Cardiac Plug I and Amulet Devices: Left Atrial Appendage Closure for Stroke Prophylaxis in Atrial Fibrillation
Akhil Parashar MD1, E. Murat Tuzcu MD, FACC, FSCAI2, Samir R Kapadia MD, FACC, FSCAI2
1Medicine Institute, Cleveland Clinic Foundation, Cleveland, OH.2Sones Cardiac Catheterization Laboratories, Heart and Vascular Institute, Cleveland Clinic Foundation, Cleveland, OH.
Percutaneous left atrial appendage (LAA) occlusion has emerged as an exciting and effective modality for stroke prophylaxis in patients with non-valvular atrial fibrillation who are deemed too high risk for anticoagulation with warfarin or newer anticoagulants. The Amplatzer devices have been used in LAA occlusion for more than a decade, starting with off label use of an atrial septal occluder device for LAA occlusion. This was followed by introduction of a dedicated Amplatzer cardiac plug (ACP) 1 for LAA occlusion, and more recently, the second generation Amulet device, with reported better stability enhancing features, has been introduced. Both these devices are widely used outside the United States, however in the US only the WATCHMAN device has been FDA approved. Unlike the WATCHMAN device, where the evidence is continuously building as the data from two pivotal randomized controlled trials are emerging, most of the evidence for ACP devices is from pooled multicenter registry data. In this article, we review the device design, implantation techniques and the most recently published evidence for both the Amplatzer cardiac plug 1 and the newer Amulet device. Our goal is to summarize the most recent literature and discuss the current role of the Amplatzer devices in the exciting and rapidly growing field of percutaneous LAA occlusion.
Corresponding Address : Samir R Kapadia, MD FACC FSCAIProfessor, Cleveland Clinic-Lerner College of Medicine,Director, Sones Cardiac Catheterization Laboratories,Department of Cardiovascular Medicine,Cleveland Clinic Foundation,9500 Euclid Avenue, J2-3Cleveland, Ohio 44195.
Atrial fibrillation (AF) is the most common cardiac arrhythmia to affect the population, with some estimates placing it’s prevalence up to 2% in the general population.1 Stroke is the most common morbidity associated with atrial fibrillation, especially in the elderly where up to a third of strokes may be attributed to atrial fibrillation.2,3 Data from the European Community Stroke project showed that strokes associated with atrial fibrillation have a significantly worse outcome both in terms of quality of life and mortality.4
Hence, stroke prophylaxis is a cornerstone of management of atrial fibrillation. Anti-coagulation with warfarin has been the benchmark of stroke prevention in atrial fibrillation. Current use of warfarin as a stroke prevention agent in patients with AF is associated with a decreased rate of residual stroke or systemic embolism (1.6% per year).5 However, over the past few years, newer oral anti-coagulants (NOAC) – factor Xa inhibitors and direct thrombin inhibitors have emerged as exciting alternatives to warfarin to achieve the same goal.6 The most recent guidelines recommend anti-coagulation therapy for everyone with non-valvular atrial fibrillation with prior stroke, transient ischemic attack (TIA), or a CHA2DS2-VASc score of 2 or greater (CHA2DS2-VASc indicates Congestive heart failure, Hypertension, Age ≥75 years (doubled), Diabetes mellitus, prior Stroke or TIA or thromboembolism (doubled), Vascular disease, Age 65 to 74 years, Sex).7,8 However, all forms of anticoagulation increase the risk of both intracranial and extra cranial bleeding, and approximately 30% to 50% of patients with AF are ineligible to receive anticoagulation.9
The left atrial appendage (LAA) is an out-pouching structure in the adult left atrial chamber and persists as a remnant of the embryonic left atrium. In atrial fibrillation, the LAA acts as a site of blood stasis which is thought to significantly increase the risk of thrombogenesis and subsequent systemic embolization of a clot commonly leading to ischemic strokes.10 Current estimates put the risk of thrombus formation at 15% in patients with non-valvular atrial fibrillation and the LAA as the site for clot formation in >90% of these cases.11,12
The surgical approach of excluding the left atrial appendage from the circulation has been previously explored, and it was fraught with high rates of incomplete closure, which in turn led to increased stroke risk.13 Percutaneous left atrial appendage occlusion devices offer a viable alternative to oral anticoagulants in patients who are deemed high risk for bleeding or are otherwise ineligible to receive anti-coagulation. In the following sections we extensively review evidence behind two generations of one such device, designed exclusively for the minimally invasive endocardial approach towards closure of the left atrial appendage. The device includes the first generation Amplatzer cardiac plug (ACP) and an improved second generation Amulet device.14 In this review we summarize the indications, device
design, implantation technique, current literature on outcomes and
future scope of the Amplatzer devices.
Amplatzer Cardiac plug (ACP) 1
The system consists of a delivery catheter, a deployment wire and
the self-expanding plugging device made of nitinol mesh with two
polyester patches sewn to a lobe in the distal part and a disc in the
proximal segment which are connected via an articulated waist.
The design is aimed at sealing the body and ostium of the LAA,
respectively, using the “pacifier effect.’ The lobe is usually implanted
10 mm inside the LAA body, and the anchoring mechanism is aided
by six pairs of stabilizing wires. The device is available in 16, 18, 20,
22, 24, 26, 28 and 30 mm (with 9, 10 and 13 F sheaths) corresponding
to LAA diameters of 12.6 to 28.5 mm. The device has to be manually
loaded into the delivery cable before implantation.
The Amulet or ACP 2 is the second generation of the ACP device
which retains the basic structure of ACP 1 with some significant
improvements.14 The device has a pre-loaded system eliminating the
need for manual loading and also has stiffer stabilizing wires. It is
also available in larger disc diameters, longer lobe and waist length,thus usually implanted approximately 12 mm inside the LAA cavity.
Larger sizes of 31mm and 34mm are also available, comprised of
more stabilizing wires. The inner wire allows re-evaluation of the
device orientation thereby allowing more room for post-deployment
adjustment. The Amulet device, as of this writing, is undergoing a
modification of its delivery systems with a plan of a relaunch soon15
(Figure 1).
Figure 1. Comparison between the ACP 1 and the ACP 2. Comparison between the ACP 1 (left) and the ACP 2 (right) highlighting the greater diameter of the ACP 2, distal lobe (A and B) and waist (B), the increased number of stabilizing wires (A) and the inversion of the disc endscrew (C). (Adapted DIRECTLY from Friexa et al)14 PERMISSIONS PENDING
Guidelines for Percutaneous LAA Occlusion
There is a wide variation in the indications for Percutaneous LAA
occlusion, depending on the region of the world and the existing
regulatory framework. The recent AHA/ACC atrial fibrillation
guidelines do not mention percutaneous LAA occlusion, as the
WATCHMAN device was just recently approved by the food and
drug administration (FDA).7 All device implantations in the United
States have been experimental in patients considered high risk for
bleeding with anticoagulation. The Amplatzer Cardiac Plug device
is currently not approved for use in the USA on a commercial
basis. However, the European Society of Cardiology in its focused
update to atrial fibrillation guidelines in 2012 offered a class IIb
recommendation for percutaneous LAA closure (Level of evidence
B) in patients with high stroke risk who are otherwise ineligible
for long term oral anticoagulation.8 There are no specific data on
comparability of different percutaneous devices; hence there are
no specific guidelines to prefer one device over the other. The ACP
received the CE mark in December 2008, and the Amulet received it
in January 2013. In a recent pooled analysis of multicentric registries
evaluating the ACP device, the most common indications for LAA
occlusion were previous major bleeding (47%), high bleeding risk
(35%), stroke on Warfarin (16%), and coronary stents (22%). Notably,
most patients had >1 indication.16
Baseline Imaging For Planning The Procedure
Trans-Esophageal Echocardiography
Almost all operators prefer to define LAA anatomy via a prior
TEE. This is paramount for excluding any pre-existing LAA
thrombus and also to aid in selection of the right device size. It is
highly recommended to measure the LAA at both the long axis
(120 to 150 degrees) and the short axis (30 to 60 degrees). The
echocardiographic orifice is defined as the line from the pulmonary
vein ridge to the circumflex artery. The deployment zone (“landing
zone”) for the ACP 1 is approximately 10mm inside the orifice at a
right angle to the neck axis. However, for the Amulet, the deployment zone is approximately 12-15 mm from the orifice (Figure 2).17
Figure 2. Transesophageal echocardiogram assessment of the LAA. Evaluation of the septum-LAA ostium distance (A), angle of the appendage (B), and distance from the ostium to the first bend of the appendage (C). LA: left atrium; LAA: left atrial appendage; LV: left ventricle. (Adapted with permission from Krishnaswamy et al)19
Cardiac Computed Tomography
Cardiac CT is emerging as a promising alternative imaging
modality. It has an added advantage of a better 3 dimensional
resolution.18 However, the experience in pre-procedural imaging
is limited, as of this writing. Cardiac CT may play a vital role in
determining the relationship between the LAA and surrounding
crucial structures including the left superior pulmonary vein and the
left aortic sinus (Figure 3). Efforts are currently underway to develop
cardiac CT protocols for adjunctive imaging prior to the procedure
to aid in procedural planning.19 Autopsy studies have demonstrated
that the LAA ostium can be located at different levels relative to the
left superior pulmonary vein ostium.20 There is also ongoing work on
the role of Cardiac CT in post-procedural surveillance.21 However, at
present TEE is the modality of choice for pre-procedural planning in
most centers, with cardiac CT at best being a useful adjunct.
Figure 3. Systematic evaluation of distance along three planes from the fossa ovalis to the left atrial appendage ostium. Lateral distance from the FO (black dot) to the lateral margin (yellow dot) is measured in the axial (A) image using the coronal view (B) as a reference for the plane of the LAA ostium; (C) subsequent anterior distance to the plane of the LAA ostium (yellow dot to blue dot) and (D) subsequent superior distance to the LAA ostium (blue dot to red dot) are measured in the coronal view. FO: fossa ovalis; LA: left atrium; LAA: left atrial appendage; LV: left ventricle; RA: right atrium (Adapted with permission from Krishnaswamy et al)19
Imaging during the procedure
TEE remains the intra-procedural modality of choice for most
operators reporting implantation of ACP devices with a few
exceptions.22 The procedure is performed both under local and
general anesthesia depending on the institutional preferences. It is
generally recommended that all ACP device implantations should
be done under TEE guidance in the setting of general anesthesia,
with the exception of centers that are highly experienced with this
technique.
The femoral vein is the access site, right being preferred over the
left, in most of the reported studies for ACP device implantation.
The rationale is that approach from the right side provides a more
direct access for trans-septal puncture than the left femoral approach.
After obtaining the access, the next step is trans-septal puncture to
gain access to the left atrial cavity. However, some operators have
also used a patent foramen ovale or a preexisting atrial septal defect
to gain access to the left atrium, thereby eliminating the need for
trans-septal puncture.22 The trans-septal puncture is usually made
at the postero-inferior atrial septum at the fossa ovalis under TEE
guidance. Following the puncture, a pigtail catheter is placed in the
LAA and angiographic measurements are performed, preferably in RAO cranial projections to visualize the orifice and proximal part of
the appendage.23
Access Sheath Placement And Device Implantation
A stiff wire is passed and placed in the left upper pulmonary vein
over which the sheath (appropriately sized) is passed up to the vein
ostium. The ACP devices are usually upsized by approximately 4 mm
for ACP 1 and 3 mm for Amulet for proper anchoring and stability.
The device is advanced to the “landing zone” inside the LAA, and the
device is ready to be deployed (Figure 4). The following criteria have
been defined in the literature to ensure adequate deployment:
(1) adequate alignment of the device lobe in the LAA and adequate
(2) concave shape of the disk for good seal;
(3) separation of lobe and the disk;
(4) right angle of lobe to the neck axis at the “landing zone”;
(5) lobe position at least two-thirds distal to left circumflex.24
Once these criteria are met on TEE imaging, the device is released
by twisting the delivery cable counterclockwise.
Figure 4. A) Angiography of Amplatzer Cardiac Plug – lobe “ball” shape; B) Deployment of Amplatzer Cardiac Plug – lobe; C) Angiography of Amplatzer Cardiac Plug – lobe positioned RAO 40° – caudal 20°; D) Deployment of Amplatzer Cardiac Plug – disc. (Adapted DIRECTLY from Berti et al23) PERMISSIONS PENDING
Amplatzer Cardiac Plug 1 Device
The Amplatzer cardiac plug (ACP) 1 and the second generation
Amulet devices are among the two most commonly used
percutaneous LAA occlusion devices in the world (the other being
the WATCHMAN device). The data from randomized controlled
trials (RCTs) are only available for the WATCHMAN device.25-27
Most of the data for the ACP devices is derived from small registries
maintained at centers outside the United States17,22,28-35 [Table 1].
A prospective randomized multicenter controlled trial is presently
underway to compare this device head-to-head to long-term OAC
with warfarin or dabigatran in a 2:1 randomization strategy.36 Most
promising data for the ACP devices are derived from a pooled
analysis of 1047 consecutive patients from 22 centers in Europe,
Asia, Latin America and Canada recently published by Tzikas et
al.16 They reported pooled procedural success as 97.3% with 5%
periprocedural major adverse events. Mean follow up was 13 months
(1349 patient years), and one year all-cause mortality was reported at
4.2%. The stroke rate is reported at 0.9%, TIA’s at 0.9% (9 each) and
systemic embolism was 2.3% (31 events). Since, the data are derived
from pooled estimates from individual registries, there was no control
group.
Table 1. Summary of procedural events in studies reporting implantation of Amplatzer Cardiac Plug 1 device
| Author | Region | Duration | Number | Procedural success, % | Ischemic Stroke, % | Embolization rate, % | Severe Pericardial effusion, % |
|---|
| Park et al28 | Europe | 2008-09 | 143 | 96 | 2.1 | 1.4 | 3.5 |
| Lam et al29 | Asia-Pacific | 2009-10 | 20 | 95 | 0 | 0 | 0 |
| Damonte et al40 | L. America | 2009-12 | 60 | 100 | 0 | 1.7 | 6.6 |
| Lopez-Minguez et al30 | Europe | 2009-11 | 35 | 97 | 0 | 0 | 0 |
| Streb et al31 | Europe | 2009-12 | 21 | 95 | 0 | 0 | 4.8 |
| Bethencourt et al | Europe | 2009-11 | 213 | 93 | 0.5 | 1.9 | 1.4 |
| Nielispach et al22 | Europe | 2008-12 | 120 | 98 | 0.8 | 1.6 | 1.6 |
| Urena et al32 | Canada | 2009-11 | 52 | 98 | 0 | 1.9 | 0 |
| Plicht et al33 | Europe | 2009-11 | 20 | 100 | 0 | 0 | 5 |
| Meerkin et al34 | Israel | 2009-12 | 100 | 100 | 0 | 0 | 1 |
| Santoro et al35 | Europe | 2009-12 | 134 | 93 | 0 | 0 | 2.4 |
| Kefer et al41 | Europe | 2009-12 | 90 | 99 | 0 | 0 | 3.3 |
| Park et al42 | Europe | 2009-11 | 204 | 97 | 0 | 1.5 | 2.4 |
| Gloeckler et al17 | Europe | 2012-13 | 50 | 98 | 0 | 4 | 6 |
Sub-group analysis from another systematic review evaluating
the safety and efficacy of percutaneous LAA devices demonstrated
a stroke rate of 0.9% [95% CI :(0.7-2.4)] after implantation of an
Amplatzer Cardiac plug device. The periprocedural adverse event rate
was reported as 23.5% [95% CI: (15.9-33.2%)] for ACP devices.37
Santoro et al have recently reported up to 4 years of follow up
data on a group of 134 patients implanted with an ACP 1 device,
representing 238 patient years of follow up. They report an ischemic
stroke rate of 0.8/100 person-years, thromboembolic event rate of
2.5/100 person-years and all-cause mortality of 2.5% over the follow
up period.35
Unlike the ACP 1 device, there are very few studies evaluating the
newer Amulet device [Table 2]. Lam et al reported a case series of 17
patients with follow up data available up to 90 days.24 They reported
a procedural success of 100% and no procedural complication except
for 1 case of pericardial effusion. A larger case series of 25 patients
reported by Freixa et al reported a procedural success of 96% without any complication at up to 3 months of follow up38. A recent report by
Gloeckler et al compares the last consecutive 50 ACP 1 cases with the
first 50 consecutive Amulet cases in a non-randomized manner.17 The
study gives an interesting insight into the efficacy and safety of the
newer Amulet device versus the ACP 1, wherein patient population
and operating conditions, including the procedure specialists, are
similar between the two groups. The devices were similar in efficacy
and safety per their analysis. The authors conclude that, at least in
the early experience, the Amulet offers no significant benefit over
the ACP 1 device except for a non-significant reduction in rates of
pericardial effusion.
Table 2. Summary of procedural events in studies reporting implantation of Amplatzer Cardiac Plug 1 device
| Author | Region | Duration | N | Procedural success, % | Ischemic Stroke, % | Embolization rate, % | Severe Pericardial effusion, % |
|---|
| Freixa et al38 | Europe-Canada | 2012-2013 | 25 | 96 | 0 | 0 | 0 |
| Lam et al24 | Europe | 2013 | 17 | 100 | 0 | 0 | 6 |
| Gloekler et al17 | Europe | 2012-13 | 50 | 94 | 0 | 6 | 6 |
Left atrial appendage occlusion by percutaneous strategy is
a rapidly growing discipline in the field of structural cardiac
interventions. The technology involving multiple devices has a
potential to modify the risk of stroke in patients with non-valvular
atrial fibrillation. The Amplatzer cardiac plug 1 and the second
generation Amulet devices seem promising from the limited non
randomized controlled trial data available from centers mostly
outside the United States. In light of the promising results for the
most commonly used WATCHMAN device from two randomized
studies, it seems obvious that similar studies are also needed to assess
the efficacy and safety of the Amplatzer devices. However some
additional key questions remain unanswered before these devices regulatory framework, training requirements, role of learning curve
and most importantly selection of patient cohorts who are most
likely to benefit from these devices. Also relevant are the data on
superiority versus non-inferiority of new devices to warfarin39. The
answers to these questions can only be derived from more data on the
outcomes after implantation of these devices. Considering the recent
approval of the WATCHMAN device by FDA, it is unlikely that
the Amplatzer devices will see a head-to-head comparison with the
WATCHMAN device in any randomized study. It will be interesting
to see the results from the presently ongoing ACP trial for definite
assessment of safety and efficacy outcomes in ACP devices. The
need of the hour currently is to standardize the outcome measures
and possibly also create a nationwide registry of percutaneous LAA
occlusion devices including the Amplatzer, so that more high quality
data may be generated.
