Catheterization and Cardiovascular Interventions 85:297–303 (2015)
VALVULAR AND STRUCTURAL HEART DISEASES Original Studies Left Atrial Appendage Closure Using the Amulet Device: An Initial Experience with the Second Generation Amplatzer Cardiac Plug Simon Cheung Chi Lam, MD, Stefan Bertog, MD, Sameer Gafoor, MD, Laura Vaskelyte, MD, Patrick Boehm, Raymond Wei Jian Ho, Jennifer Franke, MD, Ilona Hofmann, MD, and Horst Sievert,* MD, PhD Objectives: Aim of this study was to demonstrate the feasibility, safety, and short-term outcome of left atrial appendage (LAA) closure with a new generation LAA closure device. Background: The Amulet device (AGA, St Jude Medical, Minneapolis, MN) is a new generation of the amplatzer cardiac plug (ACP), specifically designed for LAA closure. This new version is designed to facilitate the implantation process and minimize procedural or device-related complications. Methods: The device was implanted in 17 patients with nonvalvular atrial fibrillation (AF). Clinical data were obtained at baseline, during the procedure, at discharge, at 30 and 90 days. Results: All devices were implanted successfully. Device sizes ranged from 20 mm to 31 mm. A 12 French (Fr) or 14 Fr delivery sheath was used depending on the selected device size. Full and partial recapture was performed in 1 case and 3 cases, respectively. There was 1 procedurerelated pericardial effusion successfully managed with pericardiocentesis. There was no device embolization. The mean length of stay was 2.1 6 0.3 days. At 90 days, there were no deaths, strokes, systemic thromboembolism, or bleeding complications. There was no device-related thrombus or pericardial effusion at 90-day TEE. In 2 of the 17 patients minimal peridevice flow (smaller than 2 mm) was present. Conclusions: The Amulet device, which has new novel features as compared with the first generation ACP, is a feasible option for LAA closure. From our initial experience, implantation of the Amulet is associated with high success rate and good short-term outcome. VC 2014 Wiley Periodicals, Inc. Key words: left atrial appendage closure; structural heart intervention; new device
CardioVascular Center Frankfurt, Seckbacher Landstrasse 65, Frankfurt, Germany Conflict of interest: Drs. Lam, Bertog, Gafoor, Vaskelyte, Franke, Hofmann, Mr. Boehm and Mr. Ho have no conflict of interest to disclose. Dr. Sievert’s institution has ownership interest in or has received consulting fees, travel expenses or study honoraries from the following companies: Abbott, Access Closure, AGA, Angiomed, Arstasis, Atritech, Atrium, Avinger, Bard, Boston Scientific, Bridgepoint, Cardiac Dimensions, CardioKinetix, CardioMEMS, Coherex, Contego, CSI, EndoCross, EndoTex, Epitek, Evalve, ev3, FlowCardia, Gore, Guidant, Guided Delivery Systems, Inc., InSeal Medical, Lumen Biomedical, HLT, Kensey Nash, Kyoto Medical, Lifetech, Lutonix, Medinol, Medtronic, NDC, NMT, OAS, Occlutech, C 2014 Wiley Periodicals, Inc. V
Osprey, Ovalis, Pathway, PendraCare, Percardia, pfm Medical, Rox Medical, Sadra, SJM, Sorin, Spectranetics, SquareOne, Trireme, Trivascular, Velocimed, Veryan. *Correspondence to: H. Sievert, CardioVascular Center Frankfurt, Seckbacher Landstrasse 65, 60389 Frankfurt, Germany. E-mail: [email protected] Received 7 October 2013; Revision accepted 17 August 2014 DOI: 10.1002/ccd.25644 Published online 26 August 2014 in Wiley Online Library (wileyonlinelibrary.com)
298
Lam et al.
INTRODUCTION
Catheter-based left atrial appendage (LAA) closure was shown to be noninferior to warfarin therapy in patients with nonvalvular atrial fibrillation (AF) in the PROTECT AF Trial [1]. It can be considered as an alternative to anticoagulation therapy especially in patients with high bleeding risk. Recent studies suggested continuously improving procedural safety and implantation success [2]. It had also been reported that LAA closure may be superior to warfarin therapy with extended follow-up period [3]. Since the first successful case of LAA closure with the PLAATO system [4], a number of device designs have emerged. Currently, the Watchman Device (Boston Scientific, Natrick, MA) and the Amplatzer Cardiac Plug (ACP) (AGA, St Jude Medical, Minneapolis, MN) are the 2 devices with the largest clinical experiences worldwide. The ACP consists of two major components: a distal lobe to anchor to the body of the LAA and a proximal disc to seal the ostium of the LAA [5]. Registries on the use of ACP for LAA closure demonstrated high implantation success [6–9]. The Amulet device is the second generation of the ACP. It was specifically designed to facilitate the implantation process and to minimize procedural or device-related complications. First-in-man implantation of the Amulet device was reported by Freixa et al. in 2013 [10]. We report our initial experience with the Amulet device for LAA closure. The purpose of this study is to assess the feasibility, safety and short-term outcome of LAA closure using this new system. METHODS
LAA occlusion with the Amulet device was performed in 17 patients with nonvalvular AF between February and May 2013. Patients with high stroke risk (defined as CHADS2 equal to or greater than 1) who were not suitable for anticoagulation therapy were screened for eligibility for the procedure. All patients underwent pre-procedural transesophageal echocardiography (TEE). All patients provided written informed consent for the procedure. Clinical data were reviewed retrospectively from clinical records, procedural logs, and dedicated technical worksheets. Clinical and echocardiographic information was obtained at baseline, during the procedure, at discharge, at 30 days and at 90 days. Device and Sizing
The main concept of the Amulet device (Fig. 1) is based on the ACP. It consists of a distal lobe that anchors the device to the LAA body and a proximal
Fig. 1. The Amulet device. The Amulet device consists of a distal lobe and a proximal disc connected by a flexible waist. The whole delivery system is preloaded. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
TABLE I. Specifications and Sizing Information of the Amulet Device Device size (mm)
Lobe length (mm)
Disc diameter (mm)
Waist length (mm)
11–13 13–15 15–17 17–19
16 18 20 22
7.5
Lobe þ 6
5.5
19–22 22–25 25–28 28–31
25 28 31 34
10
Maximun landing zone width (mm)
Sheath diameter (Fr)
12
Lobe þ 7
8 14
disc that seals the LAA ostium. The device is made of a nitinol mesh with two polyester patches sewn onto the two components, which are linked by a connecting waist. The Amulet device was specifically modified for LAA closure in several aspects. First, it is preloaded within the delivery system facilitating preparation. Second, the inner 0.014 inches core wire of the delivery pusher cable is made movable and remains connected to the device when the delivery core is retracted. This enables assessment of the device position without tension while allowing recapture if the position is suboptimal. Third, compared with the ACP, the lengths of the lobe and waist are longer. This provides more flexibility and, thereby better conformation to angles between the LAA body and ostium. Fourth, the proximal disc diameter is 6 or 7 mm larger than that of the lobe
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
LAA Occlusion with the Amulet Device TABLE II.
Baseline Characteristics (N 5 17)
Age (mean) Sex Male Female History of stroke or TIA Yes No CHADS2 Score 1 2 3 4 Rhythm at implantation AF SR Pacemaker rhythm Landing zone diameter (mm) Minimum Maximum Mean Number of LAA lobes 1 2 Multiple Concomitant PFO Yes No Previous LAA occlusion Yes No HAS-BLED score 2 3 4 5 Previous antithrombotic therapy Aspirin/Clopidogrel Warfarin/NOAC DAPT None Indications for procedure Absolute contraindications to anticoagulation due to risk of bleeding Relative contraindications to anticoagulation due to risk of bleeding
TABLE III. 66.8 6 13.0 11 (65%) 6 (35%) 5 (29%) 12 (71%) 1 12 3 1
(6%) (70%) (18%) (6%)
11 (65%) 5 (29%) 1 (6%) 16.7 28.7 22.2 6 4.1 4 (24%) 8 (47%) 5 (29%) 3 (18%) 14 (82%) 1 (6%) 16 (94%) 2 11 3 1
(12%) (65%) (18%) (6%)
7 7 1 2
(41%) (41%) (6%) (12%)
6 (35%)
11 (65%)
TIA, transient ischaemic attack; AF, atrial fibrillation; SR, sinus rhythm; LAA, left atrial appendage; PFO, patent foramen ovale; NOAC, novel oral anticoagulants; DAPT, double antiplatelet therapy.
(compared with 4 or 6 mm with ACP) depending on the size of the device. This may reduce the residual leak incidence. Fifth, there are larger sizes available (31 and 34 mm). Sixth, the number of stabilizing wires surrounding the device body is increased in the larger device sizes potentially reducing the embolization risk. Finally, the proximal end-screw is recessed perhaps lowering the incidence of device associated thrombi.
299
Procedure Details (N 5 17)
Device size (mm) 16 18 20 22 25 28 31 34 Delivery sheath (Fr)—double-curved TorqVue 45 –45 12 14 Transseptal puncture Yes No Implantation success Recapture of device Full Partial Periprocedureal complication Device embolization Cardiac perforation/Pericardial effusion Stroke MI Length of stay (mean number of days)
0 (0%) 1 (6%) 2 (12%) 3 (18%) 4 (23%) 2 (12%) 5 (29%) 0 (0%) 12 (71%) 5 (29%) 14 (82%) 3 (18%) 17 (100%) 1 (6%) 3 (18%) 0 (0%) 1 (6%) 0 (0%) 0 (0%) 2.1 6 0.3
MI, myocardial infarction.
Landing zone measurement was based on fluoroscopy and TEE. Device specifications and sizing recommendations of the Amulet device are illustrated in Table I. Delivery Sheath
Amulet devices were implanted using the doublecurved TorqVue 45 –45 sheath (AGA, St Jude Medical, Minneapolis, MN). For each case we specifically modified the curvature with reference to relative positions of the transseptal puncture site and orientation of the LAA. Twelve Fr sheaths were used for devices with size 16, 18, 20, 22, 25, and 28 mm. Fourteen Fr sheaths were selected for devices with size 31 and 34 mm (Table I). Procedure
All procedures were performed under local anesthesia. Femoral venous access was obtained. Preprocedural TEE was performed to rule out intracardiac thrombi and assess for concomitant patent foramen ovale (PFO). LAA morphology and dimensions (including landing zone diameter, ostial diameter, and length of LAA) were measured in different angles (0 , 45 , 90 , and 135 ). Intravenous heparin was administered. The implantation procedure was guided by fluoroscopy and TEE. In patients with concomitant PFO, left atrial access was established via the PFO (without transseptal puncture).
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
300
Lam et al.
Fig. 2. Case example illustrating recapture of device and maneuvers to optimize device positioning. (a–c) The proximal disc was partially recaptured into the delivery sheath and redeployed in attempt to optimize the position. However, it was still suboptimal. (d–f) Slight traction was applied on the delivery cable and the disc was repositioned. The final posi-
tion was good with complete sealing of the LAA ostium. The movable core safety wire remained connected to the device and the final position of the device could be evaluated while it was free from tension of the pusher cable (red arrow) before final release. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
After confirming proper position using TEE and fluoroscopy, the Amulet Device can be deployed. Successful deployment is determined by 5 criteria: (1) proper device lobe alignment with the LAA; (2) adequate lobe compression and apposition to LAA wall; (3) presence of a concave-shaped disc; (4) separation of the disc from the lobe, and (5) lobe position by at least two thirds distal to left circumflex artery on TEE. The tug test was performed to test for device stability. Patients were monitored overnight and discharged with a regimen of clopidogrel 75 mg for 3 months and aspirin for a minimum of 6 months after implantation.
history of stroke or transient ischaemic attack (TIA). With the exception of one patient, all patients had CHADS2 scores of equal to or greater than 2. Eleven patients (65%) had AF at the time of procedure. The mean landing zone diameter was 22.2 6 4.1 mm. Transseptal puncture was performed in 14 patients (82%). Three devices were implanted via concomitant PFOs. Implantation success was 100%. Full device recapture was performed in 1 case. In this case, to optimize device position a smaller Amulet device was used. In three other cases the device was partially recaptured and redeployed to optimize the final position (Fig. 2). There was no device embolization, periprocedural stroke, or myocardial infarction (MI). Pericardial effusion occurred in 1 patient (6%) 1 hr after completion of the procedure and the patient underwent pericardiocentesis without further sequaelae. The average length of hospital stay was 2.1 6 0.3 days. All patients were admitted one day before the
RESULTS
Baseline clinical characteristics and procedural details are outlined in Tables II and III. The mean age was 66.8 6 13 years. There were 11 males (65%) and 6 females (35%). Five (29%) had a
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
LAA Occlusion with the Amulet Device
301
Fig. 3. Case example illustrating closure of previous incomplete LAA occlusion with the Amulet device. (a) A 5 Fr Pigtail catheter was passed through the residual gap for selective angiogram. Contrast injection reviewed significant residual gap and the uncovered lobule (blue triangle). (b) Deployment of the distal lobe. (c) Deployment of the proximal disc. (d) The final angiogram showed sealing of the LAA with the Amulet device.
procedure. After the procedure they would stay for observation for one night and discharged on the following day. One patient was observed for one extra night because of occurrence of pericardial effusion and another patient stayed in the hospital for one more night due to logistic reason (patient traveled from long distance for the procedure and stayed one more night before long travel). Of note, a 28-mm Amulet Device was used in a patient with incomplete LAA occlusion after previous Watchman (Boston Scientific, Natrick, MA) implantation 3 years prior (Fig. 3). In this patient, a residual uncovered lobule and a 7 mm wide leak had been present. The Amulet device was considered because of the potential advantage of a larger proximal disc diameter
to improve sealing of the residual leak and uncovered lobe. The new movable inner 0.014 inches core wire allowed a more precise assessment of the configuration and position of the device after deployment. When repositioning was required, the delivery coil of the pusher cable could be advanced towards the end screw of the device to cover the inner core wire, which facilitated recapture of the proximal disc in order to redeploy in the most optimal position. At 90-days, there were no deaths, strokes, systemic thromboembolism, myocardial infarctions, or additional bleeding complications. TEE showed no deviceassociated thrombi or pericardial effusions. A peridevice leak was identified in 2 patients (12%) with widths small than 2 mm.
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
302
Lam et al.
DISCUSSION
We report our experience with the new generation ACP for LAA closure. The implantation process is similar to that of the first generation ACP. The Amulet device was designed to improve the ease of preparation, facilitate the implantation process, and minimize procedure- or device-related complications. The new system allows a more reliable assessment of final device configuration, orientation, and position free from tension of the delivery cable while allowing recapture as long as the floppy 0.014 inches wire is still in place. In our case series 1 case required full recapture and three cases required partial recapture. All were successful. In previous reports of LAA occlusion using the ACP device, the incidence of pericardial effusion or hemorrhage varied between 1.1 and 3.5% [7–9]. One case in our series developed a postprocedural pericardial effusion requiring pericardiocentesis (1 hr after completion of the procedure) without further sequelae. The implantation had been performed through a PFO (transseptal puncture was not performed). No device recapture after first deployment had been necessary. Pericardial hemorrhage might have been related to sheath or device manipulations within the LAA during the implantation process but the exact cause remains uncertain. Theoretically, a larger number of stabilizing wires may be associated with a higher risk of pericardial hemorrhage; however, a larger registry would be necessary to determine the incidence of pericardial effusion as compared to the ACP or other devices. The reported incidence of device-related thrombus formation with the ACP ranges between 2.4 and 14% and the reported thrombus formation rate in PROTECT-AF was around 4% [2,11]. Concerns have been raised that the protruding end screw on the ACP may be thrombogenic. This screw is now embedded within the disc perhaps lowering the thrombogenic potential. In this context, though no device related thrombus was detected in our series, the number of patients was too small to allow conclusions whether the new design, indeed, prevents thrombus formation. Two of our patients had minor peridevice leaks at 90 days. The magnitude of these leaks (
VALVULAR AND STRUCTURAL HEART DISEASES Original Studies Left Atrial Appendage Closure Using the Amulet Device: An Initial Experience with the Second Generation Amplatzer Cardiac Plug Simon Cheung Chi Lam, MD, Stefan Bertog, MD, Sameer Gafoor, MD, Laura Vaskelyte, MD, Patrick Boehm, Raymond Wei Jian Ho, Jennifer Franke, MD, Ilona Hofmann, MD, and Horst Sievert,* MD, PhD Objectives: Aim of this study was to demonstrate the feasibility, safety, and short-term outcome of left atrial appendage (LAA) closure with a new generation LAA closure device. Background: The Amulet device (AGA, St Jude Medical, Minneapolis, MN) is a new generation of the amplatzer cardiac plug (ACP), specifically designed for LAA closure. This new version is designed to facilitate the implantation process and minimize procedural or device-related complications. Methods: The device was implanted in 17 patients with nonvalvular atrial fibrillation (AF). Clinical data were obtained at baseline, during the procedure, at discharge, at 30 and 90 days. Results: All devices were implanted successfully. Device sizes ranged from 20 mm to 31 mm. A 12 French (Fr) or 14 Fr delivery sheath was used depending on the selected device size. Full and partial recapture was performed in 1 case and 3 cases, respectively. There was 1 procedurerelated pericardial effusion successfully managed with pericardiocentesis. There was no device embolization. The mean length of stay was 2.1 6 0.3 days. At 90 days, there were no deaths, strokes, systemic thromboembolism, or bleeding complications. There was no device-related thrombus or pericardial effusion at 90-day TEE. In 2 of the 17 patients minimal peridevice flow (smaller than 2 mm) was present. Conclusions: The Amulet device, which has new novel features as compared with the first generation ACP, is a feasible option for LAA closure. From our initial experience, implantation of the Amulet is associated with high success rate and good short-term outcome. VC 2014 Wiley Periodicals, Inc. Key words: left atrial appendage closure; structural heart intervention; new device
CardioVascular Center Frankfurt, Seckbacher Landstrasse 65, Frankfurt, Germany Conflict of interest: Drs. Lam, Bertog, Gafoor, Vaskelyte, Franke, Hofmann, Mr. Boehm and Mr. Ho have no conflict of interest to disclose. Dr. Sievert’s institution has ownership interest in or has received consulting fees, travel expenses or study honoraries from the following companies: Abbott, Access Closure, AGA, Angiomed, Arstasis, Atritech, Atrium, Avinger, Bard, Boston Scientific, Bridgepoint, Cardiac Dimensions, CardioKinetix, CardioMEMS, Coherex, Contego, CSI, EndoCross, EndoTex, Epitek, Evalve, ev3, FlowCardia, Gore, Guidant, Guided Delivery Systems, Inc., InSeal Medical, Lumen Biomedical, HLT, Kensey Nash, Kyoto Medical, Lifetech, Lutonix, Medinol, Medtronic, NDC, NMT, OAS, Occlutech, C 2014 Wiley Periodicals, Inc. V
Osprey, Ovalis, Pathway, PendraCare, Percardia, pfm Medical, Rox Medical, Sadra, SJM, Sorin, Spectranetics, SquareOne, Trireme, Trivascular, Velocimed, Veryan. *Correspondence to: H. Sievert, CardioVascular Center Frankfurt, Seckbacher Landstrasse 65, 60389 Frankfurt, Germany. E-mail: [email protected] Received 7 October 2013; Revision accepted 17 August 2014 DOI: 10.1002/ccd.25644 Published online 26 August 2014 in Wiley Online Library (wileyonlinelibrary.com)
298
Lam et al.
INTRODUCTION
Catheter-based left atrial appendage (LAA) closure was shown to be noninferior to warfarin therapy in patients with nonvalvular atrial fibrillation (AF) in the PROTECT AF Trial [1]. It can be considered as an alternative to anticoagulation therapy especially in patients with high bleeding risk. Recent studies suggested continuously improving procedural safety and implantation success [2]. It had also been reported that LAA closure may be superior to warfarin therapy with extended follow-up period [3]. Since the first successful case of LAA closure with the PLAATO system [4], a number of device designs have emerged. Currently, the Watchman Device (Boston Scientific, Natrick, MA) and the Amplatzer Cardiac Plug (ACP) (AGA, St Jude Medical, Minneapolis, MN) are the 2 devices with the largest clinical experiences worldwide. The ACP consists of two major components: a distal lobe to anchor to the body of the LAA and a proximal disc to seal the ostium of the LAA [5]. Registries on the use of ACP for LAA closure demonstrated high implantation success [6–9]. The Amulet device is the second generation of the ACP. It was specifically designed to facilitate the implantation process and to minimize procedural or device-related complications. First-in-man implantation of the Amulet device was reported by Freixa et al. in 2013 [10]. We report our initial experience with the Amulet device for LAA closure. The purpose of this study is to assess the feasibility, safety and short-term outcome of LAA closure using this new system. METHODS
LAA occlusion with the Amulet device was performed in 17 patients with nonvalvular AF between February and May 2013. Patients with high stroke risk (defined as CHADS2 equal to or greater than 1) who were not suitable for anticoagulation therapy were screened for eligibility for the procedure. All patients underwent pre-procedural transesophageal echocardiography (TEE). All patients provided written informed consent for the procedure. Clinical data were reviewed retrospectively from clinical records, procedural logs, and dedicated technical worksheets. Clinical and echocardiographic information was obtained at baseline, during the procedure, at discharge, at 30 days and at 90 days. Device and Sizing
The main concept of the Amulet device (Fig. 1) is based on the ACP. It consists of a distal lobe that anchors the device to the LAA body and a proximal
Fig. 1. The Amulet device. The Amulet device consists of a distal lobe and a proximal disc connected by a flexible waist. The whole delivery system is preloaded. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
TABLE I. Specifications and Sizing Information of the Amulet Device Device size (mm)
Lobe length (mm)
Disc diameter (mm)
Waist length (mm)
11–13 13–15 15–17 17–19
16 18 20 22
7.5
Lobe þ 6
5.5
19–22 22–25 25–28 28–31
25 28 31 34
10
Maximun landing zone width (mm)
Sheath diameter (Fr)
12
Lobe þ 7
8 14
disc that seals the LAA ostium. The device is made of a nitinol mesh with two polyester patches sewn onto the two components, which are linked by a connecting waist. The Amulet device was specifically modified for LAA closure in several aspects. First, it is preloaded within the delivery system facilitating preparation. Second, the inner 0.014 inches core wire of the delivery pusher cable is made movable and remains connected to the device when the delivery core is retracted. This enables assessment of the device position without tension while allowing recapture if the position is suboptimal. Third, compared with the ACP, the lengths of the lobe and waist are longer. This provides more flexibility and, thereby better conformation to angles between the LAA body and ostium. Fourth, the proximal disc diameter is 6 or 7 mm larger than that of the lobe
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
LAA Occlusion with the Amulet Device TABLE II.
Baseline Characteristics (N 5 17)
Age (mean) Sex Male Female History of stroke or TIA Yes No CHADS2 Score 1 2 3 4 Rhythm at implantation AF SR Pacemaker rhythm Landing zone diameter (mm) Minimum Maximum Mean Number of LAA lobes 1 2 Multiple Concomitant PFO Yes No Previous LAA occlusion Yes No HAS-BLED score 2 3 4 5 Previous antithrombotic therapy Aspirin/Clopidogrel Warfarin/NOAC DAPT None Indications for procedure Absolute contraindications to anticoagulation due to risk of bleeding Relative contraindications to anticoagulation due to risk of bleeding
TABLE III. 66.8 6 13.0 11 (65%) 6 (35%) 5 (29%) 12 (71%) 1 12 3 1
(6%) (70%) (18%) (6%)
11 (65%) 5 (29%) 1 (6%) 16.7 28.7 22.2 6 4.1 4 (24%) 8 (47%) 5 (29%) 3 (18%) 14 (82%) 1 (6%) 16 (94%) 2 11 3 1
(12%) (65%) (18%) (6%)
7 7 1 2
(41%) (41%) (6%) (12%)
6 (35%)
11 (65%)
TIA, transient ischaemic attack; AF, atrial fibrillation; SR, sinus rhythm; LAA, left atrial appendage; PFO, patent foramen ovale; NOAC, novel oral anticoagulants; DAPT, double antiplatelet therapy.
(compared with 4 or 6 mm with ACP) depending on the size of the device. This may reduce the residual leak incidence. Fifth, there are larger sizes available (31 and 34 mm). Sixth, the number of stabilizing wires surrounding the device body is increased in the larger device sizes potentially reducing the embolization risk. Finally, the proximal end-screw is recessed perhaps lowering the incidence of device associated thrombi.
299
Procedure Details (N 5 17)
Device size (mm) 16 18 20 22 25 28 31 34 Delivery sheath (Fr)—double-curved TorqVue 45 –45 12 14 Transseptal puncture Yes No Implantation success Recapture of device Full Partial Periprocedureal complication Device embolization Cardiac perforation/Pericardial effusion Stroke MI Length of stay (mean number of days)
0 (0%) 1 (6%) 2 (12%) 3 (18%) 4 (23%) 2 (12%) 5 (29%) 0 (0%) 12 (71%) 5 (29%) 14 (82%) 3 (18%) 17 (100%) 1 (6%) 3 (18%) 0 (0%) 1 (6%) 0 (0%) 0 (0%) 2.1 6 0.3
MI, myocardial infarction.
Landing zone measurement was based on fluoroscopy and TEE. Device specifications and sizing recommendations of the Amulet device are illustrated in Table I. Delivery Sheath
Amulet devices were implanted using the doublecurved TorqVue 45 –45 sheath (AGA, St Jude Medical, Minneapolis, MN). For each case we specifically modified the curvature with reference to relative positions of the transseptal puncture site and orientation of the LAA. Twelve Fr sheaths were used for devices with size 16, 18, 20, 22, 25, and 28 mm. Fourteen Fr sheaths were selected for devices with size 31 and 34 mm (Table I). Procedure
All procedures were performed under local anesthesia. Femoral venous access was obtained. Preprocedural TEE was performed to rule out intracardiac thrombi and assess for concomitant patent foramen ovale (PFO). LAA morphology and dimensions (including landing zone diameter, ostial diameter, and length of LAA) were measured in different angles (0 , 45 , 90 , and 135 ). Intravenous heparin was administered. The implantation procedure was guided by fluoroscopy and TEE. In patients with concomitant PFO, left atrial access was established via the PFO (without transseptal puncture).
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
300
Lam et al.
Fig. 2. Case example illustrating recapture of device and maneuvers to optimize device positioning. (a–c) The proximal disc was partially recaptured into the delivery sheath and redeployed in attempt to optimize the position. However, it was still suboptimal. (d–f) Slight traction was applied on the delivery cable and the disc was repositioned. The final posi-
tion was good with complete sealing of the LAA ostium. The movable core safety wire remained connected to the device and the final position of the device could be evaluated while it was free from tension of the pusher cable (red arrow) before final release. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
After confirming proper position using TEE and fluoroscopy, the Amulet Device can be deployed. Successful deployment is determined by 5 criteria: (1) proper device lobe alignment with the LAA; (2) adequate lobe compression and apposition to LAA wall; (3) presence of a concave-shaped disc; (4) separation of the disc from the lobe, and (5) lobe position by at least two thirds distal to left circumflex artery on TEE. The tug test was performed to test for device stability. Patients were monitored overnight and discharged with a regimen of clopidogrel 75 mg for 3 months and aspirin for a minimum of 6 months after implantation.
history of stroke or transient ischaemic attack (TIA). With the exception of one patient, all patients had CHADS2 scores of equal to or greater than 2. Eleven patients (65%) had AF at the time of procedure. The mean landing zone diameter was 22.2 6 4.1 mm. Transseptal puncture was performed in 14 patients (82%). Three devices were implanted via concomitant PFOs. Implantation success was 100%. Full device recapture was performed in 1 case. In this case, to optimize device position a smaller Amulet device was used. In three other cases the device was partially recaptured and redeployed to optimize the final position (Fig. 2). There was no device embolization, periprocedural stroke, or myocardial infarction (MI). Pericardial effusion occurred in 1 patient (6%) 1 hr after completion of the procedure and the patient underwent pericardiocentesis without further sequaelae. The average length of hospital stay was 2.1 6 0.3 days. All patients were admitted one day before the
RESULTS
Baseline clinical characteristics and procedural details are outlined in Tables II and III. The mean age was 66.8 6 13 years. There were 11 males (65%) and 6 females (35%). Five (29%) had a
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
LAA Occlusion with the Amulet Device
301
Fig. 3. Case example illustrating closure of previous incomplete LAA occlusion with the Amulet device. (a) A 5 Fr Pigtail catheter was passed through the residual gap for selective angiogram. Contrast injection reviewed significant residual gap and the uncovered lobule (blue triangle). (b) Deployment of the distal lobe. (c) Deployment of the proximal disc. (d) The final angiogram showed sealing of the LAA with the Amulet device.
procedure. After the procedure they would stay for observation for one night and discharged on the following day. One patient was observed for one extra night because of occurrence of pericardial effusion and another patient stayed in the hospital for one more night due to logistic reason (patient traveled from long distance for the procedure and stayed one more night before long travel). Of note, a 28-mm Amulet Device was used in a patient with incomplete LAA occlusion after previous Watchman (Boston Scientific, Natrick, MA) implantation 3 years prior (Fig. 3). In this patient, a residual uncovered lobule and a 7 mm wide leak had been present. The Amulet device was considered because of the potential advantage of a larger proximal disc diameter
to improve sealing of the residual leak and uncovered lobe. The new movable inner 0.014 inches core wire allowed a more precise assessment of the configuration and position of the device after deployment. When repositioning was required, the delivery coil of the pusher cable could be advanced towards the end screw of the device to cover the inner core wire, which facilitated recapture of the proximal disc in order to redeploy in the most optimal position. At 90-days, there were no deaths, strokes, systemic thromboembolism, myocardial infarctions, or additional bleeding complications. TEE showed no deviceassociated thrombi or pericardial effusions. A peridevice leak was identified in 2 patients (12%) with widths small than 2 mm.
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
302
Lam et al.
DISCUSSION
We report our experience with the new generation ACP for LAA closure. The implantation process is similar to that of the first generation ACP. The Amulet device was designed to improve the ease of preparation, facilitate the implantation process, and minimize procedure- or device-related complications. The new system allows a more reliable assessment of final device configuration, orientation, and position free from tension of the delivery cable while allowing recapture as long as the floppy 0.014 inches wire is still in place. In our case series 1 case required full recapture and three cases required partial recapture. All were successful. In previous reports of LAA occlusion using the ACP device, the incidence of pericardial effusion or hemorrhage varied between 1.1 and 3.5% [7–9]. One case in our series developed a postprocedural pericardial effusion requiring pericardiocentesis (1 hr after completion of the procedure) without further sequelae. The implantation had been performed through a PFO (transseptal puncture was not performed). No device recapture after first deployment had been necessary. Pericardial hemorrhage might have been related to sheath or device manipulations within the LAA during the implantation process but the exact cause remains uncertain. Theoretically, a larger number of stabilizing wires may be associated with a higher risk of pericardial hemorrhage; however, a larger registry would be necessary to determine the incidence of pericardial effusion as compared to the ACP or other devices. The reported incidence of device-related thrombus formation with the ACP ranges between 2.4 and 14% and the reported thrombus formation rate in PROTECT-AF was around 4% [2,11]. Concerns have been raised that the protruding end screw on the ACP may be thrombogenic. This screw is now embedded within the disc perhaps lowering the thrombogenic potential. In this context, though no device related thrombus was detected in our series, the number of patients was too small to allow conclusions whether the new design, indeed, prevents thrombus formation. Two of our patients had minor peridevice leaks at 90 days. The magnitude of these leaks (
