Cardiovascular Anesthesiology: Echo Rounds
A 51-year-old man presented to our hospital with increasing shortness of breath and lower extremity edema. He was status post St. Jude mechanical mitral valve replacement for endocarditis 4 years before admission. Transesophageal echocardiography (TEE) revealed a severe paravalvular leak around the inferolateral aspect of the sewing ring. Left ventricular systolic function was reduced (ejection fraction of 35%–45%). His mean diastolic pressure gradient across the mitral valve was 6 mm Hg. A small secundum atrial septal defect (ASD) with bidirectional shunting was observed.
The patient was referred for percutaneous closure of the paravalvular leak with an Amplatzer occluder. TEE was used to help guide the catheter-based procedure. Real-time 3-dimensional (RT3D) TEE (X7-2t transducer, Philips Healthcare, Andover, MA) was used to create color full-volume gated reconstructions demonstrating a severe inferolateral crescentic paravalvular leak (Fig. 1). (Video 1, see Supplemental Digital Content 1, http://links.lww.com/AA/A112. See Appendix for Video captions.) The feasibility of closure with placement of a circular device on a curvilinear defect was discussed, but the decision was made to proceed. Initially, an 8-mm Amplatzer Septal occluder (AGA Medical, Plymouth, MN) was deployed across the paravalvular defect using RT3D TEE as an aid to catheter navigation (Fig. 2). With 2D color flow Doppler (CFD) interrogation, the leak was unchanged so that the occluder was extracted and a larger 12-mm device was deployed. Repeat 2D CFD, despite the presence of acoustic shadowing artifact along the path of regurgitation, suggested a severe persistent leak on the basis of vena contracta width (6.9 mm). (Video 2, see Supplemental Digital Content 2, http://links.lww.com/AA/A113. See Appendix for Video captions.) An RT3D “en face” view of the mitral valve confirmed the severity of the leak and the inability of the circular Amplatzer to span the curvilinear defect (Fig. 3). (Video 3, see Supplemental Digital Content 3, http://links.lww.com/AA/A114. See Appendix for Video captions.) The procedure was terminated and the Amplatzer left in place. The patient's symptoms were not improved, and he was referred for operative intervention. He underwent a successful redo-sternotomy, explantation of a normal-appearing Amplatzer device, primary closure of the inferolateral paravalvular defect, and ASD closure. Three weeks after his operation, he was seen in follow-up and was asymptomatic.
Amplatzer occluders are routinely used for percutaneous closure of ASDs under fluoroscopy. Although percutaneous paravalvular leak closure with Amplatzer occluders is a suitable option for patients with increased perioperative risk, beneficial outcomes have not been consistently reported.1–3 Two-dimensional TEE has been shown to complement fluoroscopy for catheter-based procedures, but there are technical limitations. The success of the procedure may depend on the ability of 2D TEE to accurately identify the location and extent of the leak.1 The addition of RT3D TEE offers unique advantages, which can facilitate the technical procedure and improve appropriate patient selection before the intervention.
Two-dimensional TEE is not ideal for visualizing an intracardiac catheter.4 Locating the catheter tip within surrounding anatomic structures is difficult and requires multiple imaging planes. Minor adjustments of the wire can alter the curvature of the catheter and shift the tip outside the plane of a 2D TEE sector. Each adjustment of the catheter requires reassessment with subtle manipulations of the TEE probe to reconfirm its location within intracardiac structures. This can slow the procedure and impair communication with an interventionalist who may be less familiar with nonstandard TEE views, causing them to rely more heavily on fluoroscopy for navigation, increasing overall radiation exposure. RT3D TEE permits complete visualization of the catheter within a familiar anatomic landscape facilitating communication with the operator. Enhanced ability of catheter navigation and unequivocal confirmation of the defect location may improve overall success rates.1,4 For multiple defects in close proximity, RT3D TEE can confirm that the catheter passes through the targeted orifice before deployment of the device.4
RT3D TEE allows for unique perspectives that may improve patient selection for percutaneous paravalvular leak closure and other catheter-based procedures. Dehiscence of the convex sewing ring from the annulus often renders a defect crescent shaped. The extent of dehiscence can be assessed using multiple 2D imaging planes through mental reconstruction. RT3D TEE can generate an anatomic en face view of the mitral valve dehiscence, which may simplify interrogation and improve diagnostic accuracy.4 The extent of dehiscence can guide patient selection in cases of geometric mismatch between crescentic paravalvular defects and circular Amplatzer occluders. Deployment in such a case as ours may fail to abolish the leak and can even fracture the sewing ring from the annulus, exacerbating paravalvular regurgitation.5
Some disadvantages of RT3D TEE must be considered as well. The 3D pyramidal data set is subject to the same artifacts as 2D imaging. Unobstructed visualization is not always possible; therefore, cropping features may be needed to remove obstructive anatomy to an en face view, a time-consuming process. A major diagnostic concern is tissue dropout secondary to an undergained image. Dropout may give the impression of a false anatomic defect, leading to speculation of nonexistent pathology, a hole where a hole should not be.4 Anatomic detail is often less precise secondary to reduced lateral resolution compared with 2D imaging, especially in assessing deeper structures. RT3D images with large zoomed-in pyramids have lower frame rates (<10 Hz) compared with 2D TEE and fluoroscopy, resulting in choppy images and visual delays after catheter manipulation. For 3D CFD imaging, RT acquisition is not yet available, instead relying on multibeat gated reconstructions, another time-consuming process.
Despite the limitations, the addition of RT3D TEE to the arsenal of 2D TEE and fluoroscopy for catheter-based closure of paravalvular defects serves a complementary role in terms of aiding procedural efficiency and guiding patient selection.
1. Alfirevic A, Koch CG. Failed closure of paravalvular leak with an amplatzer occluder device after mitral valve replacement. Anesth Analg 2009;108:439–40
2. Phillips SA, Thompson A, Abu-Halimah A, Crenshaw MH, Zhao DX, Pretorius M. Percutaneous closure of aortic prosthetic paravalvular regurgitation with two amplatzer septal occluders. Anesth Analg 2009;108:437–8
3. Shapira Y, Hirsch R, Kornowski R, Hasdai D, Assali A, Vaturi M, Sievert H, Hein R, Battler A, Sagie A. Percutaneous closure of perivalvular leaks with amplatzer occluders: feasibility, safety, and short term results. J Heart Valve Dis 2007;16:305–13
4. Perk G, Lang RM, Garcia-Fernandez MA, Lodato J, Sugeng L, Lopez J, Knight BP, David Messika-Zeitoun D, Shah S, Slater J, Brochet E, Varkey M, Hijazi Z, Marino N, Ruiz C, Kronzon I. Use of real time three-dimensional transesophageal echocardiography in intracardiac catheter based interventions. J Am Soc Echocardiogr 2009;22:865–81
5. Hein R, Wuderlich N, Wilson N, Sievert H. New concepts in transcatheter closure of paravalvular leaks. Future Cardiol 2008;4:373–8
Video 1: Color full-volume real-time 3-dimensional (RT3D) transesophageal echocardiography (TEE) 7-beat gated reconstruction of the mitral valve (MV) and inferolateral crescentic paravalvular leak. This view replicates an anatomic view of the MV with the left atrium opened. Basal inferior and lateral segments are labeled for orientation.
Video 2: 2-dimensional color flow Doppler clips of: A, the paravalvular leak during before occlusion with an Amplatzer device; B, the paravalvular leak after occlusion with the second 12-mm Amplatzer occluder. There is a shadowing artifact along the interatrial septum, but a vena contracta width of 6.9 mm inferolaterally at the level of the sewing ring denotes a persistent severe leak. LA = left atrium; LV = left ventricle; AV = aortic valve.
Video 3: Color full-volume realtime 3-dimensional (RT3D) transesophageal echocardiography (TEE) 7-beat gated reconstruction of the mitral valve (MV) and inferolateral crescentic paravalvular after deployment of the second 12-mm Amplatzer occluder. The paravalvular leak was unchanged documenting procedural failure and inability of the circular Amplatzer to span the curvilinear defect. This view replicates an anatomic view of the MV with the left atrium opened. Basal inferior and lateral segments are labeled for orientation.
Clinician's Key Teaching Points By Martin M. Stechert, MD, Kent H. Rehfeldt, MD, and Martin J. London, MD
- Although paravalvular leaks are often crescent shaped, percutaneous closure with circular occluder devices such as the Amplatzer has been attempted with inconsistent results. Two-dimensional (2D) transesophageal echocardiography (TEE) may assist in the deployment of these devices despite difficulties relating to location of the catheter tip in single, often nonstandard imaging planes.
- Real-time 3D (RT3D) TEE offers several advantages over 2D TEE in guiding the delivery of percutaneous occluder devices including complete visualization of the catheter, improved geometric definition of the paravalvular defect, and the ability to provide a familiar, en face view of the mitral prosthesis and surrounding structures.
- RT3D ultrasound was used during percutaneous placement of an Amplatzer occluder device through a paravalvular mitral valve defect. RT3D was most effective during catheter navigation while gated 3D reconstruction was valuable illustrating a geometric mismatch between crescentic paravalvular defects and circular Amplatzer occluders.
- RT3D TEE offers advantages compared with 2D TEE during the percutaneous closure of paravalvular mitral regurgitation, and its use may result in a reduced use of fluoroscopy. However, RT3D TEE is not without limitations including the need to perform time-consuming image cropping, image dropout that may simulate anatomic defects, and the inability to obtain RT3D color Doppler imaging.