Lack of Complete Seal with Watchman Device Common, But No Cause for Alarm

Download this article's Factoid in PDF (& PPT for Gold Subscribers)

Leakage into the left atrial appendage (LAA) due to incomplete closure with the Watchman device occurs in about one-third of patients out to 12 months. However, these cases do not cause any increase in thromboembolic events, according to a PROTECT AF subanalysis published in the March 6, 2012, issue of the Journal of the American College of Cardiology.

For the PROTECT AF trial, researchers randomized 707 patients with nonvalvular atrial fibrillation in a 2:1 ratio to percutaneous closure of the LAA using the Watchman (Atritech, Plymouth, MN) device with the aim of discontinuing warfarin after 45 days (n = 463) or proceeding on warfarin treatment alone (n = 244). While the Watchman proved noninferior compared with warfarin with regards to efficacy, safety events occurred more frequently in the intervention group due primarily to periprocedural complications.

In the subanalysis, researchers led by Vivek Y. Reddy, MD, of the Mount Sinai School of Medicine (New York, NY), looked at transesophageal echocardiography (TEE) results to assess the incidence and consequences of peri-device blood flow due to incomplete LAA closure in 445 patients from PROTECT AF who had successful Watchman implantation. The concern arises from the fact that the occlusion device is round while the LAA ostium is elliptical, potentially resulting in an incomplete seal of the orifice. This, in turn, could lead to enhanced formation of thrombi, which could potentially travel past the device into the circulation.

Residual Leakage Common

The incidence of any blood flow around the device was 40.9% at 45 days, decreasing to 33.8% at 6 months and 32.1% at 12 months (P = 0.001). Of the patients with any residual flow, most had moderate (59.9%; 1-3 mm) or major (32.4%; > 3 mm) peri-device leakage, with relatively few (7.7%) showing minor (< 1 mm) blood flow around the device. These levels were maintained out to 12 months (P = 0.275). At 45 days, the mean and maximum width of the leak measured 2.8 mm and 6.2 mm, respectively. These values also did not change appreciably at 6 months (2.9 mm and 6.8 mm, respectively) or 12 months (2.9 mm and 6.0 mm, respectively; P = 0.731).

There was no difference in the primary endpoint (stroke, systemic embolism, or CV or unexplained death) in patients with and without any peri-device blood flow (2.0% vs. 2.8%; P = 0.635). There was also no significant association with the primary endpoint for each 1 mm increase in leakage size (HR 0.84; 95% CI 0.62-1.14; P = 0.256). In addition, neither peri-device leakage nor warfarin status showed any influence on risk of clinical events (table 1).

Table 1. Primary Event Rates According to Peri-Device Blood Flow and Warfarin Statusa




(95% CI)

No Leak, Discontinued Warfarin (Reference)



No Leak, Continued Warfarin




Any Leak, Discontinued Warfarin




Any Leak, Continued Warfarin




a P = 0.857 across all groups.

Dr. Reddy and colleagues conclude that there is a “lack of interaction between residual flow around the LAA closure device and clinical outcome, including thromboembolic events,” adding that “continued anticoagulation therapy did not significantly decrease the primary event rate compared to that of patients who discontinued anticoagulation.”

While the results should be interpreted with caution, the authors say, “[t]hese data suggest the safety of warfarin discontinuation after Watchman implantation regardless of the presence of the residual peri-device flow, provided that the size of the leak is ≤ 5 mm.” In PROTECT AF, LAA closure was deemed successful if the residual gap was at or below that threshold.

In a telephone interview with TCTMD, Ziyad M. Hijazi, MD, MPH, of Rush University Medical Center (Chicago, IL), agreed that clinicians “should feel okay to discontinue warfarin after 45 days even in the presence of this small, residual leak.”

Peri-Device Flow Rates Surprising

Still, Dr. Hijazi was surprised at the high incidence of peri-device flow. “I did not realize there was that much residual leakage around the device, and that shows you the inherent problem with device design,” he said. “The device may not fit all shapes and forms of left atrial appendages, so the leakage rate is higher than you would think.”

Fortunately, though, “in this paper is was reassuring that even if you don’t close the appendage 100%, the likelihood that you will have increased events is very low,” Dr. Hijazi added.

He pointed out that the leakage rates do beg the question as to the ideal device design for LAA closure. “I don’t know if you will have any device that will achieve 100% complete closure, but maybe you will find one with a better closure rate,” Dr. Hijazi said. “But that’s the name of the game with devices. You will have more companies getting into this business because every device will have some drawbacks.”

An Indirect Message?

While there are only 2 LAA closure devices with any appreciable data—Watchman and Amplatzer (St. Jude Medical, St. Paul, MN)—other methods are being explored, he noted. But in the meantime, the current study should suffice for clinicians worried about potential complications of incomplete closure with the Watchman.

“Once and if and when the device receives [US Food and Drug Administration (FDA)] approval, then translation into real-world experience will determine if we can duplicate these results or whether they will be different,” Dr. Hijazi said, adding that perhaps the subanalysis can serve as an “indirect message” to the FDA, which has been reviewing Watchman data since an FDA advisory panel narrowly recommended approval of the device in April 2009.

“Appendage closure is exciting and important because there are so many patients who don’t tolerate anticoagulation,” Dr. Hijazi said. “We need something approved as long as it’s safe and effective and eliminates the flow into the appendage. With these data, if the width of the jet is less than 5 mm, then the message to patients as well as operators is: don’t worry about it.”


Viles-Gonzales JF, Kar S, Douglas P, et al. The clinical impact of incomplete left atrial appendage closure with the Watchman device in patients with atrial fibrillation. A PROTECT AF (Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation) substudy. J Am Coll Cardiol. 2012;59:923-929.



Related Stories:

  • Dr. Reddy reports receiving clinical research grant support and consultant fees from Atritech.
  • Dr. Hijazi reports no relevant conflicts of interest.

We Recommend