Inc., Tirat Carmel, Israel^

Corresponding Author:

Timm Dickfeld, MD, PhD

Division of Cardiology

University of Maryland

22 S. Greene Str., Room N3W77

Baltimore, MD 21201

Fax: 410-328-2062

Phone: 410-328-6056

E-mail: [email protected]

Journal Subject Codes: [22] Ablation/ICD/surgery, [30] CT and MRI, [106] Electrophysiology

MMMMMMMagagagagagagagdididididididi S S S S S S Sabababababababa,a,a,a,a,a,a, M M M M M M MDDDDDDD

dy, MMMMMMMD#D#D#D#D#D#D#

of Cardiology* and Radiology#, University of Maryland, Balof Cardiology* and Radiology#, University of Maryland, BalBiosense Webster Inc., Tirat Carmel, Israel^



2

Abstract:

Background - Substrate-guided ablation of ventricular tachycardia (VT) in patients with

implanted defibrillators (ICD) relies on voltage mapping to define the scar/border zone.

An integrated three-dimensional (3D) scar reconstruction from late gadolinium-enhanced

(LGE) MRI could facilitate VT ablations.

Methods and Results - Twenty-two patients with ICD underwent contrast-enhanced

cardiac MRI (CE-CMR) with a specific absorption rate <2.0W/kg prior to VT ablation.

Device interrogation demonstrated unchanged ICD parameters immediately before, after

or at 68±21 days follow-up (p>0.05). ICD imaging artifacts were most prominent in the

anterior wall and allowed full and partial assessment of LGE in 9±4 and 12±3 of 17

segments, respectively.

In fourteen patients with LGE a 3D scar model was reconstructed and successfully

registered with the clinical mapping system (accuracy: 3.9±1.8mm). Using ROC-curves

bipolar and unipolar voltages of 1.49mV and 4.46mV correlated best with endocardial

MRI scar.

Scar visualization allowed the elimination of falsely-low voltage recordings (suboptimal

catheter contact) in 4.1±1.9% of <1.5mV mapping points.

Display of scar border zone allowed identification of excellent pacemapping sites with

only limited voltage mapping in 64% of patients. Viable endocardium of >2mm resulted

in >1.5mV voltage recordings despite up to 63% transmural midmyocardial scar

successfully ablated with MRI guidance. All successful ablation sites demonstrated LGE

(transmurality: 68±26%) and were located within 10mm of transition zones to 0-25% scar

in 71%.

Conclusions - CE-CMR can be safely performed in selected ICD patients and allows the

integration of detailed 3D scar maps into clinical mapping systems. This provides

supplementary anatomic guidance to facilitate substrate-guided VT ablations.

Key words: MRI/ICD/Ablation/Ventricular Tachycardia

uctededededddd aaaa aa andndndndndndnd sss s sssucucucucucucuccececececececesssss

he clinical mapping s stem (accurac : 3.9±1.8mm). Using R

o n

(

in 4 1±1 9% of <1 5mV mapping points

he clinical mapping system (accuracy: 3.9±1.8mm). Using Rm

olar voltages of 1.49mV and 4.46mV correlated best with en

n allowed the elimination of falsely-low voltage recordings (f

in 4 1±1 9% of <1 5mV mapping points



3

INTRODUCTION

With expanding ICD indications an increasing amount of patients present with

appropriate ICD shocks for ventricular arrhythmias.1, 2 The side effects and lack of long-

term efficacy of antiarrhythmics have made VT ablation an increasingly attractive

treatment option.3, 4

In ~70-90% of patients a substrate-guided VT ablation must be pursued due to

hemodynamic instability or multiple VT morphologies.4, 5 During the ablation tangential

or radial lesions are placed along and transsecting the scar border in an attempt to

eliminate conducting channels and VT exit sites.

In these procedures, voltage mapping is the “gold standard” for defining

myocardial scar/border zone or surviving myocardial channels. However, voltage

mapping has multiple limitations: a single endo- (or epi-)cardial voltage measurement

representing a poor surrogate for a complex intramural three-dimensional (3D) scar

anatomy; limited spatial resolution of the electroanatomic maps; prolonged mapping

times; and falsely-low voltage measurements due to imperfect catheter contact.6

MRI has the ability to accurately delineate myocardial scar,7 but ICDs are present

in the majority of VT patients and are still considered a contraindication or impact

imaging quality.8 This study evaluated the safety/diagnostic yield of cardiac MRI in the

largest reported ICD patient cohort to-date and is the first to assess the feasibility of

registered 3D MRI scar reconstructions to facilitate VT ablations in ICD patients.

METHODS

Patient Population

ndarrd”d”ddddd f f f f fffororororororor d d d ddddefefefefefefefiiiiinini

/b o

l a

o D

border zone or surviving myocardial channels. However, vo

ltiple limitations: a single endo- (or epi-)cardial voltage mea

oor surrogate for a complex intramural three-dimensional (3D



4

Consecutive patients with an ischemic/non-ischemic cardiomyopathy and ICD

scheduled for VT ablation for frequent ICD therapies were enrolled. Exclusion criteria

consisted of device implantation <6 weeks, epicardial lead/coil system,

abandoned/capped endovascular leads, ICDs manufactured before 2002, and creatinine

clearance <30ml/min. Written consent was obtained in all patients after institutional

approval of the research protocol.

Pre- and Post MRI ICD Interrogation

ICD interrogation was performed immediately before/after the MRI as well as

during follow-up visits. During the MRI tachyarrhythmia detection and/or therapy was

disabled. In pacemaker-dependent patients, bradycardia parameters were programmed to

VOO/DOO mode (60 beats per minute). In not pacemaker-dependent patients rate-

responsiveness feature was temporarily disabled and VVI or DDD programming was left

unchanged.

ECG, blood pressure and pulse oxymetry was continuously monitored during the

MRI and clinical symptoms were recorded. Post-MRI programming was restored to its

original parameters.

MR Imaging

MRI was performed supine using a 1.5T scanner (Siemens Avanto, Princeton, NJ,

USA). Imaging sequences resulted in a specific absorption rate (SAR) <2.0W/kg. Cine

turboFlash sequences to assess anatomic and dynamic characteristics were performed to

assess wall thinning and contractility (FOV=320-400mm/TR=75-80ms/TE=3-4ms/flip

/after the MRIRIIIIII a a

ectioioooooon nn n n n n ananananananand/d/d/d/d/d/d/ororororororor t t t t tt thh

e o

d t

fe m

emaker-dependent patients, bradycardia parameters were pro

de (60 beats per minute). In not pacemaker-dependent patientt

feature was temporarily disabled and VVI or DDD programm



5

angle=15°/slice thickness=8mm/10mm spacing). First-pass perfusion using dynamic

gradient echo sequences to assess regional hypoperfusion characteristics were performed

(FOV=320-400mm/TR=160-165ms/TE=1-1.5ms/flip angle=15°/slice

thickness=8mm/10mm spacing).

Inversion recovery images were obtained ECG-gated during diastole 10-15min

after iv. injection of 0.1mg/kg gadobenate dimeglumine (Multihance, Bracco, Princeton,

NJ). Short and long axis 2D (slice thickness=8mm/no gap/TI=250-350ms/TR=725-

950ms/TE=1-4ms/flip angle=25°) and 3D (slice thickness=4-6mm/no gap/TR=700-

750ms/TE=1-1.5ms/flip angle=10°) inversion recovery sequences (Figure 1) were

obtained after optimal nulling of the inversion time to assess late gadolinium

enhancement (LGE).

Delineation of MRI-Derived 3D Myocardium and LV Scar

2D and 3D MRI scans were exported in DICOM format to an external imaging

workstation. Left and right ventricular (LV, RV) endo- and epicardial surfaces as well as

intramyocardial boundaries of myocardial LGE were determined by a radiologist expert

reader with >10 years of experience and hand-planimetered on contiguous, reformatted

2/3D slices using modified Siemens VPT (Siemens, Princeton, NJ, USA) and Amira

software (Amira, Berlin, Germany). In areas of ICD artifact an approximated

reconstruction of the LV endo- and epicardium (without scar) was performed to facilitate

image registration. Scar was only reconstructed from LGE images and if no significant

ICD artifact was present. Reconstructed 3D models of the RV/LV anatomy with

embedded myocardial scar were exported as 3D Carto readable Mesh-files and allowed

ences (Figure 1111111))

latee g g gggggadadadadadadadolololololololinininininininiuiuiuiuiuiuiumm

G

M

GE).

MRI-Derived 3D Myocardium and LV Scar



6

the identification of mapping point positions on the corresponding 2D MRI images.

Quantitative and regional scar assessment was performed with Segment v1.x software.9

Using the IPE research software module (Biosense-Webster, Diamond Barr, CA) MRI

surfaces were uploaded into the clinical CartoMERGE system.

Registration of Voltage Map and MRI-Derived LV Anatomy and Scar

If the CartoSOUND module was used an anatomic 3D reconstruction of the RV

and LV was created from sequential 2D intracardiac ultrasound slices (SoundStar,

Biosense Webster) acquired from a RV position.

Using a 3.5mm open-irrigation tip catheter (Thermocool, Biosense Webster) LV

voltage maps were created via retrograde or transseptal approach using a filling threshold

of 15mm and voltage settings of 0.5mV-1.5mV. Uni- and bipolar signals were filtered at

10-400Hz and were acquired during sinus rhythm or ventricular pacing in case of

resynchronization therapy/pacemaker dependency. Signals were defined as normal,

fractionated (FP) or diastolic potentials (DP).10

Applying an “early registration” strategy visual alignment of the 3D MRI

reconstruction was performed with the CartoMERGE software using >10 RV/>50 LV

mapping points. Alternatively, using a “CartoSOUND registration” approach the MRI

LV/scar reconstructions were registered to the anatomic 3D RV and LV ultrasound

shells. This allowed the display of the extracted LV scar during an early stage of the LV

mapping (“early registration”) or even prior to entering the arterial circulation

(“CartoSOUND registration”). “Final registration” of the 3D MRI scar map was

performed after completing RV/LV mapping.

ool, BiBiBiBiBiBiBiosososososososenenenenenenensesesesesesese W W W W W W W

e i

l r

w a

ere created via retrograde or transseptal approach using a filli

ltage settings of 0.5mV-1.5mV. Uni- and bipolar signals wer

were acquired during sinus rhythm or ventricular pacing in ca



7

For visually alignment superior mitral valve (MV) points (12 o’clock) were used

as determined by electrical signals, fluoroscopy±ultrasound. Rotational errors were

minimized by aligning the superior and inferior RV septal insertion sites of the MRI-

derived shell with the RV/LV voltage map (Figure 1). For landmark point registration LV

apex, MV and RV septal insertions were selected.

Registration accuracy was determined using the internal CARTO summation

statistics, averaging the distance of the individual mapping points to the closest surface

point of the registered MRI shell.11

VT Ablation

After registration additional MRI-guided mapping was performed at sites <1.5mV

remote from matching MRI scar to determine falsely-low voltage recordings due to

suboptimal catheter contact as well as in areas of complex MRI border zone geometry.

VT was induced by programmed electrical stimulation (PES) with up to triple

extrastimuli from two RV and up to two LV sites (drive cycle length: 350/400/600ms;

minimum coupling interval 200ms). If VT was non-sustained or not tolerated

pacemapping sites were chosen based on MRI-defined border zone (after

“early/CartoSound registration”) and the 12-lead VT morphology. Matches 11/12 were

used to determine the approximate exit sites. In case of hemodynamically tolerated VT

standard entrainment criteria were used within areas of MRI scar.12 Successful ablation

lesions (50W; 60s) were defined as terminating VT or occurring at site of pacemap

matches.

s

t

e

istration additional MRI-guided mapping was performed at s

tching MRI scar to determine falsely-low voltage recordings

eter contact as well as in areas of complex MRI border zone



8

Additional tangential/radial ablation lines were created based on scar anatomy

defined by voltage and MRI data. LV mapping was completed (unless noted otherwise)

to assess “final registration” and location of FPs/DPs/voltage-defined border zone. PES

was repeated at the end of the VT ablation. Criteria defining ablation success were the

inability to induce the clinical/presumed clinical VT or sustained monomorphic VT with

longer cycle length.3

Statistical Analysis

SPSS for Windows 16.0 was used to perform the statistical analysis. Continuous

variables are expressed as mean ±1 SD unless otherwise noted.

Comparisons between pre- and post- MRI measures were conducted using the

Wilcoxon signed rank test. The Spearman Rank Correlation Test was obtained to assess

possible correlations. Differences were considered significant at a level of P<0.05.

Receiver operating characteristic (ROC) curves were created for unipolar/bipolar voltage

recordings to identify the best cut off voltage values to predict endocardial MRI scar.

Areas under the curve (AUCs) for each voltage recordings were reported and compared.

RESULTS

Patient Population

Four of the 26 screened patients were excluded due to an epicardial ICD system

(n=1), abandoned endocardial leads (n=2) or creatinine clearance of <30ml/min (n=1).

Twenty-two patients were enrolled in the study. Baseline characteristics are shown in

Table 1.

istical analysisssssss. ... ... C

d.

s u

d e

i <

sons between pre- and post- MRI measures were conducted u

d rank test. The Spearman Rank Correlation Test was obtaine

ions. Differences were considered significant at a level of P<



9

Safety of Cardiac MRI in Selected ICD Patients

CE-CMRI was performed in twenty-two patients with single (n=9), dual (n=10)

and bi-ventricular (n=3) ICDs (Data Supplement 1). ICDs were interrogated before/after

MRI and at 68±21 days follow-up (Table 2). Five devices were equipped with wireless

telemetry. Two patients were pacemaker-dependent with ventricular escape rates

<30bpm. No significant changes in battery voltage/impedance, atrial or ventricular

thresholds, intrinsic amplitudes or impedances were seen between pre-MRI, post MRI

and follow-up (Table 2).

No complication occurred during the MRI scanning. No inappropriate/inhibited

pacing or tachytherapy were observed.

Image Quality of Cardiac MRI in ICD Patients

Anatomic-dynamic and first-pass perfusion MR sequences demonstrated limited

artifacts and allowed detailed assessment of the septal, inferior, lateral and anterior wall

regarding anatomic-dynamic (100%/100%/100%/82%) and perfusion

(100%/100%/100%/91%) characteristics.

In LGE sequences for scar visualization ICD artifacts were more pronounced and

appeared as a central signal void (ICD generator) with a surrounding rim of increased

signal intensity (Figure 1A). LGE MRI allowed the partial/full evaluation of the septal

(50%/36%), inferior (100%/91%), lateral (100%/18%) and anterior (27%/0%) wall,

respectively. Only limited <1cm circumferential artifacts were caused by the ICD leads in

No iinananananananapppppppppppppprororororororoprprprprprprpriaiaiaiaiaiaiatet

h

f

herapy were observed.

f Cardiac MRI in ICD Patients



10

any of the imaging sequences (Figure 1A). Using the 17-segment AHA classification

LGE could be fully assessed in 9±4 segments and partially assessed in 12±3 segments.13

In ischemic patients 82±7% of scar had endocardial components with mid-

myocardial or epicardial components being predominantly seen as <1cm scar extensions

in the border zone. Of the seven patients with non-ischemic cardiomyopathy two (29%)

had exclusively epicardial or midmyocardial scar. Two non-ischemic patients (29%) had

no identifiable scar.

Extraction and Registration of 3D MRI Scar Map

CE-CMR-based scar integration into a mapping system prior to ablation was

performed in all fourteen patients with LGE that proceeded to a VT ablation (ischemic

cardiomyopathy: eleven). Of the other eight patients, three underwent placement of a left

ventricular assist device or heart transplantation for worsening heart failure, two

requested to have the MRI stopped due to claustrophobia, two had no visible scar and one

patient opted for pharmacological therapy.

Endocardial/epicardial RV/LV borders were successfully hand-planimetered on

contiguous 2D short axis slices in all patients with approximated LV wall contours in

areas of artifact. Detailed 3D scar anatomywas displayed embedded into the myocardium

(Figure 1). LV reconstruction from areas of artifact were separately reconstructed

(without scar) with different color annotation to simplify visualization if artifact and scar

were adjacent (Figure 2).

“Early registration” was performed in the first nine patients using 13±3 RV points

and 54±4 LV points with a registration accuracy of 5.7±3.9mm. In the last five patients

m ppririririririr ororororororor t t t t t to o o o o o o ababababababablalalalalalalatit

: eleven). Of the other eight patients, three underwent plac m

t t

fourteen patients with LGE that proceeded to a VT ablation

: eleven). Of the other eight patients, three underwent placem

t device or heart transplantation for worsening heart failure, tr



11

MRI-derived LV/scar maps were registered to CartoSOUND-reconstructed RV/LV shells

with a registration error of 4.6±3.1mm.

“Final registration” accuracy of the completed voltage map (58±39 RV and

190±32 LV points) was 3.9±1.8mm using visual alignment and 4.1±2.0mm with

landmark point/surface registration (Figure 1B). A full LV map was created in all nine

patients with “early registration” and two out of five patients with “CartoSOUND

registration”.

Intraprocedural MRI Guidance During VT Ablation

There was a good correlation between myocardial scar defined by CMR and

voltage.Endocardial voltage points of <0.1mV, <0.5mV and <1.5mV demonstrated MRI-

defined scar in 100%, 87% and 75%, respectively. Increasing transmurality of MRI-

derived scar correlated with decreasing endocardial bipolar voltage (r=-0.83;p<0.01;

Figure 3). ROC curves demonstrated the best cut off values for bipolar/unipolar voltage

recordings to optimally differentiate endocardial scar from non-scar defined by MRI are

1.49mV (AUC:0.86±0.01) and 4.46mV (AUC:0.78±0.02), respectively.

Procedure time for “early registration”, “CartoSOUND registration” and “final

registration” was 4±3min, 5±4min and 5±3min. The “early registration” or

“CartoSOUND registration” allowed the visualization of MRI-derived scar size, location

and transmural extent during 71% and 100% of the total LV mapping point acquisition,

respectively.

Display of the MRI scar guided the further mapping/ablation in several ways:

Detection of Suboptimal Catheter Contact

ar deeeeeefififififififinenenenenenened d d d d dd bybybybybybyby C C C CCCCMM

d

o

r

dial voltage points of <0.1mV, <0.5mV and <1.5mV demon

00%, 87% and 75%, respectively. Increasing transmurality o

related with decreasing endocardial bipolar voltage (r=-0.83;



12

Immediate identification of low voltage recordings (<1.5mV) in areas without

MRI-derived scar suggested imperfect catheter contact. Repeat mapping using

echocardiographic contact confirmation demonstrated that 78±12% of those points

(average voltage: 0.83±0.52mV) were shown to have voltages >1.5mV. This represented

4.1±1.9% of the total original <1.5mV mapping points. Locations of mislabeled points

were mid-anterior (37%), septal (29%), infero-basal (25%) and lateral (9%).

Targeted Mapping of Voltage/DP/FP

The integration of 3D MRI-defined scar provided an anatomic characterization of

the 3D scar/border zone geometry displaying varying transmurality and epi-, mid- and

endocardial scar components (Figure 4). Visualization of complex transition zones of scar

transmurality within scar or border zones allowed anatomically targeted mapping for

FPs/DPs or preserved voltages. FPs and DPs were found more frequently within 10mm of

the MRI scar border than in the scar center (69±18% vs. 31±17%, p<0.01). Additionally,

FPs and DPs were found more frequently in areas of >50% vs. <50% MRI-defined scar

transmurality (68±19% vs. 31±18%, p<0.01 and 74±28% vs. 26±28%, p<0.01,

respectively).

Targeted Pacemapping/Entrainment/Ablation

Empiric pacemap sites were selected along the displayed MRI border zone guided

by the 12-lead VT morphology prior to completing LV voltage mapping. This identified

presumptive VT exit sites with 11/12 pacemap match in 64% of patients (Figure 5A;5B).

Pacemap matches 11/12 were found within 10mm of the MRI scar border zone in 77%

of VTs. Mapping density was generally 5-10mm in these targeted locations.

urality and epipipipipipipi ,

mpleex xx x x x x trtrtrtrtrtrtranananananannsisisisisisisitititititititiononooooo

t p

e t

r A

thin scar or border zones allowed anatomically targeted map

erved voltages. FPs and DPs were found more frequently wit

rder than in the scar center (69±18% vs. 31±17%, p<0.01). A



13

MRI detected surviving papillary muscle within a large area of infarcted

myocardium in 7% of patients. Early integration of 3D-reconstructed viable myocardium

allowed identification of 11/12 pacemap match at scar/myocardial interface prior to

complete LV mapping (Figure 6).

Despite preserved endocardial voltage (>1.5mV) the display of MRI

midmyocardial scar with a transmurality of 11-63% identified a 11/12 pacemap match

(7% of patients) at the transition zone of enhanced/non-enhanced myocardium (Figure 7).

The identification of midmyocardial scar as VT substrate adjacent to alive endocardium

at the excellent pacemap location suggested a possible endocardial scar exit site,

explained the >1.5mV amplitude, and provided the rationale for ablation in “voltage-

defined” normal myocardium that successfully eliminated the clinical VT. During

bipolar, endocardial mapping a >2mm rim of viable endocardium was observed in 14%

of patients and resulted in >1.5mV voltage recordings during bipolar, endocardial

mapping. This was observed predominantly in non-ischemic patients with septal/infero-

lateral scar location, in which the mid-myocardial scar accounted for 27% and 100% of

total scar, respectively (Data Supplement 2). The preserved voltage would have

prevented the detection of adjacent intramyocardial scar based on voltage criteria alone.

Hemodynamically tolerated VT was induced in 14% of patients. Registration of

MRI-defined scar to the CartoSOUND shell enabled entrainment mapping within the scar

without prior voltage mapping and facilitated the identification of the successful ablation

site (Figure 8). In three of the five patients using the CartoSOUND registration strategy

real-time visualization of RF lesions within the MRI-defined abnormal substrate could be

confirmed by increasing local ultrasound signal intensity (Figure 8).

ardial scar exxititititititit s

for ababababababablalalalalalalatititititiitiononononononon i i i i i iin n n nnnn ““

d v

e a

myocardium that successfully eliminated the clinical VT. D

dial mapping a >2mm rim of viable endocardium was observ

esulted in >1.5mV voltage recordings during bipolar, endoca



14

Ablation Results

At PES 2.4±1.2 VTs/patient were inducible, with at least one VT matching the

presumed clinical VT by 12-lead surface ECG and/or ICD-based intracardiac

morphology/rate.3 Cycle length was 389±99ms with 5 left bundle branch, 24 right bundle

branch and 4 indeterminate morphologies (Data Supplement 2).

Twenty-two VTs were targeted for ablation. Acute procedural success was

assessed with PES including double (n=6) and triple extrastimuli (n=8). Clinical VT

remained inducible (n=1) or monomorphic VT, which was faster than the clinical VT, but

had not been previously clinically observed, was inducible and was not ablated (n=6).

All successful ablation sites demonstrated MRI scar (Figure 5C-F;7E-F). Average

scar transmurality was 68±26% [30-100%]. Eighty-one percent of ablation sites had

50% transmural scar components and were located within 10mm of transition zones to

0-25% transmural scar in 71%. Successful ablation sites were more commonly located in

the scar periphery (76%). Additional linear ablations were created a) along voltage/MRI

defined scar borders (n=12 patients); b) connecting areas of dense scar (<0.1mV/>85%

scar transmurality by MRI; n=3 patients); and c) connecting dense scar (<0.1mV/>85%

transmurality) to electrically non-conducting structures such as mitral valve ring (n=6

patients). More than one linear ablation strategy was applied in some patients.

No procedure-related complications were observed. After 15±12months follow-up

54% of patients had non-sustained or sustained VT documented by ICD interrogation

(time-to-first-occurrence 5±4 months post-ablation). Appropriate ICD shocks were seen

in 4 of the 14 patients. ICD shocks occurred in two patients during a heart failure

aster than the clclclclclclclini

nd wawawawawawaw s ss s s s s nononononononot t t t ttt ababababababablalalalalalalatt

e E

ty was 68±26% [30-100%]. Eighty-one percent of ablation i

o

essful ablation sites demonstrated MRI scar (Figure 5C-F;7Ed

ty was 68±26% [30-100%]. Eighty-one percent of ablation si

l scar components and were located within 10mm of transitio



15

exacerbation, which resulted ultimately in the patients’ deaths 1 week and 20 months

after ablation. One patient underwent a second ablation and one patient was treated

medically.

DISCUSSION

The findings of this study are a) cardiac contrast-enhanced MRI can be safely

performed in selected patients with ICD; b) ICD artifact decreases imaging quality but

allows the detection of myocardial MRI scar in the majority of patients; c) 3D LV

anatomy and reconstructed MRI-derived scar can be accurately registered with clinical

mapping systems; d) MRI-derived scar provides anatomic guidance during mapping and

VT ablation and e) in combination with 3D ICE allows real-time assessment of ablation

lesions within the MRI substrate in some patients.

Safety and Diagnostic Yield of Cardiac MRI in ICD Patients

While most patients requiring VT ablations have ICDs, defibrillators are still

considered a contraindication for MRI as reflected by a recent FDA summary statement.8

While early reports demonstrated potential soft- and hardware failure with clinical

complications more recent phantom, animal and patient series have suggested that MRI

might be able to be performed safely with appropriate ICD and patient selection.8, 14-16

However, only a minority of MRI in these studies were performed for dedicated cardiac

imaging.

To our knowledge this is the largest study of cardiac MRI (n=22) performed in

ICD patients. No complications were seen with limited SAR of <2.0W/kg.

ely registered wwwwwwwi

uidancncncncncncnce e e e ee e dudududududuuririririririringngngngngngng m m

e) in combination with 3D ICE allows real-time assessment

e MRI substrate in some patients.



16

Gimbel et al. reported a “power-on-reset” in one of 7 ICD patients undergoing

non-thoracic MRI (SAR not specified). Nazarian et al. found no complications in 13

patients with cardiac MRI out of 24 ICD patients with thoracic/extra-thoracic MRI

(SAR<2W/kg). Naehle et al. reported recently on MRI in 18 ICD patients (4 patients with

cardiac MRI; SAR 2W/kg), which resulted in an acute decrease of battery voltage that

recovered at follow-up.

LGE MRI has been shown to accurately determine areas of myocardial necrosis7,

but imaging quality is impacted by ICD-induced geometric distortion, warping and

susceptibility-induced artifacts.17

In this study ICD artifacts significantly limited the detection of myocardial scar in

the anterior wall and depending on imaging plane/low implant location parts of the septal,

lateral and rarely inferior wall. However, LGE representing MRI scar could be visualized

in all patients.

The anatomic-dynamic and perfusion imaging sequences were less susceptible to

ICD artifacts and may allow the further characterization of the anterior wall scar substrate

based on anatomic (e.g. wall thinning), dynamic (e.g. hypocontractility), and first-pass

perfusion (hypoenhancement) characteristics, but was not performed in this study.

Feasibility of Image Integration of 3D MRI Derived Scar

Reconstruction of representative 3D scar models from 2D/3D image sequences

was possible in all cases. Integration and registration of the 3D reconstructed MRI scar

into the clinical mapping system was successfully performed in all patients. The final

registration accuracy was 3.9±1.8mm, which is similar to previously reported image

tectitiiiiiiononononononn o oo o o oof f f f f f f mymymymymymymyocoooooo

y b

and depending on imaging plane/low implant location parts

y inferior wall. However, LGE representing MRI scar could b



17

integration studies.18-20 Codreanu et al.21 and Desjardin et al.19 retrospectively

reconstructed and registered post-infarct MRI scar with previously acquired voltage

maps. Bogun et al. registered MRI-scar in fourteen non-ischemic patients in preparation

for VT or PVC ablation.18 All these studies used non-ICD patients and found a good

correlation between abnormal voltage and (especially endocardial) LGE.

Supplementary MRI Substrate Characterization During VT Ablation

To our knowledge this is the first time that early registration approaches have

been used to display intramural scar location and transmurality to guide the majority of

LV mapping and ablation.

An important finding of this approach is that ~4% of mapping points with initial

voltages of <1.5mV were found to be due to suboptimal catheter contact and that

simultaneous MRI scar display allowed for instantaneous correction resulting ultimately

in a more accurate substrate characterization. This is consistent with the findings of

Codreanu et al., who reported in 3 of 12 patients (25%) a >20% larger scar by voltage

criteria (<1.5mV) than by MRI, which they attributed to likely difficult catheter contact.21

Another important result of this study is that in over 60% of our patients the early

display of the MRI border zone combined with the 12-lead VT morphology allowed the

identification of 11/12 pacemap sites prior to detailed voltage mapping (normally

required to determine the border zone and pacemap sites). This is consistent with other

studies that found that successful ablation sites in ischemic VT were located in 68% at

the scar border zone.22 Additionally, not only the display of the lateral border zone, but

also the 3D integration of alive myocardium within a larger area of scar (such as

ty to guide the e e e e e e mm

r

m d

R g

rtant finding of this approach is that ~4% of mapping points

mV were found to be due to suboptimal catheter contact and

RI scar display allowed for instantaneous correction resulting



18

surviving papillary muscle) was able to visually guide successfully pacemapping prior to

complete voltage mapping. This may enable a faster, anatomically guided ablation

approach.

Furthermore, 3D MRI scar integration provided visualization of the

midmyocardial scar, functioning as the VT reentrant substrate that was not detected by

conventional voltage criteria. The midmyocardial scar display anatomically guided

pacemapping and provided the rationale to ablate in an endocardial site with normal

voltage and signal characteristics that successfully eliminated the clinical tachycardia. In

our study a >2mm rim of surviving endocardium was able to prevent the detection of up

to ~60% transmural midmyocardial scar using conventional voltage criteria.

Correspondingly, bipolar amplitudes of midmyocardial/epicardial scar were found in

ischemic patients to be 1.52±1.41mV.21

Differences between MRI scar and voltage defined scar have also been reported in

other studies. Twenty percent larger MRI scar than voltage scar was found by Codreanu

et al. in the infero-basal LV wall.21 Desjardins et al. reported normal EGM characteristics

in areas with 5±18% transmural scar.19 Similarly, we found that endocardial MRI scar of

<25% transmurality might not be detected by voltage mapping.23 This limitation of

voltage mapping to detect thin layers of scar may also help explain successful VT

ablations in myocardium with normal voltage amplitudes in 4% of patients.22

In this study, successful ablation sites displayed a scar transmurality of 68±26%

and were frequently adjacent to areas of <25% scar transmurality. This is consistent with

previous studies in ischemic patients that found a scar transmurality of 60±38% at

successful ablation sites19 and a preferential location in the border zone area.22 Similarly,

prevent the dedeeeeeetetttttt

volttagagagagagagage e e e e e e crcrcrcrcrcrritititititititerererererereriaiaiaiaiaiaia.

y

s

ces between MRI scar and voltage defined scar have also bee

y, bipolar amplitudes of midmyocardial/epicardial scar were

s to be 1.52±1.41mV.21

ces between MRI scar and voltage defined scar have also bee



19

LGE was seen at all ablation sites of LGE-positive patients with non-ischemic

cardiomyopathy.18

In this study the final decision about the ablation site relied on the conventional

parameters like pacemapping or concealed entrainment, which likely explains the similar

results to non-image-guided VT ablation studies.4 With improving protocols predictive

MRI characteristics of successful ablation sites could potentially be developed. In a

canine-model Ashikaga et al. demonstrated that high-resolution MRI was able to

delineate areas of surviving myocardial areas within scar that correlated with VT

channels during epicardial mapping.24 Several mathematical models exist which may be

able to predict VT channels and successful ablation sites based on the infarct border zone

geometry.25

Limitations

While in this small sample of ICD patients MRI did not result in adverse events,

this needs to be confirmed in larger patient cohorts and various imaging protocols.

ICD artifacts prevented the detection of LGE in the anterior wall and possibly

other parts of the LV. Complimentary anatomic/dynamic or perfusion imaging, which is

less susceptible to those artifacts could be used to provide additional information about

those areas. Alternatively, fusion imaging (e.g. PET/MRI) could help define the presence

of anterior scar.

While careful image registration was performed, misregistrations cannot be

excluded. However, the used registration algorithms have been validated in previous

models exist wwwwwwwhh

ed on n n n n n n thththththththe e e e e e inininininininfafafafafafafarcrcrrrrr t



20

studies and the overall registration accuracy was comparable to other reports suggesting

no significant additional error.18-20

The study population included patients with predominantly ischemic, but also

non-ischemic cardiomyopathy. However, this heterogeneity would be unlikely to affect

the MRI safety profile of ICDs. While scar distribution is frequently different in both

entities, only patients with LGE scar were used for 3D MRI scar registration and LGE

was found to be a critical component of all successful ablation sites.

CONCLUSIONS

3D MRI scar maps of ICD patients can be successfully extracted/registered and

provide an improved appreciation of the complex scar anatomy than can be provided by

voltage mapping alone. The early MRI scar display can assist in more accurate scar

delineation/mapping and provide anatomic guidance for identification of critical scar

substrate in many patients. Its impact on clinical parameters such as procedure times and

long-term success will need to be assessed in future randomized-controlled trials.

Acknowledgments: We thank Dr. Tom O’Donnell, PhD (Siemens Corporate Research,

New York) for his excellent technical support and software expertise.

This study was supported by a AHA Scientist Development Grant (0635304N).

Conflict of Interest Disclosures: Timm Dickfeld – Research Collaboration/Research

Grant (Minor):Biosense Webster; Aharon Turgeman - Employer Biosense-Webster

ly exxxxxxxtrtrtrtrtrtrracacacacacaca teteteteteteted/d/d/d/d/d/d/rererererereregggigg

o p

g a

p t

oved appreciation of the complex scar anatomy than can be p

g alone. The early MRI scar display can assist in more accura

ping and provide anatomic guidance for identification of crit



21

References:

1. Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R, Domanski M,

Troutman C, Anderson J, Johnson G, McNulty SE, Clapp-Channing N, vidson-Ray

LD, Fraulo ES, Fishbein DP, Luceri RM, Ip JH. Amiodarone or an implantable

cardioverter-defibrillator for congestive heart failure. N.Engl.J.Med. 2005;352:225-

37.

2. Moss AJ, Zareba W, Hall WJ, Klein H, Wilber DJ, Cannom DS, Daubert JP,

Higgins SL, Brown MW, Andrews ML. Prophylactic implantation of a defibrillator

in patients with myocardial infarction and reduced ejection fraction. N.Engl.J.Med.

2002;346:877-83.

3. Aliot EM, Stevenson WG, mendral-Garrote JM, Bogun F, Calkins CH, Delacretaz

E, Della BP, Hindricks G, Jais P, Josephson ME, Kautzner J, Kay GN, Kuck KH,

Lerman BB, Marchlinski F, Reddy V, Schalij MJ, Schilling R, Soejima K, Wilber

D. EHRA/HRS Expert Consensus on Catheter Ablation of Ventricular Arrhythmias:

developed in a partnership with the European Heart Rhythm Association (EHRA), a

Registered Branch of the European Society of Cardiology (ESC), and the Heart

Rhythm Society (HRS); in collaboration with the American College of Cardiology

(ACC) and the American Heart Association (AHA). Heart Rhythm. 2009;6:886-

933.

4. Stevenson WG, Wilber DJ, Natale A, Jackman WM, Marchlinski FE, Talbert T,

Gonzalez MD, Worley SJ, Daoud EG, Hwang C, Schuger C, Bump TE, Jazayeri M,

Tomassoni GF, Kopelman HA, Soejima K, Nakagawa H. Irrigated radiofrequency

mplantation off aaaa ff

tionn ff f f f fraractctioioioioioioion.n.n.n.n.n.n. NNNNN N N.

8

S ,

P, Hindricks G, Jais P, Josep son ME, Kautzner J, Kay GN,

877-83.

Stevenson WG, mendral-Garrote JM, Bogun F, Calkins CH,

P, Hindricks G, Jais P, Josephson ME, Kautzner J, Kay GN,



https://www.researchgate.net/publication/11457583_Prophylactic_Implantation_of_a_Defibrillator_in_Patients_with_Myocardial_Infarction_and_Reduced_Ejection_Fraction?el=1_x_8&enrichId=rgreq-4da6c628e7ed2673b9d954aaa6de88ae-XXX&enrichSource=Y292ZXJQYWdlOzQ5Nzg4MTIwO0FTOjI2NDY1NzEwOTc3ODQzMkAxNDQwMTEwNTgwNTkw




https://www.researchgate.net/publication/224883640_Bardy_GH_Lee_KL_Mark_DB_et_al_Amiodarone_or_an_implantable_cardioverter-defibrillator_for_congestive_heart_failure?el=1_x_8&enrichId=rgreq-4da6c628e7ed2673b9d954aaa6de88ae-XXX&enrichSource=Y292ZXJQYWdlOzQ5Nzg4MTIwO0FTOjI2NDY1NzEwOTc3ODQzMkAxNDQwMTEwNTgwNTkw





22

catheter ablation guided by electroanatomic mapping for recurrent ventricular

tachycardia after myocardial infarction: the multicenter thermocool ventricular

tachycardia ablation trial. Circulation 2008;118:2773-82.

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

control of unmappable ventricular tachycardia in patients with ischemic and

nonischemic cardiomyopathy. Circulation 2000;101:1288-96.

6. Dickfeld T, Kato R, Zviman M, Lai S, Meininger G, Lardo AC, Roguin A, Blumke

D, Berger R, Calkins H, Halperin H. Characterization of radiofrequency ablation

lesions with gadolinium-enhanced cardiovascular magnetic resonance imaging. J

Am Coll Cardiol 2006;47:370-8.

7. Kim RJ, Fieno DS, Parrish TB, Harris K, Chen EL, Simonetti O, Bundy J, Finn JP,

Klocke FJ, Judd RM. Relationship of MRI delayed contrast enhancement to

irreversible injury, infarct age, and contractile function. Circulation 1999;100:1992-

2002.

8. Faris OP, Shein M. Food and Drug Administration perspective: Magnetic resonance

imaging of pacemaker and implantable cardioverter-defibrillator patients.

Circulation 2006;114:1232-3.

9. Heiberg E, Ugander M, Engblom H, Gotberg M, Olivecrona GK, Erlinge D,

Arheden H. Automated quantification of myocardial infarction from MR images by

accounting for partial volume effects: animal, phantom, and human study.

Radiology 2008;246:581-8.

q y

neticc r resesononanancecececececece ii

ieno DS, Parrish T Harris Chen EL Simonetti O, Bun y

Judd RM. Relationship of MRI delayed contrast enhance e

ardiol 2006;47:370-8.

eno DS, Parrish TB, Harris K, Chen EL, Simonetti O, Bundy

Judd RM. Relationship of MRI delayed contrast enhanceme



23

10. Josephson, M. E. Clinical Cardiac Electrophysiology. 3rd ed. 2002. Philadelphia,

Lippincott Williams & Wilkins.

11. Bertaglia E, Brandolino G, Zoppo F, Zerbo F, Pascotto P. Integration of three-

dimensional left atrial magnetic resonance images into a real-time electroanatomic

mapping system: validation of a registration method. Pacing Clin.Electrophysiol.

2008;31:273-82.

12. Stevenson WG, Khan H, Sager P, Saxon LA, Middlekauff HR, Natterson PD,

Wiener I. Identification of reentry circuit sites during catheter mapping and

radiofrequency ablation of ventricular tachycardia late after myocardial infarction.

Circulation 1993;88:1647-70.

13. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK,

Pennell DJ, Rumberger JA, Ryan T, Verani MS. Standardized myocardial

segmentation and nomenclature for tomographic imaging of the heart: a statement

for healthcare professionals from the Cardiac Imaging Committee of the Council on

Clinical Cardiology of the American Heart Association. Circulation 2002;105:539-

42.

14. Gimbel JR, Kanal E, Schwartz KM, Wilkoff BL. Outcome of magnetic resonance

imaging (MRI) in selected patients with implantable cardioverter defibrillators

(ICDs). Pacing Clin.Electrophysiol. 2005;28:270-3.

pp ggggggg

afteer r mymymyyyococarardididididididiaaaalalaa

n

M y

, Rumberger JA, Ryan T, Verani MS. Standardized myocard

n 1993;88:1647-70.

MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey

, Rumberger JA, Ryan T, Verani MS. Standardized myocard



24

15. Naehle CP, Strach K, Thomas D, Meyer C, Linhart M, Bitaraf S, Litt H, Schwab

JO, Schild H, Sommer T. Magnetic resonance imaging at 1.5-T in patients with

implantable cardioverter-defibrillators. J.Am.Coll.Cardiol. 2009;54:549-55.

16. Nazarian S, Roguin A, Zviman MM, Lardo AC, Dickfeld TL, Calkins H, Weiss

RG, Berger RD, Bluemke DA, Halperin HR. Clinical utility and safety of a protocol

for noncardiac and cardiac magnetic resonance imaging of patients with permanent

pacemakers and implantable-cardioverter defibrillators at 1.5 tesla. Circulation

2006;114:1277-84.

17. Schueler BA, Parrish TB, Lin JC, Hammer BE, Pangrle BJ, Ritenour ER,

Kucharczyk J, Truwit CL. MRI compatibility and visibility assessment of

implantable medical devices. J.Magn Reson.Imaging 1999;9:596-603.

18. Bogun FM, Desjardins B, Good E, Gupta S, Crawford T, Oral H, Ebinger M, Pelosi

F, Chugh A, Jongnarangsin K, Morady F. Delayed-enhanced magnetic resonance

imaging in nonischemic cardiomyopathy: utility for identifying the ventricular

arrhythmia substrate. J.Am.Coll.Cardiol. 2009;53:1138-45.

19. Desjardins B, Crawford T, Good E, Oral H, Chugh A, Pelosi F, Morady F, Bogun

F. Infarct architecture and characteristics on delayed enhanced magnetic resonance

imaging and electroanatomic mapping in patients with postinfarction ventricular

arrhythmia. Heart Rhythm. 2009;6:644-51.

20. Dickfeld, T., Lei, P, Dilsizian, V, Jeudy, J, Dong, J., Voudouris, A, Peters, R, Saba,

M, Shekhar, R, and Shorofsky, S. Integration of Three-Dimensional Scar Maps for

e BJJJJJJJ, , ,, RiRiRiRiRiRiRitetettttt nonoururrrrurr EEEEEEE

k t

e

g

k J, Truwit CL. MRI compatibility and visibility assessment

e medical devices. J.Magn Reson.Imaging 1999;9:596-603.

, Desjardins B, Good E, Gupta S, Crawford T, Oral H, Ebing



https://www.researchgate.net/publication/24237023_Delayed-Enhanced_Magnetic_Resonance_Imaging_in_Nonischemic_Cardiomyopathy?el=1_x_8&enrichId=rgreq-4da6c628e7ed2673b9d954aaa6de88ae-XXX&enrichSource=Y292ZXJQYWdlOzQ5Nzg4MTIwO0FTOjI2NDY1NzEwOTc3ODQzMkAxNDQwMTEwNTgwNTkw



https://www.researchgate.net/publication/6823585_Clinical_Utility_and_Safety_of_a_Protocol_for_Noncardiac_and_Cardiac_Magnetic_Resonance_Imaging_of_Patients_With_Permanent_Pacemakers_and_Implantable-Cardioverter_Defibrillators_at_15_Tesla?el=1_x_8&enrichId=rgreq-4da6c628e7ed2673b9d954aaa6de88ae-XXX&enrichSource=Y292ZXJQYWdlOzQ5Nzg4MTIwO0FTOjI2NDY1NzEwOTc3ODQzMkAxNDQwMTEwNTgwNTkw





https://www.researchgate.net/publication/13065108_MRI_compatibility_and_visibility_assessment_of_implantable_medical_devices?el=1_x_8&enrichId=rgreq-4da6c628e7ed2673b9d954aaa6de88ae-XXX&enrichSource=Y292ZXJQYWdlOzQ5Nzg4MTIwO0FTOjI2NDY1NzEwOTc3ODQzMkAxNDQwMTEwNTgwNTkw



https://www.researchgate.net/publication/26705976_Magnetic_Resonance_Imaging_at_15-T_in_Patients_With_Implantable_Cardioverter-Defibrillators?el=1_x_8&enrichId=rgreq-4da6c628e7ed2673b9d954aaa6de88ae-XXX&enrichSource=Y292ZXJQYWdlOzQ5Nzg4MTIwO0FTOjI2NDY1NzEwOTc3ODQzMkAxNDQwMTEwNTgwNTkw



25

Ventricular Tachycardia Ablation With Positron Emission Tomography-Computed

Tomography. J Am Coll Cardiol Img 1, 73-82. 2008.

21. Codreanu A, Odille F, Aliot E, Marie PY, Magnin-Poull I, Andronache M, Mandry

D, Djaballah W, Regent D, Felblinger J, de CC. Electroanatomic characterization of

post-infarct scars comparison with 3-dimensional myocardial scar reconstruction

based on magnetic resonance imaging. J.Am.Coll.Cardiol. 2008;52:839-42.

22. Verma A, Marrouche NF, Schweikert RA, Saliba W, Wazni O, Cummings J,

Abdul-Karim A, Bhargava M, Burkhardt JD, Kilicaslan F, Martin DO, Natale A.

Relationship between successful ablation sites and the scar border zone defined by

substrate mapping for ventricular tachycardia post-myocardial infarction.

J.Cardiovasc.Electrophysiol. 2005;16:465-71.

23. Tian J, Smith MF, Jeudy J, Dickfeld T. Multimodality fusion imaging using

delayed-enhanced cardiac magnetic resonance imaging, computed tomography,

positron emission tomography, and real-time intracardiac echocardiography to

guide ventricular tachycardia ablation in implantable cardioverter-defibrillator

patients. Heart Rhythm. 2009;6:825-8.

24. Ashikaga H, Sasano T, Dong J, Zviman MM, Evers R, Hopenfeld B, Castro V,

Helm RH, Dickfeld T, Nazarian S, Donahue JK, Berger RD, Calkins H, Abraham

MR, Marban E, Lardo AC, McVeigh ER, Halperin HR. Magnetic resonance-based

anatomical analysis of scar-related ventricular tachycardia: implications for catheter

ablation. Circ.Res. 2007;101:939-47.

, ,,

scarr bbbb bbborordeddddedd r r zozoooooonnnnnnene

mapping for ventricular tachycardia post-myocardial infarctio

s

ith MF, Jeudy J, Dickfeld T. Multimodality fusion imaging u

mapping for ventricular tachycardia post-myocardial infarctio

sc.Electrophysiol. 2005;16:465-71.

ith MF, Jeudy J, Dickfeld T. Multimodality fusion imaging u



26

25. Ciaccio EJ, Ashikaga H, Kaba RA, Cervantes D, Hopenfeld B, Wit AL, Peters NS,

McVeigh ER, Garan H, Coromilas J. Model of reentrant ventricular tachycardia

based on infarct border zone geometry predicts reentrant circuit features as

determined by activation mapping. Heart Rhythm. 2007;4:1034-45.

Table 1. Baseline Characteristics of Patient Population Undergoing MR Imaging

Patient Characteristics Results

Age (years) 58±15

Sex (M/F) 21/1

Weight (Kg) 95.1± 26.8

Height (cm) 177± 9

Body surface area (m2) 2.1± 0.3

VT episodes/previous 30 days 17±12

Ischemic cardiomyopathy (%) 68

Non-ischemic cardiomyopathy (%) 32

Ejection fraction (%) 34±14

Hypertension (%) 50

Diabetes (%) 23

Hyperlipidemia (%) 59

Antiarrhythmic therapy (%) 100

Amiodarone (%) 68

Sotalol 28

Quinidine (%) 4

e

v

95.1± 26.8

177± 9

ea (m2) 2.1± 0.3

vious 30 days 17±12



https://www.researchgate.net/publication/6164476_Model_of_reentrant_ventricular_tachycardia_based_on_infarct_border_zone_geometry_predicts_reentrant_circuit_features_as_determined_by_activation_mapping?el=1_x_8&enrichId=rgreq-4da6c628e7ed2673b9d954aaa6de88ae-XXX&enrichSource=Y292ZXJQYWdlOzQ5Nzg4MTIwO0FTOjI2NDY1NzEwOTc3ODQzMkAxNDQwMTEwNTgwNTkw




27

Table 2. ICD parameters. ICD parameters measured before and after MRI as well as after

68±21 days of follow-up summarizing ventricular leads (n=22) and atrial leads (n=13)

measurements. To account for devices allowing only pulse width decrements thresholds

are also displayed as (V*ms). Significance evaluated with Wilcoxon Signed Ranks Test.

Parameters Pre-MRI Post-MRI Follow-Up p-value

(pre/post)

P-wave amplitude [mV] 3.3±2.1 3.0±1.4 3.1±1.6 p=0.35

R-wave amplitude [mV] 9.9±5.2 10.2±5.4 10.1±5.9 p=0.09

Atrial lead impedance [ ] 475±78 476±69 462±83 p=0.51

Ventricular lead impedance [ ] 461±97 451±89 450±95 p=0.82

Shock impedance [ ] 45±8 45±8 44±7 p=0.34

Atrial capture [V]

[V]*[ms]

1.5±0.9

0.41±0.23

1.4±0.9

0.41±0.22

1.5±1.0

0.41±0.28

p=0.32

p=0.66

Ventricular capture [V]

[V]*[ms]

1.6±0.7

0.43±0.18

1.6±0.7

0.44±0.16

1.5±0.7

0.44±0.15

p=0.32

p=0.79

Battery [V] 2.97±0.21 2.97±0.21 2.95±0.22 p=0.08

Figure Legends:

Figure 1. MRI and Extraction of Anatomy in ICD Patients. A) Inverse recovery serial

short-axis MR images (apical-to-base: 1-6). ICD results in high intensity artifact affecting

the anterior wall. ICD lead seen as low intensity signal in inferior right ventricle (RV).

Transmural late gadolinium enhancement (LGE) seen in inferior wall of left ventricle

(LV). B) Registration of Extracted 3D MRI. LV voltage map is aligned with MRI-

extracted endocardial shell (Endo; turquoise). Rotational errors are corrected by aligning

462±83 pppp=p=p=p=p=p=p=00

450±±±±±±±95959595959595 pppppp=p=p=p=p=p=p=0000000

0

00 44±0 150 44±0 160 43±0 18[V]*[ ]

n 0

0

0

ture [V]

[V]*[ ]

1.6±0.7

0 43±0 18

1.6±0.7

0 44±0 16

1.5±0.7

0 44±0 15

pp=0

0

nce [ ] 45±8 45±8 44±7 pp=0

[V]

[V]*[ms]

1.5±0.9

0.41±0.23

1.4±0.9

0.41±0.22

1.5±1.0

0.41±0.28

pp=0

pp=0



28

RV voltage map with reconstructed MRI RV slice (red). After integration of MRI-

extracted epicardial shell (Epi; blue mesh) embedded MRI-derived transmural scar

(brown) is well visible within myocardial wall.

Figure 2. Visualization of Anterior Scar, Artifact and Viable Myocardium A) MRI

demonstrates anterior artifact (grey arrows) concealing possible anterior scar, but allows

visualization of adjacent septal and lateral scar (brown arrows) and preserved inferior

viable myocardium (purple arrows). B) Corresponding 3D reconstruction demonstrates

approximated LV wall reconstruction from area of artifact (grey solid), septal and lateral

scar (brown mesh) and viable inferior myocardium (purple solid), which correlate with

<0.5mV and >1.5mV voltage recordings in registered voltage map.

Figure 3. Correlation of MRI Scar Transmurality and Bipolar Voltage. Increasing scar

transmurality demonstrates a good correlation with decreasing bipolar voltage (Spearman

Rank Correlation Analysis).

Figure 4. Complex 3D Scar Border Geometry. Clipping plane allows visualization of

endo-/epicardium (turquoise shell, black arrows). Note embedded basal transmural scar

(blue shell, white arrows) with decreasing epicardial extension towards the apex (yellow

arrows). Endocardial voltage map normalizes (red-to-purple color; red arrow) with

increasing alive endocardial component and does not detect mid-epicardial scar. Areas of

fractionated signals (pink tags) seen close to transition zone.

rey solid), sepepppppptatattttt

olid)d)))))), , , , ,,, whwhwhwhwhwhw icicicicicicich h h h h h h cococococococorrr

5

ation of MRI Scar Transmurality and Bipolar Voltage. Inc e

5mV voltage recordings in registered voltage map.

ation of MRI Scar Transmurality and Bipolar Voltage. Incre



29

Figure 5. MRI-Guided Identification of Ablation Site at Scar Border. “Early registration”

strategy guided pacemapping (orange-brown tags) along the border of the registered scar

mesh and allowed rapid identification of a fractionated, low-voltage site (yellow tag; A)

with 11/12 match (B). Note excellent correlation between scar mesh and abnormal

voltage area after “final registration” with complete LV map (blue/pink tags:

fractionated/diastolic potentials; A). Ablation at that site eliminated the clinical VT.

Endo/epicardium, intramyocardial scar and projected RF lesion of corresponding MRI

slice were extracted (C) and used for 80-sector analysis (D). Resulting diagram (E)

displays color-coded in analogous short-axis views the scar transmurality and identifies

RF location (arrow). Local scar characteristics at ablation sector (arrow) can be identified

in subselected regions-of-interest of LV wall (F).

Figure 6. MRI-Guided Pacemapping/Ablation. Despite significant metal artifact viable,

non-enhanced postero-medial papillary muscle (PM) with alive endo-, mid and epicardial

components embedded in inferior scar seen in MRI (A; black arrows; apical-to-basal scan

plane:1-4). Pacemapping (orange-brown tags) visually guided around 3D-

reconstructed/registered papillary muscle (turquoise) identifies fractionated, low-voltage

site at lateral PM insertion site (B; yellow tag; shown with “final registration”).

Pacemapping achieved 11/12 match (C) and ablation resulted in non-inducibility of VT.

Imaging planes (1-4) corresponding to Figure A.

Figure 7: MRI-Guided Detection and Ablation of Midmyocardial Scar. Normal

endocardial voltage map with lowest bipolar voltage in infero-lateral wall. MR imaging

ransmurality aaaaaaanndnnnnn

ctor (((((((arararararararrorororororor w)w)w)w)w)w)w) c c cccccanananaanaa

G i

gions-of-interest of LV wall (F).

Guided Pacemapping/Ablation. Despite significant metal arti



30

plane (white line, panel A) demonstrates corresponding infero-lateral midmyocardial scar

(B), which was 3D reconstructed displaying the accurate intramyocardial location (C).

Pacemapping at this location resulted in 11/12 pacemap match and ablation eliminated

VT. Eighty-sector analysis demonstrates maximum scar transmurality of 63% with

ablation sites located at scar border zone (E;F).

Figure 8. Ultrasound-guided Scar Registration and Ablation Confirmation.

Reconstructed inferior 3D scar (white arrows) is registered to CartoSOUND-constructed

RV/LV shell acquired from RV (“CartoSOUND registration”). After VT induction

entrainment mapping is limited to area of MRI reconstructed scar and identifies critical

channel basal inferior wall ((B; orange arrow) where ablation terminated the clinical VT.

Creation of successful ablation site could be confirmed by increased focal echodensity

(yellow arrow).

). After VT innnnnnnddddddd

scarrrrrrr a a aaaaandndndndndndnd i i i i ii idededededededentntntntntntntifi

f

e c

ferior wall ((B; orange arrow) where ablation terminated the

essful ablation site could be confirmed by increased focal ec



Supplemental Material



P. # Etiology

Voltage Scar Location MRI Scar Location

MRI scar area endocardial [cm2]

Voltage area <0.5mV [cm2]

Voltage area <1.5mV [cm2] Pre-RFA VT

CL [ms]

Clinical VT

Post-RFA Induction

1 ischemic inferior, septal inferior, septal 70.2 19.4 69.8 RBBB 312 yes* Non-clinical VT Indeterminate 308 no

2 ischemic inferior, infero-septal, infero-lateral

inferior, infero-septal, infero-lateral 65.4 24.9 57.7 RBBB 320 yes* Non-clinical VT

RBBB 230 no 3 ischemic inferior, infero-septal inferior, infero-septal 38 13.2 39.4 RBBB 365 yes* No VT

LBBB 299 no

4 non-ischemic inferior, lateral inferior, lateral 28.9 22.5 37.1 RBBB 448 yes* No VT

5 ischemic inferior, infero-apical inferior, infero-apical 38.2 15.4 34.1 RBBB 441 yes* No VT 6 ischemic inferior inferior 24.3 11.9 23.8 RBBB 521 yes* Non-clinical VT

RBBB 345 no Indeterminate 440 no LBBB 362 no

7 ischemic anterior, septal anterior (A), septal 31.6 12.4 33.2 RBBB 532 yes* No inducible VT

8 non-ischemic

none lateral (mid/epicardial only) 6.2 (mid) 0 0 RBBB 520 yes* Non-clinical VT

RBBB 242 yes*

9 non-ischemic

inferior, infero-septal inferior, septal (mostly mid) 54.6 18.7 48.4 RBBB 280 yes*

No VT

(+20.2 mid) LBBB 278 yes* 10 ischemic inferior, infero-septal inferior, infero-septal 35.5 16.6 36.7 RBBB 422 yes* Clinical VT

RBBB 450 yes* LBBB 324 no RBBB 367 no

11 ischemic inferior, infero-lateral inferior, infero-lateral 17 7.5 24.6 RBBB 530 yes* Non-clinical VT RBBB 525 no Indeterminate 521 no

12 ischemic inferior inferior 34.4 4.2 33.1 RBBB 430 yes* No VT RBBB 595 no*

13 ischemic inferior inferior 42.7 21.4 45.5 RBBB 444 yes* No VT RBBB 412 yes*

14 ischemic anterior, septal, lateral

anterior (A), septal, lateral 60.1 51.9 63.6 RBBB 342 yes* Non-clinical VT

RBBB 360 yes* RBBB 304 yes* Indeterminate 347 no* LBBB 237 no

Data Supplement 2: Image-Guided Ventricular Tachycardia Ablation. Dickfeld et al.

by guest on February 21, 2013circep.ahajournals.org

Dow

nloaded from


Supplement 2: Individual Procedural and Imaging Patient Data. To allow valid comparison between MRI scar and voltage scar, area of ICD artifact was excluded for scar area analysis. Relevant Abbreviations: P# - patient number; mid – midmyocardial; (A) – predominant artifact; CL – cycle length of ventricular tachycardia; * - Ventricular tachycardia targeted during ablation (RFA)


Dow

nloaded from


Data Supplement 1: Image-Guided Ventricular Tachycardia Ablation. Dickfeld et al.

Lead Data # Generator Ventricular Lead Atrial Lead

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Medtronic Maximo VR 7232 Medtronic Maximo VR 7232 Guidant Livian H227 RF Guidant Ventak Prism II 1861 St Jude Current VR RF 1207-36 Medtronic Virtuoso VR D154VWC Medtronic Maximo VR 7232 Medtronic EnTrust D154VRC Medtronic Maximo VR 7232 Medtronic EnTrust D154ATG Medtronic Insync Maximo 7303 Medtronic Concerto C154DWK Guidant Vitality A155 Medtronic Intrinsic 7288 St Jude Atlas II VR V168 Medtronic EnTrust D154ATG Medtronic Secura DR D224DRG St Jude Current RV RF 1207-36 Medtronic EnTrust D154VRC Guidant Vitality A155 Medtronic Marquis DR7274 Medtronic Secura DR D224DRG

Medtronic 6949 Sprint Fidelis Medtronic 6947 Sprint Quattro Medtronic 6945 Sprint Guidant 0148 Reliance St Jude 7001 Riata ST Medtronic 6949 Sprint Fidelis Medtronic 6947 Sprint Quattro Medtronic 6949 Sprint Fidelis Medtronic 6947 Sprint Quattro Medtronic 6947 Sprint Quattro Medtronic 6943 Sprint Medtronic 4193 Attain OTW (LV) Medtronic 6947 Sprint Quattro Medtronic 4194 AttainOTW (LV) Medtronic 6945 Spring Medtronic 6947 Sprint Quattro St Jude 7000 Riata ST Medtronic 6943 Sprint Medtronic 6947 Sprint Quattro St Jude 7120 Durata Medtronic 6949 Sprint Fidelis Guidant 0184 Reliance Medtronic 6947 Sprint Secure Medtronic 6945 Sprint

___ ___ Medtronic 5076 CapsureFix Novus Guidant 4087 Flextend ____ ____ ____ ____ Medtronic 5592 CapSure SP Medtronic 5076 CapSure Fix Novus St Jude 1688T Tendril SDX Medtronic 5076 CapsureFix Novus St Jude 1388T Tendril DX ____ Medtronic 5076 CapsureFix Novus Medtronic 5076 CapsureFix Novus ____ ____ Guidant 4087 Flextend Medtronic 4568 CapsureFix Medtronic 5076 Capsure Fix Novus

Supplement 1. ICD system specifications for individual patients. Ventricular lead denotes right ventricular ICD leads. Coronary sinus leads identified with (LV).


Dow

nloaded from


MRI-Guided ventricular tachycardia ablation: integration of late gadolinium-enhanced 3D scar in...

Documents

Transcript of MRI-Guided ventricular tachycardia ablation: integration of late gadolinium-enhanced 3D scar in...