Three-Dimensional Electroanatomic Imaging for Atrial Fibrillation Ablation and Device Therapy


Rajesh Kabra, MD and Jagmeet Singh, MD, DPhil
Massachusetts General Hospital, Harvard Medical School
Boston, Massachusetts

One of the most significant innovations that have transformed the field of cardiac electrophysiology has been the evolution of three-dimensional (3-D) mapping systems. Advances in electroanatomic mapping have revolutionized interventional strategies, especially those pertinent to atrial fibrillation (AF) ablation and device therapy for heart failure.

With respect to AF, it is evident that anatomic variations in the left atrial ridges and pouches, the unpredictable numbers of veins, inconsistent ostial sizes, and variable take-offs and locations can significantly impact procedural success. Conventional imaging systems poorly define these anatomic deviations, and thereby can lead to prolonged procedural times, increased radiation exposure and inaccurate lesion delivery. With regards to device therapy (in particular, cardiac resynchronization therapy [CRT]), there continues to be debate regarding the most optimal ventricular pacing site. Electroanatomic mapping enables us to get a better perspective of the electrical activation pattern in the failing heart, thereby potentially directing us toward the region of maximal electromechanical delay.

In this article, we provide a brief overview of the commonly used electroanatomic mapping systems in the electrophysiology laboratories, namely Carto® XP (Biosense Webster, Inc., a Johnson & Johnson company, Diamond Bar, CA), EnSite Array™ noncontact mapping system (St. Jude Medical, St. Paul, MN), and EnSite NavX™ (St. Jude Medical). This article describes and compares these three modalities, highlighting their advantages and limitations with respect to mapping and ablation. In addition, we also describe the novel applications of these mapping systems pertinent to lead positioning for cardiac resynchronization therapy.

Current Techniques

The contemporary approach to catheter ablation of AF is comprised of two fundamental principles: 1) pulmonary vein (PV) isolation and eradication of focal triggers, and 2) atrial substrate modification. The latter approach at this time is usually reserved for patients with persistent AF, as this subgroup has a significantly more enlarged and remodeled atrium. Currently, the most accepted approach involves isolation of a large circumferential area around the pulmonary veins with verification of conduction block. The long-term success of ablation in patients with persistent AF can be further enhanced by incorporating adjunctive ablation strategies to PV isolation, such as targeting sites demonstrating complex atrial fractionated electrograms (CFAEs) during AF or at regions that may show ‘AF nests’ during real-time spectral mapping in sinus rhythm.

On the device therapy front, electroanatomic mapping facilitates the delineation of the electrical activation pattern of the heart. This usually involves contact mapping with the use of a mapping catheter that is dragged over the endocardium. This enables the sequential acquisition of points, while the location of the electrode tip and the local electrogram in relation to the surface QRS signal is recorded. The method uses bipolar signals and facilitates the creation of a color-coded activation map, generated from regional activation timings, which can be superimposed on 3-D chamber geometry. The electrical activation sequence in the failing heart can vary depending on the type of the conduction defect, as well as the severity and type of cardiomyopathy (i.e., ischemic and non-ischemic). Interplay between the native activation sequence and pacing-induced alterations dictate the benefit derived from CRT, and although routinely impractical, can be a useful strategy to evaluate a select group of non-responders to CRT.

Carto® XP System Overview

Initially described by Ben-Haim et al in 1996,1 the Carto® system involves point-by-point mapping using a deflectable quadripolar catheter. The tip of this catheter contains a location sensor, which is traced by an external ultra-low magnetic field generated by a unit mounted under the patient table. The location and orientation of the catheter tip is determined by a processing unit. This helps to create a 3-D map of any cardiac chamber of interest. Additionally, the local electrograms at each point can be gated to a preselected reference electrogram to create activation or propagation color-coded maps as well as a voltage map that can be superimposed on the anatomical map of the chamber. These maps can be viewed in multiple projections. Areas of interest like the His bundle, scars, fractionated electrograms, ablation points, or points close to the esophagus or phrenic nerve can be tagged in these maps. A reference patch is placed over the patient’s back to correct for minor patient movement.

The Carto® XP system provides a highly accurate geometry of a cardiac chamber, which can be visualized in multiple views. Activation maps are straightforward, and respiratory artifact is limited.

One limitation of the Carto® system is that the proprietary ablation catheter uses only a single bipolar electrode in the catheter tip to record electroanatomic data. Therefore, it can be time consuming to generate a high-density electroanatomic map of a chamber. The need for point-to-point mapping on a beat-by-beat basis limits the use of this mapping system for non-sustained or unstable rhythms. Nevertheless, voltage mapping may be useful to delineate scars and identify potential arrhythmia circuits that could facilitate a successful ablation outcome. This system is sensitive to patient motion because of the presence of the magnetic tripod underneath the patient; in cases of significant patient movement, an arrhythmia may need to be remapped. If not detected early, it can lead to inaccuracies in mapping. Also of note is that this system requires the use of the proprietary single-use catheter made by Biosense Webster, Inc. and does not support catheters from other manufacturers.

Advances in these mapping techniques include the CartoMerge® Image Integration Software Module (Biosense Webster, Inc.), in which the image generated by Carto® can be integrated with the images generated by CT scan or MRI. These have been most commonly used in atrial fibrillation ablation to correctly identify the anatomy and location of the pulmonary veins as well as other structures like the esophagus (Figure 1). Similarly, the CartoSound™ Image Integration Module (Biosense Webster, Inc.) incorporates the electroanatomic map to a map derived from intracardiac echocardiography (ICE). The CartoMerge® Module has been around for a few years and involves the identification of specific anatomical landmarks (i.e., pulmonary vein ostia) that can be identified on the segmented CT or MRI image, as well as on the generated Carto® map of the left atrium (LA). The images are then locked together to create the integrated image. This is then further refined by obtaining more endocardial points all over the left atrium, ensuring that these points are well distributed over the entire anatomy. Before commencing the ablation, we ensure the adequacy of the integration by roving the catheter through the atrium and assessing its relationship to well-recognized landmarks (i.e., veins, ridges and appendage).

The CartoSound™ Module is an innovative strategy that facilitates 3-D reconstruction of the cardiac chambers using real-time ICE (Figures 2 and 3). The ICE catheter has an embedded navigation sensor and is particularly helpful for mapping the left atrium for AF ablation. This method involves positioning the ICE catheter either in the right atrium, coronary sinus or even at times across a transseptal puncture into the left atrium. Electrocardiogram-gated echocardiographic images of the endocardial surface of the left atrium are obtained, delineated and then collated to create a 3-D volume-rendered image. This strategy is useful in that it enables detailed visualization of the left atrium and its adjacent structures. Importantly, it eliminates chamber deformity, which often occurs with contact mapping. The LA anatomical maps created are then registered to the patient’s CT or MRI using two or more fiducial points. In our center, we sometimes (depending on the image quality from the right side) cross to the left atrium through a pre-existing transseptal puncture, enabling more complete left atrial visualization. The image quality is enhanced by better tissue penetration in these circumstances, which helps with better delineation of landmarks such as the mitral valve annulus, right-sided pulmonary vein ostia, and the ridge between the LA appendage and the left superior pulmonary vein ridge. This is a safe strategy, but nevertheless requires careful manipulation of the stiff ICE catheter in the left atrium.

EnSite Array™ System Overview

The EnSite Array™ system (St. Jude Medical) provides a noncontact mapping technique to create a 3-D electroanatomic map without point-by-point contact electrograms. It involves a 9 Fr multielectrode EnSite Array™ Catheter mounted on a 7.5 ml balloon with 64 insulated filaments acting as noncontact electrodes. The EnSite Array™ is deployed in the cardiac chamber of interest and used as a reference to create 3-D anatomical geometry by sweeping any conventional catheter throughout the cardiac chamber to define the endocardial borders. After the creation of the anatomical geometry, the far-field potentials are recorded and processed to create 3,360 virtual unipolar electrograms.2,3 The computed voltage data from these electrograms can be displayed on the anatomical model of the cardiac chamber to generate an electroanatomic map.

The main advantage of this system is having the ability to map a non-sustained, hemodynamically unstable or difficult-to-induce arrhythmia, in which a single cycle of tachycardia is utilized to create an activation map. In addition, any catheter from any manufacturer can be used with this mapping system. Limitations include possible difficulty in deploying the balloon catheter in some chambers due to its size or shape; the volume occupied by the balloon may limit the movement of the ablation catheter in a small chamber. In contrast, in large chambers like a dilated left ventricle, virtual electrograms may be inaccurate if the endocardial surface is at a distance from the balloon surface. Unipolar mapping may record far-field signals, making it difficult to identify the local activation. On the other hand, low endocardial voltages may be too weak to be detected by the balloon catheter. Although this system is not particularly useful for AF ablation, on the device therapy front, it is a quicker strategy to delineate the spread of the depolarization wavefront in the heart failure patient.

EnSite NavX™ System Overview

This mapping system is based on localization of multiple electrodes using an electrical field generated by three pairs of surface electrodes placed on the patient’s body along three orthogonal axes. The patches emit a low-amplitude 5.7 kHz signal, and conventional catheters are localized by measuring the electrical potential or field strength received by them. The fundamental principle underlying EnSite NavX™ (St. Jude Medical) navigation is that of an impedance-based measure, which is dependent on the voltage gradient that exists across tissue when a current is applied through the surface electrodes. This mapping system can be used to localize up to 12 catheters and 64 electrodes. A geometric model of a cardiac chamber can be created by maneuvering any catheter with multiple electrodes to localize the endocardial borders. An electroanatomic map is generated by acquiring and displaying the activation and voltage data on this model. A reference intracavitary electrode is required, the position of which needs to be stable in order to maintain the accurate position of the electroanatomic map. The value of a secure reference cannot be overemphasized, as a significant shift can frequently lead to restarting the entire case.

The EnSite NavX™ system establishes an accurate geometry of the cardiac chamber. The graphic display of multiple catheters can significantly reduce fluoroscopic time and facilitate complex ablations, especially AF ablations. The system also has the capability to import and integrate three-dimensional CT or MRI images to facilitate anatomically-based ablation procedures.

As with Biosense Webster’s Carto® XP system, an important limitation of the EnSite NavX™ system is point-by-point and beat-to-beat mapping that is not feasible in a non-sustained or hemodynamically unstable arrhythmia. However, the use of multipolar catheters can facilitate dense electroanatomic maps and generation of endocardial surface geometry in a short period of time (Figure 4). The ability to visualize all the catheter electrodes within the CT integrated cardiac geometry assists the ablation procedure by providing 3-D information about the relative positions of the catheters with the heart. The EnSite NavX™ system is also good at providing timing and signal data, which are critical for assessing the adequacy of lines created during the AF ablation. Both EnSite NavX™ as well as Carto® XP systems allow mapping of the flutter circuits, importantly directing ablation at sites with complex fractionated electrograms and facilitating targeted substrate ablation in patients with persistent atrial fibrillation.

More recently, advances in image integration with the EnSite NavX system, known as NavX Fusion, have become available. The difference with this technique lies in the ability of the created geometry to dynamically mold into the CT. Once the left atrial geometry is created and appears to closely resemble the CT image, a field scaling algorithm is applied to the left atrial geometry. This field scaling adjusts for the non-linearity of the geometry and is based on the measured inter-electrode spacing for all the locations within the geometry. This field scaled geometry is fused into the CT in two stages, termed as primary and secondary fusion. The primary fusion is dependent on the use of four fiducial corresponding landmarks on both the CT and the created geometry. These points are used to lock together both structures, ensuring reasonable 3-D anatomic separation. Secondary fusion is applied at sites of local mismatch between the two superimposed geometries, enabling the molding process. Although this is a validated approach, it is not free from the possible errors of integration.

Both EnSite NavX™ and Carto® XP systems provide all the necessary anatomical information needed for encircling the pulmonary veins and additional lines (i.e., roof, mitral-isthmus, etc.). Both technologies (Carto® XP and Ensite NavX™) have different algorithms, but enable the operator to generate complex fractionated electrogram maps and target the appropriate regions for ablation (Figure 5). In addition, like Carto®, the EnSite NavX™ is useful in generating an activation map to identify the circuits of atrial flutter during atrial fibrillation ablation (Figure 6). However, there is a paucity of data on a head-to-head comparison of these different electroanatomic mapping technologies on safety, efficacy and outcomes.

Cardiac Resynchronization Therapy

The EnSite NavX™ system can also potentially play an important role in cardiac resynchronization therapy. CRT improves left ventricular (LV) efficiency and overall hemodynamic function by using LV or biventricular pacing and altering the sequence of electrical activation in patients with heart failure and intraventricular conduction abnormalities. However, approximately one-third of the patients receiving CRT fail to attain any benefit from it. An important determinant of response is the placement of the left ventricular lead. Commonly, it is placed in the lateral or posterolateral branches of the coronary venous system. In spite of being a useful technique, many patients still fail to respond with this anatomical approach. It is possible that the higher proportion of non-responders is due to a suboptimal pacing lead position. Studies have shown that targeting regions of delayed mechanical or electrical activation in the left ventricle may be a better strategy than a purely anatomical approach. The EnSite NavX™ system can be used intra-operatively to assess the electrical activation pattern within the branches of the coronary veins, delineate scars through voltage mapping, as well as potentially measure mechanical activation (Figure 7). This may have a role in the future toward targeting the LV lead to the most optimal coronary vein branch, to enhance response in the CRT patient.


Ablation strategies targeting the pulmonary veins and other regions of the atrium constitute the foundation for radiofrequency catheter ablation of atrial fibrillation. The advent of electrophysiological navigation systems allowing the integration of CT and MRI images of the left atrium with electroanatomic maps have been a significant advancement in the field. These maps provide accurate, real-time navigation with detailed anatomical mapping to facilitate the ablation procedure. This translates into enhanced accuracy of lesion delivery, reduced fluoroscopy time and procedural duration, with improved outcomes. Each mapping system has its own advantages and limitations (Table 1). Biosense Webster’s Carto® XP and St. Jude Medical’s EnSite NavX™ system work well for mapping sustained, stable arrhythmias. Mapping non-sustained arrhythmias is tedious with these approaches, so St. Jude Medical’s EnSite Array system may be helpful in these cases. The choice of mapping system is dependent on the skill and experience of the operator, as well as familiarity with the software. These mapping systems will continue to evolve and facilitate individualized approaches for both catheter ablation and device therapy in the future.

Add new comment