Defying Hopes, Dual Antiplatelets May Be Needed With LAA-Closure Devices in AF

July 01, 2017

VIENNA, AUSTRIA — Thrombi were sometimes found on two left-atrial-appendage (LAA) occluder devices in the weeks following implantation in two large registry studies of patients with atrial fibrillation (AF) deemed at high risk for bleeding complications on oral anticoagulation[1,2].

In the registry of about 450 patients implanted with the Amplatzer Amulet (St Jude Medical/Abbott) or Watchman (Boston Scientific) catheter-delivered LAA occluders, such thrombus was seen in about 5% of cases and was associated with an eightfold increased risk of ischemic stroke or transient ischemic attack (P=0.008) in the ensuing weeks. The registry wasn't industry-sponsored.

Device thrombosis was significantly more likely if the patient had a history of ischemic stroke, and the risk of it being detected in the months after implantation plunged if discharge medications included dual antiplatelet therapy (DAPT).

In both registry analyses, reported here at the European Heart Rhythm Association (EHRA) EUROPACE-CARDIOSTIM 2017 sessions, the overwhelming majority of patients were discharged on some form of antithrombotic therapy, sometimes oral anticoagulation (OAC) with or without antiplatelets but most often dual-antiplatelet therapy.

"It seems that the therapy can be deployed safely in most patients. There are still questions about its ability to completely replace antithrombotic therapy at all and safely prevent ischemic stroke," Prof Gerhard Hindricks (University of Leipzig Heart Center, Germany) told|Medscape Cardiology.

"And there is a key question: how do these technologies perform against the novel oral anticoagulants that have shown a significant benefit for patients with atrial fibrillation and stroke risk?" posed Hindricks, who isn't connected to either registry analysis.

"I don't think at present we can use this device as an alternative to oral anticoagulation," said Dr Boris Schmidt (Cardiovascular Center Bethanien, Frankfurt, Germany), during the question-and-answer period following his presentation of findings from the Amulet Observational Study, which was company-sponsored.

"That's how we had started, in the era of vitamin K antagonists," Schmidt said, referring to the original hopes that catheter-delivered LAA-occluder devices would cut the risk of ischemic stroke enough in patients with nonvalvular AF enough to render OAC unnecessary.

"Nowadays we consider them as an alternative for patients who are contraindicated to long-term oral anticoagulation," he said, including the new oral anticoagulants (NOACs), which are direct thrombin inhibitors or factor Xa inhibitors.

In his analysis, about three-fourths of patients discharged on OAC after Amulet implantation suffered a major bleeding event during follow-up. The device thrombosis rate was <1% overall but up to 1.5% when researchers counted only patients who underwent transesophageal echocardiography (TEE) for confirmation or rule-out.

"A Clear Dilemma"

Both analyses highlight a crucial point about the use of these LAA occluders in practice—namely, there is no agreement on discharge meds. Antiplatelet and OAC regimens varied widely in the two studies.

"It goes from no treatment to oral anticoagulation plus antiplatelets, with no randomized data available telling us what is really the best way in various cohorts," said the assigned discussant for both presentations, Dr Thorsten Lewalter (Peter Osypka Heart Center, Munich, Germany).

"There is an urgent need in this field to move forward from registries to randomized controlled trials" that include randomized postdischarge antithrombotic therapy, he said.

Prof Laurent Fauchier (Université François Rabelais, Tours, France) acknowledged that the use of these LAA occluders in patients with AF may in some ways be substituting one clinical problem for another. That is, DAPT appears necessary to prevent device thrombosis and possible stroke, yet the occluders are most appropriate for those who've previously had serious bleeding on antithrombotic therapy.

"This is a clear dilemma," Fauchier said during the Q&A following his presentation of the combined Amulet-Watchman analysis. "Knowing better how to manage antithrombotic therapy is one of the major issues in these patients."

Combined Amulet, Watchman Independent Registry

The independent registry analysis that Fauchier presented included 453 consecutive patients with AF who received either of the LAA closure devices at eight centers in France from February 2012 to December 2016. About 60% received the Watchman and about 40% the Amulet.

The two groups didn't differ significantly at baseline with respect to history of ischemic stroke, history of vascular disease, presence of diabetes, prevalence of permanent AF, or CHA2DS2VASc or HAS-BLED scores.

Nor did they differ significantly in their reasons for receiving the LAA occluder, which were previous bleeding (90.3% in the Amulet and 89.4% in the Watchman groups) and/or contraindication to OAC (77.7% and 74.8%, respectively), and history of recurrent ischemic stroke (2.9% and 6.8%, respectively).

Antithrombotic Therapy on Discharge After LAA Closure by Amulet vs Watchman

Discharge antithrombotic therapy Overall (%) Amulet (n=181) (%) Watchman (n=272) (%) P, Amulet vs Watchman
No OAC, no APT 8 14.6 3.4 <0.0001
Single APT 34 39.9 30.9 0.07
Dual APT 24 25.3 22.5 0.58
OAC, no APT 30 18.5 37.0 <0.0001
OAC with APT 5 1.7 6.1 0.045
OAC=oral anticoagulation therapy
APT=antiplatelet therapy

Over a mean follow-up of 11 months, 5.3% of patients developed device-related thrombus by TEE, 4.0% experienced major bleeding, and 3.8% had an ischemic stroke and 0.4% a TIA. Also in that time, 7.3% of patients died, although only 1.3% overall died of cardiovascular causes. None of the rates differed significantly between Amulet and Watchman.

DAPT at discharge was significantly more likely among those free of device thrombus throughout the follow-up compared with those with such thrombus.

Discharge Antithrombotic Therapy After LAA Closure by Watchman or Amulet by Detection of Device Thrombus

Discharge antithrombotic therapy Device thrombus (n=24) No device thrombus (n=429) P
No OAC, no APT 12.5 7.7 0.65
Single APT 41.7 34.2 0.60
Dual APT 4.2 24.8 0.04
OAC, no APT 41.7 28.7 0.26
OAC with APT 0 4.6 0.58
OAC=oral anticoagulation therapy
APT=antiplatelet therapy

Significant predictors of device thrombus included prior ischemic stroke; discharge DAPT was significantly protective. But CHA2DS2VASc score or discharge single antiplatelet therapy or OAC were not predictive.

Device thrombus itself was significantly predictive of ischemic stroke, along with and independently of history of vascular disease.

Independent Predictors of Device Thrombus and Stroke or TIA After LAA Closure by Watchman or Amulet

Predictors of Events During Follow-up Hazard ratio (95% CI) P
Device thrombus
Age, y 1.06 (1.004–1.13) 0.035
Previous ischemic stroke 5.90 (1.91–18.20) 0.002
DAPT 0.06 (0.006–0.63) 0.018
Stroke or TIA
Vascular disease history 11.79 (2.54–55.26) 0.002
Device thrombus 8.20 (1.73–38.94) 0.008

St Jude Amulet Registry

In Schmidt's "real-world" registry of 1077 patients with AF initially without LAA thrombus who received Amulet devices at 61 centers in Europe, the Middle East, Asia, South America, and Australia, TEE was performed in most after 1 to 3 months and reviewed at an independent core lab. Serious adverse events were also independently adjudicated.

The patients' baseline features reflected their high-risk status and general suitability for an LAA occluder device, Schmidt said. The prevalence of prior stroke was 27.5%, prior TIA 10.6%, and prior major bleed 72.4%, and their mean CHA2DS2VASc score was 4.2 and mean HAS-BLED score was 3.3. A contraindication to OAC was present in 82.8%.

Adjudicated In-Hospital Major Adverse Events With LAA Closure in the Amulet Registry

Major in-hospital events Rate (%)
Death 0.2
Major bleeding* 2.4
Major vascular complications 0.9
Periprocedural stroke 0.2
Device embolization 0.1
Total 3.2

There were 41 major adverse events in 35 patients (3.2%) at the procedure and during the index hospitalization. The rate of successful LAA closure by adjudicated TEE was 99.6% at the procedure and 98.2% at the 1- to 3-month follow-up, Schmidt said.

DAPT was the prevailing discharge antithrombotic therapy, although about 19% received OAC; only 2% were discharged on neither OAC nor antiplatelets.

Antithrombotic Therapy on Discharge After LAA Closure in the Amulet Registry

Discharge antithrombotic therapy Rate (%)
No OAC, no APT 2.0
Single APT: Aspirin 16.0
Single APT: Clopidogrel 7.0
Dual APT 54.3
OAC with or without APT 18.9
APT=antiplatelet therapy
OAC=oral anticoagulation therapy (either vitamin K antagonists or NOAC

There were 10 cases of device thrombus by 90 days by adjudicated TEE and clinical evaluation. That makes for a 0.9% device thrombus rate over all 1077 cases but a 1.5% rate for only the 673 adjudicated TEEs, Schmidt observed.

Of the 10 patients with device thrombus, four had been on OAC, three on single antiplatelet therapy, and three on DAPT. Treatment on identifying device thrombus consisted of maintenance of single antiplatelet therapy in two cases, escalation from single antiplatelet to DAPT in one, and initiation of OAC in seven.

There was one stroke related to device thrombus; it occurred 28 days after detection of the thrombus (while on single antiplatelet therapy) and 71 days after implantation, Schmidt said.

The findings "confirm that Amulet LAA occlusion is probably a safe alternative for prevention of stroke in patients with nonvalvular atrial fibrillation and contraindications to long term oral anticoagulation," Schmidt said in his presentation.

Lewalter, as discussant, agreed at least that the findings represented "a big step ahead in terms of safety" compared with the Amulet clinical-trial experience.

"At the end of the day, we have learned that having a thrombus on the device is a factor that creates more stroke during follow-up—that is, more ischemic stroke. We have a less favorable outcome," according to Lewalter.

But a big question about device thrombus remains, he said. "Is it really the source of the embolic stroke, or is it a risk factor, an additional risk factor for having thrombus formation in other areas?" To underscore the issue, he pointed out that vascular disease itself was a significant independent risk factor for ischemic stroke in the combined Amulet-Watchman registry.

The Amulet registry was sponsored by St Jude Medical/Abbott. Hindricks has disclosed receiving research grants from Boston Scientific and St Jude Medical. Fauchier discloses receiving lecture fees from Bayer, Bristol-Myers Squibb/Pfizer, Boehringer Ingelheim, and Medtronic; travel grants from Bayer, Biotronik, Bristol-Myers Squibb/Pfizer, Boehringer Ingelheim, Boston Scientific, Medtronic, and Servier; and consulting fees from Bayer, Boehringer Ingelheim, Medtronic, and Novartis. Schmidt has reported serving as a consultant for and receiving speaker honoraria from Boston Scientific and St Jude Medical. Lewalter has reported serving as a speaker or a member of a speaker's bureau for Biotronik, Medtronic, and St Jude Medical/Abbott.

Follow Steve Stiles on Twitter: @SteveStiles2. For more from, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.