- Email: [email protected]
Preclinical assessment of a modified Occlutech left atrial appendage closure device in a canine model
International Journal of Cardiology 221 (2016) 413–418
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Preclinical assessment of a modified Occlutech left atrial appendage closure device in a canine model Jung-Sun Kim a,c,1, Seul-Gee Lee b,1, Sung-Kyung Bong b, Se-Il Park c, Sung-Yu Hong c, Sanghoon Shin d, Chi Young Shim a, Geu-Ru Hong a, Donghoon Choi a,c, Yangsoo Jang a,c,⁎⁎, Jai-Wun Park e,f,⁎ a
Severance Cardiovascular Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea Graduate Program in Science for Aging, Yonsei University, Seoul, Republic of Korea c Cardiovascular Product Evaluation Center d Division of Cardiology, Department of Internal Medicine, NHIS Ilsan Hospital, Goyang, Republic of Korea e Department of Cardiology, Angiology, and Pneumology, Coburg Hospital, 96450 Coburg, Germany f Department of Cardiology, Charite Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany b
a r t i c l e
i n f o
Article history: Received 23 May 2016 Accepted 4 July 2016 Available online 05 July 2016 Keywords: Left atrial appendage Canine model Device study
a b s t r a c t Background: LAA occlusion has a similar stroke prevention efficacy compared to anticoagulation treatment for non-valvular atrial fibrillation. Objective: The objective of this study was to assess the feasibility and safety of a modified Occlutech® left atrial appendage (LAA) closure device in a canine model. Methods: The device was implanted in 10 dogs (33 ± 1 kg) using fluoroscopy and transesophageal echocardiography (TEE) guidance. The modified Occlutech® LAA occlusion device was compared with the current version, the Watchman device, and the Amplazter cardiac plug (ACP). LAA occlusion and anchoring to the LAA were evaluated. All dogs were assessed using angiography, TEE, and a gross anatomy examination. Results: The 10 LAA occlusion devices were to be implanted into 10 dogs (5 modified Occlutech devices, 3 current version of Occlutech devices, 1 Watchman, and 1 ACP). LAA implantation was not performed in one dog due to transeptal puncture failure. The three current version of Occlutech devices were embolized immediately after implantation, so three modified devices of the same size were implanted securely without embolization. The mean implant size was 20.1 ± 2.0 mm. The devices chosen were a mean of 23.3 ± 10.6% larger than the measured landing zone diameters. Post-implant angiography and TEE revealed well-positioned devices without pericardial effusion or impingement on surrounding structures. Conclusions: The results of this acute animal study suggested that a modified Occlutech® LAA occlusion device was feasible and had greater anchoring performance in canines. Additional large clinical studies are needed to evaluate safety and efficacy. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Atrial fibrillation (AF) is the most common arrhythmia in a clinical setting and is associated with cardioembolic stroke [1]. Anticoagulation has been a standard treatment modality used to effectively prevent a cardioembolic stroke during atrial fibrillation with a high stroke risk [2]. However, anticoagulation can increase the risk of bleeding events and is limited in its application to patients with a high bleeding risk, ⁎ Correspondence to: J.-W. Park, Department of Cardiology, Angiology, and Pneumology, Coburg Hospital, 96450 Coburg, Germany. ⁎⁎ Correspondence to: Y. Jang, Division of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, 250 Seongsanno, Seodaemun-gu, Seoul 120-752, Republic of Korea. E-mail addresses: [email protected] (Y. Jang), [email protected] (J.-W. Park). 1 These authors contributed equally to this study.
http://dx.doi.org/10.1016/j.ijcard.2016.07.048 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
even as novel anticoagulants are introduced [3]. A previous review of 23 studies revealed that N 90% of thrombi are detected in or originate from the left atrial appendage (LAA) in non-valvular AF [4]. In this context, LAA ligation or occlusion has been proposed as an alternative strategy to anticoagulation for stroke prevention in non-valvular AF [5]. The Watchman device (Boston Scientific, Natick, MA) and the Amplatzer cardiac plug or Amulet device (ACP, St. Jude Medical, St. Paul, MN) are widely used in clinical practice. This use is supported by evidence for efficacy and safety. However, the sharp barbs that anchor the devices to the LAA may damage the soft tissue and increase the difficulty of device retrieval [6]. Unlike previous designs, rounded loops are present at the distal rim side of the Occlutech device, and are used to anchor it to the LAA [7]. Studies in human subjects have revealed that implantation of the presently used Occlutech LAA occlusion device configuration is limited
414
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
to certain LAA anatomies. There are also safety issues related to anchoring the device to the LAA. Recently, the device design has been modified to shorten the length during implantation and improve anchorage to the LAA. The objectives of this study were to further evaluate the safety and efficacy of the Occlutech LAA occlusion device in a canine model and to compare the performance of the previous device design with the performance of the new design and that of approved LAA occlusion devices (Watchman and ACP).
2.3. Interventional procedure
The new design of the Occlutech LAA occlusion device requires evaluation using an invivo model before implantation into humans. The protocol was reviewed and approved by the Animal Ethics Committee and the Institutional Animal Care and Use Committee (IACUC) at the Cardiovascular Product Evaluation Center of the Yonsei University College of Medicine (Seoul, Korea, CPEC-IACUC-151,007). All animals received humane care in compliance with the Animal Welfare Act and the “Principles of Laboratory Animal Care” formulated by the Institute of Laboratory Animal Resources (National Research Council, NIH Publication No. 85–23, revised 1996). The primary objective was to evaluate the performance (deployment and successful implantation) of the modified design of the Occlutech LAA occlusion device immediately after implantation. The performance (deployment and successful implantation) of the modified design was also compared with the current version of the Occlutech LAA occlusion device, the Watchman, and the ACP device. The assessment included evaluation of device protrusion into the left atrium (assessed using imaging), device penetration of the LAA (assessed using gross examination of explanted hearts), and sealing of the LAA. Closure was defined as a leak that was b3 mm (assessed using imaging).
After induction of anesthesia, a groin incision was used to expose the femoral vein. After sheath placement, 3000 IU (100 units/kg) of unfractionated heparin was given through the sheath after successful septal puncture. Additional heparin was administered if a thrombus was detected in the catheter or sheath during the implantation. The left atrium was accessed via standard transeptal puncture using a Brockenbrough needle (BRK™ Transseptal Needle, St. Jude, St. Paul, MN) and an 8 Fr transeptal sheath (SL1, St Jude Medical, St. Paul, MN). The procedure was performed using fluoroscopic and transesophageal echocardiography (TEE) guidance. The devices were delivered to the LAA using appropriately-sized catheters and pushers (delivery system) and a standard implantation procedure. The LAA was measured using TEE and contrast angiography. The LAA was imaged in multiple planes (0°, 45°, 90°, and 135°) to define the maximum LAA width and length. D1 was measured from the circumflex artery to slightly below the “ridge” between the left superior pulmonary vein and the LAA (Fig. 2, white arrow). D2 (the landing zone) was measured along a plane parallel to D1 (Fig. 2). The device was approximately 10% oversized compared to the measured maximum size of D2. The LAA occlusion device was mounted on the delivery system and advanced through the sheath into the left atrium. Correct placement of the device and occlusion was confirmed using a gentle tug test, TEE, and contrast fluoroscopy (Fig. 3). The device was (re)positioned until occlusion was seen (contrast fluoroscopy). If satisfactory closure could not be achieved, the device was recaptured and replaced with a device that was the same model but of a different size. The sealing rate was graded using a 0 to 3 color Doppler scale (0 if no leak, 1 if a trivial leak (b3 mm), 2 if a small leak (3–5 mm), and 3 if a significant leak (N5 mm), was present). If the device was embolized after detachment, a modified Occlutech device of the same size was implanted in the animal. A computed tomography assessment of LAA morphology and size was performed on one case with a modified Occlutech device before the procedure, and for the position of the device after the procedure (Fig. 4).
2.1. Experimental animal model
2.4. Gross specimen evaluation
A canine model (mongrel dog) was chosen because of the anatomic similarities in cardiac shape and size to the human heart. The circulatory system in these dogs is large enough to allow for heart catheterization using a 12F catheter. This model has also been used successfully for testing of other dedicated LAA occlusion devices and allows for the direct transfer of the technology to human trials [8,9]. A total of 11 male dogs (8 months of age, body weight 30 to 35 kg), were used in the study. The study was performed at an independent animal facility (Cardiovascular Product Evaluation Center, Yonsei University College of Medicine, Osong, Korea) accredited by the Korea Ministry of Food and Drug Safety. This facility was approved for evaluation of the modified Occlutech LAA design. The animals were sedated using intramuscular injections of 0.01 mg/kg of medetomidine hydrochloride (Domitor®), 0.2 mg/kg of xylazine, and 0.04 mg/kg of atropine. Once sedated, they were anesthetized using isoflurane and oxygen, which were delivered via facemask. The animals were then intubated and anesthesia was maintained using isoflurane and oxygen. After each animal was transferred to the procedure table, isoflurane was delivered through a volume-regulated respirator. End-tidal CO2 was maintained within individual physiological ranges. Medication for appropriate anesthetic management was also available, and was used if indicated. At the end of the study, each animal was anesthetized using isoflurane and was euthanized using an intravenous overdose of potassium.
Each heart was carefully explanted. Care was taken to ensure that the device did not penetrate the LAA wall during the explantation procedure. The LAA exterior was examined macroscopically for device perforation. After this examination, the left atrium was cut open without damage to the LAA, and the placement of the occluder was exposed (Fig. 5). A tug test was performed, and the force needed to remove the device from the LAA was recorded using a manual manometer. To document the results, photographs were taken of the entire explanted heart, and of opened hearts in which the LAA and the occluder were visible.
2. Methods
2.2. Left atrial appendage device description The Occlutech LAA occlusion device (Occlutech®, Jena, Germany) was described previously [7]. Briefly, this device has a self-expanding, flexible, nitinol mesh structure with a cylindrical shape. Unlike the previous version, the distal end of the device has an inverted floor instead of a flat floor; this structural change results in reduced elongation during deployment and enhanced anchoring stability (Fig. 1).
Fig. 1. Design of device. The left atrial appendage (LAA) closure device (Occlutech® LAA occlusion device) is a flexible nitinol-based, self-expanding device consisting of an outer surface covered with a non-woven, bio-stable poly (carbonate) urethane layer. The device's proximal section has a larger diameter to seal the orifice and a distal loop rim that helps to maintain the implanted device in position.
2.5. Statistical analysis Statistical analysis was performed using SPSS (version 20.0.0, IBM, Armonk, NY, USA). The results were expressed as mean ± standard deviation, or percentage (%) values. Comparisons were made using Chi-square statistics or Fischer's exact test for the categorical data and Student's t-test for the continuous variables. If a distribution was skewed, a non-parametric test was used. A p-value b0.05 was considered to indicate a statistically significant result.
3. Results The characteristics of the implanted devices are presented in Table 1. Of the 11 dogs that were included in this study, 5 dogs received a modified Occlutech LAA occlusion device, 3 received a current version of an Occlutech device, 1 received a Watchman, and 1 received an ACP device. One animal did not receive a device because septal puncture failure and cardiac rupture occurred during the procedure. The modified version of the Occlutech device was successfully implanted, retrieved, repositioned, and re-implanted into five animals. Among these 5 dogs, 3 received devices that were embolized immediately after the implantation. Three additional modified devices of the same size were then implanted securely without embolization. The mean implant size was 20.1 ± 2.0 mm and the device chosen was 23.3 ± 10.6% larger than the measured landing zone diameter. Post-implant contrast angiography confirmed proper and stable implantation into the LAA. Device migration, significant peri-implant leakage, or impingement on surrounding cardiac structures did not occur in any of the dogs. TEE performed immediately after implantation revealed no mitral valve dysfunction or pulmonary venous obstruction. The TEE results indicated that successful occlusion of the appendages in all of the implanted devices had been accomplished (sealing rate 0: 3 modified Occlutech, sealing rate 1: 5 modified Occlutech and 1 ACP, and sealing rate 2: 1 Watchman). Each modified Occlutech device had an acceptable closure
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
415
Fig. 2. Measurement of the landing zone and the orifice of the left atrial appendage. Multiplanar transesophageal echocardiography facilitated measurement of the size of the landing zone and of the orifice of the left atrial appendage. The TEE images were used for the measurement of the left atrial appendage width to the orifice (D1) and the landing zone (D2) widths.
rate and each had b 3 mm peridevice leakage. Computed tomographic evaluation results were available for two modified Occlutech devices; they were adequately seated, and no contrast was visible inside the LAA (Fig. 4). 3.1. Gross anatomical evaluation of the heart Embedding of the various devices in the LAA of the canine heart was also examined. The modified Occlutech device was better-occluded compared with the other devices. All of the devices were securely
embedded in the LAA. The smaller-sized devices were deeply seated into the LAA, and the larger devices tended to protrude into the LAA ostium. Cardiac perforation and device perforation of the LAA were not detected for the modified Occlutech devices, or for the other LAA occlusion devices, implanted into any of the animals (Fig. 5-A). The modified Occlutech device was well-seated inside the LAA and completely closed the LAA ostium (Fig. 5-B and C), but larger devices tended to protrude over the ostium. The forces required to remove the devices from the LAA were 0.9 ± 0.4 N/m2 for the modified Occlutech devices (N = 8),
Fig. 3. Step-by-step illustration of the modified Occlutech device deployment procedure. The sheath was placed on the proximal portion of the left atrial appendage (LAA) (A) and the device was partially deployed (B) by slowly pushing it out of the sheath. The entire system then progressed to the desirable LAA landing zone and engagement of the loops into the LAA walls (C). Contrast was then injected to check for LAA sealing (D). Finally, the delivery catheter was carefully released from the device (E).
416
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
Fig. 4. CT images of insertion of the device. CT images of device before (A) and after (B) insertion; the arrow indicates the device.
0.7 N/m2 for the Watchman device, and 1.3 M/m2 for the ACP device (Fig. 5-E). There were no macroscopic signs of damage to the intact endocardium surrounding the device (Fig. 5-F). Thrombotic attachments to the device surface were not visible.
the atrial septum due to the rotated anatomy of the canine heart, and was not related to the device.
3.2. Complications
This study was designed to evaluate the efficacy of the modified Occlutech LAA occlusion device and to compare it with the current version of the Occlutech LAA occlusion device and the Watchman and ACP devices. The device was found to be easy to deploy and adjusted very well to the LAA, compared with the previous Occlutech version, and the Watchman and ACP devices.
There was one procedural complication, which was caused by heart perforation that occurred during trans-septal puncture. The implantation of the occlusion device failed; the autopsy revealed that tamponade had occurred. This complication was the result of difficulty visualizing
4. Discussion
Fig. 5. Left atrial appendage-modified Occlutech occlusion devices explanted immediately after implantation. Gross anatomical views of the left atrial appendage (LAA) immediately after implantation (A), small-sized device well-seated inside the LAA (B), large-sized device protruding over the LAA ostium (C), the device separated from the LAA (D), the tug test and manual manometer measurement (E), and overturn of the LAA to confirm (F) after occlusion by the device.
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
417
Table 1 Summary of the results. Case no
Weight kg
Device size D1/D2, mm
Length
Occluder
Implanted device
Sealing rate
Oversize (%)
1 2 3 4 5 6 7 8 9 10 11
32 33 35 33 33 34 35 31 33 33 35
19.0/16.0 18.5/17.2 14.5/13.9 18.0/16.0 16.3/15.2 17.8/16.4 19.2/16.2 19.9/18.5 18.5/17.2
22 7 15 17 18 22 23 22 21
21 21 18 18 18 24 18 21 21
0 0 1 1 1 1 0 0 0
31.3 22.3 28.9 12.5 18.4 46.2 11.1 13.5 22.1
18.0/16.6
21
Modified OCL occluder Modified OCL occluder Modified OCL occluder Modified OCL occluder Watchman Amplanz cardiac plug Modified OCL occluder Modified OCL occluder Modified OCL occluder Septal puncture failure Modified OCL occluder
21
0
26.5
no = number; kg = kilograms; mm = millimeters; LAA = left atrial appendage; OCL = Occlutech. LAA landing zone diameter used for selection of device size. Cases 8, 9, and 11 were tried with same-sized previous version of OCL device, but all three devices were embolized. Then, three of the same-sized modified devices were applied.
Percutaneous LAA occlusion has been proposed as an alternative treatment option to anticoagulation, even given the novel anticoagulants, in patients with non-valvular AF and at a high risk of bleeding [5]. In clinical practice, the Watchman and the ACP or Amulet devices are widely used and have favorable clinical outcomes. The long-term follow-up PROTEC AF study revealed significant improvement in clinical outcomes (e.g., stroke) and reduction in mortality, compared with warfarin treatment in patients with AF [10]. Review of a large registry of patients with the ACP revealed that the risks of systemic embolization and bleeding are markedly reduced based on the CHAD2VSc and HAS-BLED scores (59% and 61%, respectively) [11]. Although conceptually the devices are acceptable for LAA closure, two devices use sharp barbs to stabilize the device to the LAA wall. These barbs may cause damage during the acute phase of implantation and during the long-term follow-up period. Patients have experienced cardiac tamponade caused by pulmonary artery injury related to the sharp ACP anchoring hooks or the metallic struts of the Watchman device [12,13]. The Occlutech device has been developed to replace previous LAA occlusion devices. Compared with the two devices that are currently used (ACP or Amulet and Watchman), the Occlutech® device has rounded loops at its distal rim to fix the device at the landing zone in the LAA; this change allows for early reuptake/repositioning of the device, and for a device exchange when the size is not appropriate [7]. The soft nitinol body of the modified device has a low radial force, which results in better modeling and adaptation to the LAA wall and complete LAA closure. A swine model experimental study revealed that the Occlutech® LAA occlusion device achieves complete sealing of the LAA in all animals by 6 weeks post-implantation [7]. This same study, which also tested current versions of the Occlutech device, found that the current version's loops had penetrated the LAA tissue in three pigs, but pericardial effusion or injury to the LAA wall was not apparent. The loop appearance in the surrounding LAA tissue might have occurred during the explantation procedure and this possibility is an outstanding critical issue that should be considered for the current version of the Occlutech device. In our acute study, there was no evidence that the modified version's loops caused tissue damage, but long-term evaluation is needed. Human clinical studies have revealed that the current version of the Occlutech LAA occlusion device is difficult to apply when specific LAA anatomy is present. This limitation is considered to be attributable to a temporary elongation that renders the occluder “too long” during its deployment into the LAA from within the delivery catheter. To circumvent this temporary elongation, a minor design modification has been made at the distal end of the device (i.e., inverted floor versus a previously flat floor). Floor inversion also results in a “stiffer” distal end, which is expected to further enhance the safety of the device as a result of improved anchoring. Results of bench-top and ex-vivo implantation tests have indicated that this modification reduces the elongation and enhances anchoring. These results were supported by the results of this study to evaluate the feasibility and safety of implanting the modified
Occlutech LAA occlusion device system into a canine model. The modified device had adequate anchoring to the LAA and there were no serious safety issues during this acute animal experiment. Unfortunately, a direct between-model comparison for histology and the degree of anchoring could not be performed for the current and modified versions because three of the current version devices were embolized immediately after deployment. Thereafter, the same-sized modified version of the LAA occlusion device was applied in the same canine model to assess the efficacy of the new version. The modified device was more stably anchored to the LAA and all of the modified devices well-seated into the LAA. The results of the assessment using TEE suggested that overall, the modified Occlutech system had a greater degree of acute sealing of the LAA, compared with the ACP or Watchman devices. 4.1. Study limitations This experimental animal study had a small sample size and only data on acute results were collected. Follow-up outcomes should be tested and investigated using additional studies. However, the results indicated that the modified version of Occlutech device had better safety because of more stable anchoring without tissue damage. Only one ACP and one Watchman device were tested for the comparison between Occlutech and other LAA occlusion devices. Therefore, the sample sizes were too small for a complete evaluation of differences between the types of LAA occlusion devices. However, the modified Occlutech device had better performance in terms of secure anchoring compared with the current Occlutech version. It also had similar stability compared to the ACP and Watchman devices. Lastly, the morphology and structure of the canine LAA is not identical to the human LAA, which should be considered before the results of this study are applied in clinical practice. 5. Conclusions The results of our acute animal study suggested that the modified Occlutech LAA occlusion device was a feasible option and had greater anchoring performance in canines. The Occlutech® LAA occlusion device has rounded loops at its distal rim, which make it easy to exchange and reduce the risk of LAA injury. The nitinol body promotes better remodeling and adaptation to the LAA. Additional studies are needed to evaluate the safety and efficacy of this modified device. Funding This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2014R1A1A2055584), and by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry
418
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
of Health & Welfare, Republic of Korea (grant number HI15C1277), and by academic grant funded by Cardiovascular Research Center, Seoul, Korea and the Occlutech, Jena, Germany. Conflict of interest Dr. Park is a consultant for Occlutech, Jena, Germany. References [1] P.A. Wolf, R.D. Abbott, W.B. Kannel, Atrial fibrillation as an independent risk factor for stroke: the Framingham study, Stroke 22 (1991) 983–988. [2] G.Y. Lip, D.A. Lane, Stroke prevention in atrial fibrillation: a systematic review, JAMA 313 (2015) 1950–1962. [3] G.Y. Lip, F. Andreotti, L. Fauchier, K. Huber, E. Hylek, E. Knight, et al., Bleeding risk assessment and management in atrial fibrillation patients. Executive summary of a position document from the European heart rhythm association [EHRA], endorsed by the European Society of Cardiology [ESC] working group on thrombosis, Thromb. Haemost. 106 (2011) 997–1011. [4] J.L. Blackshear, J.A. Odell, Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation, Ann. Thorac. Surg. 61 (1966) 755–759. [5] U. Landmesser, D.R. Holmes Jr., Left atrial appendage closure: a percutaneous transcatheter approach for stroke prevention in atrial fibrillation, Eur. Heart J. 33 (2012) 698–704.
[6] I. Cruz-Gonzalez, B.P. Yan, Y.Y. Lam, Left atrial appendage exclusion: state-of-the-art, Catheter. Cardiovasc. Interv. 75 (2010) 806–813. [7] J.W. Park, M.A. Sherif, K. Zintl, Y.Y. Lam, M. Goedde, T. Scharnweber, et al., Percutaneous left atrial appendage closure with a novel self-modelizing device: a pre-clinical feasibility study, Int. J. Cardiol. 177 (2014) 957–963. [8] R.S. Schwartz, D.R. Holmes, R.A. Van Tassel, R. Hauser, T.D. Henry, M. Mooney, et al., Left atrial appendage obliteration: mechanisms of healing and intracardiac integration, J. Am. Coll. Cardiol. Cardiovasc. Interv. 3 (2010) 870–877. [9] Y.Y. Lam, B.P. Yan, S.K. Doshi, A. Li, D. Zhang, M.G. Kaya, et al., Preclinical evaluation of a new left atrial appendage occluder (Lifetech LAmbre device) in a canine model, Int. J. Cardiol. 168 (2013) 3996–4001. [10] V.Y. Reddy, H. Sievert, J. Halperin, S.K. Doshi, M. Buchbinder, P. Neuzil, et al., Percutaneous left atrial appendage closure vs warfarin for atrial fibrillation: a randomized clinical trial, JAMA 312 (2014) 1988–1998. [11] A. Tzikas, S. Shakir, S. Gafoor, H. Omran, S. Berti, G. Santoro, et al., Left atrial appendage occlusion for stroke prevention in atrial fibrillation: multicentre experience with the AMPLATZER cardiac plug, EuroIntervention 10 (2015) 1170–1179. [12] G. Bianchi, M. Solinas, T. Gasbarri, S. Bevilacqua, K.K. Tiwari, S. Berti, et al., Pulmonary artery perforation by plug anchoring system after percutaneous closure of left appendage, Ann. Thorac. Surg. 96 (2013) e3–e5. [13] A. Sepahpour, M.K. Ng, P. Storey, M.A. McGuire, Death from pulmonary artery erosion complicating implantation of percutaneous left atrial appendage occlusion device, Heart Rhythm. 10 (2013) 1810–1811.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Preclinical assessment of a modified Occlutech left atrial appendage closure device in a canine model Jung-Sun Kim a,c,1, Seul-Gee Lee b,1, Sung-Kyung Bong b, Se-Il Park c, Sung-Yu Hong c, Sanghoon Shin d, Chi Young Shim a, Geu-Ru Hong a, Donghoon Choi a,c, Yangsoo Jang a,c,⁎⁎, Jai-Wun Park e,f,⁎ a
Severance Cardiovascular Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea Graduate Program in Science for Aging, Yonsei University, Seoul, Republic of Korea c Cardiovascular Product Evaluation Center d Division of Cardiology, Department of Internal Medicine, NHIS Ilsan Hospital, Goyang, Republic of Korea e Department of Cardiology, Angiology, and Pneumology, Coburg Hospital, 96450 Coburg, Germany f Department of Cardiology, Charite Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany b
a r t i c l e
i n f o
Article history: Received 23 May 2016 Accepted 4 July 2016 Available online 05 July 2016 Keywords: Left atrial appendage Canine model Device study
a b s t r a c t Background: LAA occlusion has a similar stroke prevention efficacy compared to anticoagulation treatment for non-valvular atrial fibrillation. Objective: The objective of this study was to assess the feasibility and safety of a modified Occlutech® left atrial appendage (LAA) closure device in a canine model. Methods: The device was implanted in 10 dogs (33 ± 1 kg) using fluoroscopy and transesophageal echocardiography (TEE) guidance. The modified Occlutech® LAA occlusion device was compared with the current version, the Watchman device, and the Amplazter cardiac plug (ACP). LAA occlusion and anchoring to the LAA were evaluated. All dogs were assessed using angiography, TEE, and a gross anatomy examination. Results: The 10 LAA occlusion devices were to be implanted into 10 dogs (5 modified Occlutech devices, 3 current version of Occlutech devices, 1 Watchman, and 1 ACP). LAA implantation was not performed in one dog due to transeptal puncture failure. The three current version of Occlutech devices were embolized immediately after implantation, so three modified devices of the same size were implanted securely without embolization. The mean implant size was 20.1 ± 2.0 mm. The devices chosen were a mean of 23.3 ± 10.6% larger than the measured landing zone diameters. Post-implant angiography and TEE revealed well-positioned devices without pericardial effusion or impingement on surrounding structures. Conclusions: The results of this acute animal study suggested that a modified Occlutech® LAA occlusion device was feasible and had greater anchoring performance in canines. Additional large clinical studies are needed to evaluate safety and efficacy. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Atrial fibrillation (AF) is the most common arrhythmia in a clinical setting and is associated with cardioembolic stroke [1]. Anticoagulation has been a standard treatment modality used to effectively prevent a cardioembolic stroke during atrial fibrillation with a high stroke risk [2]. However, anticoagulation can increase the risk of bleeding events and is limited in its application to patients with a high bleeding risk, ⁎ Correspondence to: J.-W. Park, Department of Cardiology, Angiology, and Pneumology, Coburg Hospital, 96450 Coburg, Germany. ⁎⁎ Correspondence to: Y. Jang, Division of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, 250 Seongsanno, Seodaemun-gu, Seoul 120-752, Republic of Korea. E-mail addresses: [email protected] (Y. Jang), [email protected] (J.-W. Park). 1 These authors contributed equally to this study.
http://dx.doi.org/10.1016/j.ijcard.2016.07.048 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
even as novel anticoagulants are introduced [3]. A previous review of 23 studies revealed that N 90% of thrombi are detected in or originate from the left atrial appendage (LAA) in non-valvular AF [4]. In this context, LAA ligation or occlusion has been proposed as an alternative strategy to anticoagulation for stroke prevention in non-valvular AF [5]. The Watchman device (Boston Scientific, Natick, MA) and the Amplatzer cardiac plug or Amulet device (ACP, St. Jude Medical, St. Paul, MN) are widely used in clinical practice. This use is supported by evidence for efficacy and safety. However, the sharp barbs that anchor the devices to the LAA may damage the soft tissue and increase the difficulty of device retrieval [6]. Unlike previous designs, rounded loops are present at the distal rim side of the Occlutech device, and are used to anchor it to the LAA [7]. Studies in human subjects have revealed that implantation of the presently used Occlutech LAA occlusion device configuration is limited
414
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
to certain LAA anatomies. There are also safety issues related to anchoring the device to the LAA. Recently, the device design has been modified to shorten the length during implantation and improve anchorage to the LAA. The objectives of this study were to further evaluate the safety and efficacy of the Occlutech LAA occlusion device in a canine model and to compare the performance of the previous device design with the performance of the new design and that of approved LAA occlusion devices (Watchman and ACP).
2.3. Interventional procedure
The new design of the Occlutech LAA occlusion device requires evaluation using an invivo model before implantation into humans. The protocol was reviewed and approved by the Animal Ethics Committee and the Institutional Animal Care and Use Committee (IACUC) at the Cardiovascular Product Evaluation Center of the Yonsei University College of Medicine (Seoul, Korea, CPEC-IACUC-151,007). All animals received humane care in compliance with the Animal Welfare Act and the “Principles of Laboratory Animal Care” formulated by the Institute of Laboratory Animal Resources (National Research Council, NIH Publication No. 85–23, revised 1996). The primary objective was to evaluate the performance (deployment and successful implantation) of the modified design of the Occlutech LAA occlusion device immediately after implantation. The performance (deployment and successful implantation) of the modified design was also compared with the current version of the Occlutech LAA occlusion device, the Watchman, and the ACP device. The assessment included evaluation of device protrusion into the left atrium (assessed using imaging), device penetration of the LAA (assessed using gross examination of explanted hearts), and sealing of the LAA. Closure was defined as a leak that was b3 mm (assessed using imaging).
After induction of anesthesia, a groin incision was used to expose the femoral vein. After sheath placement, 3000 IU (100 units/kg) of unfractionated heparin was given through the sheath after successful septal puncture. Additional heparin was administered if a thrombus was detected in the catheter or sheath during the implantation. The left atrium was accessed via standard transeptal puncture using a Brockenbrough needle (BRK™ Transseptal Needle, St. Jude, St. Paul, MN) and an 8 Fr transeptal sheath (SL1, St Jude Medical, St. Paul, MN). The procedure was performed using fluoroscopic and transesophageal echocardiography (TEE) guidance. The devices were delivered to the LAA using appropriately-sized catheters and pushers (delivery system) and a standard implantation procedure. The LAA was measured using TEE and contrast angiography. The LAA was imaged in multiple planes (0°, 45°, 90°, and 135°) to define the maximum LAA width and length. D1 was measured from the circumflex artery to slightly below the “ridge” between the left superior pulmonary vein and the LAA (Fig. 2, white arrow). D2 (the landing zone) was measured along a plane parallel to D1 (Fig. 2). The device was approximately 10% oversized compared to the measured maximum size of D2. The LAA occlusion device was mounted on the delivery system and advanced through the sheath into the left atrium. Correct placement of the device and occlusion was confirmed using a gentle tug test, TEE, and contrast fluoroscopy (Fig. 3). The device was (re)positioned until occlusion was seen (contrast fluoroscopy). If satisfactory closure could not be achieved, the device was recaptured and replaced with a device that was the same model but of a different size. The sealing rate was graded using a 0 to 3 color Doppler scale (0 if no leak, 1 if a trivial leak (b3 mm), 2 if a small leak (3–5 mm), and 3 if a significant leak (N5 mm), was present). If the device was embolized after detachment, a modified Occlutech device of the same size was implanted in the animal. A computed tomography assessment of LAA morphology and size was performed on one case with a modified Occlutech device before the procedure, and for the position of the device after the procedure (Fig. 4).
2.1. Experimental animal model
2.4. Gross specimen evaluation
A canine model (mongrel dog) was chosen because of the anatomic similarities in cardiac shape and size to the human heart. The circulatory system in these dogs is large enough to allow for heart catheterization using a 12F catheter. This model has also been used successfully for testing of other dedicated LAA occlusion devices and allows for the direct transfer of the technology to human trials [8,9]. A total of 11 male dogs (8 months of age, body weight 30 to 35 kg), were used in the study. The study was performed at an independent animal facility (Cardiovascular Product Evaluation Center, Yonsei University College of Medicine, Osong, Korea) accredited by the Korea Ministry of Food and Drug Safety. This facility was approved for evaluation of the modified Occlutech LAA design. The animals were sedated using intramuscular injections of 0.01 mg/kg of medetomidine hydrochloride (Domitor®), 0.2 mg/kg of xylazine, and 0.04 mg/kg of atropine. Once sedated, they were anesthetized using isoflurane and oxygen, which were delivered via facemask. The animals were then intubated and anesthesia was maintained using isoflurane and oxygen. After each animal was transferred to the procedure table, isoflurane was delivered through a volume-regulated respirator. End-tidal CO2 was maintained within individual physiological ranges. Medication for appropriate anesthetic management was also available, and was used if indicated. At the end of the study, each animal was anesthetized using isoflurane and was euthanized using an intravenous overdose of potassium.
Each heart was carefully explanted. Care was taken to ensure that the device did not penetrate the LAA wall during the explantation procedure. The LAA exterior was examined macroscopically for device perforation. After this examination, the left atrium was cut open without damage to the LAA, and the placement of the occluder was exposed (Fig. 5). A tug test was performed, and the force needed to remove the device from the LAA was recorded using a manual manometer. To document the results, photographs were taken of the entire explanted heart, and of opened hearts in which the LAA and the occluder were visible.
2. Methods
2.2. Left atrial appendage device description The Occlutech LAA occlusion device (Occlutech®, Jena, Germany) was described previously [7]. Briefly, this device has a self-expanding, flexible, nitinol mesh structure with a cylindrical shape. Unlike the previous version, the distal end of the device has an inverted floor instead of a flat floor; this structural change results in reduced elongation during deployment and enhanced anchoring stability (Fig. 1).
Fig. 1. Design of device. The left atrial appendage (LAA) closure device (Occlutech® LAA occlusion device) is a flexible nitinol-based, self-expanding device consisting of an outer surface covered with a non-woven, bio-stable poly (carbonate) urethane layer. The device's proximal section has a larger diameter to seal the orifice and a distal loop rim that helps to maintain the implanted device in position.
2.5. Statistical analysis Statistical analysis was performed using SPSS (version 20.0.0, IBM, Armonk, NY, USA). The results were expressed as mean ± standard deviation, or percentage (%) values. Comparisons were made using Chi-square statistics or Fischer's exact test for the categorical data and Student's t-test for the continuous variables. If a distribution was skewed, a non-parametric test was used. A p-value b0.05 was considered to indicate a statistically significant result.
3. Results The characteristics of the implanted devices are presented in Table 1. Of the 11 dogs that were included in this study, 5 dogs received a modified Occlutech LAA occlusion device, 3 received a current version of an Occlutech device, 1 received a Watchman, and 1 received an ACP device. One animal did not receive a device because septal puncture failure and cardiac rupture occurred during the procedure. The modified version of the Occlutech device was successfully implanted, retrieved, repositioned, and re-implanted into five animals. Among these 5 dogs, 3 received devices that were embolized immediately after the implantation. Three additional modified devices of the same size were then implanted securely without embolization. The mean implant size was 20.1 ± 2.0 mm and the device chosen was 23.3 ± 10.6% larger than the measured landing zone diameter. Post-implant contrast angiography confirmed proper and stable implantation into the LAA. Device migration, significant peri-implant leakage, or impingement on surrounding cardiac structures did not occur in any of the dogs. TEE performed immediately after implantation revealed no mitral valve dysfunction or pulmonary venous obstruction. The TEE results indicated that successful occlusion of the appendages in all of the implanted devices had been accomplished (sealing rate 0: 3 modified Occlutech, sealing rate 1: 5 modified Occlutech and 1 ACP, and sealing rate 2: 1 Watchman). Each modified Occlutech device had an acceptable closure
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
415
Fig. 2. Measurement of the landing zone and the orifice of the left atrial appendage. Multiplanar transesophageal echocardiography facilitated measurement of the size of the landing zone and of the orifice of the left atrial appendage. The TEE images were used for the measurement of the left atrial appendage width to the orifice (D1) and the landing zone (D2) widths.
rate and each had b 3 mm peridevice leakage. Computed tomographic evaluation results were available for two modified Occlutech devices; they were adequately seated, and no contrast was visible inside the LAA (Fig. 4). 3.1. Gross anatomical evaluation of the heart Embedding of the various devices in the LAA of the canine heart was also examined. The modified Occlutech device was better-occluded compared with the other devices. All of the devices were securely
embedded in the LAA. The smaller-sized devices were deeply seated into the LAA, and the larger devices tended to protrude into the LAA ostium. Cardiac perforation and device perforation of the LAA were not detected for the modified Occlutech devices, or for the other LAA occlusion devices, implanted into any of the animals (Fig. 5-A). The modified Occlutech device was well-seated inside the LAA and completely closed the LAA ostium (Fig. 5-B and C), but larger devices tended to protrude over the ostium. The forces required to remove the devices from the LAA were 0.9 ± 0.4 N/m2 for the modified Occlutech devices (N = 8),
Fig. 3. Step-by-step illustration of the modified Occlutech device deployment procedure. The sheath was placed on the proximal portion of the left atrial appendage (LAA) (A) and the device was partially deployed (B) by slowly pushing it out of the sheath. The entire system then progressed to the desirable LAA landing zone and engagement of the loops into the LAA walls (C). Contrast was then injected to check for LAA sealing (D). Finally, the delivery catheter was carefully released from the device (E).
416
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
Fig. 4. CT images of insertion of the device. CT images of device before (A) and after (B) insertion; the arrow indicates the device.
0.7 N/m2 for the Watchman device, and 1.3 M/m2 for the ACP device (Fig. 5-E). There were no macroscopic signs of damage to the intact endocardium surrounding the device (Fig. 5-F). Thrombotic attachments to the device surface were not visible.
the atrial septum due to the rotated anatomy of the canine heart, and was not related to the device.
3.2. Complications
This study was designed to evaluate the efficacy of the modified Occlutech LAA occlusion device and to compare it with the current version of the Occlutech LAA occlusion device and the Watchman and ACP devices. The device was found to be easy to deploy and adjusted very well to the LAA, compared with the previous Occlutech version, and the Watchman and ACP devices.
There was one procedural complication, which was caused by heart perforation that occurred during trans-septal puncture. The implantation of the occlusion device failed; the autopsy revealed that tamponade had occurred. This complication was the result of difficulty visualizing
4. Discussion
Fig. 5. Left atrial appendage-modified Occlutech occlusion devices explanted immediately after implantation. Gross anatomical views of the left atrial appendage (LAA) immediately after implantation (A), small-sized device well-seated inside the LAA (B), large-sized device protruding over the LAA ostium (C), the device separated from the LAA (D), the tug test and manual manometer measurement (E), and overturn of the LAA to confirm (F) after occlusion by the device.
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
417
Table 1 Summary of the results. Case no
Weight kg
Device size D1/D2, mm
Length
Occluder
Implanted device
Sealing rate
Oversize (%)
1 2 3 4 5 6 7 8 9 10 11
32 33 35 33 33 34 35 31 33 33 35
19.0/16.0 18.5/17.2 14.5/13.9 18.0/16.0 16.3/15.2 17.8/16.4 19.2/16.2 19.9/18.5 18.5/17.2
22 7 15 17 18 22 23 22 21
21 21 18 18 18 24 18 21 21
0 0 1 1 1 1 0 0 0
31.3 22.3 28.9 12.5 18.4 46.2 11.1 13.5 22.1
18.0/16.6
21
Modified OCL occluder Modified OCL occluder Modified OCL occluder Modified OCL occluder Watchman Amplanz cardiac plug Modified OCL occluder Modified OCL occluder Modified OCL occluder Septal puncture failure Modified OCL occluder
21
0
26.5
no = number; kg = kilograms; mm = millimeters; LAA = left atrial appendage; OCL = Occlutech. LAA landing zone diameter used for selection of device size. Cases 8, 9, and 11 were tried with same-sized previous version of OCL device, but all three devices were embolized. Then, three of the same-sized modified devices were applied.
Percutaneous LAA occlusion has been proposed as an alternative treatment option to anticoagulation, even given the novel anticoagulants, in patients with non-valvular AF and at a high risk of bleeding [5]. In clinical practice, the Watchman and the ACP or Amulet devices are widely used and have favorable clinical outcomes. The long-term follow-up PROTEC AF study revealed significant improvement in clinical outcomes (e.g., stroke) and reduction in mortality, compared with warfarin treatment in patients with AF [10]. Review of a large registry of patients with the ACP revealed that the risks of systemic embolization and bleeding are markedly reduced based on the CHAD2VSc and HAS-BLED scores (59% and 61%, respectively) [11]. Although conceptually the devices are acceptable for LAA closure, two devices use sharp barbs to stabilize the device to the LAA wall. These barbs may cause damage during the acute phase of implantation and during the long-term follow-up period. Patients have experienced cardiac tamponade caused by pulmonary artery injury related to the sharp ACP anchoring hooks or the metallic struts of the Watchman device [12,13]. The Occlutech device has been developed to replace previous LAA occlusion devices. Compared with the two devices that are currently used (ACP or Amulet and Watchman), the Occlutech® device has rounded loops at its distal rim to fix the device at the landing zone in the LAA; this change allows for early reuptake/repositioning of the device, and for a device exchange when the size is not appropriate [7]. The soft nitinol body of the modified device has a low radial force, which results in better modeling and adaptation to the LAA wall and complete LAA closure. A swine model experimental study revealed that the Occlutech® LAA occlusion device achieves complete sealing of the LAA in all animals by 6 weeks post-implantation [7]. This same study, which also tested current versions of the Occlutech device, found that the current version's loops had penetrated the LAA tissue in three pigs, but pericardial effusion or injury to the LAA wall was not apparent. The loop appearance in the surrounding LAA tissue might have occurred during the explantation procedure and this possibility is an outstanding critical issue that should be considered for the current version of the Occlutech device. In our acute study, there was no evidence that the modified version's loops caused tissue damage, but long-term evaluation is needed. Human clinical studies have revealed that the current version of the Occlutech LAA occlusion device is difficult to apply when specific LAA anatomy is present. This limitation is considered to be attributable to a temporary elongation that renders the occluder “too long” during its deployment into the LAA from within the delivery catheter. To circumvent this temporary elongation, a minor design modification has been made at the distal end of the device (i.e., inverted floor versus a previously flat floor). Floor inversion also results in a “stiffer” distal end, which is expected to further enhance the safety of the device as a result of improved anchoring. Results of bench-top and ex-vivo implantation tests have indicated that this modification reduces the elongation and enhances anchoring. These results were supported by the results of this study to evaluate the feasibility and safety of implanting the modified
Occlutech LAA occlusion device system into a canine model. The modified device had adequate anchoring to the LAA and there were no serious safety issues during this acute animal experiment. Unfortunately, a direct between-model comparison for histology and the degree of anchoring could not be performed for the current and modified versions because three of the current version devices were embolized immediately after deployment. Thereafter, the same-sized modified version of the LAA occlusion device was applied in the same canine model to assess the efficacy of the new version. The modified device was more stably anchored to the LAA and all of the modified devices well-seated into the LAA. The results of the assessment using TEE suggested that overall, the modified Occlutech system had a greater degree of acute sealing of the LAA, compared with the ACP or Watchman devices. 4.1. Study limitations This experimental animal study had a small sample size and only data on acute results were collected. Follow-up outcomes should be tested and investigated using additional studies. However, the results indicated that the modified version of Occlutech device had better safety because of more stable anchoring without tissue damage. Only one ACP and one Watchman device were tested for the comparison between Occlutech and other LAA occlusion devices. Therefore, the sample sizes were too small for a complete evaluation of differences between the types of LAA occlusion devices. However, the modified Occlutech device had better performance in terms of secure anchoring compared with the current Occlutech version. It also had similar stability compared to the ACP and Watchman devices. Lastly, the morphology and structure of the canine LAA is not identical to the human LAA, which should be considered before the results of this study are applied in clinical practice. 5. Conclusions The results of our acute animal study suggested that the modified Occlutech LAA occlusion device was a feasible option and had greater anchoring performance in canines. The Occlutech® LAA occlusion device has rounded loops at its distal rim, which make it easy to exchange and reduce the risk of LAA injury. The nitinol body promotes better remodeling and adaptation to the LAA. Additional studies are needed to evaluate the safety and efficacy of this modified device. Funding This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2014R1A1A2055584), and by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry
418
J.-S. Kim et al. / International Journal of Cardiology 221 (2016) 413–418
of Health & Welfare, Republic of Korea (grant number HI15C1277), and by academic grant funded by Cardiovascular Research Center, Seoul, Korea and the Occlutech, Jena, Germany. Conflict of interest Dr. Park is a consultant for Occlutech, Jena, Germany. References [1] P.A. Wolf, R.D. Abbott, W.B. Kannel, Atrial fibrillation as an independent risk factor for stroke: the Framingham study, Stroke 22 (1991) 983–988. [2] G.Y. Lip, D.A. Lane, Stroke prevention in atrial fibrillation: a systematic review, JAMA 313 (2015) 1950–1962. [3] G.Y. Lip, F. Andreotti, L. Fauchier, K. Huber, E. Hylek, E. Knight, et al., Bleeding risk assessment and management in atrial fibrillation patients. Executive summary of a position document from the European heart rhythm association [EHRA], endorsed by the European Society of Cardiology [ESC] working group on thrombosis, Thromb. Haemost. 106 (2011) 997–1011. [4] J.L. Blackshear, J.A. Odell, Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation, Ann. Thorac. Surg. 61 (1966) 755–759. [5] U. Landmesser, D.R. Holmes Jr., Left atrial appendage closure: a percutaneous transcatheter approach for stroke prevention in atrial fibrillation, Eur. Heart J. 33 (2012) 698–704.
[6] I. Cruz-Gonzalez, B.P. Yan, Y.Y. Lam, Left atrial appendage exclusion: state-of-the-art, Catheter. Cardiovasc. Interv. 75 (2010) 806–813. [7] J.W. Park, M.A. Sherif, K. Zintl, Y.Y. Lam, M. Goedde, T. Scharnweber, et al., Percutaneous left atrial appendage closure with a novel self-modelizing device: a pre-clinical feasibility study, Int. J. Cardiol. 177 (2014) 957–963. [8] R.S. Schwartz, D.R. Holmes, R.A. Van Tassel, R. Hauser, T.D. Henry, M. Mooney, et al., Left atrial appendage obliteration: mechanisms of healing and intracardiac integration, J. Am. Coll. Cardiol. Cardiovasc. Interv. 3 (2010) 870–877. [9] Y.Y. Lam, B.P. Yan, S.K. Doshi, A. Li, D. Zhang, M.G. Kaya, et al., Preclinical evaluation of a new left atrial appendage occluder (Lifetech LAmbre device) in a canine model, Int. J. Cardiol. 168 (2013) 3996–4001. [10] V.Y. Reddy, H. Sievert, J. Halperin, S.K. Doshi, M. Buchbinder, P. Neuzil, et al., Percutaneous left atrial appendage closure vs warfarin for atrial fibrillation: a randomized clinical trial, JAMA 312 (2014) 1988–1998. [11] A. Tzikas, S. Shakir, S. Gafoor, H. Omran, S. Berti, G. Santoro, et al., Left atrial appendage occlusion for stroke prevention in atrial fibrillation: multicentre experience with the AMPLATZER cardiac plug, EuroIntervention 10 (2015) 1170–1179. [12] G. Bianchi, M. Solinas, T. Gasbarri, S. Bevilacqua, K.K. Tiwari, S. Berti, et al., Pulmonary artery perforation by plug anchoring system after percutaneous closure of left appendage, Ann. Thorac. Surg. 96 (2013) e3–e5. [13] A. Sepahpour, M.K. Ng, P. Storey, M.A. McGuire, Death from pulmonary artery erosion complicating implantation of percutaneous left atrial appendage occlusion device, Heart Rhythm. 10 (2013) 1810–1811.