Current Cardiology Reports (2018) 20:42

https://doi.org/10.1007/s11886-018-0981-z

STRUCTURAL HEART DISEASE (RJ SIEGEL AND NC WUNDERLICH, SECTION EDITORS)

Current State of Left Atrial Appendage Closure

Ricardo Kosturakis 1 & Matthew J. Price 1

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
Purpose of Review Atrial fibrillation (AF), the most common sustained arrhythmia, is a major cause of stroke and systemic
embolism, and is increasing in prevalence. Device closure of the left atrial appendage (LAA) represents a non-pharmacologic
approach to stroke prevention in AF patients. This review presents the rationale for LAA closure (LAAC), describes current
transcatheter approaches to LAAC, and summarizes the current evidence for LAAC for stroke prevention, highlighting the main
randomized trials and the most recent data available.
Recent Findings Meta-analysis of randomized clinical trials demonstrates similar rates of all-cause stroke with transcatheter
LAAC compared with vitamin K antagonist therapy and significantly less bleeding with LAAC after cessation of mandated post-
procedure pharmacology. Recent prospective observational studies, including those evaluating outcomes after commercial
approval in the USA, show significantly improved procedure safety compared with earlier experiences.
Summary LAAC appears to be an attractive alternative strategy for stroke prevention in AF patients, particularly in those who
can take short-term oral anticoagulation (OAC), but are not optimal candidates for long-term OAC. Recent data suggests the
procedure can be safely performed in patients with contraindications to OAC. Further, robust studies are needed to evaluate safety
and efficacy in OAC-contraindicated patients, to compare outcomes with non-vitamin K antagonist OACs, and to explore the
relative safety and efficacy of different LAAC devices.

Keywords Atrial fibrillation . Stroke prevention . Left atrial appendage . Watchman

Introduction medical expenditures than strokes not related to AF [4].

With respect to stroke risk, the LAA appears to be of
Atrial fibrillation (AF) is a common arrhythmia, and is esti- great importance in AF patients: the LAA has been shown
mated to affect 33.5 million individuals worldwide and 5 mil- to be the source of more than 90% of thrombi among
lion in the USA alone [1]. Advancing age is a significant risk cases of stroke in which thrombus could be identified
factor for AF, and due to the aging population and improve- [5], likely because fibrillation of the left atrium results
ments in life-expectancy, the prevalence of AF in the USA is in failure of function, dilation of the atrial wall, and stasis
expected to increase to greater than 15 million by 2050. The within the LAA which results in thrombus formation. For
introduction of sensitive diagnostic tools such as mobile car- this reason, this vestigial organ has been labeled “our
diac telemetry devices, pacemakers and implantable monitors most lethal attachment” [6].
is also contributing to increased recognition of this entity [2].
AF is an important cause of stroke, as approximately 20% of
all strokes are attributed to AF [3]. Stroke is one of the most
common causes of disability in the elderly, and AF-related Rationale for LAAC
strokes tend to be more incapacitating and result in larger
Stroke prevention is an essential part of AF management.
This article is part of the Topical Collection on Structural Heart Disease The traditional strategy for stroke prevention in AF patients
is oral anticoagulation (OAC) with vitamin K antagonists
* Matthew J. Price (VKA) or non-vitamin K antagonist oral anticoagulants
[email protected] (NOACs) [7]. Patients at high thromboembolic risk derive
1 a clear benefit from OAC treatment in terms of stroke pre-
Division of Cardiovascular Diseases, Scripps Clinic, 9898 Genesee
Avenue, AMP-200, La Jolla, CA 92037, USA vention and mortality [8, 9].
42 Page 2 of 9 Curr Cardiol Rep (2018) 20:42

There are several barriers that limit the use of OAC, and as development of less percutaneous LAAC therapy in AF pa-
a result, only about half of patients at high thromboembolic tients for stroke prevention.
risk are receive OAC therapy [10] (Fig. 1). VKA has multiple
drug-drug and drug-food interactions and a narrow therapeutic Watchman Left Atrial Appendage Occluder
window. Studies have shown that patients taking VKA remain
in the therapeutic window only 60% of the time, and only 60% Randomized Trials
will continue therapy after four years [11, 12]. Long-term
treatment with VKA is also associated with an increased risk The Watchman occluder (Boston Scientific, Natick, MA) is
of bleeding, which depends on a variety of established risk the only device to date whose safety and efficacy has been
factors [13]. Rates of major bleeding, including intracranial evaluated in randomized trials (Table 1). It is the most widely
hemorrhage are increased by as much as 0.5% per year [14]. studied device and the only one approved for commercial use
This risk is increased 4–5-fold with the use of antiplatelet in the USA. It consists of a self-expanding nitinol frame cov-
agents [15]. NOACs have recently become a preferred treat- ered on its atrial surface by a polyethylene membrane. The
ment option, as they not only result in lower rates of fatal device is secured in place by its radial strength and by evenly
bleeding and intracranial hemorrhage compared to VKA, they spaced tines in the distal portion of the device which interdig-
are also better tolerated and result in greater patient compli- itate within the trabeculation of the LAA. In March 2015, the
ance [16]. However, high cost, concerns over bleeding risk, FDA approved the Watchman for LAAC to reduce the risk of
side effects, renal failure, and lack of an available reversal stroke in patients with non-vavular AF based upon data from
agent limit the use of NOACs [17]. Even with the introduction two randomized trials of Watchman LAAC versus VKA and
of NOACs, approximately 40% of patients who are indicated two continuing access registries [21, 22•].
for oral anticoagulation are not treated [18•]. Single antiplate- PROTECT AF (Percutaneous closure of the left atrial ap-
let therapy in combination with a NOAC, which is often re- pendage versus warfarin therapy for prevention of stroke in
quired in patients with coronary artery disease and/or those patients with atrial fibrillation: a randomized non-inferiority
with a history of percutaneous coronary intervention, in- trial) was the first randomized clinical trial of the safety and
creases the relative risk of bleeding by 40 to 50% [19]. As a efficacy of the Watchman LAA occluder. This was a
consequence, nonpharmacological therapies for stroke pre- Bayesian, non-inferiority trial of 707 nonvalvular AF patients
vention address an important unmet clinical need. at high risk of stroke who were eligible for long-term warfarin
therapy. Subjects were randomized 2:1 to receive either device
closure of the LAA or warfarin therapy. The primary efficacy
Current Evidence endpoint was the composite of stroke of any cause, systemic
embolism, or cardiovascular/unexplained death. At 18-month
Surgical LAA excision was first proposed more than half a follow-up, non-inferiority was achieved for the device [21]. At
century ago [20]. Recent technological advances have led to 4-year follow-up, Watchman LAAC led to a significant reduc-
tion in the primary efficacy endpoint compared with warfarin
(hazard ratio [HR] 0.61 [95% CI, 0.38-0.97], P = 0.04) [23••].
Among the components of the primary endpoint, Watchman
LAAC led to significantly fewer hemorrhagic stroke and low-
er cardiovascular mortality than warfarin, and no significant
difference in ischemic stroke. Similar results were also seen at
5-year follow-up [24••]. Device closure also resulted in better
quality-of-life scores [25]. However, these favorable out-
comes came at the expense of an 8% rate of serious
periprocedural complications including a 4.8% rate of pericar-
dial effusion requiring percutaneous or surgical drainage.
The following trial, PREVAIL (Prospective randomized
evaluation of the Watchman left atrial appendage closure de-
vice in patients with atrial fibrillation versus long-term warfa-
rin therapy), was a Bayesian, non-inferiority trial of 407 sub-
jects randomized 2:1 to Watchman LAAC or to warfarin ther-
Fig. 1 Prevalence of treatment strategies across the spectrum of apy. The main purpose of the study was to evaluate procedural
CHA2DS2-VASc scores in patients in the NCDR-PINNACLE registry.
(Adapted with permission from: Hsu JC, et al. JAMA Cardiology
safety among sites and operators who were not experienced in
2016;1(1):55–62. Copyright ©(2016) American Medical Association. LAAC, and to further explore the efficacy of LAAC. Similarly
All rights reserved) [10] to PROTECT-AF, the primary efficacy endpoint was a
Curr Cardiol Rep (2018) 20:42 Page 3 of 9 42

Prospective Observational
composite of stroke, systemic embolism, and CV or unex-

Post approval registry

plained death. A second, co-primary endpoint was stroke

Operator discretion
and systemic embolism occurring > 7 days from procedure,
which was meant to test the mechanistic hypothesis of LAAC
[22•]. A third co-primary endpoint was procedural safety,

N = 3822
using a pre-defined performance criterion set by the FDA.

4.0
In PREVAIL, the procedural complication rate was 2.2%,
which met the pre-specified performance criterion and which
N = 1025 73% OAC ineligible

DAPT 15% warfarin 11%

was significantly lower compared with that observed in the
Prospective Observational

NOAC + ASA lifelong

Operator discretion: 60%
PROTECT AF trial. At the final 5-year analysis, device clo-
47 centers in Europe,
Russia, Middle East
sure achieved non-inferiority to warfarin for the primary end-
point of stroke or systemic embolism after 7 days but did not
EWOLUTION

achieve non-inferiority for the primary efficacy endpoint of

cardiovascular death, stroke, or systemic embolism at
18 months [24••].
ASA aspirin, CI contraindication, DAPT dual antiplatelet therapy, NOAC non-vitamin K antagonist oral anticoagulant, OAC oral anticoagulation
4.5

A pooled patient-level meta-analysis of the 5-year follow-

up of PROTECT AF and PREVAIL was recently reported. A
N = 150 OAC ineligible 4
Prospective Observational

DAPT × 6 months ASA

total of 4,343 patient-years of follow-up were analyzed.

LAAC with the Watchman occluder provided similar stroke
European centers

prevention to warfarin, with reductions in hemorrhagic stroke

and mortality (Fig. 2) [24••]. The rate of ischemic stroke or
lifelong

systemic embolism was numerically greater among patients

ASAP

treated with Watchman compared with warfarin, but this dif-

4.4

ference did not reach statistical significance. Contributing fac-

tors to this difference may include the peri-procedural strokes
warfarin × 45 days DAPT ×
5 months ASA lifelong

observed in the early experience of PROTECT AF and the

Summary of the clinical studies of the Watchman device for stroke prevention in AF patients

N = 269 OAC eligible

Non-inferiority versus

substantially lower-than-expected rate of ischemic stroke in

the warfarin-treated patients in PREVAIL. Furthermore, oral
randomization
warfarin, 2:1

anticoagulation may provide some benefit in reducing ische-

mic stroke unrelated to AF in patients with clinical character-
PREVAIL

istics reflected by high CHA2DS2-VASc score (e.g., diabetes,

4.0

elderly age, hypertension, etc.)

A key rationale to pursue LAAC is to obviate the need for
warfarin × 45 days DAPT ×

long-term oral anticoagulation and reduce the risk of bleeding.

5 months ASA lifelong
Prospective Observational

A pooled patient-level analysis of the 1,114 patients from the

N = 566 OAC eligible

PROTECT AF and PREVAIL trials compared the risk of

bleeding with Watchman LAAC compared with long-term
OAC with warfarin. Landmark analysis demonstrated that
the rate of major bleeding was significantly lower among the
CAP

device group beyond the post-procedural medical regimen of

3.9

aspirin and warfarin (1.0 events vs. 3.5 events per 100 patient-
years; RR: 0.28; 95% CI 0.16 to 0.49; p < 0.001) (Fig. 3). This
N = 463 OAC eligible
Non-inferiority versus

DAPT × 5 months
warfarin × 45 days

long-term benefit must be balanced with the risk of procedural

randomization

ASA lifelong
warfarin, 2:1

complications, which have decreased significantly over time

PROTECT AF

[26••].

Observational Data
3.2
Post implantation

PROTECT AF and PREVAIL enrolled patients eligible for

CHA2DS2VASc

long-term warfarin, and patients assigned to device therapy

Population

therapy

received a 6-week course of warfarin and aspirin post-proce-

Table 1

Design

dure. Therefore, patients with an absolute or perceived contra-

indication to OAC were not evaluated.
42 Page 4 of 9 Curr Cardiol Rep (2018) 20:42

Fig. 2 Primary outcomes with

LAA closure compared with
warfarin according to a patient-
level meta-analysis of the
PROTECT AF and PREVAIL
randomized trials. HR = Hazard
ratio; CI = confidence interval;
CV = cardiovascular; SE =
systemic embolism; MRS =
modified Rankin Score. (Adapted
from: Reddy VY, et al. J Am Coll
Cardiol. 2017;pii:S0735-1097)
[24••]

The ASAP (Left atrial appendage closure with the embolism was 1.8% [95% CI, 0.9% to 3.3%], approximately
Watchman device in patients with a contraindication for oral 75% lower than the expected rate in a hypothetical patient
anticoagulation: ASA Plavix feasibility study with Watchman population with a similar CHADS2 score treated with aspirin
left atrial appendage closure technology) study investigated alone [27•, 28].
the safety and efficacy of Watchman LAAC in patients with The EWOLUTION (Registry on Watchman Outcomes in
contraindications to or ineligibility for OAC. This prospective, Real-Life Utilization) registry followed 1,025 patients under-
observational study enrolled 150 patients deemed ineligible going commercial Watchman implantation outside of the
for OAC. Patients underwent LAAC with the Watchman and USA. A total of 73% of patients in the registry were not
were subsequently treated with dual antiplatelet therapy (aspi- eligible for OAC and received dual antiplatelet therapy post
rin and clopidogrel) for 6-month post-implantation, and then implantation. The mean CHA2DS2-VASc score of the study
aspirin monotherapy thereafter. At a median follow-up of population were higher than that of previous trials [24••]. The
55.4 months, the annual rate of ischemic stroke or systemic implantation success rate was 98% and the rate of procedure-

Fig. 3 Landmark analysis of the freedom from major bleeding with post-randomization, during which device-treated patients received
Watchman LAA closure compared with warfarin. Shown is freedom warfarin and aspirin followed by dual antiplatelet therapy; and beyond
from major bleeding from a pooled, patient-level analysis from the 6 months, when device-treated patients were eligible to receive aspirin
PROTECT AF and PREVAIL trials, divided into three intervals: from alone. (Adapted from: Price MJ, et al. JACC Cardiovasc Interv. 2015.
randomization to day 7 representing the periprocedural period for 8(15):1925–1932) [26••]
patients randomly assigned to LAA closure; from 8-day to 6-month
Curr Cardiol Rep (2018) 20:42 Page 5 of 9 42

related complication was 2.8%, which was lower than that trial [32•]. Despite the initially high procedural compli-
observed in the randomized clinical trials of the device. cation rate observed in the PROTECT AF trial, subse-
Device thrombus was identified in 3.7% of patients indepen- quent studies and observational experiences demonstrate
dent of the type post-implantation antithrombotic therapy, al- that the procedural risk has significantly decreased with
though the study was not powered to detect differences in proper training and standardization of implantation tech-
safety between the post-implant medical regimens. The rate niques and preoperative imaging (Fig. 4).
of stroke, TIA or systemic embolism was 1.5 per 100 patient-
years, 84% lower than the expected rate based on
CHA2DS2VASc score [29••]. Amulet LAA Occluder
The results of ASAP and EWOLUTION suggest that
LAAC with the Watchman is a safe and effective alternative The Amulet LAA occluder (St Jude Medical, Minneapolis,
in patients with and without contraindications to OAC. The MN) is a second-generation iteration of the Amplatzer
safety and efficacy of Watchman LAAC in patients with Cardiac Plug. The device consists of a self-expanding nitinol
strong contraindications to OAC will be more robustly evalu- mesh that forms a distal lobe and proximal disk, each with a
ated in the ongoing ASAP-TOO trial (Assessment of the sewn polyester patch, connected by a short central waist. The
Watchman Device in Patients Unsuitable for Oral device has a short length relative to its diameter. The distal
Anticoagulation) trial (clinicaltrials.gov identifier, lobe has retention hooks around its circumference that anchors
NCT02928497). This prospective, multicenter clinical trial the device within the LAA, while the proximal disk covers the
will enroll approximately 900 AF patients who are deemed mouth of the LAA from within the left atrium. Therefore, its
unsuitable to oral anticoagulation and randomly assign them mechanism of closure differs from that of the Watchman
to either medical therapy (single or no antiplatelet therapy per device.
operator discretion) or Watchman LAAC followed by a The Amulet global, prospective, observational study en-
modified post-implant drug regimen of dual antiplatelet ther- rolled 1,088 patients with nonvalvular AF and high stroke risk
apy that does not included OAC. [30] at 61 sites outside of the USA. [34•]. The study population had
Procedural outcomes of Watchman implantation in the a mean CHA2DS2VASc score of 4.2 ± 1.6, and the HAS-
commercial setting were recently reported [31•]. A total of BLED score was 3 or more in 78%. A total of 82% of patients
3,822 patients were treated by 382 operators at 169 centers. had a contraindication to OAC. The post-implantation medi-
Operators who were not involved in the clinical study experi- cal therapy varied according to physician preference. Post-
ence performed approximately half the cases. The rate of peri- implantation therapy consisted of OAC in 19% of patients,
procedural complications was less than 1.5%: procedure- DAPT in 54% of patients, and single anti-platelet therapy in
related death (all due to pericardial tamponade) occurred in
0.08%, procedure-related stroke in 0.08%, pericardial
tamponade requiring intervention in 1.0%, and device embo-
lization in 0.25% (two thirds of which required surgical re-
trieval). These rates of procedural complications are substan-
tially lower than those observed in the Watchman clinical trial
experience. However, the study design may have led to under-
reporting of events, peri-procedural events were collected and
reported by the manufacturer’s clinical specialists, and events
that occurred outside the catheterization laboratory may not
have been fully captured [32•]. The National Cardiovascular
Data Registry (NCDR) Left Atrial Appendage Occlusion
(LAAO) registry will provide important insights into the safe-
ty of transcatheter LAAC for stroke prevention as it is current-
ly performed in the USA.
The totality of data suggests that LAAC with the
Fig. 4 Rates of serious procedure- and device-related events in the
Watchman provides a benefit similar to warfarin for PROTECT-AF, Continued Access to PROTECT AF (CAP), PREVAIL,
preventing stroke, systemic embolism, and cardiovascu- and EWOLUTION. The available data suggests the procedural safety of
lar death. Compared with warfarin, Watchman LAAC is transcatheter LAA occlusion has improved over time. Adapted from
associated with lower rates of hemorrhagic stroke and Boersma LV, Schmidt B, Betts TR et al. Implant success and safety of
left atrial appendage closure with the WATCHMAN device: peri-
non-procedural bleeding. The effects appear to apply procedural outcomes from the EWOLUTION registry. (Adapted from:
even to those ineligible to OAC, although this needs Boersma LV, European Heart Journal. 2016; 37: 2465–2474, by
to be confirmed in the setting of a randomized clinical permission of Oxford University Press) [33]
42 Page 6 of 9 Curr Cardiol Rep (2018) 20:42

23% of patients. The implant success rate was 99%. The rate study may provide insights regarding the relative benefit
of major procedural complications rate was 3.2% (95% CI, of NOACs or LAAC, but is limited given its small sample
2.2% to 4.4%), including pericardial effusion or tamponade in size and the use of a primary endpoint that combines mea-
1.2%, vascular complication in 0.9%, periprocedural stroke in sures of safety and efficacy. At present, LAAC should not
0.2%, and death in 0.2%. Among the 62% of patients with be considered as an alternative in patients that are other-
echocardiographic follow-up, 3-month TEE confirmed ade- wise good candidates for long-term NOAC therapy.
quate occlusion (peri-device leak < 3 mm) in 98% of all cases
and device thrombus in 1.5%. Further follow-up is required to Patient Selection Careful patient selection for LAAC is essen-
assess efficacy with respect to stroke reduction, which will tial. The ideal candidates for LAAC are those who can tolerate
need to be interpreted through imputed placebo comparisons, short-term oral anticoagulation (i.e., can tolerate the post-
with the attendant methodological limitations of such an ap- procedural medical regimen used in randomized clinical trials)
proach. In sum, this global observational experience suggests but are not optimal candidates for long-term anticoagulation.
that Amulet LAAC is associated with a relatively low rate of Such patients might include those with elevated HAS-BLED
procedural events and device-related thrombosis, with com- score (> = 3), prior gastrointestinal bleeding, having a need for
plication rates generally similar or less than those observed long-term antiplatelet therapy (e.g., prior myocardial infarc-
with the Watchman device. This will be further tested in a tion or multiple coronary stents), renal failure, significant frail-
more robust fashion in the Amulet LAA Occluder trial ty or those with recurrent falls. Patients who have absolute
(clinicaltrials.gov identifier, NCT02879448). This is a contraindications to OAC should be enrolled in randomized
prospective, randomized, non-inferiority trial of the Amulet clinical trials. Those patients who are at low-bleeding risk, are
versus the Watchman LAA occluder in approximately 1600 compliant, and can otherwise safely tolerate long-term
AF patients at high thromboembolic risk who are deemed anticoagulation should be treated with warfarin or a NOAC.
suitable for short-term warfarin but are not good candidates A proposed schema for treatment of patients within the USA
for long-term oral anticoagulation. is presented in (Fig. 5).

Current Indications Percutaneous LAAC is covered by the

Further Issues Centers for Medicare and Medicaid Services and approved
by the FDA for use as an alternative for stroke prevention in
NOACs The randomized clinical trials of LAAC to date did patients with AF and high risk of stroke who have an appro-
not compare device therapy with NOACs, which are easier priate rational for avoiding long-term OAC following a shared
to use and generally safer than warfarin. NOACs are asso- decision process with their physician. The European Society
ciated with lower rates of stroke compared with VKA, of Cardiology has included percutaneous LAAC in their most
driven by reduction in hemorrhagic stroke [8, 9, 35]. recent guidelines as a good alternative for those with contra-
NOACs do not appear to reduce the risk of ischemic indications to OAC (Class IIB, level of evidence B). To date,
stroke, with the exception of high-dose dabigatran, which percutaneous LAAC is not included in the 2014 American
provided an approximately 24% reduction in ischemic College of Cardiology and American Heart Association’s
stroke in the RE-LY trial [36]. NOACs have similar to guidelines for the management of AF [7].
lower rates of major bleeding, and similar or greater rates
of gastrointestinal bleeding, compared with VKA. A net- Device-Associated Thrombus A recent systematic review of
work meta-analysis suggested that Watchman LAAC and events reported with Watchman, Amulet and ACP offered
NOAC therapy were both superior to warfarin for some insight into the predisposing risk factors and clinical
preventing hemorrhagic strokes, and that there were no significance of device-associated thrombus DAT [38].
significant differences in clinical outcomes between Outcomes of 30 different studies including 2,118 implanta-
Watchman LAAC and NOACs, although there was a trend tions that described DAT events in their results were reviewed.
for more ischemic stroke and less hemorrhagic stroke with The overall incidence of DAT was 3.9%. Most were diagnosed
Watchman LAAC [37]. PRAGUE-17 (Left Atrial with transesophageal echocardiogram with a median time to
Appendage Closure vs. Novel Anticoagulation Agents in diagnosis of 1.5 months after procedure. Most cases of DAT
Atrial Fibrillation) will randomize approximately 400 AF were asymptomatic: of 82 reported cases, only 4 presented
patients to transcatheter LAAC with the Amulet or with ischemic stroke and 2 with a TIA. The overall incidence
Watchman device or NOAC therapy (clinicaltrials.gov of a neurological event presumably related to DAP was only
identifier, NCT02426944). The primary endpoint is a 0.28%. The treatment success rate in resolving DAP with
composite of stroke, other systemic cardioembolic event, anticoagulation was 95% at a median treatment duration of
clinically significant bleeding, cardiovascular death, or 45 days although there was large variation in treatment strat-
procedure/device-related complications at 1 year. This egies and data missing in many cases.
Curr Cardiol Rep (2018) 20:42 Page 7 of 9 42

Fig. 5 Proposed treatment

strategy for patients with atrial
fibrillation, incorporating the
option of left atrial appendage
closure

Residual Leaks The shape of the LAA and its ostium is of whom had small leaks into the LAA (< 5mm in diam-
variable from patient to patient. LAAC devices have a eter) [43•]. Transcatheter occlusion of post-Lariat LAA
circular profile, whereas the LAA ostium is usually oval leaks with occluder devices is feasible, but the long-term
and sometimes irregularly shaped. This geometric vari- efficacy of this approach in terms of stroke prevention is
ability can result in incomplete closure of the LAA. unknown [44].
Theoretically, peri-device flow into and out of the LAA
could result in thrombus formation and embolization. In
general, a major device leak is defined as peri-device flow Conclusion
into the LAA ≥ 5 mm at its smallest diameter. A sub-study
of the PROTECT AF trial demonstrated that minor peri- LAAC with the Watchman device appears to be a safe and
device leaks were quite common, as they were identified effective strategy for stroke prevention in patients with
approximately 40% of patients at 45-day follow-up TEE. nonvalvular AF who are at high risk of stroke. Taken together,
Results of the study suggest that these residual leaks are the PROTECT AF and PREVAIL randomized clinical trials
of little clinical significance and not associated with in- demonstrate similar rates of cardiovascular death, stroke, or
creased risk of cardioembolic events [39]. However, in systemic embolism with Watchman LAAC compared with
this study the event rate was considerably low, which warfarin, with reductions in hemorrhagic stroke, mortality,
limits the overall power of the results. An analysis of and non-procedural bleeding. Recent observational studies
observational data among patients that received an suggest that procedural complication rates have improved
Amplatzer Cardiac Plug (ACP) demonstrated a relatively since the clinical trial experience, and that alternative post-
lower rate of peri-device leak (12.5%), which may result implant medication regimens other than short-duration warfa-
from the different method of closure between Watchman rin might be safe and effective. Ongoing randomized trials are
and the ACP. This study also demonstrated that these testing in a robust fashion whether Watchman LAAC can be
leaks do not carry an increased risk of cardio embolic extended to those patients who are ineligible for even short-
events [40]. term OAC, and whether the Amulet LAA occluder provides
The mechanism of residual leak—i.e., peridevice or non-inferior outcomes compared with the Watchman. There is
central—may impact upon subsequent thromboembolic not a “one size fits all” approach to stroke prevention in pa-
risk. The Lariat device (SentreHeart, Redwood City, CA) tients with AF. A large body of data supports the safety and
enables the percutaneous ligation of the LAA through the efficacy of NOACs to prevent all-cause stroke in patients who
delivery of a surgical suture via a combined transseptal can tolerate therapy. Careful consideration of a particular in-
and sub-xiphoid approach [41]. Frequent early and late dividual’s thromboembolism risk off of OAC, bleeding risk
leaks have been reported [42•]. In a retrospective, multi- on long-term OAC, and procedural complication rate with
center study of 98 patients undergoing Lariat LAA liga- LAAC is required to identify the optimal management strate-
tion leaks were present in 20% of cases there were 3 gy. At present, transcatheter LAAC should be considered
patients with late stroke (> 6-month post-procedure), all “first-line second-line therapy”.
42 Page 8 of 9 Curr Cardiol Rep (2018) 20:42

Compliance with Ethical Standards 11. Albers GW, Diener HC, Drind M, Nevinson M, al w. Ximelagatran
vs warfarin for stroke prevention in patients with atrial fibrillation: a
randomized trial. J Am Med Assoc. 2005;293(6):690–8.
Conflict of Interest Matthew J. Price has received consulting honoraria
12. Olsson SB. Executive Steering Committee of the SPORTIF III
from Abbott Vascular, Boston Scientific, W.L. Gore, Medtronic, and
Investigators. Stroke prevention with the oral direct thrombin in-
AstraZeneca; has received speaker honoraria from Abbott Vascular,
hibitor ximelagatran compared with warfarin in patients with non-
Medtronic, Terumo, and AstraZeneca; has received research grants from
valvular atrial fibrillation (SPORTIF III): randomised controlled
Daiichi Sankyo; has served as a site investigator for trials involving the
trial. Lancet. 2003;22;362(9397):1691–8.
Watchman device; has received honoraria as a proctor for Abbott
13. Pisters R, Lane DA, Nieuwlaat R, de Vos CB, Crijns HJGM, Lip
Vascular and Boston Scientific; and is a member of the Steering
GYHA. Novel user-friendly score (HAS-BLED) to assess 1-year
Committee for the NCDR LAAO registry.
risk of major bleeding in patients with atrial fibrillation. Chest.
Ricardo Kosturakis has nothing to disclose.
2010;1;138(5):1093–100.
14. Schulman S, Beyth RJ, Kearon C, Levine MN. Hemorrhagic com-
Human and Animal Rights and Informed Consent This article does not plications of anticoagulant and thrombolytic treatment. Chest.
contain any studies with human or animal subjects performed by any of 2008;133(6):257S–98S.
the authors. 15. Sørensen R, Hansen ML, Abildstrom SZ, Hvelplund A, Andersson
C, Jørgensen C, et al. Risk of bleeding in patients with acute myo-
cardial infarction treated with different combinations of aspirin,
clopidogrel, and vitamin K antagonists in Denmark: a retrospective
References analysis of nationwide registry data. Lancet. 2009;12;374(9706):
1967–74.
Papers of particular interest, published recently, have been 16. Adam SS, McDuffie JR, Ortel TL, Williams JW. Comparative ef-
fectiveness of warfarin and new oral anticoagulants for the manage-
highlighted as:
ment of atrial fibrillation and venous thromboembolism: a system-
• of Importance atic review. Ann Intern Med. 2012;4;157(11):796–807.
•• of Major Importance 17. Kirley K, GouthamRao BV, Masi C. The role of NOACs in atrial
fibrillation management: a qualitative study. J Atr Fibrillation.
1. Chugh SS, Havmoeller R, Narayanan K, Singh D, Rienstra M, 2016;9(1):1416.
Benjamin EJ, et al. Worldwide epidemiology of atrial fibrillation: 18.• Marzec LN, Wang J, Shah ND, Chan PS, Ting HH, Gosch KL, et al.
a global burden of disease 2010 study. Circulation. 2014;129(8): Influence of direct oral anticoagulants on rates of oral
837–47. anticoagulation for atrial fibrillation. J Am Coll Cardiol.
2. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Borden 2017;69(20):2475–84. This paper describes the impact of newer
WB, et al. Heart disease and stroke statistics—2013 update: a report oral anticoagulants on treatment rates, and further highlights
from the American Heart Association. Vol. 127, Circulation 2013. the underutilization of oral anticoagulation in atrial fibrillation
6–245 p. patients at risk for thromboembolism.
3. De Backer O, Arnous S, Ihlemann N, Vejlstrup N, Jørgensen E, 19. Xu H, Ruff CT, Giugliano RP, Murphy SA, Nordio F, Patel I, et al.
Pehrson S, et al. Percutaneous left atrial appendage occlusion for Concomitant use of single antiplatelet therapy with edoxaban or
stroke prevention in atrial fibrillation: an update. Open Hear. warfarin in patients with atrial fibrillation: analysis from the engage
2014;1(1):e000020. af-timi48 trial. J Am Heart Assoc. 2016;5(2):1–9.
20. Madden JL. Resection of the left auricular appendix. J Am Med
4. Lloyd-jones DM, Benjamin EJ, Jarett D, Borden WB, Bravata DM,
Assoc. 1949;2;140(9):769.
Dai S, et al. Heart disease and stroke statistics—2012 update. A
21. Holmes DR, Reddy VY, Turi ZG, Doshi SK, Sievert H, Buchbinder
report from the American Heart Association. Circulation.
M, et al. Percutaneous closure of the left atrial appendage versus
2015;125(1):2–22.
warfarin therapy for prevention of stroke in patients with atrial
5. Blackshear JLOJA. Appendage obliteration to reduce stroke in car-
fibrillation: a randomised non-inferiority trial. Lancet.
diac surgical patients with atrial fibrillation. Ann Thorac Surg.
2009;15;374(9689):534–42.
1996;61(2):755–9.
22.• Holmes DR, Kar S, Price MJ, Whisenant B, Sievert H, Doshi SK,
6. Johnson WD, Ganjoo AK, Stone CD, Srivyas RC, Howard M. The et al. Prospective randomized evaluation of the watchman left atrial
left atrial appendage: our most lethal human attachment! Surgical appendage closure device in patients with atrial fibrillation versus
implications. Eur J Cardio-thoracic Surg. 2000;17(6):718–22. long-term warfarin therapy: the PREVAIL trial. J Am Coll Cardiol.
7. January CT, Wann LS, Alpert JS, Calkins H, Cigarroa JE, 2014;8;64(1):1–12. This report describes the outcomes of the
Cleveland JC, et al. 2014 AHA/ACC/HRS guideline for the man- second randomized trial of the WATCHMAN LAA closure
agement of patients with atrial fibrillation: executive summary: a compared with warfarin, and confirms improved procedural
report of the American college of cardiology/American heart asso- safety compared with earlier trial experience.
ciation task force on practice guidelines and the heart rhythm soci- 23.•• Reddy VY, Sievert H, Halperin J, Doshi SK, Buchbinder M, Neuzil
ety. J Am Coll Cardiol. 2014;64(21):2245–80. P, et al. Percutaneous left atrial appendage closure vs warfarin for
8. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. atrial fibrillation a randomized clinical trial. J Am Med Assoc.
Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N 2014;19;312(19):1988–98. This first, randomized study demon-
Engl J Med. 2011;8;365(10):883–91. strates the non-inferiority of superiority of LAA closure com-
9. Granger CB, Alexander JH, McMurray JJV, Lopes RD, Hylek EM, pared with warfarin in patients who are candidates for long-
Hanna M, et al. Apixaban versus warfarin in patients with atrial term oral anticoagulation.
fibrillation. N Engl J Med. 2011;15;365(11):981–92. 24.•• Reddy VY, Doshi SK, Kar S, Gibson DN, Price MJ, Huber K, et al.
10. Hsu JC, Maddox TM, Kennedy KF, Katz DF, Marzec LN, Lubitz 5-year outcomes after left atrial appendage closure. J Am Coll
SA, et al. Oral anticoagulant therapy prescription in patients with Cardiol. 2017;pii:S0735–1097. This paper describes the final in-
atrial fibrillation across the Spectrum of stroke risk. J Am Med dividual and pooled outcomes of the two randomized clinical
Assoc Cardiol. 2016;1(1):55. trials of WATCHMAN LAA closure compared with warfarin.
Curr Cardiol Rep (2018) 20:42 Page 9 of 9 42

25. Alli O, Doshi S, Kar S, Reddy V, Sievert H, Mullin C, et al. Quality with the WATCHMAN device: peri-procedural outcomes from the
of life assessment in the randomized PROTECT AF (percutaneous EWOLUTION registry. Eur Heart J. 2016;37:2465–74.
closure of the left atrial appendage versus warfarin therapy for pre- 34.• Landmesser U, Schmidt B, Nielsen-Kudsk JE, Lam SCC, Park J-
vention of stroke in patients with atrial fibrillation) trial of patients W, Tarantini G, et al. Left atrial appendage occlusion with the
at risk for stroke with nonvalvular atrial. J Am Coll Cardiol. AMPLATZER Amulet device: periprocedural and early clinical/
2013;61(17):1790–8. echocardiographic data from a global prospective observational
26.•• Price MJ, Reddy VY, Valderrábano M, Halperin JL, Gibson DN, study. EuroIntervention. 2017;1–10. This is the first large pro-
Gordon N, et al. Bleeding outcomes after left atrial appendage clo- spective study to evaluate the safety and efficacy of LAA closure
sure compared with long-term warfarin a pooled, patient-level anal- with the Amulet device.
ysis of the WATCHMAN randomized trial experience. J Am Coll 35. Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD,
Cardiol. Cardiovasc Interv. 2015;8(15):1925–32. This analysis of Halperin JL, et al. Edoxaban versus warfarin in patients with atrial
randomized trial data demonstrates the reduction in post- fibrillation. N Engl J Med. 2013;28;369(22):2093–104.
procedural major bleeding in patients managed with a strategy 36. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J,
of LAA closure compared with long-term oral anticoagulation Parekh A, et al. Dabigatran versus warfarin in patients with atrial
with warfarin. fibrillation. N Engl J Med. 2009;17;361(12):1139–51.
27.• Reddy VY, Möbius-Winkler S, Miller MA, Neuzil P, Schuler G, 37. Koifman E, Lipinski MJ, Escarcega RO, Didier R, Kiramijyan S,
Wiebe J, et al. Left atrial appendage closure with the watchman Torguson R, et al. Comparison of Watchman device with new oral
device in patients with a contraindication for oral anticoagulation: anti-coagulants in patients with atrial fibrillation: a network meta-
The ASAP study (ASA plavix feasibility study with watchman left analysis. Int J Cardiol. 2017;21;205:17–22.
atrial appendage closure technology). J Am Coll Cardiol. 38. Lempereur M, Aminian A, Freixa X, Gafoor S, Kefer J, Tzikas A,
2013;25;61(25):2551–6. This observational study demonstrates et al. Device-associated thrombus formation after left atrial append-
the feasibility of LAA closure to reduce stroke in patients who age occlusion: a systematic review of events reported with the
have significant contraindications to oral anticoagulation. Watchman, the Amplatzer Cardiac Plug and the Amulet. Catheter
28. Sharma D, Reddy VY, Sandri M, Schulz P, Majunke N, Hala P, Cardiovasc Interv. 2017;1:90(5):E111–21.
et al. Left atrial appendage closure in patients with contraindications 39. Viles-Gonzalez JF, Kar S, Douglas P, Dukkipati S, Feldman T,
to oral anticoagulation. J Am Coll Cardiol. 2016;67(18):2190–2. Horton R, et al. The clinical impact of incomplete left atrial append-
29.•• Boersma LV, Ince H, Kische S, Pokushalov E, Schmitz T, Schmidt age closure with the watchman device in patients with atrial fibril-
B, et al. Efficacy and safety of left atrial appendage closure with lation. J Am Coll Cardiol. 2012;59(10):923–9.
WATCHMAN in patients with or without contraindication to oral 40. Saw J, Tzikas A, Shakir S, Gafoor S, Omran H, Nielsen-Kudsk JE,
anticoagulation: 1-year follow-up outcome data of the et al. Incidence and clinical impact of device-associated thrombus
EWOLUTION trial. Hear Rhythm. 2017;14(9):1302–8. This and peri-device leak following left atrial appendage closure with the
large, observational study describes the commercial experience Amplatzer Cardiac Plug. JACC Cardiovasc Interv. 2017 Feb;10(4):
of WATCHMAN LAA closure outside of the USA, and shows 391–9.
that procedural complications have decreased since the early 41. Price MJ, Valderrábano M. Left atrial appendage closure to prevent
clinical trial experience. stroke in patients with atrial fibrillation. Circulation. 2014;8;130(2):
30. Holmes DR, Reddy VY, Buchbinder M, Stein K, Elletson M, 202–12.
Bergmann MW, et al. The assessment of the Watchman device in 42.• Price MJ, Gibson DN, Yakubov SJ, Schultz JC, Di Biase L, Natale
patients unsuitable for oral anticoagulation (ASAP-TOO) trial. Am A, et al. Early safety and efficacy of percutaneous left atrial append-
Heart J. 2017;189:68–74. age suture ligation: results from the U.S. transcatheter LAA ligation
31.• Reddy VY, Gibson DN, Kar S, O’Neill W, Doshi SK, Horton RP, consortium. J Am Coll Cardiol. 2014;12;64(6):565–72. This paper
et al. Post-approval U.S. experience with left atrial appendage clo- is the first multicenter study to examine the outcomes of trans-
sure for stroke prevention in atrial fibrillation. J Am Coll Cardiol. catheter LAA ligation, and defines the relatively high procedur-
2017;24;69(3):253–61. This study represents the first descrip- al risk associated with the procedure.
tion of the procedural outcomes of commercial implants of the 43.• Gianni C, Di Biase L, Trivedi C, Mohanty S, Göko Y, Güne MF,
WATCHMAN device in the United States. et al. Clinical implications of leaks following left atrial appendage
32.• Saw J, Price MJ. Assessing the safety of early U.S. commercial ligation with the LARIAT device. J Am Coll Cardiol. 2016;9(10):
application of left atrial appendage closure. J Am Coll Cardiol. 1051–7. This study links residual or late leaks after transcath-
2016;69(3):262–4. Important critique of the strengths and lim- eter LAA ligation with thromboembolic events.
itations of procedural outcomes in the current commercial 44. Mosley WJ, Smith MR, Price MJ. Percutaneous management of
watchman experience. late leak after lariat transcatheter ligation of the left atrial appendage
33. Boersma LV, Schmidt B, Betts TR, Sievert H, Tamburino C, Teiger in patients with atrial fibrillation at high risk for stroke. Catheter
E, et al. Implant success and safety of left atrial appendage closure Cardiovasc Interv. 2014;83(4):664–9.