J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofCardiovascularCT.com
Pictorial Essay
CT imaging for left atrial appendage closure: A review and pictorial essay Tevfik Fehmi Ismail BSc(Hons), MBBS(Lond), PhD, MRCP(UK)a,b, Sandeep Panikker BSc(Hons), MBBS(Lond), MRCP(UK)b,c, Vias Markides MD, FRCPb,c, John P. Foran MD, FRCPb,c, Simon Padley FRCRa,b, Michael B. Rubens FRCRa,b, Tom Wong MD, FRCPb,c, Edward Nicol MD, MBA, DAvMed, MRCP, FACC, FSCCTa,b,* a
Radiology Department, Royal Brompton Hospital, Sydney Street, London SW3 6NP, United Kingdom National Heart and Lung Institute, Imperial College London, London, United Kingdom c Cardiac Electrophysiology Department, Royal Brompton Hospital, London, United Kingdom b
article info
abstract
Article history:
Cardioembolic stroke is an important complication of atrial fibrillation. The thrombus
Received 18 June 2014
responsible for this arises from the left atrial appendage (LAA) in >90% of cases, providing
Received in revised form
the rationale for device-based LAA closure as a means of thromboprophylaxis. Although
4 January 2015
oral anticoagulant therapy remains the mainstay for reducing the risk of stroke in patients
Accepted 16 January 2015
with atrial fibrillation, an increasing number of patients, particularly those ineligible for
Available online 24 January 2015
conventional pharmacotherapy, are being offered percutaneous left atrial appendage closure. Cardiovascular CT can provide important information to assess the suitability of
Keywords:
patients for LAA interventions and guide device selection and approach. The high spatial
Cardiovascular CT
resolution and multiplanar capability of contemporary contrast-enhanced gated multi-
Left Atrial Appendage
detector cardiovascular CT render it an ideal modality for noninvasively evaluating pa-
Atrial Fibrillation
tients before intervention and assessing patients after intervention both for complications
Stroke
and procedural outcome.
Watchman device
Crown Copyright ª 2015 Published by Elsevier Inc. on behalf of Society of Cardiovascular Computed Tomography. All rights reserved.
Lariat Amplatzer Cardiac Plug
1.
Background
Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia and is a frequent cause of cardioembolic stroke, particularly in the elderly.1 Although anticoagulation with warfarin, and more recently with novel orally active direct thrombin or factor Xa inhibitors, remains the
principal strategy for stroke prevention in these patients; a substantial number have either relative or absolute contraindications to oral anticoagulants or experience significant bleeding complications. In patients with nonvalvular AF, 90% of the thrombi responsible for stroke are thought to originate in the left atrial appendage (LAA).2,3 This provides a strong rationale for percutaneous deviceebased closure
Conflict of interest: The authors report no conflicts of interest. * Corresponding author. E-mail address: [email protected] (E. Nicol). 1934-5925/$ e see front matter Crown Copyright ª 2015 Published by Elsevier Inc. on behalf of Society of Cardiovascular Computed Tomography. All rights reserved. http://dx.doi.org/10.1016/j.jcct.2015.01.011
90
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
of the LAA as an alternative to oral anticoagulation. Such procedures may be undertaken independently or at the same time as ablation procedures to treat the underlying AF. The high spatial resolution and noninvasive multiplanar capability of contemporary contrast-enhanced electrocardiography-gated cardiovascular multidetector CT render it an important modality for evaluating patients being considered for such procedures and for assessing patients for procedural success and complications after intervention. This review describes the relevant protocols, clinical anatomy, and procedural factors pertinent to the cardiothoracic imager and provides a pragmatic approach to the use of CT to guide and assess the success of percutaneous LAA occlusion.
2.
Clinical and radiologic anatomy
The LAA usually arises as a finger-like projection from the left atrium and forms part of the left border of the heart, superior to the left ventricle and inferior to the main pulmonary trunk (Fig. 1). The tip or apex of the LAA can vary in position but usually points anteriorly and superiorly coming into close apposition with the proximal left anterior descending coronary artery, the proximal circumflex artery, and the pulmonary trunk.4 It may point inferiorly and posteriorly or behind the aorta into the transverse pericardial sinus. A number of different shapes have been described, but for device deployment purposes, the LAA can be considered as multilobed with an obvious bend (chicken wing morphology; Fig. 2A), single lobed without a bend (windsock morphology; Fig. 2B), multilobed without an obvious bend or dominant lobe (cauliflower; Fig. 2C), or multilobed without an obvious bend but with a dominant lobe (cactus; Fig. 2D).5 Different morphologies may be associated with different risks of thromboembolism, with the chicken wing morphology thought to confer the lowest risk.6 Although the endocardial surface of the LAA os is generally smooth, the interior of the appendage is covered with pectinate muscles which may cause apparent filling defects or pseudothrombus on CT and can also mimic the appearances of thrombus on transesophageal echocardiography (TEE). The wall of the LAA in between these pectinate muscles is often paper thin, rendering it at risk of perforation during catheter manipulation and device deployment.7 The LAA os may be either circular or elliptical and is separated from the origins of the pulmonary veins by the left lateral ridge (Fig. 3). The relation of the ellipsoid type to the left lateral rim of the atrium can make measurements of os dimensions challenging, particularly with echocardiographic techniques that lack the endocardial definition offered by CT. The spatial configuration of the LAA os in relation to the left superior pulmonary vein (LSPV) is also of paramount importance given the potential for devices to impair pulmonary venous return or interfere with future ablation procedures. This can particularly be an issue in the two-thirds of patients in whom the LAA and LSPV orifices are found in close proximity at the same level.5 Deployed devices may
Fig. 1 e Volume-rendered (top) and axial images (bottom) illustrating the location and anatomy of the left atrial appendage (LAA) and its apex. In this example, the LAA (black arrow) apex lies behind the main pulmonary trunk (asterisk), the so-called “retro-PA” configuration, and is related to the left anterior descending coronary artery (white arrow).
interfere with access to the LSPV for subsequent ablation procedures. In a fifth of cases, it is superior to the LSPV, and in the remainder, inferior.5 Other surrounding structures that
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
91
Fig. 2 e Axial (top) and oblique (bottom) images illustrating the CT appearances of the different left atrial appendage (asterisk) morphologies: (A) chicken wing, (B) windsock, (C) cauliflower, and (D) cactus types.
92
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 3 e Left atrial appendage os (asterisk) and its relationship to the left lateral ridge (black arrow) and the left superior pulmonary vein (white arrow).
may be at risk from device deployment include the left circumflex artery, the sinoatrial nodal artery where this arises from the left circumflex artery, and the great cardiac vein and its tributaries.8 The more anteriorly located left phrenic nerve also courses along the surface of the pericardium in close proximity to the base of the LAA and may be liable to injury, particularly with epicardial techniques for LAA occlusion.7,8 In addition to the main LAA, up to 15% of the population may have accessory LAAs or diverticula, which may also potentially serve as a nidus for thrombus formation or interfere with catheter deployment and manipulation, particularly during AF ablation procedures where they may be mistaken for pulmonary vein ostia.9,10 These are predominantly found in the anterior roof of the left atrium but have also been described posteriorly.9 An accessory LAA is defined as a structure with a clear ostium and irregular internal contours suggestive of pectinate muscles, whereas a left atrial (LA) diverticulum is a saclike out-pouching with a broad-based ostium and smooth internal contours.10 Although some reports have suggested these structures may serve as a source of thromboembolism,11 others have not confirmed this.12 Their presence need not therefore negate the value of LAA closure.
3.
Devices and procedural aspects
A number of LAA closure devices have been developed including the following: PLAATO (Percutaneous LAA transcatheter occlusion, Appriva Medical, Sunnyvale, CA) system, which is now obsolete.
Amplatzer Cardiac Plug (ACP, St Jude Medical, St Paul, MN), which is based on the same design used for atrial septal defect closure devices. It consists of an anchoring lobe which is deployed w10 to 15 mm from the LAA os and a proximal occlusive disc which covers it. The anchoring point or “landing zone” must be 10 mm in width, and a successfully implanted ACP device is capable of covering LAA os diameters ranging from 12.6 to 28.5 mm. Watchman device (Boston Scientific, Natick, MA), which is available in 5 sizes ranging from 21 to 33 mm in diameter. It consists of a self-expanding nitinol frame with fixation barbs and a permeable polyester covering able to occlude LAA os diameters ranging from 17 to 32 mm. It is deployed w10 to 20 mm from the os. The WaveCrest device (Coherex Medical, Salt Lake City, UT), is similar in principle to the Watchman. Clinical trials to obtain regulatory approval outside Europe are ongoing. Lariat epicardial suture-snare delivery device (SentreHEART, Redwood City, CA), which differs from all the aforementioned devices in that it is delivered via a combined percutaneous pericardial and trans-septal route.
3.1.
Principles of implantation
A trans-septal access sheath is advanced through the femoral vein into the left atrium via a trans-septal puncture under TEE or intracardiac echocardiography guidance for all devices. In most cases, the optimal site of trans-septal puncture should be located in a mid-to-low and posterior-septal location. The subsequent implantation procedure varies, but all endocardially delivered devices (ACP, Watchman, WaveCrest) involve deployment of an occluder. For the ACP, the device is secured first by deploying a distal anchoring lobe, before release of the proximal occluder. The other 2 devices are secured by fixation barbs attached to the occluder itself. With the Lariat device, magnet-tipped guide wires are placed in the LAA via trans-septal access and epicardially via pericardial access. The 2 magnet-tipped guide wires are connected to act as a guide for the Lariat delivery system over the lobulated distal components of the LAA toward its base. The Lariat device is then used to snare the LAA and in so doing, delivers a polyester suture which ligates the os. Examples illustrating the CT appearance of the devices currently in widespread clinical use are presented in Figure 4.
4. Image acquisition pre-AF ablation and LAA closure The images presented in this review were obtained at our institution using a dual-source scanner (SOMATOM Definition Flash; Siemens, Erlangen, Germany), but similar results can be obtained using other platforms.13 We perform a conventional topogram scout to localize the heart, followed by a contrast test bolus (15 mL, Omnipaque 350; GE Healthcare, Hatfield, United Kingdom) to assess the time taken to reach peak enhancement in the ascending aorta at the right pulmonary artery level. Breathhold image acquisition is then undertaken with prospective gating at end inspiration after the injection of 50 mL of contrast at 5
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
93
Fig. 4 e Axial and oblique images (top and bottom) illustrating the appearance of the left atrial appendage (asterisk). (A) Before Watchman device deployment; (B) after Watchman device implantation in the same patient; (C) after Amplatzer Cardiac Plug deployment; and (D) preintervention with (E) corresponding postintervention images after Lariat. The black arrows point to a device where present.
to 6 mL/s followed by 50 mL of 50:50 contrast and 0.9% saline mix as a bolus, and 25 mL of 0.9% saline flush. The initial rapid injection of contrast affords good opacification
of the left atrium and simultaneously allows coronary assessment. The mixed contrast-saline bolus opacifies the right atrium, allowing the interatrial septum to be clearly
94
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 5 e Contrast-enhanced CT images of 2 patients illustrating left atrial appendage (LAA; asterisk) filling defects: the first patient with pseudothrombus and the second patient with true thrombus. (A) Apparent filling defect in first patient. (B) Latepass imaging on first patient after 60-second delay confirms the absence of thrombus. (C) Example of a true LAA thrombus (arrow) in second patient on transaxial and (D) vertical long-axis images. In this instance, the filling defect is very well circumscribed throughout and surrounded on all sides by dense contrast, typical of the appearances of a definite thrombus. A second or late-pass acquisition was therefore not required.
visualized and providing data on right atrial anatomy which may be of use to the interventionist. A diluted contrast mixture is used to minimize contrast load and the impact of streak artifact on right coronary artery assessment, and the saline chaser is used to ensure a compact bolus is delivered to the heart. Images are acquired 3 seconds after peak enhancement in the ascending aorta and reconstructed with a 0.75-mm collimation and 0.5-mm slice thickness. Importantly, repeat imaging of a single block covering only the LAA is performed 60 seconds after the initial contrast bolus to help exclude pseudothrombus where required (Fig. 5). The latter protocol achieves 100% sensitivity and specificity for LAA thrombus detection relative to a gold standard to TEE.14 Image acquisition is typically gated in end diastole, but this can be varied according to heart rate, with a preference for end systole at rapid heart rates, particularly if information is also required about the coronary arteries. In our institution, the overall median (range) radiation dose for this protocol is 3.5 mSv (2.1e5.2 mSv), with a dose-length product of 251 mGy$cm (150e374 mGy$cm).
5.
Preprocedural CT evaluation
This focuses on identifying contraindications for LAA closure as well as providing images which may guide AF ablation (Table 1). It may also incorporate coronary artery assessment. For both AF ablation and the deployment of any LAA device, the presence of thrombus is an absolute contraindication. Thrombus is identified by the presence of filling defects within the LAA on initial contrast-enhanced images and confirmed on delayed imaging.14 The number and position of the pulmonary veins draining into the LA are described, together with the relationship between the LSPV and the LAA os. The latter can be at the same level, superior, or inferior to the LSPV (see Section 2). Because all devices require trans-septal puncture, any anomalies of the interatrial septum such as pre-existing defects, atrial septal aneurysm, or lipomatous hypertrophy of the interatrial septum should be described. For patients receiving the ACP device, a 10-mm anchor point approximately 10 to 15 mm from the os is required. The
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Table 1 e Typical Fields for Reporting Template Pre-LAA occlusion Left atrial size Pulmonary venous anatomy (number on each side and anatomy) Interatrial septal anatomy Distance from interatrial septum to LAA orifice LAA morphology (chicken wing, windsock, cauliflower, cactus) Maximum LAA diameter at device landing zone LAA length (dominant lobe) from the center of proposed landing zone LAA relationship to left pulmonary veins Accessory left atrial appendages or diverticula Thrombus (contraindication to ablation and/or device implantation) Presence of pericardial thickening or adhesions or calcification (for Lariat) Post-LAA device deployment Type and seating of prosthesis Presence, location, and size of any rim or leak around prosthesis Contrast enhancement within LAA Pulmonary venous return? Stenosis or compromise by device LAA, left atrial appendage.
Watchman device on the other hand is implanted at or 1 to 2 cm distal to the LAA os, with care being taken to ensure the length of the appendage along a clear line of sight from the proposed landing zone is at least as long as the diameter of the device selected. This is owing to the fact that the sheath-constrained length of the Watchman device is very similar to its unconstrained diameter once deployed. Significant LAA angulation between the os and main component may preclude successful device deployment and should therefore be highlighted if it limits the usable LAA length distal to the landing zone, that is, this length is less than the diameter of the LAA at the proposed landing zone. A retropulmonary artery (retro-PA) LAA may be particularly challenging in terms of catheter-based closure (Fig. 1) and precludes the use of Lariat. Examples of LAA os measurements are presented in Figure 6. The relationship of the os and anchoring or closure points to surrounding structures such as the posteriorly located pulmonary veins, the anteriorly located mitral annulus, and in particular, the left circumflex artery (Fig. 7) should be described as these may be rendered vulnerable to obstruction or injury with device deployment if in close proximity to the os or landing zones. Common exclusion criteria for the Lariat transpericardial approach include the presence of pericardial thickening, calcification, or adhesions precluding pericardial access (Fig. 8), a large appendage (LAA width >40 mm), a retro-PA (Fig. 1), multilobed appendages in which the lobes are >40 mm apart, pectus chest wall deformity, and a posteriorly rotated heart.15 In contrast to the other devices discussed, a key part of preprocedural evaluation involves assessment of potential pericardial access routes and the relationship with surrounding structures that may be liable to injury such as the left internal mammary artery, the inferior epigastric artery, and the phrenic nerve. Previous coronary artery bypass grafting is also a contraindication to the transpericardial Lariat LAA closure technique. Volume-rendered images
95
delineating the relevant structures may be of considerable utility to the operator (Fig. 9). Cardiovascular CT can also be used to provide operators with suitable fluoroscopic projections for device implantation.
6. Postprocedural evaluation and complications Early and late procedural success may also be assessed using contrast-enhanced cardiovascular CT. It can quickly identify device embolization and thrombus adherent to the device (Fig. 10) but can also be used to assess peridevice leaks. With perfect device deployment, following on from an appropriate period of endothelialization, in theory, there should be no contrast perfusing distal to the LAA occlusion device, on either the first-pass study or delayed images (Fig. 11). However, in clinical trials, adequate LAA occlusion has been defined based on TEE as the presence of 3 2 mm of residual peridevice gap, that is, up to a 5-mm gap around the device.16,17 Initial data suggest that leaks of this magnitude may not be associated with an increased risk of stroke, but a large number of the affected patients studied continued to receive anticoagulation, which could have artificially lowered event rates.17 Gap closure may increase over time as the device endothelializes.17 CT has a greater sensitivity for peridevice leak than TEE; however, the long-term significance of this remains uncertain.18 CT can reveal both a rim around the Watchman device (Fig. 12A,B), as well as flow distal to it where there is an incomplete seal but no obvious rim (Fig. 12C). Leaks around the ACP device can be similarly detected (Fig. 13). Patients treated with the Lariat may also experience incomplete LAA closure, which can readily be identified by CT (Fig. 14). Those with a peridevice gap or leak may potentially remain unprotected from thromboembolic stroke. However, the risk of this is likely to be related to the size of the leak. Given the greater sensitivity but lower specificity of cardiovascular CT relative to TEE,18 it may therefore potentially have a role as a screening test for device leak and thereby act as a gatekeeper for more intrusive TEE evaluation. Percutaneous LAA occlusion is occasionally performed at the same time as AF ablation. The latter may be complicated by pulmonary vein stenosis (Fig. 15). It is therefore important to look for pulmonary vein stenosis and to review images for signs of injury to structures related to the LAA (see Section 3).
7.
Conclusions
Cardiovascular CT allows the anatomy of the LAA and its adjacent structures to be assessed in exquisite detail. It can provide valuable information complementary to TEE for guiding device selection, assessing early procedural success, and determining medium-to long-term outcomes.19 We predict that in the future, cardiovascular CT will become the modality of choice for planning LAA closure and assessing long-term success. It may serve as a screening tool for incomplete occlusion of the LAA and other procedural
96
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 6 e Measurement of left atrial appendage os or landing zone dimensions for Watchman. (A) The left circumflex (black arrow) is identified in a transaxial plane. An oblique section is created based on this image, and it is oriented approximately perpendicular to the left circumflex (yellow line). This produces a vertical long-axis view of the heart (B) with the circumflex (black arrow) seen in cross-section beneath the left atrial appendage (asterisk) os. (C) For Watchman sizing, the landing zone is identified 10 to 20 mm from the os. (D) The plane of implantation is then measured as the distance from the circumflex to the edge of the landing zone. (E) The left atrial appendage (asterisk) is then viewed at this level in crosssection and (F) the maximum diameter is then determined for device sizing. In this example, it is 23.2 mm. (G) The distance between the landing zone and the apex of the primary lobe is measured as it represents the usable left atrial appendage length. For the Watchman device, this needs to be greater than or equal to the maximum landing zone diameter; otherwise, implantation is not technically feasible. (H) In this example, the maximum distance is 25.8 mm but implantation can also be performed in any other direction as illustrated provided that the distance from the midpoint of the landing zone plane to the tip of the appendage along a clear line of sight is greater than or equal to the maximum diameter at the landing zone. If the appendage is particularly angulated, there may be insufficient usable length to meet this requirement. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
97
Fig. 7 e Illustration of the close relationship of left circumflex coronary artery (white arrows) to the deployed Watchman device (black arrows).
complications, thereby potentially serving as a gatekeeper for more intrusive follow-up evaluation by TEE. 10.
Acknowledgments Images used in Figures 4C and 10 have been provided by Stephan Achenbach, University of Erlangen.
11.
references
12.
13. 1. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham study. Stroke. 1991;22:983e988. 2. Holmes Jr DR, Lakkireddy DR, Whitlock RP, Waksman R, Mack MJ. Left atrial appendage occlusion: opportunities and challenges. J Am Coll Cardiol. 2014;63:291e298. 3. Stoddard MF, Dawkins PR, Prince CR, Ammash NM. Left atrial appendage thrombus is not uncommon in patients with acute atrial fibrillation and a recent embolic event: a transesophageal echocardiographic study. J Am Coll Cardiol. 1995;25:452e459. 4. Ho SY, Cabrera JA, Sanchez-Quintana D. Left atrial anatomy revisited. Circ Arrhythm Electrophysiol. 2012;5:220e228. 5. Wang Y, Di Biase L, Horton RP, Nguyen T, Morhanty P, Natale A. Left atrial appendage studied by computed tomography to help planning for appendage closure device placement. J Cardiovasc Electrophysiol. 2010;21:973e982. 6. Di Biase L, Santangeli P, Anselmino M, et al. Does the left atrial appendage morphology correlate with the risk of stroke in patients with atrial fibrillation? Results from a multicenter study. J Am Coll Cardiol. 2012;60:531e538. 7. Su P, McCarthy KP, Ho SY. Occluding the left atrial appendage: anatomical considerations. Heart. 2008;94:1166e1170. 8. Cabrera JA, Saremi F, Sanchez-Quintana D. Left atrial appendage: anatomy and imaging landmarks pertinent to percutaneous transcatheter occlusion. Heart. 2014;100:1636e1650. 9. Wongcharoen W, Tsao HM, Wu MH, et al. Morphologic characteristics of the left atrial appendage, roof, and septum:
14.
15.
16.
17.
18.
19.
Implications for the ablation of atrial fibrillation. J Cardiovasc Electrophysiol. 2006;17:951e956. Lazoura O, Reddy T, Shriharan M, et al. Prevalence of left atrial anatomical abnormalities in patients with recurrent atrial fibrillation compared with patients in sinus rhythm using multi-slice CT. J Cardiovasc Comput Tomogr. 2012;6:268e273. Lee WJ, Chen SJ, Lin JL, Huang YH, Wang TD. Images in cardiovascular medicine. Accessory left atrial appendage: a neglected anomaly and potential cause of embolic stroke. Circulation. 2008;117:1351e1352. Ko JY, Kim YD, Hong YJ, et al. Lack of association between stroke and left atrial out-pouching structures: results of a case-control study. PLoS One. 2013;8:e76617. Lacomis JM, Goitein O, Deible C, et al. Dynamic multidimensional imaging of the human left atrial appendage. Europace. 2007;9:1134e1140. Lazoura O, Lindsay A, Shriharan M, et al. Dual-phase contrast enhanced CT avoids the need for TEE prior to radiofrequency ablation for atrial fibrillation. J Cardiovasc Comput Tomogr. 2013;7:S63. Bartus K, Han FT, Bednarek J, et al. Percutaneous left atrial appendage suture ligation using the lariat device in patients with atrial fibrillation: initial clinical experience. J Am Coll Cardiol. 2013;62:108e118. Landmesser U, Holmes Jr DR. Left atrial appendage closure: a percutaneous transcatheter approach for stroke prevention in atrial fibrillation. Eur Heart J. 2012;33:698e704. Viles-Gonzalez JF, Kar S, Douglas P, et al. The clinical impact of incomplete left atrial appendage closure with the Watchman device in patients with atrial fibrillation: a PROTECT AF (percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation) substudy. J Am Coll Cardiol. 2012;59:923e929. Viles-Gonzalez JF, Reddy VY, Petru J, et al. Incomplete occlusion of the left atrial appendage with the percutaneous left atrial appendage transcatheter occlusion device is not associated with increased risk of stroke. J Interv Card Electrophysiol. 2012;33:69e75. Qamruddin S, Shinbane J, Shriki J, Naqvi TZ. Left atrial appendage: structure, function, imaging modalities and therapeutic options. Expert Rev Cardiovasc Ther. 2010;8: 65e75.
98
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 8 e Pericardial thickening which may preclude the percutaneous epicardial access required with the Lariat device.
Fig. 9 e Volume-rendered images depicting the left atrial appendage (white arrow) and surrounding structures (left internal mammary artery, black arrow) in relation to skeletal landmarks. Such images may help operators plan pericardial access routes.
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
99
Fig. 10 e (A) Axial and (B) oblique images illustrating thrombus (arrow) on the surface of an Amplatzer Cardiac Plug device that has been used to close the left atrial appendage (asterisk). Images provided courtesy of Professor Stephan Achenbach.
Fig. 11 e (A) An example of a successfully deployed Watchman device (arrow) imaged 6 weeks after the procedure. The lack of contrast perfusion distal to the device suggests the absence of a leak. However, this may also reflect inadequate initial contrast equilibration. (B) Images obtained after a 60-second delay after contrast injection confirm no perfusion of the left atrial appendage. The delay between the first and second images provides time for adequate equilibration of contrast medium with any blood in the left atrial appendage, and allows definitive exclusion of a leak or incomplete endothelialization.
100
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 12 e An example of incomplete occlusion of the left atrial appendage with a clear rim of contrast (white arrow) between the device and the appendage os (A and B) and contrast enhancement in the distal appendage (C).
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
101
Fig. 13 e (A) Axial (top) and oblique (middle) images illustrating contrast enhancement in the distal LAA (asterisk) but no obvious gap or rim around the Amplatzer Cardiac Plug device (black arrows). The same patient also previously had an Amplatzer atrial septal occlusion device (bottom) deployed (white arrow) to close an interatrial septal defectda potential mechanism for stroke by paradoxical thromboembolism. (B) Axial (top) and oblique (bottom) images depicting incomplete occlusion of the LAA (asterisk) by the Amplatzer Cardiac Plug device (black arrow), with contrast seen distal to the device. The axial image reveals a clear gap between the device and the appendage os (white arrow).
102
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 14 e (A) Preintervention axial (top) and oblique images (bottom) of the LAA (asterisk) before treatment with the Lariat technique. (B) Axial (top) and oblique (bottom) images after intervention showing incomplete closure (arrow) of the LAA (asterisk).
Fig. 15 e Example of patient with successfully deployed Watchman device (white arrow) incidentally noted to have developed stenosis of the right middle pulmonary vein at follow-up (black arrow). Note also the close relationship between the device and the left circumflex artery (gray arrow).
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofCardiovascularCT.com
Pictorial Essay
CT imaging for left atrial appendage closure: A review and pictorial essay Tevfik Fehmi Ismail BSc(Hons), MBBS(Lond), PhD, MRCP(UK)a,b, Sandeep Panikker BSc(Hons), MBBS(Lond), MRCP(UK)b,c, Vias Markides MD, FRCPb,c, John P. Foran MD, FRCPb,c, Simon Padley FRCRa,b, Michael B. Rubens FRCRa,b, Tom Wong MD, FRCPb,c, Edward Nicol MD, MBA, DAvMed, MRCP, FACC, FSCCTa,b,* a
Radiology Department, Royal Brompton Hospital, Sydney Street, London SW3 6NP, United Kingdom National Heart and Lung Institute, Imperial College London, London, United Kingdom c Cardiac Electrophysiology Department, Royal Brompton Hospital, London, United Kingdom b
article info
abstract
Article history:
Cardioembolic stroke is an important complication of atrial fibrillation. The thrombus
Received 18 June 2014
responsible for this arises from the left atrial appendage (LAA) in >90% of cases, providing
Received in revised form
the rationale for device-based LAA closure as a means of thromboprophylaxis. Although
4 January 2015
oral anticoagulant therapy remains the mainstay for reducing the risk of stroke in patients
Accepted 16 January 2015
with atrial fibrillation, an increasing number of patients, particularly those ineligible for
Available online 24 January 2015
conventional pharmacotherapy, are being offered percutaneous left atrial appendage closure. Cardiovascular CT can provide important information to assess the suitability of
Keywords:
patients for LAA interventions and guide device selection and approach. The high spatial
Cardiovascular CT
resolution and multiplanar capability of contemporary contrast-enhanced gated multi-
Left Atrial Appendage
detector cardiovascular CT render it an ideal modality for noninvasively evaluating pa-
Atrial Fibrillation
tients before intervention and assessing patients after intervention both for complications
Stroke
and procedural outcome.
Watchman device
Crown Copyright ª 2015 Published by Elsevier Inc. on behalf of Society of Cardiovascular Computed Tomography. All rights reserved.
Lariat Amplatzer Cardiac Plug
1.
Background
Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia and is a frequent cause of cardioembolic stroke, particularly in the elderly.1 Although anticoagulation with warfarin, and more recently with novel orally active direct thrombin or factor Xa inhibitors, remains the
principal strategy for stroke prevention in these patients; a substantial number have either relative or absolute contraindications to oral anticoagulants or experience significant bleeding complications. In patients with nonvalvular AF, 90% of the thrombi responsible for stroke are thought to originate in the left atrial appendage (LAA).2,3 This provides a strong rationale for percutaneous deviceebased closure
Conflict of interest: The authors report no conflicts of interest. * Corresponding author. E-mail address: [email protected] (E. Nicol). 1934-5925/$ e see front matter Crown Copyright ª 2015 Published by Elsevier Inc. on behalf of Society of Cardiovascular Computed Tomography. All rights reserved. http://dx.doi.org/10.1016/j.jcct.2015.01.011
90
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
of the LAA as an alternative to oral anticoagulation. Such procedures may be undertaken independently or at the same time as ablation procedures to treat the underlying AF. The high spatial resolution and noninvasive multiplanar capability of contemporary contrast-enhanced electrocardiography-gated cardiovascular multidetector CT render it an important modality for evaluating patients being considered for such procedures and for assessing patients for procedural success and complications after intervention. This review describes the relevant protocols, clinical anatomy, and procedural factors pertinent to the cardiothoracic imager and provides a pragmatic approach to the use of CT to guide and assess the success of percutaneous LAA occlusion.
2.
Clinical and radiologic anatomy
The LAA usually arises as a finger-like projection from the left atrium and forms part of the left border of the heart, superior to the left ventricle and inferior to the main pulmonary trunk (Fig. 1). The tip or apex of the LAA can vary in position but usually points anteriorly and superiorly coming into close apposition with the proximal left anterior descending coronary artery, the proximal circumflex artery, and the pulmonary trunk.4 It may point inferiorly and posteriorly or behind the aorta into the transverse pericardial sinus. A number of different shapes have been described, but for device deployment purposes, the LAA can be considered as multilobed with an obvious bend (chicken wing morphology; Fig. 2A), single lobed without a bend (windsock morphology; Fig. 2B), multilobed without an obvious bend or dominant lobe (cauliflower; Fig. 2C), or multilobed without an obvious bend but with a dominant lobe (cactus; Fig. 2D).5 Different morphologies may be associated with different risks of thromboembolism, with the chicken wing morphology thought to confer the lowest risk.6 Although the endocardial surface of the LAA os is generally smooth, the interior of the appendage is covered with pectinate muscles which may cause apparent filling defects or pseudothrombus on CT and can also mimic the appearances of thrombus on transesophageal echocardiography (TEE). The wall of the LAA in between these pectinate muscles is often paper thin, rendering it at risk of perforation during catheter manipulation and device deployment.7 The LAA os may be either circular or elliptical and is separated from the origins of the pulmonary veins by the left lateral ridge (Fig. 3). The relation of the ellipsoid type to the left lateral rim of the atrium can make measurements of os dimensions challenging, particularly with echocardiographic techniques that lack the endocardial definition offered by CT. The spatial configuration of the LAA os in relation to the left superior pulmonary vein (LSPV) is also of paramount importance given the potential for devices to impair pulmonary venous return or interfere with future ablation procedures. This can particularly be an issue in the two-thirds of patients in whom the LAA and LSPV orifices are found in close proximity at the same level.5 Deployed devices may
Fig. 1 e Volume-rendered (top) and axial images (bottom) illustrating the location and anatomy of the left atrial appendage (LAA) and its apex. In this example, the LAA (black arrow) apex lies behind the main pulmonary trunk (asterisk), the so-called “retro-PA” configuration, and is related to the left anterior descending coronary artery (white arrow).
interfere with access to the LSPV for subsequent ablation procedures. In a fifth of cases, it is superior to the LSPV, and in the remainder, inferior.5 Other surrounding structures that
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
91
Fig. 2 e Axial (top) and oblique (bottom) images illustrating the CT appearances of the different left atrial appendage (asterisk) morphologies: (A) chicken wing, (B) windsock, (C) cauliflower, and (D) cactus types.
92
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 3 e Left atrial appendage os (asterisk) and its relationship to the left lateral ridge (black arrow) and the left superior pulmonary vein (white arrow).
may be at risk from device deployment include the left circumflex artery, the sinoatrial nodal artery where this arises from the left circumflex artery, and the great cardiac vein and its tributaries.8 The more anteriorly located left phrenic nerve also courses along the surface of the pericardium in close proximity to the base of the LAA and may be liable to injury, particularly with epicardial techniques for LAA occlusion.7,8 In addition to the main LAA, up to 15% of the population may have accessory LAAs or diverticula, which may also potentially serve as a nidus for thrombus formation or interfere with catheter deployment and manipulation, particularly during AF ablation procedures where they may be mistaken for pulmonary vein ostia.9,10 These are predominantly found in the anterior roof of the left atrium but have also been described posteriorly.9 An accessory LAA is defined as a structure with a clear ostium and irregular internal contours suggestive of pectinate muscles, whereas a left atrial (LA) diverticulum is a saclike out-pouching with a broad-based ostium and smooth internal contours.10 Although some reports have suggested these structures may serve as a source of thromboembolism,11 others have not confirmed this.12 Their presence need not therefore negate the value of LAA closure.
3.
Devices and procedural aspects
A number of LAA closure devices have been developed including the following: PLAATO (Percutaneous LAA transcatheter occlusion, Appriva Medical, Sunnyvale, CA) system, which is now obsolete.
Amplatzer Cardiac Plug (ACP, St Jude Medical, St Paul, MN), which is based on the same design used for atrial septal defect closure devices. It consists of an anchoring lobe which is deployed w10 to 15 mm from the LAA os and a proximal occlusive disc which covers it. The anchoring point or “landing zone” must be 10 mm in width, and a successfully implanted ACP device is capable of covering LAA os diameters ranging from 12.6 to 28.5 mm. Watchman device (Boston Scientific, Natick, MA), which is available in 5 sizes ranging from 21 to 33 mm in diameter. It consists of a self-expanding nitinol frame with fixation barbs and a permeable polyester covering able to occlude LAA os diameters ranging from 17 to 32 mm. It is deployed w10 to 20 mm from the os. The WaveCrest device (Coherex Medical, Salt Lake City, UT), is similar in principle to the Watchman. Clinical trials to obtain regulatory approval outside Europe are ongoing. Lariat epicardial suture-snare delivery device (SentreHEART, Redwood City, CA), which differs from all the aforementioned devices in that it is delivered via a combined percutaneous pericardial and trans-septal route.
3.1.
Principles of implantation
A trans-septal access sheath is advanced through the femoral vein into the left atrium via a trans-septal puncture under TEE or intracardiac echocardiography guidance for all devices. In most cases, the optimal site of trans-septal puncture should be located in a mid-to-low and posterior-septal location. The subsequent implantation procedure varies, but all endocardially delivered devices (ACP, Watchman, WaveCrest) involve deployment of an occluder. For the ACP, the device is secured first by deploying a distal anchoring lobe, before release of the proximal occluder. The other 2 devices are secured by fixation barbs attached to the occluder itself. With the Lariat device, magnet-tipped guide wires are placed in the LAA via trans-septal access and epicardially via pericardial access. The 2 magnet-tipped guide wires are connected to act as a guide for the Lariat delivery system over the lobulated distal components of the LAA toward its base. The Lariat device is then used to snare the LAA and in so doing, delivers a polyester suture which ligates the os. Examples illustrating the CT appearance of the devices currently in widespread clinical use are presented in Figure 4.
4. Image acquisition pre-AF ablation and LAA closure The images presented in this review were obtained at our institution using a dual-source scanner (SOMATOM Definition Flash; Siemens, Erlangen, Germany), but similar results can be obtained using other platforms.13 We perform a conventional topogram scout to localize the heart, followed by a contrast test bolus (15 mL, Omnipaque 350; GE Healthcare, Hatfield, United Kingdom) to assess the time taken to reach peak enhancement in the ascending aorta at the right pulmonary artery level. Breathhold image acquisition is then undertaken with prospective gating at end inspiration after the injection of 50 mL of contrast at 5
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
93
Fig. 4 e Axial and oblique images (top and bottom) illustrating the appearance of the left atrial appendage (asterisk). (A) Before Watchman device deployment; (B) after Watchman device implantation in the same patient; (C) after Amplatzer Cardiac Plug deployment; and (D) preintervention with (E) corresponding postintervention images after Lariat. The black arrows point to a device where present.
to 6 mL/s followed by 50 mL of 50:50 contrast and 0.9% saline mix as a bolus, and 25 mL of 0.9% saline flush. The initial rapid injection of contrast affords good opacification
of the left atrium and simultaneously allows coronary assessment. The mixed contrast-saline bolus opacifies the right atrium, allowing the interatrial septum to be clearly
94
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 5 e Contrast-enhanced CT images of 2 patients illustrating left atrial appendage (LAA; asterisk) filling defects: the first patient with pseudothrombus and the second patient with true thrombus. (A) Apparent filling defect in first patient. (B) Latepass imaging on first patient after 60-second delay confirms the absence of thrombus. (C) Example of a true LAA thrombus (arrow) in second patient on transaxial and (D) vertical long-axis images. In this instance, the filling defect is very well circumscribed throughout and surrounded on all sides by dense contrast, typical of the appearances of a definite thrombus. A second or late-pass acquisition was therefore not required.
visualized and providing data on right atrial anatomy which may be of use to the interventionist. A diluted contrast mixture is used to minimize contrast load and the impact of streak artifact on right coronary artery assessment, and the saline chaser is used to ensure a compact bolus is delivered to the heart. Images are acquired 3 seconds after peak enhancement in the ascending aorta and reconstructed with a 0.75-mm collimation and 0.5-mm slice thickness. Importantly, repeat imaging of a single block covering only the LAA is performed 60 seconds after the initial contrast bolus to help exclude pseudothrombus where required (Fig. 5). The latter protocol achieves 100% sensitivity and specificity for LAA thrombus detection relative to a gold standard to TEE.14 Image acquisition is typically gated in end diastole, but this can be varied according to heart rate, with a preference for end systole at rapid heart rates, particularly if information is also required about the coronary arteries. In our institution, the overall median (range) radiation dose for this protocol is 3.5 mSv (2.1e5.2 mSv), with a dose-length product of 251 mGy$cm (150e374 mGy$cm).
5.
Preprocedural CT evaluation
This focuses on identifying contraindications for LAA closure as well as providing images which may guide AF ablation (Table 1). It may also incorporate coronary artery assessment. For both AF ablation and the deployment of any LAA device, the presence of thrombus is an absolute contraindication. Thrombus is identified by the presence of filling defects within the LAA on initial contrast-enhanced images and confirmed on delayed imaging.14 The number and position of the pulmonary veins draining into the LA are described, together with the relationship between the LSPV and the LAA os. The latter can be at the same level, superior, or inferior to the LSPV (see Section 2). Because all devices require trans-septal puncture, any anomalies of the interatrial septum such as pre-existing defects, atrial septal aneurysm, or lipomatous hypertrophy of the interatrial septum should be described. For patients receiving the ACP device, a 10-mm anchor point approximately 10 to 15 mm from the os is required. The
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Table 1 e Typical Fields for Reporting Template Pre-LAA occlusion Left atrial size Pulmonary venous anatomy (number on each side and anatomy) Interatrial septal anatomy Distance from interatrial septum to LAA orifice LAA morphology (chicken wing, windsock, cauliflower, cactus) Maximum LAA diameter at device landing zone LAA length (dominant lobe) from the center of proposed landing zone LAA relationship to left pulmonary veins Accessory left atrial appendages or diverticula Thrombus (contraindication to ablation and/or device implantation) Presence of pericardial thickening or adhesions or calcification (for Lariat) Post-LAA device deployment Type and seating of prosthesis Presence, location, and size of any rim or leak around prosthesis Contrast enhancement within LAA Pulmonary venous return? Stenosis or compromise by device LAA, left atrial appendage.
Watchman device on the other hand is implanted at or 1 to 2 cm distal to the LAA os, with care being taken to ensure the length of the appendage along a clear line of sight from the proposed landing zone is at least as long as the diameter of the device selected. This is owing to the fact that the sheath-constrained length of the Watchman device is very similar to its unconstrained diameter once deployed. Significant LAA angulation between the os and main component may preclude successful device deployment and should therefore be highlighted if it limits the usable LAA length distal to the landing zone, that is, this length is less than the diameter of the LAA at the proposed landing zone. A retropulmonary artery (retro-PA) LAA may be particularly challenging in terms of catheter-based closure (Fig. 1) and precludes the use of Lariat. Examples of LAA os measurements are presented in Figure 6. The relationship of the os and anchoring or closure points to surrounding structures such as the posteriorly located pulmonary veins, the anteriorly located mitral annulus, and in particular, the left circumflex artery (Fig. 7) should be described as these may be rendered vulnerable to obstruction or injury with device deployment if in close proximity to the os or landing zones. Common exclusion criteria for the Lariat transpericardial approach include the presence of pericardial thickening, calcification, or adhesions precluding pericardial access (Fig. 8), a large appendage (LAA width >40 mm), a retro-PA (Fig. 1), multilobed appendages in which the lobes are >40 mm apart, pectus chest wall deformity, and a posteriorly rotated heart.15 In contrast to the other devices discussed, a key part of preprocedural evaluation involves assessment of potential pericardial access routes and the relationship with surrounding structures that may be liable to injury such as the left internal mammary artery, the inferior epigastric artery, and the phrenic nerve. Previous coronary artery bypass grafting is also a contraindication to the transpericardial Lariat LAA closure technique. Volume-rendered images
95
delineating the relevant structures may be of considerable utility to the operator (Fig. 9). Cardiovascular CT can also be used to provide operators with suitable fluoroscopic projections for device implantation.
6. Postprocedural evaluation and complications Early and late procedural success may also be assessed using contrast-enhanced cardiovascular CT. It can quickly identify device embolization and thrombus adherent to the device (Fig. 10) but can also be used to assess peridevice leaks. With perfect device deployment, following on from an appropriate period of endothelialization, in theory, there should be no contrast perfusing distal to the LAA occlusion device, on either the first-pass study or delayed images (Fig. 11). However, in clinical trials, adequate LAA occlusion has been defined based on TEE as the presence of 3 2 mm of residual peridevice gap, that is, up to a 5-mm gap around the device.16,17 Initial data suggest that leaks of this magnitude may not be associated with an increased risk of stroke, but a large number of the affected patients studied continued to receive anticoagulation, which could have artificially lowered event rates.17 Gap closure may increase over time as the device endothelializes.17 CT has a greater sensitivity for peridevice leak than TEE; however, the long-term significance of this remains uncertain.18 CT can reveal both a rim around the Watchman device (Fig. 12A,B), as well as flow distal to it where there is an incomplete seal but no obvious rim (Fig. 12C). Leaks around the ACP device can be similarly detected (Fig. 13). Patients treated with the Lariat may also experience incomplete LAA closure, which can readily be identified by CT (Fig. 14). Those with a peridevice gap or leak may potentially remain unprotected from thromboembolic stroke. However, the risk of this is likely to be related to the size of the leak. Given the greater sensitivity but lower specificity of cardiovascular CT relative to TEE,18 it may therefore potentially have a role as a screening test for device leak and thereby act as a gatekeeper for more intrusive TEE evaluation. Percutaneous LAA occlusion is occasionally performed at the same time as AF ablation. The latter may be complicated by pulmonary vein stenosis (Fig. 15). It is therefore important to look for pulmonary vein stenosis and to review images for signs of injury to structures related to the LAA (see Section 3).
7.
Conclusions
Cardiovascular CT allows the anatomy of the LAA and its adjacent structures to be assessed in exquisite detail. It can provide valuable information complementary to TEE for guiding device selection, assessing early procedural success, and determining medium-to long-term outcomes.19 We predict that in the future, cardiovascular CT will become the modality of choice for planning LAA closure and assessing long-term success. It may serve as a screening tool for incomplete occlusion of the LAA and other procedural
96
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 6 e Measurement of left atrial appendage os or landing zone dimensions for Watchman. (A) The left circumflex (black arrow) is identified in a transaxial plane. An oblique section is created based on this image, and it is oriented approximately perpendicular to the left circumflex (yellow line). This produces a vertical long-axis view of the heart (B) with the circumflex (black arrow) seen in cross-section beneath the left atrial appendage (asterisk) os. (C) For Watchman sizing, the landing zone is identified 10 to 20 mm from the os. (D) The plane of implantation is then measured as the distance from the circumflex to the edge of the landing zone. (E) The left atrial appendage (asterisk) is then viewed at this level in crosssection and (F) the maximum diameter is then determined for device sizing. In this example, it is 23.2 mm. (G) The distance between the landing zone and the apex of the primary lobe is measured as it represents the usable left atrial appendage length. For the Watchman device, this needs to be greater than or equal to the maximum landing zone diameter; otherwise, implantation is not technically feasible. (H) In this example, the maximum distance is 25.8 mm but implantation can also be performed in any other direction as illustrated provided that the distance from the midpoint of the landing zone plane to the tip of the appendage along a clear line of sight is greater than or equal to the maximum diameter at the landing zone. If the appendage is particularly angulated, there may be insufficient usable length to meet this requirement. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
97
Fig. 7 e Illustration of the close relationship of left circumflex coronary artery (white arrows) to the deployed Watchman device (black arrows).
complications, thereby potentially serving as a gatekeeper for more intrusive follow-up evaluation by TEE. 10.
Acknowledgments Images used in Figures 4C and 10 have been provided by Stephan Achenbach, University of Erlangen.
11.
references
12.
13. 1. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham study. Stroke. 1991;22:983e988. 2. Holmes Jr DR, Lakkireddy DR, Whitlock RP, Waksman R, Mack MJ. Left atrial appendage occlusion: opportunities and challenges. J Am Coll Cardiol. 2014;63:291e298. 3. Stoddard MF, Dawkins PR, Prince CR, Ammash NM. Left atrial appendage thrombus is not uncommon in patients with acute atrial fibrillation and a recent embolic event: a transesophageal echocardiographic study. J Am Coll Cardiol. 1995;25:452e459. 4. Ho SY, Cabrera JA, Sanchez-Quintana D. Left atrial anatomy revisited. Circ Arrhythm Electrophysiol. 2012;5:220e228. 5. Wang Y, Di Biase L, Horton RP, Nguyen T, Morhanty P, Natale A. Left atrial appendage studied by computed tomography to help planning for appendage closure device placement. J Cardiovasc Electrophysiol. 2010;21:973e982. 6. Di Biase L, Santangeli P, Anselmino M, et al. Does the left atrial appendage morphology correlate with the risk of stroke in patients with atrial fibrillation? Results from a multicenter study. J Am Coll Cardiol. 2012;60:531e538. 7. Su P, McCarthy KP, Ho SY. Occluding the left atrial appendage: anatomical considerations. Heart. 2008;94:1166e1170. 8. Cabrera JA, Saremi F, Sanchez-Quintana D. Left atrial appendage: anatomy and imaging landmarks pertinent to percutaneous transcatheter occlusion. Heart. 2014;100:1636e1650. 9. Wongcharoen W, Tsao HM, Wu MH, et al. Morphologic characteristics of the left atrial appendage, roof, and septum:
14.
15.
16.
17.
18.
19.
Implications for the ablation of atrial fibrillation. J Cardiovasc Electrophysiol. 2006;17:951e956. Lazoura O, Reddy T, Shriharan M, et al. Prevalence of left atrial anatomical abnormalities in patients with recurrent atrial fibrillation compared with patients in sinus rhythm using multi-slice CT. J Cardiovasc Comput Tomogr. 2012;6:268e273. Lee WJ, Chen SJ, Lin JL, Huang YH, Wang TD. Images in cardiovascular medicine. Accessory left atrial appendage: a neglected anomaly and potential cause of embolic stroke. Circulation. 2008;117:1351e1352. Ko JY, Kim YD, Hong YJ, et al. Lack of association between stroke and left atrial out-pouching structures: results of a case-control study. PLoS One. 2013;8:e76617. Lacomis JM, Goitein O, Deible C, et al. Dynamic multidimensional imaging of the human left atrial appendage. Europace. 2007;9:1134e1140. Lazoura O, Lindsay A, Shriharan M, et al. Dual-phase contrast enhanced CT avoids the need for TEE prior to radiofrequency ablation for atrial fibrillation. J Cardiovasc Comput Tomogr. 2013;7:S63. Bartus K, Han FT, Bednarek J, et al. Percutaneous left atrial appendage suture ligation using the lariat device in patients with atrial fibrillation: initial clinical experience. J Am Coll Cardiol. 2013;62:108e118. Landmesser U, Holmes Jr DR. Left atrial appendage closure: a percutaneous transcatheter approach for stroke prevention in atrial fibrillation. Eur Heart J. 2012;33:698e704. Viles-Gonzalez JF, Kar S, Douglas P, et al. The clinical impact of incomplete left atrial appendage closure with the Watchman device in patients with atrial fibrillation: a PROTECT AF (percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation) substudy. J Am Coll Cardiol. 2012;59:923e929. Viles-Gonzalez JF, Reddy VY, Petru J, et al. Incomplete occlusion of the left atrial appendage with the percutaneous left atrial appendage transcatheter occlusion device is not associated with increased risk of stroke. J Interv Card Electrophysiol. 2012;33:69e75. Qamruddin S, Shinbane J, Shriki J, Naqvi TZ. Left atrial appendage: structure, function, imaging modalities and therapeutic options. Expert Rev Cardiovasc Ther. 2010;8: 65e75.
98
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 8 e Pericardial thickening which may preclude the percutaneous epicardial access required with the Lariat device.
Fig. 9 e Volume-rendered images depicting the left atrial appendage (white arrow) and surrounding structures (left internal mammary artery, black arrow) in relation to skeletal landmarks. Such images may help operators plan pericardial access routes.
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
99
Fig. 10 e (A) Axial and (B) oblique images illustrating thrombus (arrow) on the surface of an Amplatzer Cardiac Plug device that has been used to close the left atrial appendage (asterisk). Images provided courtesy of Professor Stephan Achenbach.
Fig. 11 e (A) An example of a successfully deployed Watchman device (arrow) imaged 6 weeks after the procedure. The lack of contrast perfusion distal to the device suggests the absence of a leak. However, this may also reflect inadequate initial contrast equilibration. (B) Images obtained after a 60-second delay after contrast injection confirm no perfusion of the left atrial appendage. The delay between the first and second images provides time for adequate equilibration of contrast medium with any blood in the left atrial appendage, and allows definitive exclusion of a leak or incomplete endothelialization.
100
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 12 e An example of incomplete occlusion of the left atrial appendage with a clear rim of contrast (white arrow) between the device and the appendage os (A and B) and contrast enhancement in the distal appendage (C).
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
101
Fig. 13 e (A) Axial (top) and oblique (middle) images illustrating contrast enhancement in the distal LAA (asterisk) but no obvious gap or rim around the Amplatzer Cardiac Plug device (black arrows). The same patient also previously had an Amplatzer atrial septal occlusion device (bottom) deployed (white arrow) to close an interatrial septal defectda potential mechanism for stroke by paradoxical thromboembolism. (B) Axial (top) and oblique (bottom) images depicting incomplete occlusion of the LAA (asterisk) by the Amplatzer Cardiac Plug device (black arrow), with contrast seen distal to the device. The axial image reveals a clear gap between the device and the appendage os (white arrow).
102
J o u r n a l o f C a r d i o v a s c u l a r C o m p u t e d T o m o g r a p h y 9 ( 2 0 1 5 ) 8 9 e1 0 2
Fig. 14 e (A) Preintervention axial (top) and oblique images (bottom) of the LAA (asterisk) before treatment with the Lariat technique. (B) Axial (top) and oblique (bottom) images after intervention showing incomplete closure (arrow) of the LAA (asterisk).
Fig. 15 e Example of patient with successfully deployed Watchman device (white arrow) incidentally noted to have developed stenosis of the right middle pulmonary vein at follow-up (black arrow). Note also the close relationship between the device and the left circumflex artery (gray arrow).
