LAAO Registry: Safety of Watchman Looks Good in First Years of Rollout

In the first 3 years of the LAAO Registry, the Watchman left atrial appendage (LAA) closure device (Boston Scientific) was implanted with rates of in-hospital safety outcomes that were comparable to those seen in the pivotal trials, despite being used in higher-risk patients, new data show.

Any major complication in the hospital was seen in 2.16% of cases. The most common were pericardial effusion requiring intervention (1.39%) and major bleeding (1.25%). There were low rates of cardiac arrest (0.24%), death (0.19%), stroke/TIA (0.17%), major vascular complications (0.15%), device embolizations (0.07%), and MI (0.04%).

“Both PROTECT AF and PREVAIL reported higher rates of pericardial effusion, major bleeding, procedure-related stroke, and device embolization compared to what we’re reporting, albeit over a slightly longer time horizon [7 days],” James Freeman, MD (Yale University School of Medicine, New Haven, CT), reported during the virtual American College of Cardiology (ACC) 2020 Scientific Session.

The data, published simultaneously online in the Journal of the American College of Cardiology, are “reassuring and, I think, hopefully will address some of the concerns from the regulatory bodies and those writing the guidelines around the safety of the procedure that particularly come up in the context of PROTECT AF and that are raised as a procedure is rolled out to more centers around the country,” Freeman said.

The LAAO Registry

After being subjected to intense scrutiny that included three appearances before a US Food and Drug Administration advisory committee, the Watchman device was approved in March 2015 for the reduction of thromboembolism in high-risk patients with nonvalvular A-fib. That decision was supported by data from the PROTECT AF and PREVAIL trials, with a patient-level meta-analysis of 5-year data from both trials showing that Watchman provided comparable stroke protection compared with warfarin and lower rates of hemorrhagic stroke and major nonprocedural bleeding.

The LAAO Registry, set up through a collaboration between the ACC, the Society for Cardiovascular Angiography and Interventions, the FDA, the Centers for Medicare and Medicaid Services (CMS), and Boston Scientific and part of the National Cardiovascular Data Registry (NCDR), started enrolling patients in January 2016. Participation in the registry is required for CMS reimbursement of Watchman implantation and supports the formal postmarket surveillance study mandated by the FDA. Freeman said it is estimated that more than 95% of Watchman procedures performed the United States are included in the registry.

He presented data from the first 3 years of the registry, 2016 to 2018. During that span, 38,158 patients (mean age 76; 41% women) were implanted with a Watchman by 1,318 physicians working in 495 hospitals, representing “by far the largest experience published to date.”

On average, Freeman said, patients in the LAAO Registry were much older and sicker—with a heavier burden of comorbidities—than those treated in the randomized trials. The mean CHA2DS2-VASc score was 4.6, higher than in PROTECT AF (3.4) and PREVAIL (3.8), and the mean HAS-BLED score was 3.0, higher than in the European EWOLUTION registry (2.3).

Also, most patients (69.4%) had prior clinically relevant bleeding. “So the vast majority of the patients included in the registry did have some relative contraindication to anticoagulation in accordance with the guidance from CMS,” Freeman said. The national coverage determination for percutaneous LAA closure from CMS states that patients should be considered suitable for short-term warfarin “but deemed unable to take long-term oral anticoagulation following the conclusion of shared decision-making.”

Median annual case volumes were relatively low for both hospitals (30) and physicians (12), although there was substantial variability at both levels.

The device was deployed in 93% of attempted procedures—3% of implantations were cancelled prior to venous access and 4% were aborted after access but before device deployment. Roughly half of the cancelled procedures were due to detection of LAA thrombus on the day of the procedure. Moreover, major adverse events were more frequent among patients with cancelled or aborted procedures, “suggesting that they were cancelled or aborted because of complications,” Freeman said.

Acute procedural success, defined as the rate of success among procedures in which a device was deployed was 98.3%, which compares favorably with prior Watchman studies, Freeman said. In 70 cases (0.2%), there was a residual leak of 5 mm or more.

Freeman reported that major in-hospital adverse event rates in the LAAO Registry tended to be lower than in the pivotal trials but were “roughly comparable” to 7-day rates seen in EWOLUTION. Procedure-related adverse event rates at 1 day, however, were lower in EWOLUTION, he said.

‘Encouraging’ Results

Discussing the study after Freeman’s presentation, Mark Estes, MD (UPMC Heart and Vascular Institute, Pittsburgh, PA), said, “I think one of things that we need to be mindful of with this closure device, the Watchman, is that it’s been through a very robust validation procedure,” referring both to the pivotal trials and the long regulatory process that culminated in approval.

Requirements for training, patient selection, operator performance, and registry participation “were really key to getting the kinds of outcomes that have been shown here, which are improved compared to the clinical trials,” Estes said, noting, however, that these are only acute results.

He pointed out that practice guidelines say the Watchman and other LAA occlusion devices can be considered in patients with A-fib who have an increased stroke risk and contraindications to long-term anticoagulation, making it a class IIb indication, before asking Freeman whether these new data might influence changes to clinical practice.

Freeman responded that the primary finding of this analysis “is that the patient populations overall were really in accordance with the CMS guidance, which is that we’re not seeing a lot of patients who are getting this device for lifestyle indications or whatnot. Most of these patients are really stuck between a rock and a hard place with a high risk for ischemic stroke and a high risk for bleeding. . . . And so the indications seem to be quite strong.”

When moving beyond clinical trials and rolling an intervention out to a large number of centers, one worries about a higher risk of complications, Freeman said, but in this study, “the rates of adverse events were quite reassuring.”

But there’s more work to be done, he acknowledged, pointing to the need for longer-term follow-up, an assessment of the relationship between procedural volumes and outcomes, and studies of optimal periprocedural antithrombotic strategies.

In an editorial accompanying the JACC paper, Dhanunjaya Lakkireddy, MD (Kansas City Heart Rhythm Institute & Research Foundation, Overland Park, KS), and Mohit Turagam, MD (Icahn School of Medicine at Mount Sinai, New York, NY), say “these short-term results from the largest Watchman postapproval LAAO registry are encouraging and align with the safety data reported previously in other prospective registries. It is actually very encouraging to see that the rate of complications did not peak, as is typically anticipated in postmarket release studies.”

They note that several other devices targeting the LAA are being studied. “Although registry data is not a substitute for RCTs, this well-designed prospective database with collaboration between various professional societies, the FDA, CMS, and industry will serve as a useful vehicle in refining patient selection and tracking the safety and efficacy of LAAO devices for the next several decades,” they write.