Atrial fibrillation (AF), the most common cardiac arrhythmia, is a significant cause of embolic stroke. Although systemic anticoagulation is the primary strategy for preventing the thromboembolic complications of AF, anticoagulants carry major bleeding risks, and many patients have contraindications to their use. Because thromboembolism typically arises from a clot in the left atrial appendage (LAA), local therapeutic alternatives to systemic anticoagulation involving surgical or percutaneous exclusion of the LAA have been developed. Surgical exclusion of the LAA is typically performed only as an adjunct to other cardiac surgeries, thus limiting the number of eligible patients. Furthermore, surgical exclusion of the LAA is frequently incomplete, and thromboembolism may still occur. Percutaneous LAA exclusion includes two approaches: transseptal delivery of an occlusion device to the LAA and epicardial suture ligation of the LAA, the LARIAT procedure. In the LARIAT procedure, a pretied snare is placed around the epicardial surface of the LAA orifice via pericardial access. Proper snare placement is achieved with epicardial and endocardial magnet-tipped guidewires. The endocardial wire is advanced transvenously to the LAA apex after transseptal puncture. The epicardial wire, introduced into the pericardial space, achieves end-to-end union with the endocardial wire at the LAA apex. The snare is then placed over the LAA, tightened, and sutured. On the basis of early clinical experience, the LARIAT procedure has a high success rate of LAA exclusion with low risk for complications. The authors describe the indispensable role of real-time transesophageal echocardiography in the guidance of LAA epicardial suture ligation with the LARIAT device.
Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in the world and is estimated to affect >3 million people in the United States. The increased prevalence of AF with age, combined with an aging population, creates a projected increased incidence of AF in the United States to 7.56 million by 2050.
Systemic thromboembolism is the major complication of both valvular and nonvalvular AF. The left atrial appendage (LAA) is the most common site of thrombus formation, accounting for 91% of left heart thrombi in patients with nonrheumatic AF and 57% of thrombi in patients with rheumatic AF.
Systemic anticoagulation is the primary means of preventing thromboembolism in patients with AF. Antiplatelet agents are an alternative to systemic anticoagulation but with inferior efficacy. Adequate anticoagulation with oral warfarin has been demonstrated to cut the risk for stroke and systemic embolism by 67% compared with placebo and by 45% compared with aspirin. Newer anticoagulants (such as dabigatran, apixaban, and rivaroxaban) have been shown to be at least noninferior to warfarin.
However, all anticoagulants have significant bleeding risk; the risk for major bleeding (generally defined as a reduction in the hemoglobin level of ≥20 g/L, transfusion of ≥2 U of packed red cells, or symptomatic bleeding occurring at a critical site or resulting in death) with either warfarin or newer agents is estimated at 1.4% to >3% per year. Because AF-associated systemic thromboembolism typically arises from a clot confined to the LAA, local therapeutic alternatives to systemic antithrombotic and antiplatelet therapy have been developed. These alternatives include either surgical or percutaneous exclusion of the LAA from the systemic circulation.
Surgical techniques of LAA exclusion have included ligation, clipping, stapling, and amputation. However, only a small number of patients with AF are eligible for these procedures because surgical LAA exclusion is typically performed only as an adjunct to other cardiac operative interventions.
Although prophylactic exclusion of the LAA in patients undergoing mitral valve surgery and/or maze procedure is recommended to reduce systemic thromboembolic events, such exclusion is frequently incomplete, and residual communication with the body of the left atrium due to incomplete LAA exclusion can paradoxically increase the risk for thrombus formation in the LAA.
Percutaneous alternatives to surgical LAA exclusion include transseptal delivery of various LAA occlusion devices, such as the PLAATO (eV3, Plymouth, MN), Watchman (Boston Scientific, Maple Grove, MN) or Amplatzer Cardiac Plug (St. Jude Medical, Minneapolis, MN), and transpericardial suture ligation of the LAA using the LARIAT device (SentreHEART, Palo Alto, CA).
Outcomes data are most numerous for the Watchman device, which was found to be noninferior to chronic warfarin therapy in a randomized trial. However, Watchman device implantation was associated with procedural complications, including pericardial effusion, device embolization, and procedure-related stroke. Furthermore, after device implantation, patients typically require warfarin therapy for 45 days and dual-antiplatelet therapy (with aspirin and clopidogrel) for 6 months to prevent clot formation during device endothelialization.
In contrast, percutaneous LAA closure with the LARIAT device, which includes an epicardial suture, does not leave any device in contact with the bloodstream and thus does not typically require postprocedural warfarin therapy. The LARIAT procedure, which may also be referred to as the permanent ligation, approximation, closure, and exclusion procedure, is an option in patients with contraindications or intolerance to anticoagulation. On the basis of short-term observational data, the LARIAT procedure is feasible, but proof of its long-term efficacy in a randomized trial is still lacking.
We emphasize that LARIAT Suture Delivery Device is not specifically approved for LAA ligation and that there is a paucity of outcomes data. Its approved indication is defined as facilitating suture placement and knot tying for use in surgical applications in which soft tissue is being approximated and/or ligated with a pretied polyester suture. However, the LARIAT procedure has entered clinical practice.
In this review, we discuss the crucial role of echocardiography (including real-time three-dimensional [3D] transesophageal echocardiographic [TEE] imaging) in periprocedural guidance of the LARIAT procedure and the relationship of echocardiography to other imaging modalities, such as fluoroscopy and computed tomographic (CT) imaging.
Clinical Experience with the LARIAT Procedure
Thus far, there are reports of three observational studies with the LARIAT device. Patients were enrolled in these studies if they had AF, had CHADS 2 scores of ≥1 or ≥2, and demonstrated contraindications to or failure of anticoagulation. Patients were excluded if they had prior cardiac surgery, a myocardial infarction within 3 months, embolic events within 30 days, or histories of pericarditis. Additional exclusion criteria were related to LAA anatomy: superior orientation of the LAA with the LAA apex positioned behind the main pulmonary artery and/or LAA width > 40 mm.
In an initial nonrandomized single-center trial, LAA ligation with the LARIAT device was successful in 96% of patients (85 of 89) when assessed by TEE imaging immediately after the procedure and in 98% at 1 year of the 65 patients who completed 1-year TEE follow-up. Another trial reported successful LAA exclusion in all 20 patients who underwent the procedure. A third study had an acute procedural success rate of 92.6% (25 of 27 subjects); the LAA remained excluded in all 22 patients who completed 45-day TEE follow-up. Although these are encouraging results, long-term data on the complete LAA closure rate by the LARIAT procedure in a larger group of patients are still unavailable.
These observational studies showed low rates of periprocedural complications, the most common being pericarditis and pericardial effusion. Two cases of intraprocedural right ventricular perforation and one of LAA perforation were also reported in these studies.
The LARIAT Procedure in a Nutshell
The LARIAT procedure consists of two parts based on access: (1) an endocardial (transvenous) portion and (2) an epicardial (transpericardial) portion. The endocardial portion entails transvenous access (typically through a femoral vein) to the right atrium with subsequent transseptal puncture and delivery of the endocardial magnet-tipped wire across the interatrial septum into the tip of the LAA.
The epicardial portion involves transthoracic pericardial access in the subxiphoid region and delivery of the epicardial magnet-tipped wire to the apex of the LAA to create an end-to-end magnetic union with the endocardial wire. There is no direct physical contact between the two wires because there is interposition of the LAA wall between them. The procedure ends with the placement of a pretied epicardial suture over the ostium of the LAA using the LARIAT device. It is important to emphasize that the LAA ostium is here defined from the procedural point of view and refers to the location of LAA ligation by the LARIAT procedure. It is similar to the location of the LAA orifice occluded by percutaneous closure devices such as the Watchman. This orifice is more distal than the true anatomic LAA orifice, because the area of the ligament of Marshall (the “Coumadin ridge”) typically cannot be ligated. The procedural LAA orifice is located at the level of the left circumflex artery and the coronary sinus.
In animal studies, postmortem histologic examination revealed complete endothelialization of the sutured LAA orifice as early as 7 days after the procedure.
The LARIAT Procedure in a Nutshell
The LARIAT procedure consists of two parts based on access: (1) an endocardial (transvenous) portion and (2) an epicardial (transpericardial) portion. The endocardial portion entails transvenous access (typically through a femoral vein) to the right atrium with subsequent transseptal puncture and delivery of the endocardial magnet-tipped wire across the interatrial septum into the tip of the LAA.
The epicardial portion involves transthoracic pericardial access in the subxiphoid region and delivery of the epicardial magnet-tipped wire to the apex of the LAA to create an end-to-end magnetic union with the endocardial wire. There is no direct physical contact between the two wires because there is interposition of the LAA wall between them. The procedure ends with the placement of a pretied epicardial suture over the ostium of the LAA using the LARIAT device. It is important to emphasize that the LAA ostium is here defined from the procedural point of view and refers to the location of LAA ligation by the LARIAT procedure. It is similar to the location of the LAA orifice occluded by percutaneous closure devices such as the Watchman. This orifice is more distal than the true anatomic LAA orifice, because the area of the ligament of Marshall (the “Coumadin ridge”) typically cannot be ligated. The procedural LAA orifice is located at the level of the left circumflex artery and the coronary sinus.
In animal studies, postmortem histologic examination revealed complete endothelialization of the sutured LAA orifice as early as 7 days after the procedure.
Role of Multimodality Imaging Including Two-Dimensional and 3D TEE during the LARIAT Procedure
The preparations and stages of the procedure have been previously described. Briefly, after clinical evaluation, potential candidates for the LARIAT procedure undergo contrast-enhanced chest CT imaging. If eligible, they then undergo the LARIAT procedure under fluoroscopic and TEE guidance.
CT scanning is used to ascertain the LAA anatomy, including its orientation and orifice size ( Figure 1 , Video 1 ; available at www.onlinejase.com ). Unfavorable LAA anatomy ( Figure 2 , Video 2 ; available at www.onlinejase.com ) includes a large LAA size (diameter > 40 mm) and a superiorly oriented LAA with the apex positioned behind the pulmonary trunk. Such LAA anatomy may make passage of the LARIAT snare over the LAA difficult.
In addition to standard cross-sectional two-dimensional (2D) imaging, 3D CT reconstruction is performed to provide a more detailed view of LAA morphology and its relationship to surrounding structures. This CT reconstruction will also help in guiding subsequent pericardial access during the LARIAT procedure; in particular, 3D reconstructions display the relationship between the sternum and the myocardium. This information will allow the clinician to determine how steeply and how far the pericardial needle should be inserted.
Once a patient is deemed eligible, the LARIAT procedure is typically done under general anesthesia to minimize patient discomfort. In a sterile fashion, the subxiphoid region (for epicardial access) and the femoral vein region (for transseptal access) are prepped and draped.
TEE guidance in conjunction with fluoroscopy is essential for successful completion of the LARIAT procedure, as well as for monitoring for any periprocedural complications. Any modern ultrasound system with a 2D TEE multiplane probe may be used to monitor the LARIAT procedure. If available, 3D TEE imaging provides additional details of LAA anatomy and enhances visualization of wires, balloons, and catheters. Three-dimensional TEE imaging may overcome many of the limitations of 2D TEE imaging related to the tomographic nature of 2D imaging: intracardiac wires, balloons, and catheters move in a 3D space, and frequently their courses are outside 2D imaging planes. In general, compared with 2D TEE imaging, 3D TEE imaging provides better visualization of the intracardiac course of the procedural hardware, particularly the catheter and wire tips.
For 3D TEE imaging, a commercially available ultrasound system using a matrix-array 3D TEE probe may be used. Of the 3D TEE imaging modalities, biplane and 3D zoom imaging appear the most helpful. Biplane imaging is particularly useful during and may enhance the safety of the transseptal puncture compared with 2D TEE. Three-dimensional zoom imaging provides intuitive en face views of cardiac structures, facilitating procedural guidance aside from transseptal puncture. We have not found that full-volume and so-called live 3D imaging are essential for LARIAT procedural guidance.
Examination of LAA Anatomy
TEE imaging is used to confirm the findings of the preprocedural CT scan with respect to LAA orifice diameter, LAA apex orientation, and the number of LAA lobes. TEE imaging is the modality of choice for visualization of LAA thrombi, as previously described. As previously noted, the following are the current exclusion criteria for the LARIAT procedure: LAA width > 40 mm, superiorly oriented LAA with the apex behind the pulmonary trunk, and LAA thrombus. All 2D and 3D echocardiographic LAA measurements are done at ventricular end-systole.
On 2D TEE imaging, the LAA should be imaged at multiple angles, typically 0°, 45°, 90°, and 135°. Measurements of the LAA orifice are performed at all imaging angles to determine the maximum diameter. Because of the tomographic nature of 2D imaging, one cannot be certain that 2D TEE orifice diameter measurements are done in the same plane. This can be overcome by 3D TEE imaging, either by multiplane reconstruction or using the en face 3D zoom technique. Using the multiplane reconstruction mode ( Figure 3 ), the two long axes of the LAA are aligned to visualize the short-axis plane of the LAA, in which precise measurements of LAA diameters are performed.
Compared to multiplane reconstruction, the 3D TEE zoom technique provides a simpler way to directly visualize the short-axis en face view of the LAA. Images may be cropped along the x, y, and z axes to eliminate nonrelevant structures and more clearly visualize the LAA. Details of the 3D zoom techniques are provided elsewhere. With the newest generation of 3D TEE software, LAA diameter can be measured online from en face images of the LAA orifice.
Two-dimensional and 3D TEE imaging can also provide information on the shape and the area of the LAA orifice, as well as the depth of the LAA. Interestingly, orifice shape and area are not as important for the LARIAT procedure as they are for LAA closure techniques using LAA occlusion devices such as the Watchman.
Two-dimensional imaging in multiple planes and careful cropping of 3D images also provide detailed information on the number and orientation of LAA lobes.
Pericardial Access
Pericardial access is the next step. It is achieved as previously described under fluoroscopic guidance, typically using a 17-gauge Touhy epidural needle (Hakko, Nagano, Japan). The needle is inserted retrosternally in an anterolateral direction along the anterior epicardial surface of the heart and pointed toward the apex of the LAA ( Figure 4 A). During needle insertion, a minimal amount of radiographic contrast is injected to confirm fluoroscopically the needle’s location in the pericardium versus the anterior mediastinum. A 0.035-inch wire (FindrWIRZ; SentreHEART) is then advanced through the needle into the anterior pericardial space. Fluoroscopy is again used to confirm the presence of the wire in the pericardial space ( Figure 4 B). This wire is left in the pericardial space while transseptal catheterization is achieved.
