EHRA/EAPCI expert consensus statement on catheter-based left atrial appendage occlusion
EHRA/EAPCI expert consensus statement on catheter-based left atrial appendage occlusion
Bernhard Meier (EAPCI Chairperson) (Switzerland) 2 3
Yuri Blaauw (The Netherlands) 1 3
Ahmed A. Khattab (Switzerland) 2 3
Torsten Lewalter (Germany) 0 3
Horst Sievert (Germany) 3 6
Claudio Tondo (Italy) 3 5
Michael Glikson (EHRA Chairperson) (Israel) 3 4
0 Isar Medical Centre , 80331 Munich , Germany
1 Department of Cardiology, Maastricht University Medical Center , 6281 Maastricht , The Netherlands
2 Cardiology, Bern University Hospital , 3010 Bern , Switzerland
3 Document Reviewers: Gregory Y. H. Lip (UK) , Jose Lopez-Minguez (Spain), Marco Roffi (Switzerland), Carsten Israel (Germany), Dariusz Dudek (Poland), Irene Savelieva, on behalf of EP-Europace , UK
4 Davidai Arrhythmia Center, Sheba Medical Center , 52621 Tel Hashomer , Israel
5 Cardiac Arrhythmia Research Center, Centro Cardiologico Monzino, IRCCS , 20138 Milan , Italy
6 Cardiovascular Center Frankfurt , 60389 Frankfurt , Germany
* Corresponding author. Tel: +972 52 6667128. E-mail address: Document is endorsed by the European Heart Rhythm Association (EHRA) and the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Developed in partnership with EHRA and EAPCI. The article has been co-published with permission in EP-Europace, HeartRhythm and Eurointervention. All rights reserved in respect of Eurointervention. & The Authors 2014. For EP-Europace, & The Author 2014.
Atrial fibrillation (AF) is the most common clinically relevant cardiac
arrhythmia. The estimated prevalence in the general population is
1 – 2% and increases with age.1 – 8
Patients with AF are at increased risk of thromboembolism, in
particular ischaemic stroke. The risk of stroke in patients with
nonvalvular (essentially non-rheumatic) AF is 5% per year.9 Moreover,
strokes related to AF are associated with a higher mortality and
morbidity when compared with non-AF strokes, emphasizing the need
for more effective stroke prevention in these patients.10
The CHADS2 score (cardiac failure, hypertension, age, diabetes,
stroke counted double) was established to assess the risk of
thromboembolic events in patients with AF of non-valvular origin.11 Although
there is a clear relationship between the CHADS2 score and stroke
rate, the CHA2DS2-VASc score has recently been introduced and
adopted by the European Society of Cardiology (ESC) as well as by
American Heart Association, American College of Cardiology, and
Heart Rhythm Society and other national bodies’ guidelines for AF in
an attempt to improve risk stratification in the low-risk group by
considering additional stroke risk factors (gender, vascular disease) in
addition to old factors including cardiac failure, hypertension, age (divided
to two risk classes) diabetes, and stroke that may influence a decision
for anticoagulation therapy.12
Prospective and randomized studies show that oral
anticoagulation (OAC) significantly reduces the risk of thromboembolism.13
However, this treatment is underutilized in patients with AF due to
poor patient compliance, contraindications, and potential bleeding
complications.14 – 18
The pathogenesis of thrombogenesis in AF is multifactorial and
includes the Virchow triad of events leading to thrombus formation,
i.e. endothelial or endocardial damage or dysfunction, abnormal blood
stasis, and altered haemostasis, platelet function, and fibrinolysis.19
There is evidence for endothelial damage, as well as intense fibrosis
and inflammation in the left atrium (LA) in patients with AF.20 These
changes are especially prominent in the left atrial appendage (LAA), a
particularly low flow area, and may enhance to thrombus formation
by their effect on the endocardial surface.21,22
There is also strong evidence for the presence of a prothrombotic
and hypercoagulable state in AF, as manifested by increased blood
levels of markers reflecting coagulation activity (prothrombin
fragments 1 and 2, fibrinopeptide A, thrombin – antithrombin complexes,
and D dimer).23,24
The LAA is the remnant of the embryonic LA. The LAA is a tubular
blind-ended structure with different lobes and variable morphology. Its
complex structure with areas of relative low flow predisposes to stasis,
especially during AF when blood flow velocity decreases, as can be
visualized on transoesophageal echocardiography (TOE) examination
with spontaneous contrast (smoke) or on pulsed-wave Doppler
during paroxysms of AF.25– 28 It has been shown that in patients with
non-valvular AF, 90% of thrombi are located in the LAA.29 Thrombi
detected in the LAA as well as a reduced LAA peak flow velocity
were identified as independent predictors of an increased
thromboembolic risk,30,31 and also for recurrence of stroke among non-valvular
AF patients recovering from ischaemic stroke.32
Moreover, patients with certain LAA morphologies have been
shown to have different levels of thrombo-embolic risk further
supporting the role of LAA in embolization.33
Left atrial appendage occlusion or exclusion in AF34 – 50 is based on
the concept that only 10% of clinically relevant emboli in non-valvular
AF do not originate in the LAA.51– 61 The rationale is that, after
excluding the LAA as an embolic source, the remaining small risk does not
longer warrant OAC with its inherent risk for major bleeds. The risk
of embolism from the LAA or the LA increases with age, but so does
the risk of bleeding under OAC. Various surgical and catheter-based
methods have been developed to exclude the LAA and the success
of catheter-based methods attests to the validity of this concept.62
This document reviews the catheter-based methods and their results.
Surgical techniques for left atrial appendage exclusion
Incidental surgical LAA exclusions during heart surgery have been
performed for decades.63 A first report by Madden dates back to
1949.34 The popularity of this procedure remained low as it did
prolong the surgical procedure and required some specific
techniques. Moreover, follow-up TOE often detected residual flow in the
LAA in case of a simple suture64 and in a large number of these
patients the need for lifelong OAC with a vitamin K antagonist
(VKA) remained due to indications unrelated to AF, most commonly
mechanical valve prostheses in the mitral position. An
electrocardiographically guided thoracoscopic technique for isolated surgical LAA
occlusion65 and a percutaneous endocardial/epicardial approach
(Lariat, SentreHeart)46,47,50 where an epicardial sling suture is
guided by a magnet inside the LAA have been introduced more
recently. In addition, a number of other minimally invasive surgical
and percutaneous devices including the AtriClip, Cardioablate, and
Aegis, are at various stages of advanced animal studies or first in
Catheter-based left atrial appendage occlusion
The electrophysiologist Michael Lesh conceived a device called
PLAATO (Percutaneous Left Atrial Appendage Transcatheter
Occlusion) for percutaneous plugging of the LAA, intrigued by the fact that
during ablation of AF the LAA was easily accessible. He assisted Horst
Sievert’s first such intervention on 30 August 2001.38 The PLAATO
device (Medtronic) had a number of significant drawbacks and the
implantation technique was fairly difficult and perilous. The device
was pulled off the market although clinical results were favourable.43
On 15 June 2002, percutaneous LAA occlusion without general
anaesthesia or echocardiographic guidance in awake patients was
introduced by Bernhard Meier using the technically simpler
Amplatzer approach41 and taking advantage of the double-disc devices
routinely used for occlusion of an atrial septal defect (ASD) or a patent
foramen ovale (PFO). The disc destined for the right side of the
interatrial septum in ASD or PFO occlusion covered the entrance
of the LAA not unlike the plate of a pacifier outside a toddler’s
mouth (pacifier principle). Subsequently, the Amplatzer devices
and introducer sheaths (St Jude) were adapted for LAA occlusion.
The dedicated Amplatzer Cardiac Plug (ACP) and LAA sheath
were introduced in 2008.
On 12 August 2002, the Watchman device (Boston Scientific) was
introduced into clinical practice by Eugen Hauptmann and Eberhard
Grube. It has since undergone several modifications and is approved
in many countries worldwide. It remains the only device studied in
randomized trials, such as PROTECT AF45 and the PREVAIL.66 In
December 2013, an Food and Drug Administration (FDA) advisory
committee voted favourably for approval of this device for use in
the USA as an alternative to warfarin.
The WaveCrest device (Johnson and Johnson) has recently
received CE mark, as well. It was developed with separately applicable
fixation anchors and a different design intended to provide more
superficial deployment at the entrance to LAA with little or no
manipulation within the LAA body.
Since 2010, a percutaneously inserted intra-LAA patch has been
used by a group around Eleftherios Sideris.49 Other devices,
currently in early animal or human trials, have been developed by, Occlutech,
Gore, and Lifetech.
The feasibility of the mentioned Lariat non-surgical combined
endocardial/epicardial suture ligation of the LAA was first
demonstrated in animals by Lee et al. in 2010,46 then in humans by Bartus
et al.47 in 2011, and subsequently evaluated in clinical routine. The
device has CE mark and is approved by the FDA.
Currently available devices and techniques including some surgical techniques
A variety of surgical approaches have been examined mainly in
observational studies and with mixed results.37,40,42,64,67 Two alternative
concepts to achieve LAA occlusion are obstruction of the LAA
orifice with an occlusion device41,45,48,68 or percutaneous suture
ligation using an endocardial/epicardial approach.47
Currently, three entirely catheter-based devices are commercially
used for mechanical orifice obstruction, the Watchman and
WaveCrest devices and the ACP. The Lariat device is used for
percutaneous endocardial/epicardial suture ligation.46,47,50 They all have
obtained CE mark.
The Watchman device consists of a nitinol cage (Figure 1) with a
160 mm polyethylene terephthalate(PTFE) membrane covering the
surface facing the LA. Fixation barbs are attached to the portion
facing the circumference of the appendage minimizing the risk of
dislodgement and embolization. It is attached to a delivery cable and
delivered via a 14 French (F, outer diameter 4.7 mm) access sheath.
A single curve or double curve configuration sheath can be used
depending on the appendage orientation. After transseptal puncture (a
low posterior puncture location is preferred to allow coaxial
alignment with the appendage), intravenous heparin is administered
maintaining an activated clotting time (ACT) .250 s and a pigtail catheter
is positioned into the LAA over a soft J-tipped 0.035 inch wire.
Angiography of the LA focusing on the LAA is performed in several views
[right anterior oblique (RAO) caudal and cranial projections typically
outline the LAA best], delineating shape and size. Sizing of the device,
taking advantage of both cine angiography and TOE, is discussed
under the ‘Imaging for left atrial appendage occlusion’ section. The
device size is typically chosen 10 – 20% larger than the diameter of
the landing zone (measured from the area of the left circumflex
coronary artery across the LAA to 1 cm inward from the tip of the
ridge separating LAA and left upper pulmonary vein). Subsequently,
an extra-stiff J tipped 0.035 inch wire is advanced into the distal LAA
and the pigtail catheter and transseptal sheath are exchanged for the
access sheath while maintaining wire position. Some operators
introduce a catheter into the left upper pulmonary vein first instead of
aiming at the LAA. In this case, after transseptal puncture, an
extrastiff 0.035 inch wire is positioned into the left upper pulmonary
vein and the transseptal sheath is exchanged over the wire for the
access catheter. Subsequently, a pigtail catheter is advanced
through the access sheath to the LAA. The access sheath has three
markers corresponding to device size and is advanced into the
LAA until the marker aligns with the ostial plane of the appendage.
After purging, the device is advanced via a delivery catheter to the
distal end of the access sheath. Finally, the access sheath and delivery
catheter are slowly withdrawn while maintaining device position,
allowing it to unfold. Once deployed, appropriate position is
confirmed by both angiography and TOE. A tug test is performed
under fluoroscopy or TOE demonstrating simultaneous movement
of the device and appendage. Optimally, the device should not
protrude .4 – 7 mm beyond the LAA ostium (depending on device
size outlined in the manufacturer’s instructions for use manual) and
should cover the entire ostium with no or minimal (,5 mm by
colour Doppler) residual flow and a compression grade of 8 – 20%
(some recommend a higher compression grade of 15 – 30%). The
compression grade is expressed in per cent comparing the diameter
of the implanted device with the unconstricted diameter indicated by
the manufacturer in the size label. When optimal positioning is
confirmed, the device is released. If position or size appears suboptimal,
the device can be retrieved and exchanged or repositioned. Table 1
lists the basic steps of LAA device implantation.
The Amplatzer ACP consists of a cylindrical nitinol cage (lobe)
securing the device in the LAA body connected by a short flexible
waist to a nitinol plate (disc) covering the appendage ostium. Both
are laid in with polyester fabric (Figure 2). Similar to the Watchman
device, the cage is surrounded by fixation hooks. The flexible waist
facilitates positioning and conformation to variable and complex
appendage shapes. Of note, contrary to the Watchman device, the
length of the ACP is shorter than its diameter. Therefore, whereas
the Watchman device cannot be implanted in appendages shorter
than wide, the ACP may be an option under those circumstances.
In fact, ACP implantation can be attempted in virtually all appendages.
The more recently introduced Amulet generation of that device may
Transoesophageal or intracardiac echocardiography immediately prior to the procedure to rule out LAA thrombus (contraindication for the procedure)
) Femoral venous access
) Transseptal puncture (typically in an inferoposterior location)
Heparin via the sheath or intravenously with maintenance of a goal ACT of .250 s (some prefer heparin administration prior to transseptal puncture)
) Sheath access to the LAA (either option a or b)
(a) A pigtail catheter is advanced into the LAA (and angiography performed) and subsequently exchanged over a stiff guidewire for the delivery sheath
(b) A catheter (typically a multipurpose catheter) is advanced into the left upper pulmonary vein, and exchanged over a stiff guidewire for the delivery
sheath. A pigtail catheter is advanced via the delivery sheath into the LAA (and angiography performed) and subsequently the delivery sheath is
advanced over the pigtail catheter into the LAA
) LAA measurements are made and the appropriate device size is chosen (10– 20% larger than the landing zone diameter)
) The sheath is advanced over a stiff guidewire (in case of option 3a) or the pigtail catheter (in case of option 3b) until the proximal marker corresponding
with the device size matches the LAA ostium
) The stiff guidewire (in case of option 3a) or pigtail catheter (in case of option 3b) is removed
) Blood is allowed to exit the sheath while holding the sheath hub and flush line as low as possible (below the patient’s anticipated midline of the chest) to
eliminate any air trapped in the sheath
) Device preparation (generous flushing of the device within the delivery catheter)
) The delivery catheter and device are advanced until the distal marker of the delivery catheter and delivery sheath match
) The sheath is gradually pulled back as a unit while maintaining delivery cable position to allow the device to unfold
) Position is confirmed via echocardiography and fluoroscopy and a tug test is performed
) Device release (e.g. delivery cable turned counter clockwise)
ACT, activated clotting time; LAA, left atrial appendage.
improve ease and safety of use and further expands the size range of
appendages that can fit the device.69 It also has dimensional changes
that are intended to improve stability and occlusion of the LAA os by
the disc. The device fits landing zones from 11 to 31 mm. Femoral
venous access (sheath sizes 9 – 14 F inner diameter depending on
the device size and type), transseptal puncture, LAA angiography,
and TOE imaging as well as delivery sheath positioning are as
previously outlined (Table 1). The sheath may alternatively be directly
advanced into the LAA. Like the Watchman device, the ACP is
retrievable prior to disconnection from the pusher cable. With the
delivery sheath at least 15 mm inside the LAA, the first half of the device
(lobe) is delivered by sheath retraction and the second half by pushing
it out. Then, the disc is produced by further retracting the sheath
while still gently pushing on the device. With optimal positioning,
the lobe should be visibly compressed (tire shape) with an
appreciable distance to the disc, connected by a stretchable waist. The
disc should assume a slightly concave shape and cover the entire
LAA ostium or at least most of it (pacifier principle). After a sustained
tug test and confirmation of an optimal position, the ACP is released.
The WaveCrest device consists of a nitinol structure without
exposed metal hub and with a foam layer facing the LAA to
promote rapid organization and a PTFE layer facing the LA to
reduce thrombus formation (Figure 3). It is conformable to LAA
anatomy and fixation anchors are separately actionable and radially
positioned to provide effective fixation at the appendage once the
desired position in attained. The WaveCrest delivery sheath is
designed to optionally position the occluder in the LAA ostium
during deployment and anchoring. Of note, the delivery sheath is
not intended for deep access and manipulation inside the appendage
as the device is designed and intended for proximal placement.
Should the Watchman or the ACP devices be deemed too large
for very short appendages, the WaveCrest may provide an
alternative. As previously described for the Watchman and ACP devices
(Table 1), LA access is provided by a 12 F sheath in the femoral vein
and a preferably posterior transseptal puncture. The measurements
of the projected landing zone on TOE include the distance from the
left circumflex coronary artery to 10 mm distal to the apex of the
lateral ridge (coumadin ridge). Most importantly, measurements
should include the widest part of the ostium, since the positioning
has to be in the proximal end of the LAA mouth, allowing headroom
for anchors. Measurement at 08, 458, 908, and 1358 are
recommended to capture the long axis and short axis of the ostium and
the 1358 view usually shows the widest diameter. The more proximal
positioning of the WaveCrest device is related to the concept that
distal deployment may compress the device itself and the anchors.
Over-compressed anchors may become entangled. Therefore, a
position proximal to all lobes guarantees best occlusion and the
risk of pericardial effusion is minimized. Before detaching the
device, the sheath needs to be pulled back 2 cm from the occluder
and contrast medium is injected through the delivery system port to
visualize the distal LAA. A tug on the delivery catheter is performed
until movement is seen (device and tissue move as a unit). In case
repositioning appears necessary, the hooks are withdrawn before
moving the device. After these steps, the device is set free.
The endocardial/epicardial Lariat approach to LAA occlusion
leaving no foreign material in the heart is more complicated. A
lassolike suture, (snare), is positioned by a percutaneous technique
epicardially at the base of the LAA and tightened followed by suture ligation.
First, epicardial access is obtained similar to epicardial access for
electrophysiological ablations70 and an epicardial soft tipped 14 F access
cannula inserted into the pericardial space. Secondly, femoral venous
access is established and transseptal puncture performed. Via an 8.5 F
delivery sheath (e.g. SL 1 transseptal catheter by St Jude), a specially
designed magnet tipped 0.025 inch endocardial guide wire is
advanced into the LAA apex followed by a balloon mounted
(compliant 20 mm balloon) catheter. The position of the endocardial guide
wire is confirmed via contrast medium injection through the
balloon catheter lumen. Via the percutaneous epicardial access
sheath, a second 0.035 inch magnet tipped epicardial wire is advanced
towards the LAA and aligned with a magnet located at the distal end
of the endocardial wire already located in the LAA apex. The balloon
in the LAA is inflated to help identify the appendage ostium and allow
a lasso delivered via the epicardial sheath over the epicardial wire to
grab the LAA ostium. Finally, the lasso is tightened. Appendage
occlusion is confirmed by TOE and fluoroscopic imaging and a suture is
deployed. The epicardial and endocardial delivery systems are
Imaging for left atrial appendage occlusion
Adequate implementation of various imaging modalities is essential
for developing a successful LAA occlusion programme. Imaging is
important for pre-procedural and periprocedural assessment of the
LAA and for follow-up. The LAA can occasionally be visualized
with transthoracic echocardiography (TTE) but usually requires
TOE, intracardiac echocardiography (ICE), cardiac magnetic
resonance imaging (MRI), or computerized tomography (CT).
Transoesophageal echocardiography is an integral part for guidance in most but
not all41 LAA occlusion publications. Imaging modalities continue to
evolve. The value of newer modalities should be compared against
TOE, the gold standard for imaging the LAA and guiding LAA
It is important to confirm the absence of LAA thrombi prior to LAA
occlusion. The presence of mobile thrombi is a contraindication for
percutaneous LAA occlusion, since dislodgement of thrombus may
occur with manipulation of sheaths or devices in the LAA. Currently,
TOE is considered the reference technique for the detection of
thrombi in the LAA.71 In most patients, the LAA can be adequately
visualized using TOE. Yet, in some patients there may be difficulties
in obtaining unequivocal images, as for example, in patients with
prominent pectinate muscles which may be falsely interpreted as
LAA thrombus. The incidence of LAA thrombus on TOE among
patients undergoing AF ablation who have been adequately
anticoagulated was found to be very low and in those patients an elevated
CHADS2 score was the strongest predictor of LAA thrombus.72
The prevalence of LA or LAA thrombus or sludge (dynamic
gelatinous, precipitous echodensity without a discrete mass) in patients
undergoing TOE examination for pulmonary vein isolation increased
from 0% in patients with CHADS2 score of 0 – 11% in patients with
CHADS2 score of 4 – 6.72 Of note, the prevalence of LAA thrombus
may be higher in patients scheduled for LAA occlusion, since patients
may not be anticoagulated because of previous bleeding
complications. The diagnostic performance of a dual-enhanced cardiac CT
protocol for detection of LAA thrombi was studied in patients with
stroke.73 – 76 The overall sensitivity and specificity of CT for the
detection of thrombi in the LAA were 96 and 100%, respectively.74
The role of cardiac MRI in management pathways for diagnosing
LAA thrombus77 is not well enough defined and further studies are
required. Ad hoc LAA occlusion using LA angiography for thrombus
exclusion has been described in a small series.78
Pre-procedural TOE already hints to the device size or should reveal
if the LAA appears difficult or impossible to occlude. The LAA is best
imaged from the mid-oesophageal view. Using the multiplane function,
the LAA is interrogated in multiple views (08, 458, 908, and 1358). The
morphology and presence of multiple lobes of the LAA are usually only
appreciated at an angle beyond 1008. Characterization of the LAA
shape and the presence of multiple lobes can be facilitated by
threedimensional (3D) TOE or pre-procedural MRI or CT.
Watchman and ACP devices require specific TOE measurements
necessary for choosing the appropriate device sizes. The maximal
width of the LAA ostium and depth of the LAA are first measured
(Figure 4). The maximum LAA ostium width is measured from the
level of the left circumflex coronary artery to a point 1 – 2 cm from
the tip of the left superior pulmonary vein limbus (at 08) and from
the mitral annulus to a point 1 – 2 cm from the limbus (458, 908,
and 1358). The ostium of the LAA usually has an oval shape. It is
recommended to use the diameter of the longest axis (generally
superoinferior). The depth of the LAA is measured from the
ostium line to the apex of the LAA. The Watchman device can be
used if the maximum LAA ostium is .17 or ,31 mm, the ACP if it
is ,28 mm (for larger diameters, deeper placement may be an
alternative or the new Amulet device may be used for landing zones up to
31 mm). For both, the Watchman and ACP devices, sizing tables are
available. In general, the device size should at least be 10 – 20% larger
than the measured diameter, although some operators may prefer up
to 30%. If the depth of the LAA is smaller than the width of the ostium,
placement of a Watchman device may result in unstable position with
unacceptable device protrusion into the LA. For further information
regarding the sizes and choice of the device, refer to the section
Implantation techniques and to Table 2.
Procedural imaging to guide left atrial appendage occlusion
Real-time visualization of the LAA for device positioning and
deployment is a key for successful implantation. As long as the total
procedure time can be kept short, deep sedation may not be necessary.
However, the majority of centres perform the procedure in
general anaesthesia with 2D TOE in combination with X-ray
guidance, while some operators close LAAs under fluoroscopic guidance
alone to facilitate the logistics (less personnel and no sedation or
intubation required).41,78,79 Personnel performing TOE during LAA
occlusion should be well trained and familiar with the procedure
and the required measurements for the type of device used.
Limited data are available on LAA occlusion using ICE, but this may
allow the procedure to be performed under local anaesthesia.80
Reports showed superior visualization of the LAA with the ICE
probe positioned in the LA or pulmonary artery compared with a
position in the right atrium or coronary sinus. To place the ICE
probe in the LA in the absence of a PFO, a second transseptal
passage is required. Left atrial appendage occlusion with ICE should
only be performed by operators with experience in ICE catheter
handling and interpretation of ICE images.
Only 3D TOE can provide a real-time full view of the LAA and
importantly, the shape of the LAA ostium. A recent report demonstrated
that 3D TOE-derived measurements of LAA orifice area were closely
related with CT measurements. In this study, 2D TOE significantly
underestimated LAA dimension and orifice size, as compared with
3D TOE.81 Future studies demonstrating the feasibility and accuracy
of 3D TOE during LAA occlusion procedures are required.
Transoesophageal echocardiography or ICE also facilitate
transseptal puncture. Most operators prefer an inferoposterior, others a
mid to superior and posterior puncture. This illustrates how variable
the anatomy is. A puncture site superior and anterior is usually
suboptimal. Therefore, LA access via a PFO (which results in a superior
and anterior path) is avoided by most operators. After positioning a
sheath or pigtail catheter in the LAA, selective contrast injection
under fluoroscopy in RAO caudal and cranial (
108 – 308
gives excellent views of the LAA not overlapping the LA. Final
decision on device size is based on information collected with both
echocardiography and fluoroscopy. Real-time TOE provides direct
information on the position of the delivery sheath in the LAA and
helps during device deployment. Following successful device
deployment, the pericardium is evaluated for effusion. Some experts
recommend another echocardiography (TTE usually) to confirm device
position and exclude pericardial effusion prior to discharge. We
are not aware of any information on the yield of such an examination.
Rare device embolizations upon first mobilization of the patient
with change of the heart position have been observed.
Transoesophageal echocardiography is the most revealing technique.
Alternatively, post-procedural imaging to assess device position,
peridevice residual flow in the LAA, and thrombus formation on
the device consists of chest X-ray (position only) or CT. Magnetic
resonance imaging is hampered by device artefacts. The timing of a
follow-up TOE varies between institutions. Most operators use
early or follow-up echocardiographic findings, i.e. the absence of
large residual flow into the LAA or thrombus, as a guide for
prescribing antithrombotic drugs. In the PROTECT AF (Watchman Left Atrial
Appendage System for Embolic Protection in Patients with Atrial
Fibrillation) trial, serial TOE imaging was performed at 45 days,
6 months, and 1 year following implant.45 The logic behind the
timing of the first two examinations was based on changes in
medications subsequent to the examination, i.e. warfarin and clopidogrel
were discontinued if TOE showed the absence of thrombus,
occlusion of the LAA, or residual peridevice flow of ,5 mm width
(assessed by colour Doppler) at 45 days and 6 months. A sizable
series with the ACP used a regimen of clopidogrel for 1 and
acetylsalicylic acid (ASA) for 3 to 6 months.82 Follow-up TOE was performed
at variable times between 3 and 6 months following implantation.
For patients treated with coumadin and antiaggregants according
to the PROTECT AF protocol, it is prudent to follow the imaging
protocol of the trial, as it serves as a guide to changes in medications.
In other patients who are treated with antiaggregants only, it makes
sense to image prior to cessation of clopidogrel and again if ASA
cessation is planned. It is recommended to perform a TOE at 45 days to
6 months after implantation, since most adverse events including
device dislodgement and thrombi were so far documented at the
45-day TOE. The value of additional TOE investigations at later
time points is unclear. In case of a new embolic event, a repeat
TOE is indicated as is the case if a TOE demonstrated a significant
leak or a thrombus on the device (see chapter on Anticoagulation).
Residual peridevice flow is a common echocardiographic finding in
patients treated with the Watchman device. There is concern that
this could potentially lead to thrombo-embolic events, since new
thrombi may form in the distal LAA pouch as a result of low flow
velocities. In 41% of patients of the PROTECT AF study, this was
observed during TOE at 45 days.83 This decreased to 32% at the
1-year follow-up. The majority of patients had flow jet widths of
1 – 3 mm. Of note, in patients with peridevice flow who discontinued
warfarin, the clinical outcome was not affected. Amplatzer devices
have less residual flow because the disc of the device typically
covers the entire LAA ostium (pacifier principle). For further
recommendations on antithrombotic treatment in patients with peridevice
flow, we refer to the section on Anticoagulation.
Information is scarce about the role of imaging in Lariat device
implantation and follow-up. Transoesophageal echocardiography is
used to verify LAA occlusion during the procedure and for follow-up,
but no clear protocol has been put forward.
Suggested standards for operators and centres and required information for registries and studies
Requirements for operator and centres
Training and knowledge of physicians performing the
Extensive knowledge of cardiac anatomy, particularly of the LA and
LAA and the surrounding structures is required for operators who
embark on LAA occlusion. Operators must be acquainted with
transseptal puncture techniques and with pericardiocentesis as a
prerequisite to start LAA occlusion training. Experience with other
procedures performed transseptally for an interventional
cardiologist or with LA ablation procedures for an electrophysiologist
should be required to be sufficiently cognizant of LA anatomy and
the anatomical position of LAA in relation to the surrounding
structures.84 – 86 Operators must be aware of the several LAA anatomical
variations, in terms of size, angulation, and mobility. The procedural
skill is also based upon the operator’s ability to put LAA morphology
in relation to the technique and the outcome of the procedure. Left
atrial appendage anatomy plays a key role for the selection of the
most adequate occlusion device. The success of the procedure is
closely related to the level of knowledge and experience of each
individual belonging to the team, including the echocardiographer
supporting the procedure. The learning curve has to be expected rather
flat in light of the intricacy of the procedure.
A key factor for procedural success is a structured training process
before becoming an independent operator. The training process,
currently provided by the device manufacturer, includes basic
principles, specific device features, and the performance of the procedure.
We believe that a training process for implantation of a specific device
should include the following:
) Theoretical course, often taken on line, teaching anatomy,
clinical data, and the theory of device implantation including
interactive cases. This training stage should also include critical
issues, such as patient and device selection, possible
complications, and detection and management of major adverse events
) Practical training including bench training with handling of the
equipment, and attending live cases at experienced centres, or
interactive during congresses or online. The addition of
hands-on simulator training on several virtual cases is helpful
but should not replace attendance at real cases.
) During their first procedures, operators are to be properly
proctored at their sites by experienced operators.
Importantly, in addition to the implanters, echocardiographers
involved in patient evaluation for the procedure and
echocardiographic support during procedures should be specifically trained
for the respective aspects of the procedure.
Centre and laboratory requirements
Since the procedure is usually performed under general anaesthesia
and TOE guidance, an anaesthesiologist and an experienced
echocardiographer who had specific training in supporting LAA occlusion
procedures should be part of the procedural team. Nursing and technical
personnel should also be familiar with every procedural step, well
accustomed to interventional techniques, and prepared to manage
adverse events and emergency situations. Site readiness for the
procedure necessitates not only a knowledgeable operator but a thorough
team understanding of the procedure and of the individual role of each
member of the team. According to current practice in house
cardiovascular surgery in centres performing LAA occlusion procedures is
not deemed mandatory but arrangements for rapid transfer to a
centre with available cardiac surgery should be available, with a
maximum time of 60 min to reach the operating room.
Data collection for registries and studies
In light of the worldwide increasing number of LAA occlusion
procedures, there is a need to collect extensive information regarding the
number of procedures, criteria of patient selection, acute and
longterm clinical outcome, and the occurrence of any type of
complications.87 Initial clinical data are available from the PROTECT AF45
study and Continued Access Protocol (CAP) study analysis88,89
using the Watchman device. They are, in part, offering randomized
comparisons to warfarin, whereas data about success and
complications rates with the ACP device are entirely based on registries with
indirect comparison to OAC.90 Due to the fact that not all patients
with AF and contraindications to OAC or with serious untoward
effects with OAC are suitable for LAA occlusion, there is a need
for extensive data to guide patient selection. In this regard, studies
should be including multiple centres with proven experience of
each operator with at least 10 proctored and 10 main operator
procedures. While new studies are being designed, there is also a need
for large registries of LAA occlusion. To create large registries,
defined inclusion criteria among the different centres (e.g.
CHA2DS2VASC and HAS-BLED scores) are the basis. This allows assessment of
acute and long-term clinical outcome with respect to several issues,
such as successful implantation, periprocedural or late complications,
rate of neurological cardiovascular events, and safety of the approach.
For the purpose of uniformity and ability to compare and group
together results of different registries and studies, Table 3 lists
recommended parameters to be collected in LAA occlusion registries.
Several observational human studies using the no longer available
PLAATO device have been published.38,43,91 –93 In all reports,
antiplatelet therapy only was used after implantation. Invariably, the
postprocedural stroke incidence was lower than projected by the
CHADS2 score for patients treated with ASA only. According to data
from three studies including a total of 359 patients (mean follow-up of
9.6 months, 9.8 months, and 24 months, respectively), the annualized
stroke risks were 2.3, 2.2, and 0.7%, respectively, substantially lower
than the risks predicted based on the CHADS2 score (6.6, 6.3, and
4.9%, respectively).43,92,93 At long-term follow-up (61 patients, mean
follow-up of 5 years), the annual stroke or transient ischaemic attack
rate (3.8%) remained lower than predicted by the CHADS2 score
(6.6%).92 Including data from 364 patients (three studies) with
attempted or successful PLAATO implantation, procedural mortality
and the incidence of pericardial effusion requiring drainage, device
embolization, and periprocedural stroke were 1, 2.5, and 0.5%
respectively.43,92,93 Hence, though associated with a small risk of major
periprocedural events, the PLAATO device appeared to be effective
in reducing the stroke risk equally or better than OAC in patients
with AF when compared with historical controls of equivalent stroke
risk in a non-randomized fashion.
Table 4 summarizes current published and presented clinical
experience with the Watchman device.
First, clinical experience with the device was published in 2007.44
Thereafter, high-risk patients have been enroled into trials and
registries providing data from 1139 patients with over 1500 patient-years
of follow-up. The results demonstrate the safety and efficacy of this
device in preventing thrombo-embolic events compared with
warfarin therapy.45,66,88,89,94 – 96
The Watchman device is the only LAA occlusion device that
has been evaluated in prospective, controlled, randomized trials
examining its efficacy and safety, e.g. in 707 patients with non-valvular
AF.45 The PROTECT AF study was designed to assess the
noninferiority of the device against chronic warfarin therapy. The first
publication included follow-up of 1065 patient-years,45 but data on
1500 patient-years are currently available97 of whom 87%
discontinued OAC at 45 days and 94% after 2 years of follow-up. Although
there was a higher rate of adverse safety events in the intervention
group than in the control group, due mainly to periprocedural
complications (pericardial effusion and procedural stroke typically
related to air embolism), most events were without long-term
sequelae. Safety events in the Watchman group occurred primarily on the
day of the procedure, while the event rate was lower than that of
the control group after the periprocedural period. Importantly,
when follow-up was extended from 600 to 1500 patient-years, there
was a 46% reduction in relative risk from 2.85 to 1.53. Longer follow-up
results were recently presented with a further decrease in primary
event rate and, for the first time, a survival benefit for the Watchman
group when compared with the control warfarin group.96
The effect of increased operator experience is demonstrated in the
CAP registry with shorter implant time, higher implant success rate,
lower complication rates, and higher warfarin discontinuation rate.88
The Randomized Trial of LAA Closure vs. Warfarin for Stroke/
Thromboembolic Prevention in Patients with Nonvalvular Atrial
Fibrillation (PREVAIL) study66 was designed similarly to strengthen the
results of the PROTECT AF trial in more patients at somewhat
higher risk treated by centres with variable experience. Its
preliminary results, presented but not yet published, demonstrated low early
and long-term primary and safety event rates.
In the PROTECT AF study (average CHADS2 score 2.2 and
CHA2DS2-VASc score 3.4), all patients were treated with warfarin for 45 days
after device implantation to facilitate device endothelialization.
Warfarin was stopped if TOE examination (performed after 45 days, 6
months, and 1 year) showed either complete occlusion of the LAA
or if there was residual peridevice flow of ,5 mm in width.
However, recent data support the safety and efficacy of LAA occlusion
in patients with contraindications to even temporary anticoagulation
treated with antiplatelet therapy only after device implantation.95
Peridevice flow is common after Watchman implantation. In a
retrospective analysis evaluating the clinical impact of incomplete
LAA sealing in patients undergoing percutaneous LAA occlusion
with the Watchman device, some degree of peridevice flow was
reported in 47% at 45 days and 33% at 12 months. However, the
overwhelming majority of leaks were small. Peridevice flow .3 mm was
seen in only 12% of patients at 12 months. Most importantly,
compared to patients with complete occlusion there was no difference
in thrombo-embolic events in those with any peridevice flow
regardless of whether or not anticoagulation was continued. Hence, small
amounts of residual flow do not appear to impact safety and clinical
efficacy of Watchman implantation.83 Given the small number of
ASA, acetylsalicylic acid. Other abbreviations as in Table 2.
. ce bo
. v m
.. e e
patients with large residual leaks, however, the safety of
discontinuation of anticoagulation under these circumstances remains unclear.
Gangireddy et al.89 performed an analysis of the net clinical benefit
(difference between the annualized rate of serious events in the
Watchman group and the rate in the warfarin group, assigning different weight
to the events according to severity) of Watchman implantation in
PROTECT AF and CAP. This analysis demonstrated an increased net
clinical benefit with higher CHADS2 scores, especially when the
Watchman device was used for secondary prevention in patients with previous
events.89 The device was also effective in improving quality of life
compared with warfarin therapy.97 It was cost-effective when compared
with warfarin but only marginally so when compared with dabigatran.98
Two ongoing studies are evaluating the device in several hundred
Amplatzer Cardiac Plug and other
Amplatzer devices implanted into the human atria look back at more
than a 20-year history in over a million patients. They have a low
propensity for device thrombosis (,1%) in patients with sinus rhythm.
This track record together with their ease of use led to their
utilization for percutaneous LAA occlusion41 only a few months after the
first percutaneous LAA occlusion procedure was performed with
the PLAATO device.38 With the PLAATO device withdrawn, the
Amplatzer devices have the longest clinical follow-up of currently
available LAA occluders.79
Non-dedicated Amplatzer devices
Experience with ASD occlusion permitted the use of Amplatzer
devices for LAA occlusion under fluoroscopic guidance only.41,79
However, the lack of retaining hooks and suboptimal sheath
configurations resulted in a high embolization rate (6% in the
aforementioned studies). Most embolized devices were retrieved
percutaneously, but in some cases surgical removal (combined
with LAA occlusion) was carried out. None of the available
Amplatzer devices designed for occlusion of atrial or ventricular septal
defects, patent ductus arteriosus, or vascular shunts proved adequate
for LAA occlusion. Therefore, a specific mould was developed
consisting of a hooked lobe, a thin connector, and a proximal disc. The
latter is a remnant of the initial double-disc devices and comes in
handy to cover the orifice of the LAA (pacifier principle). The clinical
outcome in patients with technically successful Amplatzer LAA
occlusion was rewarding with 0.5 events per 100 patient-years
compared with the expected 5.5 events without anticoagulation or 1.8
events with anticoagulation according to the CHADS2 score.99 All
patients in this series were discharged on antiplatelet therapy only.
These results have to be put into perspective of a rate of 1% no
device implanted and 4% device embolization. Interestingly and
somewhat surprisingly, there were no device embolizations in
patients with sinus rhythm at the time of implantation.
Amplatzer Cardiac Plug
Since 2008, the dedicated ACP device with lobe sizes of 16 – 30 mm
diameter (disc diameter slightly larger) and a dedicated double
curve sheath with a modified pusher cable have been used almost
exclusively. Initial registry data reflected the technical improvements
with a reduction of the embolization rate to 2%48,68,79,99 – 105
(Table 5). Pericardial effusion leading to cardiac tamponade requiring
interventions occurred in 2% as did neurological events. These
figures are comparable with those obtained with the PLAATO43 or
the Watchman devices.45,88 In contrast to the Watchman device,
there are hardly any anatomical contraindications (with the
exception of visible mobile thrombus) for an attempt at LAA occlusion
with an ACP. Technical success in the first 200 registry patients was
97% and a relevant thrombus on the device during follow-up TOE
was seen in 3%. The complete occlusion rate at 6-month TOE
was 99%. This is considerably higher than what was achieved with
the Watchman device. The difference can be explained by the ACP
disc occluding the mouth of the LAA in addition to the plug in the
neck (feature shared with the Watchman device). Long-term
followup data are lacking but design and material of the ACP are so close to
that of non-dedicated Amplatzer devices that clinical outcomes can
be expected to be superimposable to those mentioned in the
paragraph above and perhaps even competitive to non vitamin K oral
Even with the user-friendly Amplatzer technique, LAA occlusion
remains challenging and the learning curve is everything but steep,
much in contrast to that of occlusion of the PFO. Notwithstanding,
with growing experience the results improve. Technical
modifications (more and stronger hooks, deeper lobe) and the tendency to
implant larger devices more deeply should lead to improved technical
results. Generally ACPs are implanted without OAC thereafter. The
observed thrombosis rate still leaves room for improvement so that a
brief period of (N)OACs in patients without a contraindication may
be part of future protocols.
There are but preliminary data on the use of Lariat technique. A
recently published series described the results in 89 patients with
96% implant success, with 3 access related complications. Long-term
follow-up revealed severe pericarditis, late stroke, and sudden death
in two patients each and late pericardial effusion in one patient.106
Indications for left atrial appendage occlusion
The recommended indications for the use of LAA occluders are
summarized in Figure 5. In accordance with ESC guidelines, we use the
CHA2DS2-VASc risk score (.1) as the threshold value for LAA
occlusion, despite the fact that some of the evidence (mainly
related to the Watchman device) Is based on CHADS2 scores
≥1.107 We believe that both these definitions may be used.
As alternative to oral anticoagulation when oral anticoagulation is possible
Although this population constitutes a small minority of LAA
occlusion recipients today, this is the only indication that is currently based
on randomized controlled data and was recently recommended by
an FDA panel for approval. Contrary to FDA opinion and emerging
report on the cost-effectiveness of LAA occlusion,98,108 the British
National Health Service Commissioning Board ruled that the
costeffectiveness and clinical effectiveness of the device are not
established enough yet, and therefore the device is not funded in the UK.109
Alae5bT ilI-tsahnpo .............. itsrygeR .................... ilItaan101ittsrgyeR ,ltraeeunCD ragubHm79renB tsEPPoACU trakeM89itsrgyeR isaSnhp99itsrgyeR iilItan raeEunpo48irceeeExnp reLnBAA ilsccunoO010itsrgyeR iiilItsaannA68irceeeExnp iaaanndC501itsrgyeR 45FEPRTTACO
OAC, preferable NOAC
Mention LAA occlusion
1. HAS-BLED score ≥ 3
2. Need for a prolonged triple anticoagulation therapy (e.g.
recent coronary stents)
3. Increased bleeding risk not reflected by the HAS-BLED score
(e.g. thrombopenia, cancer, or risk of tumour associated
bleeding in case of systemic OAC)
4. Renal failure (severe) as contraindication to NOAC
Individual risk/benefit evaluation for
(N)OAC vs. alternative methods
Acceptable risk for systemic
No treatment vs. LAA occlusion
Patient refusal of OAC
1. Contraindication for
2. Refusing systemic
and physicians advice
(includes the need
Atrial fibrillation patient with indication for OAC for stroke/embolism prevention (CHA2DS2-VASc > 1)*
*In all: adequate and intensified rhythm control (ablation or amiodarone) in combination with continuous rhythm control by implanted devices with remote monitoring.
When patients are eligible for OAC and do not exhibit an
increased risk for bleeding, it is the consensus of the writing
committee that the option of LAA occlusion should be mentioned to the
patient while OAC currently remains the standard of therapy. Left
atrial appendage occlusion should not be presented as superior
treatment at this stage. Instead, the advantages and disadvantages of both
treatments should be explained in detail emphasizing that
randomized data currently are limited to two studies with a single device
comparing it with a single agent (warfarin), an oral VKA.45,66 As far
as devices other than Watchman are concerned, they are based
exclusively on observational studies. Long-term outcome after LAA
occlusion (taking into account periprocedural adverse events) was
shown equivalent or according to the 4-year results of the
PROTECT AF study even superior (in terms of stroke prevention
and survival) to anticoagulation with warfarin. Yet, serious
complications related to the procedure itself (including, but not limited to, the
risk of death, stroke, and emergency surgery) occur. Finally, patients
should be educated that NOACs are available that, compared with
oral VKAs, has at least equivalent and probably improved efficacy.
All tested NOACs compounds have a lower rate of intracranial and
some also of overall risk for haemorrhage and they are free of the
logistical challenges associated with surveillance of therapeutic levels. It
should be emphasized that none of them has so far been compared
with LAA occluder devices while there are ample data to show
that they are equal or better than VKAs. Hence, they should be
considered and discussed as an important preventive treatment
alternative with much more supporting evidence than LAA occluders.
Ultimately, the decision should be made by a well-informed patient
in collaboration with the treating physician(s). It should be mentioned
that for this indication the only device that has evidence-based
support for its use is the Watchman device, whereas other devices
were not systematically studied in a randomized controlled fashion.
As replacement for anticoagulation when
anticoagulation is not possible
Patients with a contraindication to anticoagulation
Patients with a high thrombo-embolic risk (CHA2DS2-VASc score
of .2) but contraindication to oral and systemic anticoagulation
(e.g. history of a significant bleeding event such as intracranial or
lifethreatening bleeding, the source of which cannot be eliminated)
represent the most accepted clinical indication for LAA occlusion,
albeit by having to extrapolate the results of the PROTECT AF
study to that specific cohort.110 So far, no randomized data targeting
this specific group of patients are available. Hence, our statement is
based on expert consensus. This is the result of several observational
studies and registries (described above in the sections dedicated to
the Watchman and Amplatzer devices) suggesting that occlusion is
safe and effective despite the absence of even temporary (N)OACs. It
should be noted that dual antiplatelet therapy (DAT) is generally
indicated for 1 – 6 months, not infrequently followed by lifelong single
antiplatelet therapy. It needs to be mentioned that DAT generates a major
bleeding risk comparable to that of warfarin.111 However, DAT
exposure following LAA occluder implantation is only short time, thus
reducing the cumulative risk of major bleeding events. Even when
single-centre experience is reporting a favourable outcome after
termination of any antiplatelet therapy, the majority of patients are
exposed to a long-lasting single antiplatelet therapy after occluder
implantation, again having the disadvantage of inducing a major and
intracranial bleeding risks while, e.g. on ASA, similar to those with warfarin
when stratified by the HAS-BLED score.112 In patients who cannot
receive any antiplatelet agent, transepicardial LAA ligation, e.g. with
the Lariat technique can be considered.
Patients with an increased bleeding risk under systemic
As depicted in the flow chart (Figure 5), we see the following three
patient groups as possible candidates for LAA occlusion as the
result of an individual risk benefit evaluation recognizing that the
primarily recommended strategy is the use of OAC:
) In general, patients with an increased HAS-BLED score should be
individually evaluated as to whether systemic OAC subjects
them to an unacceptable bleeding risk and whether this high
risk can be sufficiently reduced by the use of appropriately
dosed NOACs (discussed below) shown to be associated with
a lower bleeding risk than VKAs. Those in whom VKAs or
NOACs are still considered to pose an unacceptable bleeding
risk, but who remain at high stroke risk (CHA2DS2-VASc score
of .2), should be considered for LAA occlusion. More detailed
information to perform an individual risk evaluation were
discussed by Friberg et al.112 and Oleson et al.113
) Triple anticoagulant therapy causes a significant rise in bleeding
risk.114 – 116 Hence, in patients with the need for a prolonged
period of triple anticoagulant therapy as a result of severe
coronary artery disease treated with one or more stents and AF with a
high thrombo-embolic risk (CHA2DS2-VASc score of .2)
should be considered for percutaneous LAA occlusion.
) In clinical practice on a case-by-case basis, some patients with
high bleeding risk who are not well characterized by the
HAS-BLED score (e.g. patients with cancer or chronic
inflammatory bowel disease) but have a high risk of bleeding with OAC
may also be considered for LAA occlusion provided even
NOACs be deemed to be associated with an unacceptable
) In patients with end-stage renal failure, high stroke risk, and high
bleeding risk, the implantation of an LAA occluder is a debatable
alternative. In those patients, all NOACs are contraindicated
at a creatinine clearance ,15 mL/min. The benefit of VKA or
NOACs in renal failure with creatinine clearance ,15 – 30 mL/
min is questioned due to elevated bleeding risks. The use of
VKAs in patients with renal failure is controversially discussed
due to an increase in tissue calcification and enhanced
Importantly, for all four above groups, we recommend the
performance of an individualized risk/benefit analysis for NOACs and to
consider LAA occlusion as an alternative to anticoagulation. For
this analysis, it should to be taken into account that at least 1 – 6
months of either OAC or DAT are warranted after LAA occlusion.
Thereafter, patients are typically treated with at least one
antiplatelet agent. Therefore, the bleeding risk for ASA (as documented, for
example, in the Apixaban versus Acetylsalicylic Acid to Prevent
Strokes (AVERROES) trial)117 has to be included into the LAA
occlusion strategy discussion. However, the notion that, beyond the
post-procedural period, indefinite single antiplatelet therapy
prevents thrombo-embolic events related to the device itself is not
evidence-based, but merely the result of the assumption that
many patients have concomitant risk factors for atherosclerosis
and stroke irrespective of AF and any foreign body continues to
pose some risk of thrombus formation even beyond expected
incorporation into the surrounding tissue.
As a complement to anticoagulation
The combination of LAA occlusion and OAC is discussed and
occasionally performed in patients with embolic events despite adequate
OAC provided no other plausible cause (e.g. carotid disease,
severe mobile aortic arch atheromata) can be identified. The ESC
guidelines107 recommended approach is increasing the international
normalized ratio (INR) target 2.5 – 3.5 in this situation, when it occurs
while taking warfarin. Another discussed option is the switch from
VKA to one of the NOACs.118 – 121 Adding an antiplatelet agent to
OAC is performed in the clinical arena, especially when embolism
occurred at elevated INRs or while taking NOACs; however,
there are no data available demonstrating a positive effect on
embolic events which would support this approach. Left atrial
appendage occlusion could be debated as an alternative treatment
in those patients, especially when AF-related embolism occurs
while taking VKA with documented elevated INRs or switching to
NOACs is not possible due to a NOACs contraindication like
severe renal impairment.
As adjunct to ablation of atrial fibrillation
So far, few data on the combination of LAA occlusion and AF ablation
in a single session have been published.122 Additional personal
communications about limited single-centre experiences still do not allow
a general recommendation. However, as long as no randomized data
to support a significant reduction in thrombo-embolic events after
successful ablation are available, in very select cases, this combination
seems to be a valuable and practical approach: patients with a
significant risk of thrombo-embolic events (CHA2DS2-VASc score of .2)
undergoing an ablation procedure to treat symptomatic AF, who, in
addition, have a strict or relative contraindication to (N)OACs, might
be acceptable candidates. Under these circumstances, the ablation
itself is associated with the risk of transseptal puncture, perhaps
general anaesthesia, and anticoagulation and the incremental
procedural risk of LAA occlusion is substantially lower than if performed as a
standalone procedure even though the overall procedure is
becoming longer. However, once again, patient preference after thorough
discussion pointing out the absence of data supporting this strategy
must be integral part of the decision-making process.
In the era of new anticoagulants
There are no scientific data available directly comparing LAA
occlusion to NOACs. Though intracranial haemorrhage may be lower with
dabigatran 150 mg orally twice daily121 or rivaroxaban 20 mg orally
once daily,120 the overall incidence of major bleeding remains
similar to VKAs. Therefore, in the absence of further data, it is the
opinion of this consensus panel that, at the aforementioned doses,
contraindications to VKAs apply equally to dabigatran and
rivaroxaban, and other NOACs (Figure 5). As an exception, low-dose (110 mg
orally twice daily) dabigatran or apixaban119 have been associated
with lower rates of overall major bleeding as well as intracranial
haemorrhage compared with VKAs while maintaining equivalent
efficacy in stroke prevention.121 Therefore, consideration of low-dose
dabigatran or apixaban may be reasonable in patients at increased
bleeding risk provided inclusion criteria for pivotal trials examining
dabigatran or apixaban are met. A decision should be made on
case-by-case basis carefully evaluating potential bleeding sources
and risks and recognizing limitations (among others, the unclear
safety in patients with renal dysfunction and the absence of a
difference in gastrointestinal haemorrhage). In addition, currently the
safety of triple therapy using any of the NOACs compared with
warfarin remains to be determined. In fact, the bleeding risk reduction
of dabigatran in combination with clopidogrel and ASA was only
minor compared with triple therapy including warfarin.121 Though
there are clear advantages both in the logistics of administration
and surveillance as well as the safety and efficacy with NOACs, a
head-to-head comparison to LAA occlusion has not yet been done
and final conclusions favouring NOACs over LAA occlusion or vice
versa cannot be made.
Although LAA occlusion is originally meant as a substitute for chronic
OAC among AF patients, the selective application of anticoagulants
including antithrombotics and antiplatelets for various
procedureand device-related indications remains essential. Since a majority of
patients subjected to LAA occlusion are at high risk for bleeding,
the anticoagulation regimen should be tailored individually.
Mobile thrombi visualized by screening TOE are considered a
contraindication to catheter-based LAA occlusion. In such cases, ≥4 weeks
of (N)OACs may allow thrombus resolution to be documented on
repeated TOE before catheter-based LAA occlusion is attempted.
Femoral venous puncture by itself does not necessitate anticoagulant
therapy withdrawal. Nevertheless, most operators aim for a normal
INR at the time of the procedure and use intravenous antithrombotic
agents (mostly unfractionated heparin) during the procedure
(Table 6). The antithrombotic protocol of the PROTECT AF
study,45 mandated an INR , 2.0 at the time of procedure.
Acetylsalicylic acid 81 – 325 mg was begun at least 1 day before the procedure
and weight-adjusted heparin (70 – 100 IU/kg) was administered after
Load 300 – 600 mg prior
to procedure if not
continue 1 – 6
Load 300 – 600 mg prior
to procedure if not
continue 1 – 6
Clopidogrel often given for shorter time in extremely high-risk situations.
ASA if better
Clopidogrel often given
for shorter time in
ASA if better
Watchman/ High bleeding risk
Prior to or
Prior to or
ACT, activated clothing time; INR, international normalized ratio.
aLess than 5 mm leak.
transseptal puncture to maintain an ACT . 200 s for the duration of
the procedure.123 However, some operators perform the procedure
while patients are on OAC with a therapeutic INR, an approach that
can neither be supported nor condemned by currently available data.
Intravenous antithrombotics are generally administered at the latest
immediately after traversing the interatrial septum. A
weightadjusted bolus of unfractionated heparin (70 – 100 IU/kg) is most
commonly used, which should maintain an ACT ≥ 250 s. Left atrial
appendage occlusion may be performed as part of a combined
procedure for which another antithrombotic is being used, e.g.
bivalirudin. This requires no further anticoagulation.
Post-procedural anticoagulation with warfarin is recommended
for the Watchman device to avoid thrombus formation on the
device until completion of endocardialization, provided there are
no contraindications to anticoagulation. The anticoagulation
protocol of the PROTECT AF trial (OAC for 6 weeks, DAT for 6
months, and ASA for life) was adopted in the instructions for use
of the Watchman device. All patients enroled in the PROTECT AF
study had to be eligible for warfarin to enable randomization to
either a Watchman device occlusion or chronic warfarin therapy.
However, this circumstance is rather atypical for patients considered
for LAA occlusion in clinical practice. A considerable proportion of
them have had a bleeding complication or have a contraindication
to chronic anticoagulation (VKA or NOACs), under which
circumstances operators refrain from implementing a drug regimen
including an oral antithrombotic and directly prescribe DAT for at least
1 month or until a 6-month TOE follow-up, modifying the
anticoagulant therapy upon its result. A satisfactory result on TOE (complete
LAA occlusion or small residual shunt ,5 mm jet width in the
absence of device surface thrombi) justifies withdrawing at least
one antiplatelet agent, unless otherwise indicated. Usually the
other antiplatelet agent is continued indefinitely, as most patients
are elderly with evidence of atherosclerotic disease, although the
bleeding risk of ASA even as a standalone therapy must be
considered. This treatment rationale of DAT was mainly derived from
previous experience with the PLAATO device as well as ASD and PFO
device occlusions and was recently confirmed by the results of the
ASA-Plavix (ASAP) registry.95 In patients who underwent Watchman
implantation in the ASAP registry and received clopidogrel for
6 months and ASA indefinitely without OAC, the ischaemic stroke
rate was only 1.7% compared with 2.2% in the PROTECT AF
device group. By arbitrary practice, it is usual to load ASA or
clopidogrel na¨ıve patients accordingly (Table 6).
The ACP banks on the good record regarding low
thrombogenicity of the Amplatzer device family,124 and indicates in its instructions
for use DAT only without an oral anticoagulant. The safety and
feasibility of this drug regimen was shown in initial registry data for the
In a recently published study using the Lariat device among 89
patients, those with a contraindication to warfarin remained off
warfarin, while patients with a CHADS2 score of 2 who could tolerate
warfarin but had been non-compliant or had labile INR continued
warfarin. Warfarin use in patients with a CHADS2 score of 1 was
left to the discretion of the referring physician. For patients not on
warfarin, ASA was recommended. At 1-year follow-up, 55% of
patients were still on warfarin.106
Transoesophageal echocardiography follow-up performed after 4 – 6
months is highly recommended to verify outcome and define the
further anticoagulant regimen. In addition, in patients with clear
contraindications to warfarin it is often not possible to use warfarin even
for short periods of time and they undergo device implantation
followed by antiplatelet therapy only. Therefore, consideration
should be given to TOE follow-up at 1 month as, theoretically, this
is the crucial period for device-associated thrombus formation.
In case of thrombus on the device
Device-associated thrombus was observed in 20 of 478 successfully
implanted with a Watchman device (4.2%) in the PROTECT AF
trial.88 Of these patients, however, only three had ischaemic strokes
(thrombus-associated annualized stroke rate of 0.3/100 patient-years).
The remainder were asymptomatic. The device-associated thrombus
was mobile in 4 patients (3 pedunculated and 1 laminar) and
nonmobile in 10. In the remaining patients, the thrombus was not
further characterized. Of the three patients with a device thrombus
and a stroke, one occurred in a patient with a mobile, pedunculated
thrombus. For the ACP, individual cases of device-related thrombi
seen on routine TOE have been described.125 For both devices,
OAC for a period of weeks to months effect thrombus resolution in
most cases. Therefore, anticoagulant therapy is recommended in all
patients with device-associated thrombus regardless of symptoms
until thrombus resolution is confirmed by follow-up TOE.
In case of incomplete occlusion of the left atrial appendage
Incomplete LAA occlusion could create a thrombus containing
pocket allowing emboli to enter the systemic circulation potentially
causing strokes. Based on PROTECT AF data, small residual shunts
with a jet diameter ,5 mm are usually deemed irrelevant and may
close spontaneously with time. They do not warrant further drug
or device interventions. At 45 days, 14% of the PROTECT AF
cohort had residual flow .5 mm, in whom warfarin was continued
per protocol. At the 6-month TOE, the prevalence of large
(≥5 mm) residual leaks decreased to 8%. Persistent large shunts
are becoming rare with increasing operator experience and can
usually be avoided by proper initial sizing and implantation
techniques. Generally, when all patients with residual shunts are included
into one subgroup, the stroke risk is no different compared with
patients in whom the LAA is completely occluded regardless of
whether or not anticoagulant therapy is continued.83 However,
whether persistent large (≥5 mm) shunts deserve long-term
(N)OACs or second occlusion attempts using dedicated or
nondedicated occlusion devices remains at the operator’s discretion.
Summary, conclusions, and recommendations
Left atrial appendage occlusion as a means to prevent
thromboembolism in AF is based on the observation that the majority of
thrombi in non-valvular AF form in this cul de sac structure. Left atrial
appendage occlusion using interventional techniques has been
demonstrated to be equivalent to oral VKAs in reducing
Current commercially available devices include the ACP and the
Watchman and WaveCrest devices as well as the Lariat technique
for percutaneous endocardial/epicardial ligation. All have shown
efficacy and relative safety in achieving the goal of preventing
thromboembolism in AF patients who do not wish or cannot receive OAC.
The Watchman device has demonstrated non-inferiority and later
superiority when compared with warfarin in a controlled randomized
trial (PROTECT AF).45 The results seem to improve with increasing
operator experience. Patients who are at high risk or cannot receive
OAC have been treated successfully with LAA occlusion followed by
antiplatelet therapy only in observational studies and registries with
results comparable to the PROTECT AF trial.
Standards of performance and documentation
The procedure requires training and knowledge acquired by a
structured mentoring process. Moreover, institutions entering this field
should have appropriate echocardiography and anaesthesia
support and, preferably also surgical backup. Operators must have
experience with pericardiocentesis and autotransfusion to manage
pericardial haemorrhage and tamponade. Registries are required to
document implantation results, complications, and follow-up.
Two-dimensional TOE is currently the standard imaging technique in
selection of patients and follow-up assessments and, as a
complement to fluoroscopy, in device sizing and selection and procedural
guidance. The purpose of follow-up TOE (recommended at, e.g. 45
days, or 3 – 6 months) is to detect leaks and thrombi on the device.
Whether other modalities such as 3D TOE or ICE, and CT or MRI
will partially replace 2D TOE in the future for these purposes
remains to be determined.
The procedure is performed under full anticoagulation. Standard
medical treatment following Watchman device implantation includes
a VKA for at least 6 weeks followed by DAT for 6 months and a single
antiplatelet drug thereafter. However, recent data suggest that
Watchman implantation followed by antiplatelet therapy only (in
the absence of OAC) can also be performed safely in patients at
high bleeding risk. Most ACPs were implanted with DAT for
several weeks to months and a single antiplatelet drug or nothing
thereafter. Prolonged (N)OACs are indicated in case of a
device-associated thrombus or large (≥5 mm) leak.
Despite the results of the PROTECT AF trial, OAC (with VKA or
NOACs) remains the standard therapy when there is no special
risk or contraindication to (N)OACs. However, the option of LAA
occlusion should be discussed with the patient, including risks of
the procedure and limited proof of superiority. Patients who refuse
(N)OACs after thorough discussion of current data including
limitations may be considered for LAA occlusion. The main indication for
LAA occlusion today is a relative or absolute contraindication to
(N)OACs in patients with AF and a CHADS2 score of ≥1 or
CHA2DS2-VASc score ≥2. It is important to realize that this
recommendation is based on observational studies and registries only. With
increasing thrombo-embolic risk, the use of LAA occlusion
becomes more attractive. To be a candidate for LAA occlusion,
patients should be able to receive at least several weeks of DAT
followed in most cases by lifelong single antiplatelet drug therapy. If
antiplatelet therapy is not an option, percutaneous endocardial/
epicardial or minimally invasive surgical epicardial LAA occlusion
may be alternatives. The use of LAA occlusion as adjunct to AF
ablation and as a supplement OAC seems to be reasonable but remains to
Need for future research
Several issues regarding LAA occlusion remain to be studied. The
ACP and the WaveCrest device have yet to be studied in a
randomized controlled trial. No device has been adequately studied in a
randomized controlled fashion in a population at high risk that
cannot receive (N)OACs. None of the devices has been directly
compared with N NOACs. These knowledge gaps should stimulate
further trials to clarify best practice under these circumstances. An
LAA occlusion registry should be established by the European
Heart Rhythm Association (EHRA) and European Association of
Percutaneous Coronary Interventions (EAPCI) of the ESC to document
real-world results of implantation and follow-up and to complete
data on indications that are not well represented in current trials.
The authors wish to thank the following colleagues for their help in
reviewing scientific material and assistance in writing: Stefan Bertog
MD, CardioVascular Center Frankfurt, Frankfurt, Germany; Gaetano
Fassini MD, Cardiac Arrhythmia Research Center, Centro
Cardiologico Monzino, IRCCS, Milan, Italy; Avishay Grupper MD, Sheba
Medical Center, Tel Hashomer, Israel.
Conflict of interest: A.A.K.: Direct Personal payment: St Jude
Medical; Payment to your Institution: St Jude Medical. B.M.: Direct
Personal payment: Astra Zeneca, Bayer, Biosensors, BMS,
Boehringer-Ingelheim, Lilly St Jude Medical; Payment to your Institution:
Abbott Vascular, Biotronik, Boston Scientific, Cordis, Edwards
Lifesciences, Guerbet, Medtronic, St Jude Medical. C.T.: Direct Personal
payment: Biosense Webster, Biotronik, Boston Scientific, Medtronic,
St Jude Medical. H.S.: Direct Personal payment: none; Payment to
your Institution: Access-Closure, AGA medical, Angiomed, Ardian,
Arstasis, Atrium, Avinger, Bard, Boston Scientific, Bridgepoint,
Cardiokinetix, Cardiomems, Coherex, Contego, CSI, CVRx, EndoCross,
Endotex, Epitek, Ev3, Evalve, FlowCardia, Gardia Medical, GDS,
Gore, Guidant, HLT, InSeal, Kensey Nash, Kyoto Medical, Lifetech,
Lumen Biomedical, Lutonix, Maya Medical, Medinol, Medtronic,
NDC, OAS, Occlutech, Osprey, Ovalis, Pathway Medical,
PendraCare, Percardia, PFM, Recor, Rox Medical, Sadra Medical,
SentreHeart, Sorin Group, Spectranetics, SquareOne, Trireme,
Trivascular, Veryan, Viacor, Vessix. M.G.: Direct Personal payment:
Boston Scientific; Payment to your Institution: Biotronik, Medtronic.
T.L.: None. Y.B.: Direct Personal payment: Astra Zeneca,
1. Stewart S , Hart C , Hole D , McMurray J . Population prevalence, incidence, and predictors of atrial fibrillation in the Renfrew/Paisley study . Heart 2001 ; 86 : 516 - 21 .
2. Go A , Hylek E , Phillips K , Chang Y , Henault L , Selby J et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study . JAMA 2001 ; 285 : 2370 - 5 .
3. Kirchhof P , Auricchio A , Bax J , Crijns H , Camm J , Diener H et al. Outcome parameters for trials in atrial fibrillation: executive summary . Europace 2007 ; 9 : 1006 - 23 .
4. Miyasaka Y , Barnes M , Gersh B , Cha S , Bailey K , Abhayaratna W et al. Secular trends in incidence of atrial fibrillation in Olmsted County , Minnesota, 1980 to 2000, and implications on the projections for future prevalence . Circulation 2006 ; 114 : 119 - 25 .
5. Heeringa J , van der Kuip D , Hofman A , Kors J , van Herpen G , Stricker B et al. Prevalence, incidence and lifetime risk of atrial fibrillation: the Rotterdam study . Eur Heart J 2006 ; 27 : 949 - 53 .
6. Naccarelli G , Varker H , Lin J , Schulman K. Increasing prevalence of atrial fibrillation and flutter in the United States . Am J Cardiol 2009 ; 104 : 1534 - 9 .
7. Lloyd-Jones D , Wang T , Leip E , Larson M , Levy D , Vasan R et al. Lifetime risk for development of atrial fibrillation: the Framingham Heart Study . Circulation 2004 ; 110 : 1042 - 6 .
8. Stefansdottir H , Aspelund T , Gudnason V , Arnar DO . Trends in the incidence and prevalence of atrial fibrillation in Iceland and future projections . Europace 2011 ; 13 : 1110 - 7 .
9. Atrial Fibrillation Investigators: Atrial Fibrillation , Aspirin, Anticoagulation Study; Boston Area Anticoagulation Trial for Atrial Fibrillation Study; Canadian Atrial Fibrillation Anticoagulation Study; Stroke Prevention in Atrial Fibrillation Study; Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Study . Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials . Arch Intern Med 1994 ; 154 : 1449 - 57 .
10. Lin HJ , Wolf PA , Kelly-Hayes M , Beiser AS , Kase CS , Benjamin EJ et al. Stroke severity in atrial fibrillation . The Framingham Study. Stroke 1996 ; 27 : 1760 - 4 .
11. Gage BF , Waterman AD , Shannon W , Boechler M , Rich MW , Radford MJ . Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation . JAMA 2001 ; 285 : 2864 - 70 .
12. Lip G , Nieuwlaat R , Pisters R , Lane D , Crijns H. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the Euro Heart Survey on atrial fibrillation . Chest 2010 ; 137 : 263 - 72 .
13. Hart RG , Benavente O , McBride R , Pearce LA . Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis . Ann Intern Med 1999 ; 131 : 492 - 501 .
14. Waldo AL , Becker RC , Tapson VF , Colgan KJ . Hospitalized patients with atrial fibrillation and a high risk of stroke are not being provided with adequate anticoagulation . J Am Coll Cardiol 2005 ; 46 : 1729 - 36 .
15. Bungard TJ , Ghali WA , Teo KK , McAlister FA , Tsuyuki RT . Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med 2000 ; 160 : 41 - 6 .
16. Levine MN , Raskob G , Landefeld S , Kearon C . Hemorrhagic complications of anticoagulant treatment . Chest 2001 ; 119 : 108S - 21S .
17. Kirchhof P , Ammentorp B , Darius H , De Caterina R , Le Heuzey JY , Schilling RJ et al. Management of atrial fibrillation in seven European countries after the publication of the 2010 ESC Guidelines on atrial fibrillation: primary results of the PREvention oF thromboemolic events-European Registry in Atrial Fibrillation (PREFER in AF) . Europace 2014 ; 16 : 6 - 14 .
18. Lip GY , Laroche C , Dan GA , Santini M , Kalarus Z , Rasmussen LH et al. A prospective survey in European Society of Cardiology member countries of atrial fibrillation management: baseline results of EURObservational Research Programme Atrial Fibrillation (EORP-AF) Pilot General Registry . Europace 2014 ; 16 : 308 - 19 .
19. Watson T , Shantsila E , Lip GY . Mechanisms of thrombogenesis in atrial fibrillation: Virchow's triad revisited . Lancet 2009 ; 373 : 155 - 66 .
20. Boldt A , Wetzel U , Lauschke J , Weigl J , Gummert J , Hindricks G et al. Fibrosis in left atrial tissue of patients with atrial fibrillation with and without underlying mitral valve disease . Heart 2004 ; 90 : 400 - 5 .
21. Frustaci A , Chimenti C , Bellocci F , Morgante E , Russo M , Maseri A . Histological substrate of atrial biopsies in patients with lone atrial fibrillation . Circulation 1997 ; 96 : 1180 - 4 .
22. Goldsmith I , Kumar P , Carter P , Blann A , Patel R , Lip G . Atrial endocardial changes in mitral valve disease: a scanning electron microscopy study . Am Heart J 2000 ; 140 : 777 - 84 .
23. Kahn SR , Solymoss S , Flegel K. Nonvalvular atrial fibrillation: evidence for a prothrombotic state . CMAJ 1997 ; 157 : 673 - 81 .
24. Inoue H , Nozawa T , Okumura K , Jong-Dae L , Shimizu A , Yano K. Prothrombotic activity is increased in patients with nonvalvular atrial fibrillation and risk factors for embolism . Chest 2004 ; 126 : 687 - 92 .
25. Handke M , Harloff A , Hetzel A , Olschewski M , Bode C , Geibel A. Left atrial appendage flow velocity as a quantitative surrogate parameter for thromboembolic risk: determinants and relationship to spontaneous echocontrast and thrombus formation-a transesophageal echocardiographic study in 500 patients with cerebral ischemia . J Am Soc Echocardiogr 2005 ; 18 : 1366 - 72 .
26. Pollick C , Taylor D . Assessment of left atrial appendage function by transesophageal echocardiography. Implications for the development of thrombus . Circulation 1991 ; 84 : 223 - 31 .
27. Mugge A , Kuhn H , Nikutta P , Grote J , Lopez J , Daniel W. Assessment of left atrial appendage function by biplane transesophageal echocardiography in patients with nonrheumatic atrial fibrillation: identification of a subgroup of patients at increased embolic risk . J Am Coll Cardiol 1994 ; 23 : 599 - 607 .
28. Sahin T , Ural D , Kilic T , Bildirici U , Kozdag G , Agacdiken A et al. Evaluation of left atrial appendage functions according to different etiologies of atrial fibrillation with a tissue Doppler imaging technique by using transesophageal echocardiography . Echocardiography 2009 ; 26 : 171 - 81 .
29. Blackshear J , Odell J . Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation . Ann Thorac Surg 1996 ; 61 : 755 - 9 .
30. Zabalgoitia M , Halperin J , Pearce L , Blackshear J , Asinger R , Hart R . Transesophageal echocardiographic correlates of clinical risk of thromboembolism in nonvalvular atrial fibrillation . Stroke Prevention in Atrial Fibrillation III Investigators. J Am Coll Cardiol 1998 ; 31 : 1622 - 6 .
31. Takada T , Yasaka M , Nagatsuka K , Minematsu K , Yamaguchi T. Blood flow in the left atrial appendage and embolic stroke in nonvalvular atrial fibrillation . Eur Neurol 2001 ; 46 : 148 - 52 .
32. Tamura H , Watanabe T , Nishiyama S , Sasaki S , Wanezaki M , Arimoto T et al. Prognostic value of low left atrial appendage wall velocity in patients with ischemic stroke and atrial fibrillation . J Am Soc Echocardiogr 2012 ; 25 : 576 - 83 .
33. Di Biase L , Santangeli P , Anselmino M , Mohanty P , Salvetti I , Gili S et al. Does the left atrial appendage morphology correlate with the risk of stroke in patients with atrial fibrillation? Results from a multicenter study . J Am Coll Cardiol 2012 ; 60 : 531 - 8 .
34. Madden J . Resection of the left auricular appendix; a prophylaxis for recurrent arterial emboli . J Am Med Assoc 1949 ; 140 : 769 - 72 .
35. Landymore R , Kinley C . Staple closure of the left atrial appendage . Can J Surg 1984 ; 27 : 144 - 5 .
36. Johnson WD , Ganjoo AK , Stone CD , Srivyas RC , Howard M. The left atrial appendage: our most lethal human attachment! Surgical implications . Eur J Cardiothorac Surg 2000 ; 17 : 718 - 22 .
37. Katz E , Tsiamtsiouris T , Applebaum R , Schwartzbard A , Tunick P , Kronzon I. Surgical left atrial appendage ligation is frequently incomplete: a transesophageal echocardiograhic study . J Am Coll Cardiol 2000 ; 36 : 468 - 71 .
38. Sievert H , Lesh M , Trepels T , Omran H , Bartorelli A , Della Bella P et al. Percutaneous left atrial appendage transcatheter occlusion to prevent stroke in high-risk patients with atrial fibrillation: early clinical experience . Circulation 2002 ; 105 : 1887 - 9 .
39. Blackshear J , Johnson W , Odell J , Baker V , Howard M , Pearce L et al. Thoracoscopic extracardiac obliteration of the left atrial appendage for stroke risk reduction in atrial fibrillation . J Am Coll Cardiol 2003 ; 42 : 1249 - 52 .
40. Garcia-Fernandez M , Perez-David E , Quiles J , Peralta J , Garcia-Rojas I , Bermejo J et al. Role of left atrial appendage obliteration in stroke reduction in patients with mitral valve prosthesis: a transesophageal echocardiographic study . J Am Coll Cardiol 2003 ; 42 : 1253 - 8 .
41. Meier B , Palacios I , Windecker S , Rotter M , Cao QL , Keane D et al. Transcatheter left atrial appendage occlusion with Amplatzer devices to obviate anticoagulation in patients with atrial fibrillation . Catheter Cardiovasc Interv 2003 ; 60 : 417 - 22 .
42. Healey J , Crystal E , Lamy A , Teoh K , Semelhago L , Hohnloser S et al. Left Atrial Appendage Occlusion Study (LAAOS): results of a randomized controlled pilot study of left atrial appendage occlusion during coronary bypass surgery in patients at risk for stroke . Am Heart J 2005 ; 150 : 288 - 93 .
43. Ostermayer S , Reisman M , Kramer P , Matthews R , Gray W , Block P et al. Percutaneous left atrial appendage transcatheter occlusion (PLAATO system) to prevent stroke in high-risk patients with non-rheumatic atrial fibrillation: results from the international multi-center feasibility trials . J Am Coll Cardiol 2005 ; 46 : 9 - 14 .
44. Sick P , Schuler G , Hauptmann K , Grube E , Yakubov S , Turi Z et al. Initial worldwide experience with the Watchman left atrial appendage system for stroke prevention in atrial fibrillation . J Am Coll Cardiol 2007 ; 49 : 1490 - 5 .
45. Holmes D , Reddy V , Turi Z , Doshi S , Sievert H , Buchbinder M et al. Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation: a randomised non-inferiority trial . Lancet 2009 ; 374 : 534 - 42 .
46. Lee R , Bartus K , Yakubov S . Catheter-based left atrial appendage (LAA) ligation for the prevention of embolic events arising from the LAA: initial experience in a canine model . Circ Cardiovasc Interv 2010 ; 3 : 224 - 9 .
47. Bartus K , Bednarek J , Myc J , Kapelak B , Sadowski J , Lelakowski J et al. Feasibility of closed-chest ligation of the left atrial appendage in humans . Heart Rhythm 2011 ; 8 : 188 - 93 .
48. Park J , Bethencourt A , Sievert H , Santoro G , Meier B , Walsh K et al. Left atrial appendage closure with Amplatzer Cardiac Plug in atrial fibrillation: initial European experience . Catheter Cardiovasc Interv 2011 ; 77 : 700 - 6 .
49. Toumanides S , Sideris E , Agricola T , Moulopoulos S. Transcatheter patch occlusion of the left atrial appendage using surgical adhesives in high-risk patients with atrial fibrillation . J Am Coll Cardiol 2011 ; 58 : 2236 - 40 .
50. Shetty R , Leitner J , Zhang M. Percutaneous catheter-based left atrial appendage ligation and management of periprocedural left atrial appendage perforation with the LARIAT suture delivery system . J Invasive Cardiol 2012 ; 24 : E289 - 93 .
51. Tsai L , Chen J , Lin L , Yang Y . Role of transesophageal echocardiography in detecting left atrial thrombus and spontaneous echo contrast in patients with mitral valve disease or non-rheumatic atrial fibrillation . J Formos Med Assoc 1990 ; 89 : 270 - 4 .
52. Manning W , Weintraub R , Waksmonski C , Haering J , Rooney P , Maslow A et al. Accuracy of transesophageal echocardiography for identifying left atrial thrombi. A prospective, intraoperative study . Ann Intern Med 1995 ; 123 : 817 - 22 .
53. Stoddard M , Dawkins P , Prince C , Ammash N. Left atrial appendage thrombus is not uncommon in patients with acute atrial fibrillation and a recent embolic event: a transesophageal echocardiographic study . J Am Coll Cardiol 1995 ; 25 : 452 - 9 .
54. Brown J , Sadler D. Left atrial thrombi in non-rheumatic atrial fibrillation: assessment of prevalence by transesophageal echocardiography . Int J Card Imaging 1993 ; 9 : 65 - 72 .
55. Collins L , Silverman D , Douglas P , Manning W. Cardioversion of nonrheumatic atrial fibrillation. Reduced thromboembolic complications with 4 weeks of precardioversion anticoagulation are related to atrial thrombus resolution . Circulation 1995 ; 92 : 160 - 3 .
56. Tsai L , Lin L , Teng J , Chen J . Prevalence and clinical significance of left atrial thrombus in nonrheumatic atrial fibrillation . Int J Cardiol 1997 ; 58 : 163 - 9 .
57. Stoddard M , Singh P , Dawn B , Longaker R . Left atrial thrombus predicts transient ischemic attack in patients with atrial fibrillation . Am Heart J 2003 ; 145 : 676 - 82 .
58. Bernhardt P , Schmidt H , Hammerstingl C , Hackenbroch M , Sommer T , Luderitz B et al. Fate of left atrial thrombi in patients with atrial fibrillation determined by transesophageal echocardiography and cerebral magnetic resonance imaging . Am J Cardiol 2004 ; 94 : 801 - 4 .
59. Tang R , Dong J , Zhang Z , Li Z , Liu X , Kang J et al. Comparison of contrast enhanced 64-slice computed tomography and transesophageal echocardiography in detection of left atrial thrombus in patients with atrial fibrillation . J Interv Card Electrophysiol 2008 ; 22 : 199 - 203 .
60. Yamashita E , Takamatsu H , Tada H , Toide H , Okaniwa H , Takemura N et al. Transesophageal echocardiography for thrombus screening prior to left atrial catheter ablation . Circ J 2010 ; 74 : 1081 - 6 .
61. McCready J , Nunn L , Lambiase P , Ahsan S , Segal O , Rowland E et al. Incidence of left atrial thrombus prior to atrial fibrillation ablation: is pre-procedural transoesophageal echocardiography mandatory ? Europace 2010 ; 12 : 927 - 32 .
62. Lewalter T , Ibrahim R , Albers B , Camm AJ . An update and current expert opinions on percutaneous left atrial appendage occlusion for stroke prevention in atrial fibrillation . Europace 2013 ; 15 : 652 - 6 .
63. Bonow R , Carabello B , Chatterjee K , de Leon AJ , Faxon D , Freed M et al. ACC/ AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing Committee to Revise the 1998 guidelines for the management of patients with valvular heart disease) developed in collaboration with the Society of Cardiovascular Anesthesiologists endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons . J Am Coll Cardiol 2006 ; 48 : e1 - 148 .
64. Kanderian AS , Gillinov AM , Pettersson GB , Blackstone E , Klein AL . Success of surgical left atrial appendage closure: assessment by transesophageal echocardiography . J Am Coll Cardiol 2008 ; 52 : 924 - 9 .
65. Friedman PA , Asirvatham SJ , Dalegrave C , Kinoshita M , Danielsen AJ , Johnson SB et al. Percutaneous epicardial left atrial appendage closure: preliminary results of an electrogram guided approach . J Cardiovasc Electrophysiol 2009 ; 20 : 908 - 15 .
66. Holmes DR , Kar S , Price M , Whisenant B , Sievert H , Doshi S et al. Prospective randomized evaluation of the Watchman left atrial appendage closure device in patients with atrial fibrillation versus long-term Warfarin therapy (the PREVAIL Trial) . JACC 2014 ; 64 : 1 - 12 .
67. Dawson A , Asopa S , Dunning J . Should patients undergoing cardiac surgery with atrial fibrillation have left atrial appendage exclusion? Interact Cardiovasc Thorac Surg 2010 ; 10 : 306 - 11 .
68. Lam Y , Yip G , Yu C , Chan W , Cheng B , Yan B et al. Left atrial appendage closure with Amplatzer cardiac plug for stroke prevention in atrial fibrillation: initial Asia-Pacific experience . Catheter Cardiovasc Interv 2012 ; 79 : 794 - 800 .
69. Freixa X , Chan JL , Tzikas A , Garceau P , Basmadjian A , Ibrahim R. The Amplatzer Cardiac Plug 2 for left atrial appendage occlusion: novel features and first-in-man experience . EuroIntervention 2013 ; 8 : 1094 - 8 .
70. Sosa E , Scanavacca M , d'Avila A , Pilleggi F. A new technique to perform epicardial mapping in the electrophysiology laboratory . J Cardiovasc Electrophysiol 1996 ; 7 : 531 - 6 .
71. Klein A , Grimm R , Murray R , Apperson-Hansen C , Asinger R , Black I et al. Use of transesophageal echocardiography to guide cardioversion in patients with atrial fibrillation . N Engl J Med 2001 ; 344 : 1411 - 20 .
72. Puwanant S , Varr B , Shrestha K , Hussain S , Tang W , Gabriel R et al. Role of the CHADS2 score in the evaluation of thromboembolic risk in patients with atrial fibrillation undergoing transesophageal echocardiography before pulmonary vein isolation . J Am Coll Cardiol 2009 ; 54 : 2032 - 9 .
73. Hur J , Kim Y , Nam J , Choe K , Choi E , Shim C et al. Thrombus in the left atrial appendage in stroke patients: detection with cardiac CT angiography-a preliminary report . Radiology 2008 ; 249 : 81 - 7 .
74. Hur J , Kim Y , Lee H , Nam J , Hong Y , Kim H et al. Cardioembolic stroke: dual-energy cardiac CT for differentiation of left atrial appendage thrombus and circulatory stasis . Radiology 2012 ; 263 : 688 - 95 .
75. Kim Y , Klein A , Halliburton S , Popovic Z , Kuzmiak S , Sola S et al. Left atrial appendage filling defects identified by multidetector computed tomography in patients undergoing radiofrequency pulmonary vein antral isolation: a comparison with transesophageal echocardiography . Am Heart J 2007 ; 154 : 1199 - 205 .
76. Shapiro M , Neilan T , Jassal D , Samy B , Nasir K , Hoffmann U et al. Multidetector computed tomography for the detection of left atrial appendage thrombus: a comparative study with transesophageal echocardiography . J Comput Assist Tomogr 2007 ; 31 : 905 - 9 .
77. Ohyama H , Hosomi N , Takahashi T , Mizushige K , Osaka K , Kohno M et al. Comparison of magnetic resonance imaging and transesophageal echocardiography in detection of thrombus in the left atrial appendage . Stroke 2003 ; 34 : 2436 - 9 .
78. Nietlispach F , Krause R , Khattab A , Gloekler S , Schmid M , Wenaweser P et al. Ad hoc percutaneous left atrial appendage closure . J Invasive Cardiol 2013 ; 25 : 683 - 6 .
79. Nietlispach F , Gloekler S , Krause R , Shakir S , Schmid M , Khattab AA et al. Amplatzer left atrial appendage occlusion: single center 10-year experience . Catheter Cardiovasc Interv 2013 ; 82 : 283 - 9 .
80. Blendea D , Heist E , Danik S , Barrett C , Ruskin J , Mansour M. Analysis of the left atrial appendage morphology by intracardiac echocardiography in patients with atrial fibrillation . J Interv Card Electrophysiol 2011 ; 31 : 191 - 6 .
81. Nucifora G , Faletra F , Regoli F , Pasotti E , Pedrazzini G , Moccetti T et al. Evaluation of the left atrial appendage with real-time 3-dimensional transesophageal echocardiography: implications for catheter-based left atrial appendage closure . Circ Cardiovasc Imaging 2011 ; 4 : 514 - 23 .
82. Guerios E , Schmid M , Gloekler S , Khattab A , Wenaweser P , Windecker S et al. Left atrial appendage closure with the Amplatzer cardiac plug in patients with atrial fibrillation . Arq Bras Cardiol 2012 ; 98 : 528 - 36 .
83. Viles-Gonzalez J , Kar S , Douglas P , Dukkipati S , Feldman T , Horton R et al. The clinical impact of incomplete left atrial appendage closure with the Watchman device in patients with atrial fibrillation: a PROTECT AF (Percutaneous Closure of the Left Atrial Appendage Versus Warfarin Therapy for Prevention of Stroke in Patients With Atrial Fibrillation) substudy . J Am Coll Cardiol 2012 ; 59 : 923 - 9 .
84. De Ponti R , Cappato R , Curnis A , Della Bella P , Padeletti L , Raviele A et al. Transseptal catheterization in the electrophysiology laboratory: data from a multicenter survey spanning 12 years . J Am Coll Cardiol 2006 ; 47 : 1037 - 42 .
85. Fagundes R , Mantica M , De Luca L , Forleo G , Pappalardo A , Avella A et al. Safety of single transseptal puncture for ablation of atrial fibrillation: retrospective study from a large cohort of patients . J Cardiovasc Electrophysiol 2007 ; 18 : 1277 - 81 .
86. Lacomis J , Goitein O , Deible C , Moran P , Mamone G , Madan S et al. Dynamic multidimensional imaging of the human left atrial appendage . Europace 2007 ; 9 : 1134 - 40 .
87. Lip GY , Dagres N , Proclemer A , Svendsen JH , Pison L , Blomstrom-Lundqvist C ; Scientific Initiative Committee, European Heart Rhythm Association . Left atrial appendage occlusion for stroke prevention in atrial fibrillation in Europe: results of the European Heart Rhythm Association survey . Europace 2013 ; 15 : 141 - 3 .
88. Reddy V , Holmes D , Doshi S , Neuzil P , Kar S. Safety of percutaneous left atrial appendage closure: results from the Watchman Left Atrial Appendage System for Embolic Protection in Patients with AF (PROTECT AF) clinical trial and the Continued Access Registry . Circulation 2011 ; 123 : 417 - 24 .
89. Gangireddy SR , Halperin JL , Fuster V , Reddy VY . Percutaneous left atrial appendage closure for stroke prevention in patients with atrial fibrillation: an assessment of net clinical benefit . Eur Heart J 2012 ; 33 : 2700 - 8 .
90. Hylek E , Evans-Molina C , Shea C , Henault L , Regan S. Major hemorrhage and tolerability of warfarin in the first year of therapy among elderly patients with atrial fibrillation . Circulation 2007 ; 115 : 2689 - 96 .
91. Block P , Burstein S , Casale P , Kramer P , Teirstein P , Williams D et al. Percutaneous left atrial appendage occlusion for patients in atrial fibrillation suboptimal for warfarin therapy: 5-year results of the PLAATO (Percutaneous Left Atrial Appendage Transcatheter Occlusion) Study . J Am Coll Cardiol Cardiovasc Interv 2009 ; 2 : 594 - 600 .
92. Bayard Y , Omran H , Neuzil P , Thuesen L , Pichler M , Rowland E et al. PLAATO (Percutaneous Left Atrial Appendage Transcatheter Occlusion) for prevention of cardioembolic stroke in non-anticoagulation eligible atrial fibrillation patients: results from the European PLAATO study . EuroIntervention 2010 ; 6 : 220 - 6 .
93. Park J , Leithauser B , Gerk U , Vrsansky M , Jung F. Percutaneous left atrial appendage transcatheter occlusion (PLAATO) for stroke prevention in atrial fibrillation: 2-year outcomes . J Invasive Cardiol 2009 ; 21 : 446 - 50 .
94. Sick P , Turi Z , Grube E , Hauptmann K , Moebius Winkler S , Schuler G et al. Stroke prevention in non-valvular atrial fibrillation: long-term results after 6 years of the WATCHMAN Left Atrial Appendage Occlusion Pilot Study Presented at the ESC Congress 2012 , Munich, Germany, 25 - 29 August , 2012 . 2012 .
95. Reddy V . Watchman LAA closure device reduces the risk of ischemic stroke in patients with AF entirely without anticoagulation. The ASA-Plavix (ASAP) Registry. Presentation of Transcatheter Therapeutics (TCT), Miami , Florida, USA. November 2012 .
96. Reddy V . Late Breaking Clinical Trials . Annual Heart Rhythm Society Scientific Session 2013 .
97. Holmes D , Reddy V , Turi Z , Doshi S , Sievert H , Buchbinder M et al. Quality of life assessment in the randomized PROTECT AF trial of patients at risk for stroke with non-valvular atrial fibrillation . J Am Coll Cardiol 2012 ; 59 : E240 .
98. Singh SM , Micieli A , Wijeysundera HC . Economic evaluation of percutaneous left atrial appendage occlusion, dabigatran, and warfarin for stroke prevention in patients with nonvalvular atrial fibrillation . Circulation 2013 ; 127 : 2414 - 23 .
99. Schmid M , Gloeckler S , Saguner A , Wahl A , Fischer U , Arnold M et al. Transcatheter left atrial appendage closure in patients with atrial fibrillation: comparison between non-dedicated Amplatzer devices . Cardiovasc Med 2013 ; 16 : 123 - 30 .
100. Nietlispach F , Gloekler S , Khattab A , Pilgrim T , Schmid M , Wenaweser P et al. Percutaneous left atrial appendage closure . Eur J Geriatr Med 2012 ; 3 : 308 - 11 .
101. Santoro . Oral presentation . Progress in Clinical Pacing , Rome 2010 .
102. Park J , Sievert H , Schillinger W , Lickfett L , Lopez-Minguez J , Heyder O et al. Acute safety results from ACP European multicenter post-market observational study . EuroIntervention 2012 ; 8 ( Suppl ): N163 .
103. Lo´ pez-Minguez J , Eldoayen-Gragera J , Gonza˜ les-Ferna˜ndez R , Ferna˜ ndez-Vegas C , Fuentes-Can˜amero M, Milla˜n-Nun˜ez V et al . Immediate and one-year results in 35 consecutive patients after closure of left atrial appendage with the Amplatzer cardiac plug . Rev Esp Cardiol 2013 ; 66 : 90 - 7 .
104. Gloekler S , Schmid M , Gue´rios E, Khattab A , Wenaweser P , Windecker S et al. Transcatheter left atrial appendage occlusion in atrial fibrillation: a 100 cases single center experience . EuroIntervention 2012 ; 8 ( Suppl ): N162 .
105. Urena M , Rodes-Cabau J , Freixa X , Saw J , Webb JG , Freeman M et al. Percutaneous left atrial appendage closure with the AMPLATZER cardiac plug device in patients with nonvalvular atrial fibrillation and contraindications to anticoagulation therapy . J Am Coll Cardiol 2013 ; 62 : 96 - 102 .
106. Bartus K , Han F , Bednarek J , Myc J , Kapelak B , Sadowski J et al. Percutaneous left atrial appendage suture ligation using the LARIAT device in patients with atrial fibrillation: initial clinical experience . J Am Coll Cardiol 2013 ; 62 : 108 - 18 .
107. Camm AJ , Lip GY , De Caterina R , Savelieva I , Atar D , Hohnloser SH et al. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation. Developed with the special contribution of the European Heart Rhythm Association . Europace 2012 ; 14 : 1385 - 413 .
108. Amorosi SL , Armstrong S , Da Deppo L , Garfield S , Stein K. The budget impact of left atrial appendage closure compared with adjusted-dose warfarin and dabigatran etexilate for stroke prevention in atrial fibrillation . Europace 2014 ; 16 : 1131 - 6 .
109. Clinical commissioning policy statement: Left atrial appendage (LAA) occlusion . Commissioning Board, National Health Service 2013 ;NHSCB/A09/PS/c.
110. Lewalter T , Kanagaratnam P , Schmidt B , Rosenqvist M , Nielsen-Kudsk JE , Ibrahim R et al. Ischaemic stroke prevention in patients with atrial fibrillation and high bleeding risk: opportunities and challenges for percutaneous left atrial appendage occlusion . Europace 2014 ; 16 : 626 - 30 .
111. Healey J , Hart R , Pogue J , Pfeffer M , Hohnloser S , De Caterina R et al. Risks and benefits of oral anticoagulation compared with clopidogrel plus aspirin in patients with atrial fibrillation according to stroke risk: the atrial fibrillation clopidogrel trial with irbesartan for prevention of vascular events (ACTIVE-W) . Stroke 2008 ; 39 : 1482 - 6 .
112. Friberg L , Rosenqvist M , Lip G . Evaluation of risk stratification schemes for ischaemic stroke and bleeding in 182 678 patients with atrial fibrillation: the Swedish Atrial Fibrillation cohort study . Eur Heart J 2012 ; 33 : 1500 - 10 .
113. Olesen JB , Lip GY , Lindhardsen J , Lane DA , Ahlehoff O , Hansen ML et al. Risks of thromboembolism and bleeding with thromboprophylaxis in patients with atrial fibrillation: a net clinical benefit analysis using a 'real world' nationwide cohort study . Thromb Haemost 2011 ; 106 : 739 - 49 .
114. Lip G , Piotrponikowski P , Andreotti F , Anker S , Filippatos G , Homma S et al. Thromboembolism and antithrombotic therapy for heart failure in sinus rhythm: an executive summary of a joint consensus document from the ESC Heart Failure Association and the ESC Working Group on Thrombosis. Thromb Haemost 2012 ; 108 : 1009 - 22 .
115. Faxon D , Eikelboom J , Berger P , Holmes D , Bhatt D , Moliterno D et al. Consensus document: antithrombotic therapy in patients with atrial fibrillation undergoing coronary stenting. A North-American perspective . Thromb Haemost 2011 ; 106 : 572 - 84 .
116. Huber K , Airaksinen K , Cuisset T , Marin F , Rubboli A , Lip G. Antithrombotic therapy in patients with atrial fibrillation undergoing coronary stenting: similarities and dissimilarities between North America and Europe . Thromb Haemost 2011 ; 106 : 569 - 71 .
117. Connolly S , Eikelboom J , Joyner C , Diener H , Hart R , Golitsyn S et al. Apixaban in patients with atrial fibrillation . N Engl J Med 2011 ; 364 : 806 - 17 .
118. Connolly S , Ezekowitz M , Yusuf S , Eikelboom J , Oldgren J , Parekh A et al. Dabigatran versus warfarin in patients with atrial fibrillation . N Engl J Med 2009 ; 361 : 1139 - 51 .
119. Granger CB , Alexander JH , McMurray JJ , Lopes RD , Hylek EM , Hanna M et al. Apixaban versus warfarin in patients with atrial fibrillation . N Engl J Med 2011 ; 365 : 981 - 92 .
120. Patel M , Mahaffey K , Garg J , Pan G , Singer D , Hacke W et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation . N Engl J Med 2011 ; 365 : 883 - 91 .
121. Eikelboom J , Wallentin L , Connolly S , Ezekowitz M , Healey J , Oldgren J et al. Risk of bleeding with 2 doses of dabigatran compared with warfarin in older and younger patients with atrial fibrillation: an analysis of the randomized evaluation of longterm anticoagulant therapy (RE-LY) trial . Circulation 2011 ; 123 : 2363 - 72 .
122. Swaans M , Post M , Rensing B , Boersma L . Ablation for atrial fibrillation in combination with left atrial appendage closure: first results of a feasibility study . J Am Heart Assoc 2012 ; 1 : e002212 .
123. Fountain R , Holmes D , Chandrasekaran K , Packer D , Asirvatham S , Van Tassel R et al. The PROTECT AF (WATCHMAN Left Atrial Appendage System for Embolic PROTECTion in Patients with Atrial Fibrillation) trial . Am Heart J 2006 ; 151 : 956 - 61 .
124. Krumsdorf U , Ostermayer S , Billinger K , Trepels T , Zadan E , Horvath K et al. Incidence and clinical course of thrombus formation on atrial septal defect and patient foramen ovale closure devices in 1,000 consecutive patients . J Am Coll Cardiol 2004 ; 43 : 302 - 9 .
125. Cruz-Gonzalez I , Martin Moreiras J , Garcia E. Thrombus formation after left atrial appendage exclusion using an Amplatzer cardiac plug device . Catheter Cardiovasc Interv 2011 ; 78 : 970 - 3 .