Leaks After LAAO Do Matter, Experts Agree

NEW YORK, NY—Last week, a session billed as a debate on the importance of peridevice leaks following left atrial appendage occlusion (LAAO) at New York Valves 2024 left one thing clear: everyone agreed that leaks are a problem and need to be addressed moving forward.

Just how big a leak needs to be to pose problems and the best way to manage residual flow, however, remain unclear.

Thomas Waggoner, DO (Tucson Medical Center, AZ), was selected to argue the position that peridevice leaks are “much ado about nothing,” but he came to the opposite conclusion. “Leaks do matter,” he said, adding that the size of the leak appears to be relevant.

Indeed, Robert Sommer, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), his opponent, said, “There’s accumulating evidence that peridevice leaks are actually important,” he said. “What the cutoff is for what becomes unimportant is completely unknown. I think 3 mm is fairly reasonably well supported as a clinically important cutoff.”

He noted that the historical definition of a relevant leak—larger than 5 mm—came from the design of the PROTECT AF trial involving the first-generation Watchman 2.5 device (Boston Scientific). “Frankly, somebody made it up,” he said. “There was no basis in clinical medicine.”

Waggoner, addressing that initial threshold, said, “We got it wrong . . . in my opinion.”

Leaks Tied to Worse Clinical Outcomes

Waggoner kicked off the discussion by highlighting data from the PINNACLE FLX trial showing that LAAO with the Watchman FLX device (Boston Scientific) performed well, with effective closure at 1 year in 100% of patients who had the device successfully implanted. That was accompanied by a low rate of primary safety events (0.5%).

But, even though LAAO has emerged as an alternative to long-term oral anticoagulation for the prevention of stroke and systemic embolism in patients with atrial fibrillation, residual flow around the device has continued to be an issue. That can be related to several issues, according to Waggoner, including morphological mismatch between the device and the LAA anatomy, suboptimal device sizing or positioning, incomplete endothelialization, device migration, the contractile nature of the appendage, and anatomic remodeling of the LAA after the procedure.

Whatever the cause, residual flow around the device has been associated with poorer outcomes. Waggoner pointed to an analysis of more than 51,000 patients from the National Cardiovascular Data Registry (NCDR) LAAO Registry showing that the risk of stroke, TIA, or systemic embolism was modestly but significantly greater among patients with leaks smaller than 5 mm present on 45-day imaging than among those without leaks.

A subsequent study affirmed the relevance of leaks, demonstrating that patients with residual flow from the appendage detected a year after the procedure had a significantly higher rate of ischemic stroke or systemic embolism through 5 years of follow-up; the result was driven by an increase in nondisabling strokes and not by more serious endpoints including disabling/fatal stroke and cardiovascular or all-cause death.

Sommer pointed that even though initial analyses from PROTECT-AF and other early LAAO studies supported the idea that residual flow around the appendage was not raising the risk of subsequent thromboembolic events, there are certain issues that complicate interpretation of those data. He cited the difficulty of measuring peridevice leaks in three dimensions, the evolution of leaks over time, and the infrequency of clinical neurologic events, which limits the statistical power of research investigating the issue.

More recent studies support a link between peridevice leaks and adverse clinical outcomes, Sommer said, highlighting the studies described by Waggoner as well as others, including an analysis published this past April in Circulation: Cardiovascular Interventions.

It’s known that thrombi form spontaneously in the LAA and that after closure of the appendage, thrombi will form behind the device, Sommer said. “So common sense clearly tells us that any residual communication from the appendage to the left atrium could potentially allow thromboembolization.”

What to Do About Leaks

Though most people would agree that residual leaks matter after LAAO, there remains uncertainty about what to do about it, Sommer said. “We know that the risks of a recurrent stroke or TIA are small, and therefore the idea of putting these patients back on oral anticoagulants or going in and doing a secondary procedure where you’re going to be mucking around in an appendage that is full of clot, I think has substantial risks,” he said. “Therefore, we have to judge each patient’s risks individually and be very careful and have a very high bar about going in to close these leaks.”

During a panel discussion, Vivek Reddy, MD (Icahn School of Medicine at Mount Sinai, New York, NY), said he has some ambivalence about the issue of leaks. “I think the major cause of post-appendage closure strokes is going to be leaks. That’s what the data seems to suggest,” he said. “On the other hand, these tend to be small strokes.”

Taking all of that into consideration, he said, “I tend to close the leaks, but it would be nice if we had some data. The reality is that closing is not a harmless procedure. You can have complications—pericardial effusions, etc.”

He indicated that the issue of leaks and how to respond to them will require additional research moving forward. “I think leaks are going to be around for quite some time,” Reddy said. “Watchman FLX did not solve the problem with leaks, and maybe some of the new devices will, but it would be an area to study.”

Waggoner stressed the need to obtain a good result during the index procedure. “The majority of these are very, very elective cases, so you want to have the best result. You don’t want to leave the lab without getting the best seal possible,” he said, alluding to the possibility of trying different catheters, devices, and techniques to achieve it.

The issue hasn’t been resolved, but advancements in closure devices and delivery systems appear to be moving the field in the right direction, Waggoner said, citing higher rates of complete closure in studies of the newer Watchman FLX device than in studies of older devices. “So we’re getting there,” he said.