Dhanunjaya R. Lakkireddy, MD, FACC, FHRS

SWISS-APERO was in line with what we found in the Amulet IDE study. The obvious advantage of this study is that the researchers tested the Watchman FLX device, as a criticism of Amulet IDE was that Watchman 2.5 was used, and not many people use it in clinical practice anymore. The argument was that single-mechanism FLX closure device is an improved version of the 2.5, and is much better in its closure rates.

That argument is partly correct, but mechanistically, SWISS-APERO showed that whether it is Watchman FLX or Watchman 2.5, a single-mechanism closure device still has inherent limitations. For immediate closure post-procedure, Watchman had an 11% leak rate vs. 3% for Amulet. The Amulet’s disc design gives a distinct advantage of compensating for unusual anatomies such as proximal lobes that a single-mechanism closure device cannot compensate for, no matter how good it is. We see those anatomies in 25% to 30% of cases, and for those, the outer disc of the Amulet provides flexibility and freedom that a single-mechanism closure device will never give.

In the 45-day follow-up data, there were 27% leaks in the Watchman arm compared with 13% in the Amulet arm. That again is exactly in line with what Amulet IDE has shown. In the 12-month data of Amulet IDE, 22% of the Watchman arm had clinically significant leaks, defined as more than 3 mm, compared with 9% of the Amulet arm.

In terms of efficacy, the study was too short to determine stroke benefit, but I was not surprised at the device-related thrombosis data, which was in line with Amulet IDE and prior observational studies. With the Amulet’s disc design, the opportunity for forming a clot on a smoother surface is much lower, and perhaps the long-term endothelialization process may be much better in a dual-mechanism device with a disc rather than a single-mechanism closure device that creates nooks and crannies where there is an opportunity for thrombus to form.

The complication rates were interesting: 9% in Amulet vs. 2.7% in Watchman, mostly related to bleeding. This needs more clarification in order to make a judgment on it. A lot of the operators in the trial were novices with the Amulet. We had a similar issue in Amulet IDE, with operators who had less than six cases; when they sent people home on novel oral anticoagulants, they had a much higher risk for complications. That happens because when you are a novel operator, you do not yet have the best ability to judge the size of the device and to position it properly. If you are redeploying the device multiple times, that could cause microperforations that can result in delayed leaks and pericardial effusions.

Also interesting about this study was that it allowed crossovers. Most of those happened when an operator deployed a Watchman device and did not like it, and proceeded to close the LAA with an Amulet device; this also opens up opportunities for microperforations.

Of note, the definition used for bleeding was not standard. It was basically all-comers bleeding. I wish a better definition of a severe bleeding complication was used. That is a reason to take the 9% complication rate with Amulet with a grain of salt. I have asked the principal investigator to do a deeper dive on that so we can get a better sense of what that rate actually means.

The trial provided evidence to show that whether Watchman 2.5 or Watchman FLX, a single-mechanism closure device comes with inherent issues due to the nature of the design. By contrast, a dual-mechanism closure device like Amulet compensates for those shortcomings and is able to provide much better closure options for difficult anatomies. As operators become more facile with the devices, it will be interesting to see if they will prefer one for all cases or one for certain kinds of anatomies and another for others. Each operator will have to decide that on their own.

It is good to have two choices. I like the way the field is evolving.