- Email: [email protected]
Real-time integration of 2D intracardiac echocardiography and 3D electroanatomical mapping to guide ventricular tachycardia ablation
Real-time integration of 2D intracardiac echocardiography and 3D electroanatomical mapping to guide ventricular tachycardia ablation Yaariv Khaykin, MD* Allan Skanes, MD,† Bonnie Whaley, BSc,* Carol Hill, RN,* Marianne Beardsall, RN, MSN,* Catherine Seabrook, RN,* Zaev Wulffhart, MD,* Richard Oosthuizen, BSc,* Lorne Gula, MD,† Atul Verma, MD* From the *Southlake Regional Health Center, Newmarket, Ontario, Canada, and the †London Health Science Center, London, Ontario, Canada. BACKGROUND: Ablation of left ventricular tachycardia (LV VT) involves point-by-point reconstruction of the three-dimensional (3D) virtual anatomy. It is time consuming and requires substantial fluoroscopy exposure. Two-dimensional (2D) intracardiac echocardiography (ICE) affords real-time imaging of the cardiac structures. OBJECTIVE: This study sought to evaluate a mapping system integrating ICE with 3D mapping to guide VT ablation.
contours. Regional WMA corresponded to low bipolar voltage (⬍0.5 mV). Procedure time was 240 ⫾ 77 min, with fluoroscopy time of 25 ⫾ 12 min. LV volume by ICE was 172 ⫾ 119 cm3 versus 164 ⫾ 112 cm3 for the point-by-point maps (P ⫽ .5). Scar area by ICE was 33 ⫾ 32 cm2 versus 36 ⫾ 33 cm2 for voltage mapping (P ⫽ .4). At 5 ⫾ 4 months, 12 patients (71%) were free of VT.
METHODS: Seventeen patients (16 men, 62 ⫾ 11 years, LV ejection fraction 40% ⫾ 15%) had ablation of nonidiopathic VT guided using a system integrating 3D mapping and ICE. ICE probe with a location sensor tracked by the mapping system was positioned in the right heart. Endocardial contours traced on gated images of the LV were used to generate a registered 3D map. Regional wall motion abnormalities (WMA) were tagged.
CONCLUSION: A system combining 2D ICE and 3D mapping can reconstruct a 3D shell of the LV, including a substrate map based on regional WMA without the need to enter the LV. VT ablation guided using this approach is safe and effective.
RESULTS: 3D maps were created in 26 ⫾ 8 min, before entering the LV and without fluoroscopy. Maps were built from 23 ⫾ 7
(Heart Rhythm 2008;5:1396 –1402) © 2008 Heart Rhythm Society. All rights reserved.
Introduction
medical therapy in patients who have received implantable cardioverter-defibrillator therapies. Results from several nonrandomized registries suggesting initial ablation success in 70% to 80% of the patients with 20% to 25% long-term ventricular tachycardia (VT) recurrence rates.6 – 8 Ablation is typically guided by three-dimensional (3D) mapping of the left ventricle (LV) to target areas of early electrical activation, low voltage, or sites within the reentrant circuit as defined by entrainment, depending on the hemodynamic consequences of the VT. Ablation of VT guided by two-dimensional (2D) intracardiac echocardiography (ICE) and electroanatomical mapping has been previously reported without integrating the 2 techniques.9 Although the ICE-guided technique affords real-time imaging of the cardiac structures and monitoring for complications, this technique does not allow registration of the ablation lesions or other sites of interest during the study. 3D electroanatomical mapping, on the other hand, allows recording of special site positions in space but relies on point-by-point reconstruction of 3D virtual anatomy or registration of preacquired computed tomography (CT) or
Use of implantable defibrillators to prevent sudden cardiac death related to ventricular tachyarrhythmia in select patient populations1–3 has become the standard of care. Many patients are psychologically traumatized by receiving appropriate life-saving shocks.4 Antiarrhythmic medications, notably amiodarone, have been shown to reduce the likelihood of a patient receiving a shock; unfortunately these are not always effective or tolerated.5 Catheter ablation for ventricular tachycardia is a promising alternative or adjunct to
Dr. Khaykin has received speaker’s honoraria (Biosense Webster, St Jude Medical, Medtronic) and is a member of the Physician Advisory Board (Biosense Webster). Dr. Skanes has received speaker’s honoraria and is a member of the Physician Advisory Board (Biosense Webster). Dr. Verma has received speaker’s honoraria (Biosense Webster, St Jude Medical, Medtronic) and is a member of Physician Advisory Boards (Biosense Webster, St Jude Medical). Address reprint requests and correspondence: Dr. Yaariv Khaykin, Heart Rhythm Program, Division of Cardiology, Southlake Regional Health Center, 105-712 Davis Drive, Newmarket, Ontario, Canada, L3Y 8C3. E-mail address: [email protected]. (Received May 18, 2008; accepted June 25, 2008.)
KEYWORDS Ventricular tachycardia; Catheter ablation; Intracardiac echocardiography; Electroanatomical mapping
1547-5271/$ -see front matter © 2008 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2008.06.025
Khaykin et al
ICE and 3D Mapping to Guide VT Ablation
magnetic resonance imaging (MRI) studies. Unfortunately, registration of preacquired CT and MRI images is difficult and may be imprecise.10,11 In addition, currently available software allows for registration of the LV anatomy but not substrate. Intracardiac echocardiography provides both an anatomical and a functional assessment of the LV, allowing for real-time identification of wall motion abnormalities. Superior accuracy of the approach integrating 2D ICE with electroanatomical mapping has been recently shown.11 We report the first human experience with integration between ICE and electroanatomical mapping using a novel mapping system in a series of 17 patients referred for catheter ablation of recurrent symptomatic nonidiopathic VT refractory to amiodarone.
Methods Between February 2007 and April 2008, 17 patients presenting with nonidiopathic VT refractory to amiodarone were offered ablation guided using a system integrating 3D mapping and ICE.
ICE-generated LV geometry A 3D map of the LV was constructed using novel CARTOSOUND technology (Biosense Webster, Diamond Bar, California). An ICE probe with a CARTO navigation sensor imbedded close to the phased array (SoundStar) was positioned in the right heart and allowed sequential acquisition of electrocardiogram-gated 2D images of the LV from base to apex. Three-second video clips were recorded and imported into the CARTO system. Image acquisition and substrate mapping took place during right ventricular apical stimulation in 13 patients and during sinus rhythm in 4 patients. Images were acquired in end expiration and gated to the R-wave or the pacing spike. Gated 2D snapshots were then selected from the video clips. Endocardial surface of the LV was traced on each image manually by the operator or defined using the edge detection tool provided with the
1397 software (Figure 1). Each contour was displayed by the system in 3D space as a series of points. A family of contours could then be used to reconstruct the chamber of interest, thereby generating a registered 3D shell of the LV. Contours acquired for myocardial segments that seemed to be thinned, hypokinetic, akinetic, or dyskinetic, suggestive of scarred or ischemic myocardium, were tagged red. LV outflow tract, the aortic root, and the mitral valve were mapped with the probe positioned in the right atrium (Figures 2A and 2B), whereas the body of the LV was mapped with the probe positioned in the right ventricle, against the interventricular septum (Figures 2C and 2D). The latter position afforded a longitudinal view of the LV cavity. The probe was retroflexed and rotated toward the tricuspid valve before entering the right ventricle to improve image quality. Lateral tilt was applied to allow base-to-apex scanning through the body of the LV with deeper insertion, withdrawal, or rotation of the probe used as necessary to complete the map. In 4 patients, short-axis cross-sectional views of the LV were used during mapping with the ICE probe positioned in the right ventricular outflow tract.
Substrate mapping Substrate mapping of the LV and catheter ablation was carried out using a 3.5-mm irrigated-tip catheter (Navistar Thermocool, Biosense Webster) using a retrograde aortic approach in 13 patients and a transseptal approach in 6 patients. Epicardial access was required in 1 of the 2 patients with dilated cardiomyopathy. The patients were systemically anticoagulated once the left heart was instrumented with a target activated clotting time (ACT) of 350 to 400 seconds. Standard thresholds were used to define the scar border zone12 with scar defined as local bipolar voltage ⬍0.5 mV. Late, fractionated, and double potentials were tagged on the substrate map. Once the substrate map was complete, programmed electrical stimulation from the right ventricular apex at 2 drive-cycle lengths with up to 3
Figure 1 Acquisition of real-time registered intracardiac echocardiography (ICE) contours and 3-dimensional (3D) rendering. This figure illustrates a 2-dimensional intracardiac ultrasound image (left) registered as part of the CARTO shell of the left ventricle (middle and right). Tip of the ICE catheter positioned in the right ventricle is seen projecting an ultrasound “fan” through the body of the left ventricle. The middle panel shows a family of ICE-defined contours that are summed to create a 3D anatomical shell of the left ventricle (right). Hypokinetic segments identified on the high septum are tagged red (middle).
1398
Heart Rhythm, Vol 5, No 10, October 2008
Figure 2 Intracardiac echocardiography (ICE) position during mapping. A and B: ICE probe was placed in the right atrium to image the left ventricular outflow tract and the aortic root as well as the mitral annulus; A represents the LAO view, B represents the RAO view. C and D: ICE probe was positioned against the septum in the right ventricular inflow tract to image the left ventricle in the long axis view; C represents the LAO view, D represents the RAO view. LAO ⫽ left anterior oblique; RAO ⫽ right anterior oblique.
premature extra stimuli or burst pacing was used to induce ventricular tachycardia. If the tachycardia was hemodynamically stable, limited activation mapping was used to identify the area of earliest activation. Pacemapping close to the tachycardia cycle length in the scar border zone was carried out in patients with hemodynamically unstable VT to achieve the closest 12-lead electrocardiographic match of the QRS complex to that of the clinical tachycardia. Ablation was then carried out in the scar border zone regions containing late, fractionated, and double potentials that also corresponded to the area of earliest activation during hemodynamically stable VT or had the best pace map match to the clinical unstable VT. Entrainment mapping13 was used to confirm the VT circuit in selects patients with stable VT. Radiofrequency (RF) power was titrated between 30 W and 50 W, and the temperature cut-off was set at 40°C with heparinized saline infused through the catheter at 30 ml/min during ablation and 2 ml/min during mapping. Each RF application was at least 1 min in duration. Once the tachycardia stopped, programmed electrical stimulation in the RV was resumed to ascertain whether the tachycardia was still inducible. The procedure was repeated until VT was no longer inducible.
Results Baseline characteristics of the study cohort are described in Table 1. 3D maps took 26 ⫾ 8 min to create, before entering Table 1
Baseline patient characteristics
Male gender Ejection fraction New York Heart Association class Prior revascularization Percutaneous coronary intervention Coronary artery bypass graft surgery ICD Shocks ATP CRT device Medications Beta-blocker Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker Loop diuretic Spironolactone Digoxin Prior ablation
16 (94%) 40% ⫾ 15% 2⫾1 9 (53%) 4 (24%) 6 (35%) 10 (59%) 6 (35%) 7 (41%) 2 (12%) 16 (94%) 16 (94%) 8 2 5 1
(47%) (12%) (29%) (6%)
ICD ⫽ implantable cardioverter-defibrillator; ATP ⫽ anti-tachycardia pacing; CRT ⫽ cardiac resynchronization therapy.
Khaykin et al
ICE and 3D Mapping to Guide VT Ablation
Figure 3 Correlation between the left ventricular volume estimated using point-by-point CARTO mapping and Cartosound 3-dimensional intracardiac echocardiography rendering.
the LV and without fluoroscopy. Important anatomical structures such as papillary muscles, aneurysms, and trabeculations were rendered in 3D. A complete map of the LV was built from 23 ⫾ 7 contours. Regional wall motion abnormalities were identified and tagged in all patients. These corresponded to areas of low bipolar voltage (⬍0.5 mV) during substrate mapping in each case. On average, 2 ⫾ 1 VT morphologies were mapped and ablated per patient. Average volume of the LV estimated using the ICE generated map was 172 ⫾ 119 cm3 compared with 164 ⫾ 112 cm3 for the point-by-point maps (P ⫽ .5, Figure 3). Figure 4 illustrates one of the patients with an ischemic VT substrate. A small dyskinetic area identified by ICE corresponded to the anteroseptal scar mapped using bipolar voltage criteria. Ablation at the septal border of the scar within an ideal pace map area led to initiation, acceleration, and termination of the clinical VT and rendered it noninducible in the basal state. VT was still inducible on isoproterenol. An activation map in VT identified a very early diastolic potential on the anterior aspect of the scar (on the other side of the trabeculation identified by ICE). Ablation here
1399 resulted in termination of VT and rendered it noninducible. Figure 5 illustrates a patient with a dilated cardiomyopathy and VT refractory to 2 prior ablations. A large hypokinetic area was identified on the septum, corresponding to the scar identified during point-by-point mapping. Clinical ventricular tachycardia was induced and mapped to the basal septal area of fractionated potentials. Ablation in this area rendered VT noninducible. Average scar area defined by ICE criteria was 33 ⫾ 32 cm2 compared with 36 ⫾ 33 cm2 for point-by-point substrate mapping (P ⫽ .4, Figure 6). Five patients had predominantly anteroapical scar, 5 had scar confined to the septum, 5 had inferior scar, and 2 had scar involving primarily the lateral wall. Close correlation of the scar area identified by ICE with that confirmed during point-by-point substrate mapping is illustrated in Figure 7. Catheter tip position was monitored by ICE in each case using feedback from the navigation sensor, which allowed precise navigation to such important anatomical areas as trabeculations and crypts found critical to clinical success in several patients (Figure 8). No patient suffered a major procedure-related complication. Procedural characteristics and outcomes are illustrated in Table 2. All procedures were successfully performed within 2 to 5 hours with structural mapping using ICE performed in under 30 minutes. Less than 30 minutes of fluoroscopy was used in all but 3 patients. There were no intraprocedural complications. At 5 ⫾ 4 months of follow-up, 12 patients (71%) remain free of recurrent VT. One of the patients with an implantable cardioverter-defibrillator received a shock during follow-up. Two of the patients who had a VT recurrence received anti-tachycardia pacing (ATP), successfully terminating the arrhythmia. One of the patients died during follow-up of progressive congestive heart failure.
Discussion Integration of ICE and 3D electroanatomical mapping was used to guide ablation of ventricular tachycardia refractory
Figure 4 Intracardiac echocardiography (ICE) scan through the successful VT ablation sites. This figure illustrates a 2-dimensional intracardiac ultrasound image (left) registered as part of the CARTO shell of the left ventricle (middle). Tip of the ICE catheter positioned in the right ventricle is seen projecting an ultrasound “fan” through the body of the left ventricle. Arrows 1 and 2 on the ultrasound image correspond to arrows on the CARTO image and point to the 2 successful ablation sites on either side of a trabeculation. The cursor on the CARTO map is placed at the site of earliest activation in VT with local diastolic potential preceding onset of the QRS by 109 msec.
1400
Heart Rhythm, Vol 5, No 10, October 2008
Figure 5 Correlation of septal scar identified during intracardiac echocardiography (ICE) and point-by-point mapping. This figure illustrates a 2-dimensional intracardiac ultrasound image (left) registered as part of the CARTO shell of the left ventricle (middle). Tip of the ICE catheter positioned in the right ventricle is seen projecting an ultrasound “fan” through the body of the left ventricle. Red contour on the ICE image delineates an akinetic segment. Corresponding anatomical map of the left ventricle demonstrates position of the akinetic segment on the mid-septum. The bipolar voltage substrate correlates well with scar identified by ICE in this patient. The cursor on the CARTO map is placed at the site of a late/fractionated potential (white arrow). The local electrogram is seen in the right panel.
to amiodarone in 17 patients with both ischemic and nonischemic cardiomyopathic substrates. This approach allowed rapid acquisition of real-time cross-sectional images of the LV without having to enter this chamber. The images were then integrated into a registered 3D model of the LV without the need for preprocedural imaging. ICE not only facilitated acquisition of the anatomical information but also helped map the substrate by identifying wall motion abnormalities later found to correspond to scar as defined by local bipolar voltage. Endocardial detail including papillary mus-
Figure 6 Correlation between the scar area estimated using point-bypoint CARTO mapping and Cartosound 3D intracardiac echocardiography (ICE) rendering.
cles and trabeculations could be readily identified. Volume of the LV estimated using the point-by-point map corresponded exactly to the volume of the ICE reconstruction. Moreover, the area of scar identified using ICE corresponded exactly to that defined using standard bipolar voltage settings during point-by-point mapping. This may help further limit catheter manipulation in the LV in the future by focusing effort on the potential substrate regions identified by ICE. Real-time monitoring of the endocardial– catheter interface helped direct the catheter to the important anatomical structures found to be critical to the clinical arrhythmia in several patients, such as crypts and trabeculations. Ablation guided using this technique was successful in preventing ventricular tachycardia recurrences in 71% of the patients, potentially preventing implantable cardioverter-defibrillator therapies and hospital admissions. Although point-by-point mapping alone has been the standard for VT ablation, this procedure is time consuming and may miss important anatomical structures such as trabeculations, crypts, and aneurysms critical to the arrhythmia circuit in some patients. Integration of preacquired images may help solve this shortcoming of the point-by-point technique. Use of the preprocedural CT scans for image inte-
Khaykin et al
ICE and 3D Mapping to Guide VT Ablation
1401
Figure 7 Correlation of the inferior scar area identified during intracardiac echocardiography (ICE) and point-by-point mapping. This figure illustrates an anatomical shell of the left ventricle built from the ICE contours with hypokinetic or akinetic segments labeled red. A corresponding bipolar voltage map built using the ICE shell is seen in the middle panel. The low bipolar voltage area corresponds to the region defined by the red ICE contours. The cursor is placed at the site of a late potential during RV apical pacing, corresponding to a successful ablation site. The local electrogram is seen in the right panel.
gration may result in substantial radiation exposure for the patient, and some facilities may lack appropriate equipment and expertise to generate high-spatial-resolution cardiac MRI images. In a recent study in which a combination of CT and positron emission tomography (PET) imaging was merged with the 3D map of the LV, LV substrate data were successfully registered to guide ablation.14 In this study, scar was identified as a “hole in the wall” because of current software limitations. Although scar registration was very precise in patients with ⬍15% scar burden, registration in patients with more extensive scar was less accurate. Scar as evidenced by wall motion abnormalities was well defined by ICE in our study in all patients, corresponding precisely to the scar defined using standard criteria during point-bypoint mapping. While preacquired images may not accu-
rately represent real-time cardiac anatomy, ICE has been shown effective in guiding a variety of electrophysiology procedures by providing real-time information necessary to improve the accuracy of ablation, to monitor and prevent complications. This is a small study providing a very preliminary assessment of this new technology. Some operators proficient in VT ablation but unfamiliar with ICE may have a rather steep learning curve with this approach. Although identification of the regional wall motion abnormalities using ICE accurately defines the general area of low endocardial voltage, high-density point-by-point mapping is still required to precisely define the border zone and identify areas of late, fractionated, and double potentials. Automation of ICE image acquisition and wall motion analysis in the future may
Figure 8 Monitoring position of the catheter tip. This figure illustrates a real-time two-dimensional (2D) intracardiac echocardiography (ICE) image of a cross section of the left ventricle taken through the tip of the ablation catheter (left), while delivering radiofrequency energy at the successful ablation site. A 3-dimensional reconstruction of the left ventricle is seen on the right. The exact location of the ablation catheter tip based on the CARTO navigation sensor is overlaid on the 2D ultrasound image (green contour).
1402 Table 2
Heart Rhythm, Vol 5, No 10, October 2008 Procedure characteristics and outcomes
Procedure time, min Fluoroscopy time, min Radiofrequency delivery time, min Number of contours Follow-up duration Recurrence ATP Shock Death
240 ⫾ 77 min 25 ⫾ 12 min 28 ⫾ 13 min 23 ⫾ 7 contours 5 ⫾ 4 months 5 (29%) 2 (12%) 1 (6%) 1 (6%)
ATP ⫽ Anti-tachycardia pacing.
help address this issue. Further studies assessing utility of real-time registered integration of ICE and 3D electroanatomical mapping are necessary to define its place in the armamentarium of the invasive cardiac electrophysiologists.
Conclusion A mapping system combining ICE and 3D electroanatomical mapping can feasibly reconstruct a 3D shell of the LV, including a limited substrate map based on regional wall motion abnormalities without the need to enter the left heart chambers. Ablation guided using this approach seems safe and effective in a small group of patients with a wide variety of substrates.
Acknowledgement Dr. Offer Klemm, Biosense Webster Israel, helped with the early implementation of this technology in our laboratory.
References 1. A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. The Antiarrhyth-
2. 3.
4. 5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
mics versus Implantable Defibrillators (AVID) Investigators. N Engl J Med 1997;337:1576 –1583. Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverterdefibrillator for congestive heart failure. N Engl J Med 2005;352:225–237. Moss AJ, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877– 883. Irvine J, Dorian P, Baker B, et al. Quality of life in the Canadian Implantable Defibrillator Study (CIDS). Am Heart J 2002;144:282–289. Connolly SJ, Dorian P, Roberts RS, et al. Comparison of beta-blockers, amiodarone plus beta-blockers, or sotalol for prevention of shocks from implantable cardioverter defibrillators: the OPTIC Study: a randomized trial. JAMA 2006;295:165–171. Della Bella P, De Ponti R, Uriarte JA, et al. Catheter ablation and antiarrhythmic drugs for haemodynamically tolerated post-infarction ventricular tachycardia; long-term outcome in relation to acute electrophysiological findings. Eur Heart J 2002;23:414 – 424. Borger van der Burg AE, de Groot NM, van Erven L, et al. Long-term follow-up after radiofrequency catheter ablation of ventricular tachycardia: a successful approach? J Cardiovasc Electrophysiol 2002;13:417– 423. Calkins H, Epstein A, Packer D, et al. Catheter ablation of ventricular tachycardia in patients with structural heart disease using cooled radiofrequency energy: results of a prospective multicenter study. Cooled RF Multi Center Investigators Group. J Am Coll Cardiol 2000;35:1905–1914. Callans DJ, Ren JF, Michele J, et al. Electroanatomic left ventricular mapping in the porcine model of healed anterior myocardial infarction. Correlation with intracardiac echocardiography and pathological analysis. Circulation 1999;100: 1744 –1750. Fahmy TS, Mlcochova H, Wazni OM, et al. Intracardiac echo-guided image integration: optimizing strategies for registration. J Cardiovasc Electrophysiol 2007;18:276 –282. Okumura Y, Henz BD, Johnson SB, et al. Three-Dimensional Ultrasound for Image-Guided Mapping and Intervention: Methods, Quantitative Validation and Clinical Feasibility of a Novel Multi-Modality Image Mapping System. Circulation Arrhythmia Electrophysiol 2008;1:110 –119. Marchlinski FE, Callans DJ, Gottlieb CD, et al. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation 2000;101:1288 –1296. Stevenson WG, Khan H, Sager P, et al. Identification of reentry circuit sites during catheter mapping and radiofrequency ablation of ventricular tachycardia late after myocardial infarction. Circulation 1993;88:1647–1670. Dickfeld T, Lei P, Dilsizian V, et al. Integration of Three-Dimensional Scar Maps for Ventricular Tachycardia Ablation With Positron Emission Tomography-Computed Tomography. J Am Coll Cardiol Img 2008;1:73– 82.
contours. Regional WMA corresponded to low bipolar voltage (⬍0.5 mV). Procedure time was 240 ⫾ 77 min, with fluoroscopy time of 25 ⫾ 12 min. LV volume by ICE was 172 ⫾ 119 cm3 versus 164 ⫾ 112 cm3 for the point-by-point maps (P ⫽ .5). Scar area by ICE was 33 ⫾ 32 cm2 versus 36 ⫾ 33 cm2 for voltage mapping (P ⫽ .4). At 5 ⫾ 4 months, 12 patients (71%) were free of VT.
METHODS: Seventeen patients (16 men, 62 ⫾ 11 years, LV ejection fraction 40% ⫾ 15%) had ablation of nonidiopathic VT guided using a system integrating 3D mapping and ICE. ICE probe with a location sensor tracked by the mapping system was positioned in the right heart. Endocardial contours traced on gated images of the LV were used to generate a registered 3D map. Regional wall motion abnormalities (WMA) were tagged.
CONCLUSION: A system combining 2D ICE and 3D mapping can reconstruct a 3D shell of the LV, including a substrate map based on regional WMA without the need to enter the LV. VT ablation guided using this approach is safe and effective.
RESULTS: 3D maps were created in 26 ⫾ 8 min, before entering the LV and without fluoroscopy. Maps were built from 23 ⫾ 7
(Heart Rhythm 2008;5:1396 –1402) © 2008 Heart Rhythm Society. All rights reserved.
Introduction
medical therapy in patients who have received implantable cardioverter-defibrillator therapies. Results from several nonrandomized registries suggesting initial ablation success in 70% to 80% of the patients with 20% to 25% long-term ventricular tachycardia (VT) recurrence rates.6 – 8 Ablation is typically guided by three-dimensional (3D) mapping of the left ventricle (LV) to target areas of early electrical activation, low voltage, or sites within the reentrant circuit as defined by entrainment, depending on the hemodynamic consequences of the VT. Ablation of VT guided by two-dimensional (2D) intracardiac echocardiography (ICE) and electroanatomical mapping has been previously reported without integrating the 2 techniques.9 Although the ICE-guided technique affords real-time imaging of the cardiac structures and monitoring for complications, this technique does not allow registration of the ablation lesions or other sites of interest during the study. 3D electroanatomical mapping, on the other hand, allows recording of special site positions in space but relies on point-by-point reconstruction of 3D virtual anatomy or registration of preacquired computed tomography (CT) or
Use of implantable defibrillators to prevent sudden cardiac death related to ventricular tachyarrhythmia in select patient populations1–3 has become the standard of care. Many patients are psychologically traumatized by receiving appropriate life-saving shocks.4 Antiarrhythmic medications, notably amiodarone, have been shown to reduce the likelihood of a patient receiving a shock; unfortunately these are not always effective or tolerated.5 Catheter ablation for ventricular tachycardia is a promising alternative or adjunct to
Dr. Khaykin has received speaker’s honoraria (Biosense Webster, St Jude Medical, Medtronic) and is a member of the Physician Advisory Board (Biosense Webster). Dr. Skanes has received speaker’s honoraria and is a member of the Physician Advisory Board (Biosense Webster). Dr. Verma has received speaker’s honoraria (Biosense Webster, St Jude Medical, Medtronic) and is a member of Physician Advisory Boards (Biosense Webster, St Jude Medical). Address reprint requests and correspondence: Dr. Yaariv Khaykin, Heart Rhythm Program, Division of Cardiology, Southlake Regional Health Center, 105-712 Davis Drive, Newmarket, Ontario, Canada, L3Y 8C3. E-mail address: [email protected]. (Received May 18, 2008; accepted June 25, 2008.)
KEYWORDS Ventricular tachycardia; Catheter ablation; Intracardiac echocardiography; Electroanatomical mapping
1547-5271/$ -see front matter © 2008 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2008.06.025
Khaykin et al
ICE and 3D Mapping to Guide VT Ablation
magnetic resonance imaging (MRI) studies. Unfortunately, registration of preacquired CT and MRI images is difficult and may be imprecise.10,11 In addition, currently available software allows for registration of the LV anatomy but not substrate. Intracardiac echocardiography provides both an anatomical and a functional assessment of the LV, allowing for real-time identification of wall motion abnormalities. Superior accuracy of the approach integrating 2D ICE with electroanatomical mapping has been recently shown.11 We report the first human experience with integration between ICE and electroanatomical mapping using a novel mapping system in a series of 17 patients referred for catheter ablation of recurrent symptomatic nonidiopathic VT refractory to amiodarone.
Methods Between February 2007 and April 2008, 17 patients presenting with nonidiopathic VT refractory to amiodarone were offered ablation guided using a system integrating 3D mapping and ICE.
ICE-generated LV geometry A 3D map of the LV was constructed using novel CARTOSOUND technology (Biosense Webster, Diamond Bar, California). An ICE probe with a CARTO navigation sensor imbedded close to the phased array (SoundStar) was positioned in the right heart and allowed sequential acquisition of electrocardiogram-gated 2D images of the LV from base to apex. Three-second video clips were recorded and imported into the CARTO system. Image acquisition and substrate mapping took place during right ventricular apical stimulation in 13 patients and during sinus rhythm in 4 patients. Images were acquired in end expiration and gated to the R-wave or the pacing spike. Gated 2D snapshots were then selected from the video clips. Endocardial surface of the LV was traced on each image manually by the operator or defined using the edge detection tool provided with the
1397 software (Figure 1). Each contour was displayed by the system in 3D space as a series of points. A family of contours could then be used to reconstruct the chamber of interest, thereby generating a registered 3D shell of the LV. Contours acquired for myocardial segments that seemed to be thinned, hypokinetic, akinetic, or dyskinetic, suggestive of scarred or ischemic myocardium, were tagged red. LV outflow tract, the aortic root, and the mitral valve were mapped with the probe positioned in the right atrium (Figures 2A and 2B), whereas the body of the LV was mapped with the probe positioned in the right ventricle, against the interventricular septum (Figures 2C and 2D). The latter position afforded a longitudinal view of the LV cavity. The probe was retroflexed and rotated toward the tricuspid valve before entering the right ventricle to improve image quality. Lateral tilt was applied to allow base-to-apex scanning through the body of the LV with deeper insertion, withdrawal, or rotation of the probe used as necessary to complete the map. In 4 patients, short-axis cross-sectional views of the LV were used during mapping with the ICE probe positioned in the right ventricular outflow tract.
Substrate mapping Substrate mapping of the LV and catheter ablation was carried out using a 3.5-mm irrigated-tip catheter (Navistar Thermocool, Biosense Webster) using a retrograde aortic approach in 13 patients and a transseptal approach in 6 patients. Epicardial access was required in 1 of the 2 patients with dilated cardiomyopathy. The patients were systemically anticoagulated once the left heart was instrumented with a target activated clotting time (ACT) of 350 to 400 seconds. Standard thresholds were used to define the scar border zone12 with scar defined as local bipolar voltage ⬍0.5 mV. Late, fractionated, and double potentials were tagged on the substrate map. Once the substrate map was complete, programmed electrical stimulation from the right ventricular apex at 2 drive-cycle lengths with up to 3
Figure 1 Acquisition of real-time registered intracardiac echocardiography (ICE) contours and 3-dimensional (3D) rendering. This figure illustrates a 2-dimensional intracardiac ultrasound image (left) registered as part of the CARTO shell of the left ventricle (middle and right). Tip of the ICE catheter positioned in the right ventricle is seen projecting an ultrasound “fan” through the body of the left ventricle. The middle panel shows a family of ICE-defined contours that are summed to create a 3D anatomical shell of the left ventricle (right). Hypokinetic segments identified on the high septum are tagged red (middle).
1398
Heart Rhythm, Vol 5, No 10, October 2008
Figure 2 Intracardiac echocardiography (ICE) position during mapping. A and B: ICE probe was placed in the right atrium to image the left ventricular outflow tract and the aortic root as well as the mitral annulus; A represents the LAO view, B represents the RAO view. C and D: ICE probe was positioned against the septum in the right ventricular inflow tract to image the left ventricle in the long axis view; C represents the LAO view, D represents the RAO view. LAO ⫽ left anterior oblique; RAO ⫽ right anterior oblique.
premature extra stimuli or burst pacing was used to induce ventricular tachycardia. If the tachycardia was hemodynamically stable, limited activation mapping was used to identify the area of earliest activation. Pacemapping close to the tachycardia cycle length in the scar border zone was carried out in patients with hemodynamically unstable VT to achieve the closest 12-lead electrocardiographic match of the QRS complex to that of the clinical tachycardia. Ablation was then carried out in the scar border zone regions containing late, fractionated, and double potentials that also corresponded to the area of earliest activation during hemodynamically stable VT or had the best pace map match to the clinical unstable VT. Entrainment mapping13 was used to confirm the VT circuit in selects patients with stable VT. Radiofrequency (RF) power was titrated between 30 W and 50 W, and the temperature cut-off was set at 40°C with heparinized saline infused through the catheter at 30 ml/min during ablation and 2 ml/min during mapping. Each RF application was at least 1 min in duration. Once the tachycardia stopped, programmed electrical stimulation in the RV was resumed to ascertain whether the tachycardia was still inducible. The procedure was repeated until VT was no longer inducible.
Results Baseline characteristics of the study cohort are described in Table 1. 3D maps took 26 ⫾ 8 min to create, before entering Table 1
Baseline patient characteristics
Male gender Ejection fraction New York Heart Association class Prior revascularization Percutaneous coronary intervention Coronary artery bypass graft surgery ICD Shocks ATP CRT device Medications Beta-blocker Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker Loop diuretic Spironolactone Digoxin Prior ablation
16 (94%) 40% ⫾ 15% 2⫾1 9 (53%) 4 (24%) 6 (35%) 10 (59%) 6 (35%) 7 (41%) 2 (12%) 16 (94%) 16 (94%) 8 2 5 1
(47%) (12%) (29%) (6%)
ICD ⫽ implantable cardioverter-defibrillator; ATP ⫽ anti-tachycardia pacing; CRT ⫽ cardiac resynchronization therapy.
Khaykin et al
ICE and 3D Mapping to Guide VT Ablation
Figure 3 Correlation between the left ventricular volume estimated using point-by-point CARTO mapping and Cartosound 3-dimensional intracardiac echocardiography rendering.
the LV and without fluoroscopy. Important anatomical structures such as papillary muscles, aneurysms, and trabeculations were rendered in 3D. A complete map of the LV was built from 23 ⫾ 7 contours. Regional wall motion abnormalities were identified and tagged in all patients. These corresponded to areas of low bipolar voltage (⬍0.5 mV) during substrate mapping in each case. On average, 2 ⫾ 1 VT morphologies were mapped and ablated per patient. Average volume of the LV estimated using the ICE generated map was 172 ⫾ 119 cm3 compared with 164 ⫾ 112 cm3 for the point-by-point maps (P ⫽ .5, Figure 3). Figure 4 illustrates one of the patients with an ischemic VT substrate. A small dyskinetic area identified by ICE corresponded to the anteroseptal scar mapped using bipolar voltage criteria. Ablation at the septal border of the scar within an ideal pace map area led to initiation, acceleration, and termination of the clinical VT and rendered it noninducible in the basal state. VT was still inducible on isoproterenol. An activation map in VT identified a very early diastolic potential on the anterior aspect of the scar (on the other side of the trabeculation identified by ICE). Ablation here
1399 resulted in termination of VT and rendered it noninducible. Figure 5 illustrates a patient with a dilated cardiomyopathy and VT refractory to 2 prior ablations. A large hypokinetic area was identified on the septum, corresponding to the scar identified during point-by-point mapping. Clinical ventricular tachycardia was induced and mapped to the basal septal area of fractionated potentials. Ablation in this area rendered VT noninducible. Average scar area defined by ICE criteria was 33 ⫾ 32 cm2 compared with 36 ⫾ 33 cm2 for point-by-point substrate mapping (P ⫽ .4, Figure 6). Five patients had predominantly anteroapical scar, 5 had scar confined to the septum, 5 had inferior scar, and 2 had scar involving primarily the lateral wall. Close correlation of the scar area identified by ICE with that confirmed during point-by-point substrate mapping is illustrated in Figure 7. Catheter tip position was monitored by ICE in each case using feedback from the navigation sensor, which allowed precise navigation to such important anatomical areas as trabeculations and crypts found critical to clinical success in several patients (Figure 8). No patient suffered a major procedure-related complication. Procedural characteristics and outcomes are illustrated in Table 2. All procedures were successfully performed within 2 to 5 hours with structural mapping using ICE performed in under 30 minutes. Less than 30 minutes of fluoroscopy was used in all but 3 patients. There were no intraprocedural complications. At 5 ⫾ 4 months of follow-up, 12 patients (71%) remain free of recurrent VT. One of the patients with an implantable cardioverter-defibrillator received a shock during follow-up. Two of the patients who had a VT recurrence received anti-tachycardia pacing (ATP), successfully terminating the arrhythmia. One of the patients died during follow-up of progressive congestive heart failure.
Discussion Integration of ICE and 3D electroanatomical mapping was used to guide ablation of ventricular tachycardia refractory
Figure 4 Intracardiac echocardiography (ICE) scan through the successful VT ablation sites. This figure illustrates a 2-dimensional intracardiac ultrasound image (left) registered as part of the CARTO shell of the left ventricle (middle). Tip of the ICE catheter positioned in the right ventricle is seen projecting an ultrasound “fan” through the body of the left ventricle. Arrows 1 and 2 on the ultrasound image correspond to arrows on the CARTO image and point to the 2 successful ablation sites on either side of a trabeculation. The cursor on the CARTO map is placed at the site of earliest activation in VT with local diastolic potential preceding onset of the QRS by 109 msec.
1400
Heart Rhythm, Vol 5, No 10, October 2008
Figure 5 Correlation of septal scar identified during intracardiac echocardiography (ICE) and point-by-point mapping. This figure illustrates a 2-dimensional intracardiac ultrasound image (left) registered as part of the CARTO shell of the left ventricle (middle). Tip of the ICE catheter positioned in the right ventricle is seen projecting an ultrasound “fan” through the body of the left ventricle. Red contour on the ICE image delineates an akinetic segment. Corresponding anatomical map of the left ventricle demonstrates position of the akinetic segment on the mid-septum. The bipolar voltage substrate correlates well with scar identified by ICE in this patient. The cursor on the CARTO map is placed at the site of a late/fractionated potential (white arrow). The local electrogram is seen in the right panel.
to amiodarone in 17 patients with both ischemic and nonischemic cardiomyopathic substrates. This approach allowed rapid acquisition of real-time cross-sectional images of the LV without having to enter this chamber. The images were then integrated into a registered 3D model of the LV without the need for preprocedural imaging. ICE not only facilitated acquisition of the anatomical information but also helped map the substrate by identifying wall motion abnormalities later found to correspond to scar as defined by local bipolar voltage. Endocardial detail including papillary mus-
Figure 6 Correlation between the scar area estimated using point-bypoint CARTO mapping and Cartosound 3D intracardiac echocardiography (ICE) rendering.
cles and trabeculations could be readily identified. Volume of the LV estimated using the point-by-point map corresponded exactly to the volume of the ICE reconstruction. Moreover, the area of scar identified using ICE corresponded exactly to that defined using standard bipolar voltage settings during point-by-point mapping. This may help further limit catheter manipulation in the LV in the future by focusing effort on the potential substrate regions identified by ICE. Real-time monitoring of the endocardial– catheter interface helped direct the catheter to the important anatomical structures found to be critical to the clinical arrhythmia in several patients, such as crypts and trabeculations. Ablation guided using this technique was successful in preventing ventricular tachycardia recurrences in 71% of the patients, potentially preventing implantable cardioverter-defibrillator therapies and hospital admissions. Although point-by-point mapping alone has been the standard for VT ablation, this procedure is time consuming and may miss important anatomical structures such as trabeculations, crypts, and aneurysms critical to the arrhythmia circuit in some patients. Integration of preacquired images may help solve this shortcoming of the point-by-point technique. Use of the preprocedural CT scans for image inte-
Khaykin et al
ICE and 3D Mapping to Guide VT Ablation
1401
Figure 7 Correlation of the inferior scar area identified during intracardiac echocardiography (ICE) and point-by-point mapping. This figure illustrates an anatomical shell of the left ventricle built from the ICE contours with hypokinetic or akinetic segments labeled red. A corresponding bipolar voltage map built using the ICE shell is seen in the middle panel. The low bipolar voltage area corresponds to the region defined by the red ICE contours. The cursor is placed at the site of a late potential during RV apical pacing, corresponding to a successful ablation site. The local electrogram is seen in the right panel.
gration may result in substantial radiation exposure for the patient, and some facilities may lack appropriate equipment and expertise to generate high-spatial-resolution cardiac MRI images. In a recent study in which a combination of CT and positron emission tomography (PET) imaging was merged with the 3D map of the LV, LV substrate data were successfully registered to guide ablation.14 In this study, scar was identified as a “hole in the wall” because of current software limitations. Although scar registration was very precise in patients with ⬍15% scar burden, registration in patients with more extensive scar was less accurate. Scar as evidenced by wall motion abnormalities was well defined by ICE in our study in all patients, corresponding precisely to the scar defined using standard criteria during point-bypoint mapping. While preacquired images may not accu-
rately represent real-time cardiac anatomy, ICE has been shown effective in guiding a variety of electrophysiology procedures by providing real-time information necessary to improve the accuracy of ablation, to monitor and prevent complications. This is a small study providing a very preliminary assessment of this new technology. Some operators proficient in VT ablation but unfamiliar with ICE may have a rather steep learning curve with this approach. Although identification of the regional wall motion abnormalities using ICE accurately defines the general area of low endocardial voltage, high-density point-by-point mapping is still required to precisely define the border zone and identify areas of late, fractionated, and double potentials. Automation of ICE image acquisition and wall motion analysis in the future may
Figure 8 Monitoring position of the catheter tip. This figure illustrates a real-time two-dimensional (2D) intracardiac echocardiography (ICE) image of a cross section of the left ventricle taken through the tip of the ablation catheter (left), while delivering radiofrequency energy at the successful ablation site. A 3-dimensional reconstruction of the left ventricle is seen on the right. The exact location of the ablation catheter tip based on the CARTO navigation sensor is overlaid on the 2D ultrasound image (green contour).
1402 Table 2
Heart Rhythm, Vol 5, No 10, October 2008 Procedure characteristics and outcomes
Procedure time, min Fluoroscopy time, min Radiofrequency delivery time, min Number of contours Follow-up duration Recurrence ATP Shock Death
240 ⫾ 77 min 25 ⫾ 12 min 28 ⫾ 13 min 23 ⫾ 7 contours 5 ⫾ 4 months 5 (29%) 2 (12%) 1 (6%) 1 (6%)
ATP ⫽ Anti-tachycardia pacing.
help address this issue. Further studies assessing utility of real-time registered integration of ICE and 3D electroanatomical mapping are necessary to define its place in the armamentarium of the invasive cardiac electrophysiologists.
Conclusion A mapping system combining ICE and 3D electroanatomical mapping can feasibly reconstruct a 3D shell of the LV, including a limited substrate map based on regional wall motion abnormalities without the need to enter the left heart chambers. Ablation guided using this approach seems safe and effective in a small group of patients with a wide variety of substrates.
Acknowledgement Dr. Offer Klemm, Biosense Webster Israel, helped with the early implementation of this technology in our laboratory.
References 1. A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. The Antiarrhyth-
2. 3.
4. 5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
mics versus Implantable Defibrillators (AVID) Investigators. N Engl J Med 1997;337:1576 –1583. Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverterdefibrillator for congestive heart failure. N Engl J Med 2005;352:225–237. Moss AJ, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877– 883. Irvine J, Dorian P, Baker B, et al. Quality of life in the Canadian Implantable Defibrillator Study (CIDS). Am Heart J 2002;144:282–289. Connolly SJ, Dorian P, Roberts RS, et al. Comparison of beta-blockers, amiodarone plus beta-blockers, or sotalol for prevention of shocks from implantable cardioverter defibrillators: the OPTIC Study: a randomized trial. JAMA 2006;295:165–171. Della Bella P, De Ponti R, Uriarte JA, et al. Catheter ablation and antiarrhythmic drugs for haemodynamically tolerated post-infarction ventricular tachycardia; long-term outcome in relation to acute electrophysiological findings. Eur Heart J 2002;23:414 – 424. Borger van der Burg AE, de Groot NM, van Erven L, et al. Long-term follow-up after radiofrequency catheter ablation of ventricular tachycardia: a successful approach? J Cardiovasc Electrophysiol 2002;13:417– 423. Calkins H, Epstein A, Packer D, et al. Catheter ablation of ventricular tachycardia in patients with structural heart disease using cooled radiofrequency energy: results of a prospective multicenter study. Cooled RF Multi Center Investigators Group. J Am Coll Cardiol 2000;35:1905–1914. Callans DJ, Ren JF, Michele J, et al. Electroanatomic left ventricular mapping in the porcine model of healed anterior myocardial infarction. Correlation with intracardiac echocardiography and pathological analysis. Circulation 1999;100: 1744 –1750. Fahmy TS, Mlcochova H, Wazni OM, et al. Intracardiac echo-guided image integration: optimizing strategies for registration. J Cardiovasc Electrophysiol 2007;18:276 –282. Okumura Y, Henz BD, Johnson SB, et al. Three-Dimensional Ultrasound for Image-Guided Mapping and Intervention: Methods, Quantitative Validation and Clinical Feasibility of a Novel Multi-Modality Image Mapping System. Circulation Arrhythmia Electrophysiol 2008;1:110 –119. Marchlinski FE, Callans DJ, Gottlieb CD, et al. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation 2000;101:1288 –1296. Stevenson WG, Khan H, Sager P, et al. Identification of reentry circuit sites during catheter mapping and radiofrequency ablation of ventricular tachycardia late after myocardial infarction. Circulation 1993;88:1647–1670. Dickfeld T, Lei P, Dilsizian V, et al. Integration of Three-Dimensional Scar Maps for Ventricular Tachycardia Ablation With Positron Emission Tomography-Computed Tomography. J Am Coll Cardiol Img 2008;1:73– 82.