WATCHMAN Updates Point to Better Safety, Enduring Efficacy With LAA Closure Device for Stroke

Shelley Wood

September 23, 2010

September 23, 2010 (Washington, DC)— New data on the Watchman left atrial appendage (LAA) closure device for stroke prevention in atrial-fibrillation patients suggest that, as more time passes postimplantation, the device continues to prevent strokes and deaths. Meanwhile, the safety issues that first raised flags for the procedure--pericardial effusion, device retrievals, and different types of bleeding--are, indeed, declining over time.

Dr David Holmes

Presenting extended follow-up results for the PROTECT AF trial here at TCT 2010, principal investigator for the study, Dr David Holmes (Mayo Clinic, Rochester, MN), stressed that much of the data collected over the past year should help assuage the FDA's concerns about the device. As reported by heartwire , PROTECT AF was first presented in March 2009, followed by an FDA advisory committee meeting one month later, where members voted 7 to 5 to recommend approval of the device. The agency, however, subsequently asked the manufacturer, in March 2010, to conduct a new study to provide better safety and efficacy data.

What's With the Watchman?

Warfarin has been the "cornerstone" therapy for the more than three million US patients with atrial fibrillation, who have a fivefold increased risk of stroke. As much as 90% of the thrombus accumulation leading to thromboemboli and stroke is believed to originate in the LAA, leading to the development of percutaneous therapies for occluding the LAA, ideally allowing the patient to go off warfarin therapy. The Watchman occluder is placed distal to the ostium of the LAA to occlude flow and prevent the migration of thrombus forming in the appendage.

Continued-Access Registry

Dr Saibal Kar

Among the new studies and updates provided here during a session devoted to the Watchman, Dr Saibal Kar (Cedars Sinai Medical Center, Los Angeles, CA) presented outcomes from the single-arm continued-access registry, known as CAP, which ran from the time PROTECT AF reached its target enrollment until the FDA advisory meeting. In all, 567 patients from 26 sites were enrolled, using the same criteria as the original trial. Of interest, Kar noted, the CHADS2 score for patients was higher in CAP, while procedure time was reduced, implant success was greater, and the proportion of patients who discontinued warfarin at 45 days was significantly increased.

Baseline and Procedural Variables: PROTECT AF vs CAP

Mean CHADS2 score 2.2 2.4 <0.001
Procedure time (min) 62 50 <0.001
Implant success (%) 89.5 95.0 0.001
45-day warfarin discontinuation (%) 86.6 94.9 <0.001

Even more important, Kar said, safety findings were improved in CAP, possibly reflecting greater operator skill with the device.

Major Safety End Points: PROTECT AF vs CAP

Safety end point PROTECT AF (%) CAP (%) p
All procedure/device-related events at 7 d 7.7 3.7 0.007
Serious pericardial effusions at 7 d 5.0 2.2 0.019
Procedure-related stroke 0.9 0.0 0.039

PROTECT AF: Ongoing Follow-up

Also during the session, Holmes presented 1500-patient-year follow-up data from PROTECT AF, building on the original 600-patient-year follow-up presented at the ACC 2009 meeting and the 1050-patient-year follow-up later presented at HRS 2009.

As Holmes showed, the primary efficacy results in the intention-to-treat population met the noninferiority criteria for the trial and now reflect a 29% lower relative risk reduction in the Watchman-treated patients, with 1500 patient-years of follow-up (HR 0.71, down from 0.76). Stroke rate has also continued to decline (from 0.96 to 0.77), a 23% lower relative risk. Importantly, Holmes noted, the relative risk reduction for the primary composite efficacy end point for the study (stroke, cardiac death, systemic embolization) was seen in both CHADS2 1 patients and patients with scores higher than 1 and for both patients under the age of 73 or 73 and over.

More important to physicians and regulators for the Watchman device is the device safety. Holmes showed that, as with the efficacy findings, device safety has also improved over time, with hazard ratios for the composite end point of major bleeding, pericardial effusion, and device embolization dropping as time goes on, from 2.85 at the original 600-patient-year follow-up to 1.53 at 1500 patient-years.

Addressing the FDA's concerns about pericardial infusions, Holmes cited an analysis now in press by his coinvestigator, Dr Vivek Reddy (Mount Sinai Medical Center, New York, NY), indicating that pericardial infusions for the most part occurred very early on and that rates tend to go down with more operator experience. "We're better than we used to be," Holmes said. "Training has been very important in this regard."

Commenting on the findings for heartwire , Dr Josep Rodés-Cabau (Quebec Heart and Lung Institute, QC) called the findings "really encouraging."

Rodés-Cabau said that he was particularly impressed by the reduction in ischemic strokes. "They didn't unfortunately present much data on this, but they did say that most of the ischemic strokes in the device group appeared in the first seven days. This is very important. . . . Everybody expected a significant reduction in hemorrhagic strokes [as a result of reduced warfarin usage], but the other side is, what about the cardioembolic ischemic strokes?"

What they saw, he continued, was that the rate of ischemic strokes over follow-up was "pretty low. Meaning, as they said, the concept seems to be right--that the thrombus formation in the left atrium is mostly in the LAA and that closing this will significantly reduce ischemic strokes."

One question that Rodés-Cabau believes still needs to be answered is whether the device can be used in people who can't take warfarin at all: in PROTECT AF, even patients receiving the device received several weeks of warfarin. In Canada, he says, patients unable to take warfarin are permitted to get the Watchman through a compassionate-use process. In these cases, he says, they give clopidogrel plus aspirin or sometimes just aspirin alone. But he worries that the higher risk of thrombus in the first week or so after device implantation is a hurdle that needs to be overcome and the question of which anticoagulant(s) should be used "likely needs a randomized trial."

In fact, investigators in Europe are addressing this issue in the single-arm, multicenter ASAPstudy, testing the use of the Watchman device in patients who cannot take warfarin at all and are thus getting six months of clopidogrel plus lifelong aspirin instead. While the study is still ongoing, Kar showed preliminary results for the 64 patients followed to date, noting there has been just one serious pericardial effusion and no strokes or transient ischemic attacks.

Finally, audience members were also given a glimpse of the new pivotal trial mandated by the FDA. Dubbed PREVAIL, the trial will randomize 475 patients, 2:1, at multiple US sites and will include a similar patient population as PROTECT AF. The primary end point of the study is a composite of hemorrhagic stroke, ischemic stroke, systemic embolism, and CV/unexplained death. Addressing several of the concerns raised during the advisory panel hearing, the trial will focus on enrolling patients with a CHADS2 score of 2 or higher or CHADS2 1 patients with existing heart disease. Patients already taking clopidogrel will be excluded from the trial.

If all goes according to plan, the last follow-up visit will be in September 2013, with the data submitted to the FDA one month later.

A surgical device for LAA occlusion, known as the AtriClip, has been FDA approved, but no percutaneous devices hold market approval in the US. However, the Watchman is just one of several devices already approved in Europe for LAA closure for stroke prevention.

Holmes disclosed that both he and the Mayo Clinic have a financial interest in technology related to this research, that the technology has been licensed to Atritech, and that Mayo Clinic and Holmes have contractual rights to receive future royalties from this license, although none have been received so far. Kar disclosed receiving grant/research support from Abbott, Atritech, AGA, St Jude, Circulite, and Coherex; consulting fees or honoraria from Abbott, AGA, Atritech, and Gore; and having "other financial benefits" from Coherex. Rodés-Cabau disclosed receiving research/grant support from Boston Scientific and consulting for Cordis, St Jude Medical, Edwards Lifesciences, and AGA Medical.


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.