John Mandrola, MD


June 26, 2017

A major downside for percutaneous left atrial appendage occlusion (LAAC) is that it leaves a foreign body within the left atrium. Foreign bodies in the arterial circulation raise concern over thrombus.

Two abstracts presented during the late-breaking clinical-trials sessions at the European Heart Rhythm Association (EHRA) EUROPACE-CARDIOSTIM 2017 meeting addressed this issue. Their results did not deliver reassurance.

As you read about the two studies, which were selected as featured presentations at a major medical meeting, remember that the pivotal PREVAIL-AF trial, which compared Watchman with warfarin, failed to meet either of its noninferiority efficacy end points.[1,2] And as Prof Hein Heidbuchel (University of Antwerp, Belgium) said in a debate on left atrial appendage closure here at EHRA: "Not meeting noninferiority criteria means inferior."

In PREVAIL-AF, there was a relative increase in ischemic strokes in the device arm. That could be a statistical aberration due to small numbers or it could mean that the strategy of LAAC is inferior. Two plausible reasons for increased ischemic events after LAAC are reduction in the intensity of postimplant antithrombotic therapy and thrombogenicity of the device.

The First Late-Breaking Study

Dr Laurent Fauchier (University of Tours, France) presented an observational study of LAAC performed in eight centers in France between 2012 and 2016. Of the 453 patients included, 272 had a Watchman and 183 had an Amplatzer Amulet (Abbott).

The Amulet device is a second-generation percutaneous LAAC device approved for use in Europe. Multiple colleagues from different countries have told me that the Amulet device is easier to implant than Watchman, which is why it's commonly used outside the US.

In this French study, antithrombotic regimens were left up to the operators: none were used in 8% of patients; a single antiplatelet was prescribed in 34%; dual antiplatelet therapy (DAPT) in 24%; oral anticoagulation (OAC) in 30%; and both agents in 5% of patients.

This was an older cohort (mean age 75 years); 90% of the patients had a history of major bleeding and the average CHA2DS2-VASc score was 4.4 to 4.7.

The authors reported 100 major adverse events over a mean follow-up of 11 months. The overall death rate was 7.3%; ischemic stroke occurred in 3.8% and major hemorrhage in 4%.

Transesophageal echo (TEE) within 90 days detected thrombus in 24/453 cases (5.3%)—with no statistically significant differences between the Watchman vs Amulet groups.

Older age and a previous history of ischemic stroke independently predicted thrombus. Use of DAPT associated with a markedly lower risk of having device thrombus (HR 0.06, CI 0006–0.53; P=0.018). Not surprisingly, the presence of thrombus associated with an increased risk of ischemic stroke and TIA (adjusted HR 8.20, 95%CI 1.73–38.94; P =0.008).

These results led the authors to conclude that thrombus formation on the device is "not a benign finding and should be actively investigated since it is strongly associated with a higher risk of ischemic stroke." They also suggested that DAPT might be the best antithrombotic regimen.

Second Late-Breaking Study

Dr Boris Schmidt (Cardioangiologisches Centrum Bethanien, Agaplesion Markus-Krankenhaus, Frankfurt, Germany) presented interim data from the industry-sponsored Amulet-Registry Study. The specific goal of this analysis was to evaluate the influence of postimplant antithrombotic drug therapy on the incidence of device thrombus and stroke.

The study enrolled slightly more than 1000 patients from 61 centers in Europe, Asia, South America, Australia, and the Middle East. This was also an elderly (mean age 75 years), high-risk cohort, most of whom (72%) had a prior history of major bleeding. Almost two-thirds of the group had a CHA2DS2-VASc score ≥4, and 83% of the patients were deemed ineligible for oral anticoagulation.

Major adverse events during the implant or the hospital stay included two deaths, 26 (2.4%) major bleeding events, 10 (0.9%) major vascular complications, two strokes, and one device embolization. These 41 major events in 35 patients produced an overall major complication rate of 3.8%.

More than half of the patients were discharged on DAPT, 19% on oral anticoagulation (with or without an antiplatelet), 23% on single antiplatelet therapy (mostly aspirin), and 2% got no antithrombotic therapy. Only 673 of the 1000-plus patients received a TEE within 90 days of implant.

Dr Schmidt reported 10 (1.5%) cases of device thrombus: three of these were on single antiplatelet, three on DAPT, and four on oral anticoagulation. One patient with a device-related thrombus had an ischemic stroke while taking single antiplatelet therapy.


Real-world observational data can be useful. It helps patients and doctors understand how a therapy performs outside the protective confines of a clinical trial. Most patients do not receive medical care at major referral centers.

Notably, the French study, which was not industry sponsored, revealed a 5% rate of device thrombus. That's considerable. The German registry reported a lower rate of thrombus, but nearly 400 patients in that analysis did not have a reported TEE.

The French investigators also suggest that DAPT provides significant protection against device thrombus, and the Amulet registry study did not refute this point.

If confirmed, the need for DAPT after device implant poses a problem for LAAC as a strategy. The main reason most patients are considered for LAAC is ineligibility for anticoagulation—usually because of major bleeding. Even for 3 months, DAPT is a significant amount of antithrombotic therapy to give to vulnerable patients.

The 4% rate of major procedural complications in the Amulet study also warrants attention. That's higher than was reported in the Watchman observational study—the one where device reps reported complications that occurred during the procedure.[3]

A proponent of LAAC might explain away the high rate of complications by saying it was due to multiple operators implanting the device in frail elderly patients.

But that defense adds more support for a cautious approach. Frail older patients are exactly the type of patients who are getting LAAC. And, in the US, LAAC is increasingly being performed by multiple operators of varying expertise.

The potential benefit of LAAC for stroke reduction has to be weighed against these high rates of major procedural complications and the risk of harm from drugs used to prevent thrombus. If you start with a 3% to 4% risk of major harm, future absolute risk reduction from LAAC has to be better than that. So far, it's not.[2]

These two real-world studies bolster my feeling that LAAC has to be proven superior to no therapy in anticoagulation-ineligible patients before we embrace it.


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.
Post as: