Author + information
- Received June 27, 2013
- Revision received October 2, 2013
- Accepted October 3, 2013
- Published online March 1, 2014.
- Francesco Fulvio Faletra, MD∗ (, )
- Giovanni Pedrazzini, MD,
- Elena Pasotti, MD,
- Stefano Muzzarelli, MD,
- Maria Cristina Dequarti, MD,
- Romina Murzilli, MD,
- Susanne Anna Schlossbauer, MD,
- Iveta Petrova Slater, MD and
- Tiziano Moccetti, MD
- ↵∗Reprint requests and correspondence:
Dr. Francesco Fulvio Faletra, Division of Cardiology, Fondazione Cardiocentro Ticino, Via Tesserete 48, CH-6900 Lugano, Switzerland.
Guidance of catheter-based procedures is performed using fluoroscopy and 2-dimensional transesophageal echocardiography (TEE). Both of these imaging modalities have significant limitations. Because of its 3-dimensional (3D) nature, 3D TEE allows visualizing the entire scenario in which catheter-based procedures take place (including long segments of catheters, tips, and the devices) in a single 3D view. Despite these undeniable advantages, 3D TEE has not yet gained wide acceptance among most interventional cardiologists and echocardiographists. One reason for this reluctance is probably the absence of standardized approaches for obtaining 3D perspectives that provide the most comprehensive information for any single step of any specific procedure. Therefore, the purpose of this review is to describe what we believe to be the most useful 3D perspectives in the following catheter-based percutaneous interventions: transseptal puncture; patent foramen ovale/atrial septal defect closure; left atrial appendage occlusion; mitral valve repair; and closure of paravalvular leaks.
- catheter-based percutaneous interventions
- 2-dimensional transesophageal echocardiography
- 3-dimensional transesophageal echocardiography
Advances in technology and human skill have made possible the adoption of percutaneous catheter-based procedures in a wide spectrum of structural heart diseases that over the past 2 decades would have required open-heart surgery. Typically, guidance of these catheter-based procedures is performed using fluoroscopy and 2-dimensional (2D) transesophageal echocardiography (TEE). Both of these imaging modalities have significant limitations. Fluoroscopy is limited by its 2D projections of a complex 3-dimensional (3D) heart, and its inability to delineate soft structures precisely; 2D TEE, because of its tomographic nature, needs multiple planes and adjustments to visualize the course of intracardiac catheters and their complex relationship with cardiac structures.
3D TEE has the unique ability to depict cardiac structures as they are in reality (1,2). Moreover, because of its 3D nature, long segments of catheters, tips, and the devices can easily be intersected by the pyramidal ultrasound beam and displayed without excessive probe manipulations. Finally, the entire scenario in which most of the catheter-based procedures take place (i.e., atrial septum, left atrial appendage [LAA], left atrium, and mitral valve) can be shown in a single 3D view (3–5). Theoretically, 3D TEE should be the ideal guidance-imaging tool for catheter-based procedures. Despite its undeniable advantages, 3D TEE has not yet gained wide acceptance among most interventional cardiologists and among those echocardiographists involved in catheter-based procedures.
Several reasons may explain the reluctance to shift from 2D to 3D TEE. Historically, 2D TEE was used to guide these procedures, thus interventional cardiologists and echocardiographists became accustomed working with 2D TEE imaging. Because of its tomographic nature, 2D TEE needs multiple planes and adjustments to accurately track catheters moving in a 3D environment. For this reason, its use, especially during some complex catheter-based interventions, has been carefully standardized and both echocardiographists and interventional cardiologists have a clear notion of which plane(s) must be used for each step of the procedure (6).
The use of 3D TEE as guidance imaging modality during catheter-based procedures was first described by Perk et al. (3), who collected data from 5 institutions with great expertise with this new imaging tool. However, this first experience was not followed by an effort to select (and hence standardize) those 3D views that might provide the most useful data for each step of any specific procedure. Even when 3D TEE is described as a useful tool for guiding interventional procedures (4,7–10), there are no systematic descriptions of how to acquire those perspectives that can provide the best images of catheters, devices, and their relationships with target structures. The need to provide specific 3D views derives from the fact that the “volumetric” acquisition includes many cardiac structures that, in turn, can be imaged from countless perspectives. However, only a few of them are very innovative and useful, others are redundant and useless, and some are just confusing. Moreover, from some viewpoints, target structures may be covered by surrounding tissue that needs to be removed. Finally, reverberations (i.e., multiple reflections) and shadowing (i.e., lines of dropout beyond catheters similar to tears in the tissue) both caused by the specific material of the catheters may further complicate the selection of the most appropriate perspectives. The only way to select those views that best match the requests of interventionalists and to remove the presence of redundant tissue and avoid reverberations or dropout artifacts, is to manipulate the volumetric data set (i.e., cropping, rotating, and adjusting the acquired images) during the procedure.
This review aims to provide practical guidance for echocardiographists who, working in institutions performing catheter-based interventions, are involved in these procedures. We describe how to obtain those 3D perspectives that, to our minds, are the most useful in the following procedures: transseptal puncture, patent foramen ovale (PFO)/atrial septal defect (ASD) closure, LAA occlusion, mitral valve repair (mitral clip), and closure of mitral paravalvular leaks.
Basic Technical Aspects of Imaging Acquisitions
The use of 3D TEE for guidance during catheter-based procedures depends exclusively on the generation of 3D images in real time. Currently, with the 2 available 3D TEE technologies (i.e., Philips, Medical Systems, Andover, Massachusetts; and Vivid 9, GE Healthcare, Milwaukee, Wisconsin), there are 2 modalities for real-time image acquisition:
1. 3D zoom modality: This modality can display a truncated but magnified pyramidal dataset of variable size. After sizing the zoom sector over the area of interest, the volume dataset is acquired. Minimizing sector width is important for increasing temporal resolution and image quality. Both GE and Philips 3D TEE technologies have similar 3D zoom modality acquisition.
2. Single beat: With Philips technology, a pyramidal set 60° × 30° is displayed in real time. This acquisition modality generates high-quality 3D images at a volume rate up to 25 Hz. New technical developments allow electronic steering, in both lateral and elevation planes of the pyramidal dataset, thus avoiding transducer manipulations. Single-beat acquisition of a larger area (60° × 60°) produces images at a volume rate of 9 to 10 Hz. With GE technologies, the single-beat modality has 3 pre-defined acquisition modalities with increasing angles: bird's eye view, medium modality, and large modality. The angle of these pre-configured volume datasets can also be manually changed according to specific needs.
The transseptal crossing is the common entry point for many left-side catheter-based procedures. Experienced operators may safely perform a transseptal puncture using only fluoroscopy (11,12). However, in catheter-based procedures, this maneuver is usually performed under fluoroscopy and 2D TEE. The use of an imaging guide may avoid complications related to inappropriate puncture sites, especially in high-risk patients (i.e., those with a previous transseptal crossing failure or severe kyphoscoliosis, septal aneurysm, or aortic root dilation). When the catheter is against the fossa ovalis and the interventional cardiologist applies pressure, the site of the puncture may be identified by the “tenting” seen in 2D TEE images.
3D TEE shows the “authentic” anatomical deformation of the interatrial septum (IAS) during the puncture (i.e., a configuration similar to a “conic tent”), yet many echocardiographists still prefer to use 2D TEE. Indeed, with 3D TEE, the atrial septum is usually displayed in an en face perspective (2). However, this perspective is not as effective as 2D TEE in showing the tenting. Difficulties in interpreting 3D tenting in an en face view perspective are shown in Figures 1A and 1B.
Even when echocardiographists manipulate the image online to obtain a view similar to that of 2D TEE, interventional cardiologists remain reluctant to maneuver under 3D guidance because the borders of the tenting may be difficult to distinguish against a background of a similar color (Fig. 1C).
The safest site for septal puncture is across the fossa ovalis (11). The various steps needed to acquire 3D imaging of the fossa ovale en face from a right perspective using Philips technology are described elsewhere (2). Table 1 summarizes technical details of image acquisition with both Philips and GE technologies.
The en face view from the right atrial perspective is particularly appreciated by interventionalists because it matches the fluoroscopic right anterior oblique projection. They can follow the catheter tip (part of the body usually remains out of volumetric dataset) from the superior vena cava to the fossa ovalis, associating their tactile feedback with the visualization of the movement of the catheter through these structures (Fig. 2).
Once the catheter has been moved into the fossa ovalis, tenting can best be viewed from a lateral perspective. This view can be obtained quite simply by rotating the volumetric dataset left-to-right around the y-axis (Fig. 3, Online Video 1). The source of light (created by a specific algorithm) illuminates the tenting, laterally enhancing its edges against the background and facilitating recognition. Moreover, both dropout artifacts and reverberations are covered by the tenting itself. We found that this perspective is very effective in imaging the tenting and, more importantly, was well accepted by our interventional cardiologists.
Percutaneous closure of PFO/ASD is usually performed via the right femoral vein. The septal crossing and the other steps of the procedure can be easily guided by 3D TEE visualizing the IAS from both oblique and lateral perspectives, which enhance catheter imaging and disk expansion (Fig. 4, Online Videos 2 and 3). Technical details on how to acquire these perspectives are described in Table 2.
Specific 3D perspectives are needed for guiding the correct positioning of the guide catheter inside the LAA and following the expansion of the occluder (Figs. 5 to 8, Online Video 4). Details on how to acquire these perspectives are described in Table 3. Figures 5 to 8 and Online Video 4 refer to the deployment of the Amplatzer cardiac plug (AGA Medical, Plymouth, Minnesota) (13).
Mitral Clip Procedure
The catheter-based edge-to-edge mitral clip repair (mitral clip) consists of bringing the anterior and posterior leaflets together with a metallic clip (14). The procedure is complex and embraces several steps. None of them can be made without TEE guidance. Because of the complexity, the use of 2D TEE has been strictly standardized and at least 4 key basic views are recommended, each of them crucial for any specific step (6).
The role of 3D TEE as the guidance imaging modality during a mitral clip procedure has been recently explained (5,15). During the septal crossing, the site of the septal puncture is of particular relevance: a distance not inferior to 4.0 to 4.5 cm between the septal tenting and mitral valve orifice provides an adequate space for maneuvering the mitral clip delivery system into the left atrium; a lateral perspective of IAS enables the visualization in a single image of the tenting and the mitral valve plane. How to obtain this perspective is described in detail in Table 4 and shown in Figure 9A.
The advancement of the mitral clip delivery system into the left atrium and its steering toward the mitral valve requires 3D views that display the spatial relationship of the delivery system with the atrial wall and IAS. An oblique perspective of IAS satisfies these requirements (Figs. 9 and 10). Such a view is particularly appreciated by interventionalists because they can safely maneuver (advancing, pulling back, and steering) catheters while maintaining the entire delivery system in the left atrium and avoiding hurting the atrial walls. Details on how to acquire these perspectives are described in Table 4.
Orienting the clip arms perpendicular to the coaptation line is of paramount importance for the success of the procedure because lack of perpendicularity may result in a failure to capture or inadequately grasp 1 or both leaflets. The overhead perspective, which enables imaging the open arms, and in a deeper plane, the mitral valve coaptation line, has been the first 3D view fully accepted during the procedure (3). Even the most reluctant adopters of 3D modality should admit that this unique perspective is by far preferable to the 2D TEE transgastric short-axis view (6), allowing fine adjustments until the clip arms are perfectly perpendicular to the coaptation line (Fig. 11, Online Video 5, Table 4).
The most relevant step of the procedure is the act of grasping the leaflets. Currently, this step is exclusively guided by 2D TEE because the spatial resolution of 3D imaging is not sufficiently adequate to image the thin leaflets between arms and grippers. Once captured, however, among the most challenging issues is evaluating the adequacy of the insertion of the leaflets into the clip. Because the clip is inserted from below, we found the 3D perspective from the left ventricle to be particularly valuable. This perspective allows one to evaluate the adequacy of the insertion, residual orifices, and the number and position of clip(s) (Fig. 12, Table 4). The 3D views that allow monitoring of the clip deployment and the removal of the delivery system are shown in Figure 13 and described in Table 4.
Mitral Paravalvular Leak
3D TEE allows a “panoramic” view of the suture ring from an overhead perspective (the surgical view) with the aortic valve at the top of the mitral ring (12 o'clock), and the LAA at approximately the 9-o'clock position. Within this virtual clock, the location of any mitral paravalvular leak may be reported as a single hour, if localized or, as a range of hours if the defect is larger or has a crescent-shaped configuration. The same view may be maintained during the entire procedure of mitral valve leaflet closure. The main advantage of this perspective is that it facilitates observation of the spatial relationship between catheter tip, target leak, and surrounding anatomical structures, providing a 3D anatomical environment, where interventionalists can safely maneuver the catheter, observing its movement toward the leak (Figs. 14A to 14C). Moreover, the device itself can be visualized during expansion from the same perspective (Figs. 14D and 14E); once deployed, the exact location and shape of the device can be appreciated (Fig. 14F, Table 5).
Most of the images shown in this review are obtained by a single ultrasound machine that uses its own technology (i.e., Philips technology). This is no longer the only machine in which 3D TEE is available. A second ultrasound machine that uses a different technology, acquisition modalities, and imaging processing (i.e., GE technology) is currently available. However, once the most appropriate perspectives have been established, similar perspectives can be obtained with both ultrasound machines, although with different imaging acquisition and processing algorithms (Fig. 15).
We did not include in this review the role of 3D imaging in transcutaneous aortic valve replacement. Indeed, whereas 3D TEE is useful pre-procedurally for annulus measurements, its role during the procedure is limited, being the mainstay of intraprocedural imaging fluoroscopy and angiography. Currently in our institution, 2D/3D TEE is performed in the post-implant assessment.
Certainly, we do not advocate the use of 3D TEE as the sole imaging technique. We are aware that many limitations remain. In particular, the frame rate, though acceptable (Online Videos 2 to 7), is still not optimal; the spatial resolution is inferior when compared with that of 2D TEE, and there is also the issue of the lack of real-time acquisition of the 3D color Doppler. These limitations may affect the use of 3D TEE in several steps of any interventional procedure and create difficulties in interpreting 3D images. The most frequent limitations and difficulties that we experienced in interpreting 3D images are listed in Table 6 and shown in Figure 16.
Ongoing technological improvements (such as the 3D color Doppler in real time with an adequate frame-rate and better temporal and spatial resolutions in large panoramic images) will further facilitate the use of this novel technique.
Specific 3D artifacts may occur when the pyramidal ultrasound beam intersects the metallic structures of catheters and devices. These artifacts when displayed in 3D format may appear more “realistic” and may lead to misinterpretation. Dropout artifacts in the atrial septum, for instance, resembling “real” holes, may cause misinterpretation in patients scheduled for ASD closure. Shadowing from catheters may create the impression of real tears in the cardiac tissue behind catheters. Selecting those specific perspectives that are effective in guiding the procedure and, at the same time, minimize these artifacts (thus avoiding that they might have any impact on the procedure) requires experience and practice.
The main goal of this review is to provide practical suggestions on how to obtain specific views that, in our minds, may have additional value over conventional 2D TEE in specific steps of the above-mentioned catheter-based procedures. Searching for these perspectives during the procedure requires time that many interventionalists may be unwilling to concede. However, in our experience, recognizing in advance those perspectives that would meet the needs of interventionalists could eventually speed acquisition and online processing. In the future, the most effective perspectives might be pre-defined to be immediately available during the procedure.
It should be emphasized that our findings are based on the experience of a single center. Other echocardiographists working in hemodynamic laboratories performing structural interventions, may have found different but equally (or even more) effective perspectives. Table 7 shows the most significant manuscripts on the use of 3D TEE for interventional procedures and the perspectives shown by different investigators. Because in the future the technique is expected to be widely used during catheter-based procedures, this review might be considered an encouragement for professional associations such as the American and European Society of Echocardiography to generate appropriate recommendations in this specific field. Finally, following in real time what interventional cardiologists are doing in a 3D panoramic environment may be very attractive. But to be completely accepted by interventional cardiologists, we need to provide clear evidence that the use of 3D TEE instead of 2D TEE generates practical advantages such as avoiding potential complications, reducing radiation exposure, and/or shortening procedural times.
For supplementary videos and their legends, please see the online version of this article.
All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial septal defect
- interatrial septum
- left atrial appendage
- patent foramen ovale
- transesophageal echocardiography
- Received June 27, 2013.
- Revision received October 2, 2013.
- Accepted October 3, 2013.
- American College of Cardiology Foundation
- Faletra F.F.,
- Ho S.Y.,
- Auricchio A.
- Altiok E.,
- Becker M.,
- Hamada S.,
- et al.
- Faletra F.F.,
- Pedrazzini G.,
- Pasotti E.,
- et al.
- Lee A.P.,
- Lam Y.Y.,
- Yip G.W.,
- Lang R.M.,
- Zhang Q.,
- Yu C.M.
- Cavalcante J.L.,
- Rodriguez L.L.,
- Kapadia S.,
- Tuzcu E.M.,
- Stewart W.J.
- Earley M.J.
- Feldman T.,
- Wasserman H.S.,
- Herrmann H.C.,
- et al.
- Faletra F.F.,
- Grimaldi A.,
- Pasotti E.,
- et al.