Small Leaks After LAAO Linked to Thromboembolism, Bleeding

Patrice Wendling

April 06, 2022

Small leaks after left atrial appendage occlusion (LAAO) with the Watchman 2.5 device were common and associated with more thromboembolic and bleeding events, new research shows.

One in four patients (25.8%) in a national registry had a small leak of 5 mm or less detected on imaging 45 days after the procedure.

At 1-year, their risk-adjusted rates of stroke, transient ischemic attack (TIA), or systemic embolism were 15% higher than for patients with no leaks (hazard ratio [HR], 1.15; 95% CI, 1.03 - 1.29).

The absolute incidence was low but remained significantly higher in patients with small leaks than in those with large leaks (3.5% vs 3.0%), Mohamad Alkhouli, MD, reported at the American College of Cardiology (ACC) 2022 Scientific Session.

Leaks larger than 5 mm were not associated with an excess risk for thromboembolic complications, likely because the group was small (0.7%) and the majority were on anticoagulants, as per current practice, he said.

Surprisingly, patients with small leaks also had an 11% higher rate of major bleeding than those with no leaks (HR, 1.11; 95% CI, 1.03 - 1.20), even though they had similar rates of anticoagulant use, noted Alkhouli, Mayo Clinic, Rochester, Minnesota.

The risk for any major adverse event (death, cardiac arrest, stroke, TIA, systemic embolism, major bleeding, major vascular complications, myocardial infarction, pericardial effusion requiring intervention, or device embolization) was also higher with small than with no leaks (HR, 1.10; 95% CI, 1.05 - 1.16).

The all-cause death rate at 1 year was numerically higher with large leaks, but not significantly different across the three groups.

The study, published simultaneously online in JACC: Clinical Electrophysiology, is the largest thus far to investigate residual leaks after LAAO.

It involved 51,333 patients in the National Cardiovascular Data Registry (NCDR) LAAO registry implanted with the Watchman 2.5 device from 2016 to 2019 who had a transesophageal echocardiogram at 45 days and were followed for a year.

At baseline, patients with large leaks were significantly more likely than those with small or no leaks to have nonparoxysmal atrial fibrillation (53.8% vs 48.2% vs 43.3%), cardiomyopathy (24.0% vs 22.1% vs 19.8%), and a larger LAA orifice diameter (23.7 vs 22.3 vs 21.1 mm). Procedural and in-hospital complications were similar across groups.

Is LAA Closure a "Misnomer?"

Alkhouli opened the presentation with the provocative question of whether the name LAA occlusion is a "misnomer," pointing to rates of residual flow at 45 days ranging from 10% to 50% across studies, depending on the device or imaging modality used. "It does decrease a little bit over time but remains not infrequent."

There is also no consensus as to what defines a large leak, with some studies using a 3 mm threshold and others 5 mm. "But we know that the average diameter of the middle cerebral artery is 3 mm, so that is a side note for you to decide what is important," he said.

As to whether residual leaks matter clinically, data are very limited, observed Alkhouli. Leaks were not associated with thromboembolic events in substudies from PROTECT AF and the Amulet Observational Study, although both were underpowered, with only 16 and seven thromboembolic events, respectively.

A 2022 Vanderbilt Registry study, however, showed that leaks larger than 3 mm were associated with a combined endpoint that included stroke and TIA, he said. A study of more than 1000 patients presented at the American Heart Association 2021 meeting by Vivek Reddy, MD, showed leaks smaller than 5 mm at 1 year, but not 45 days, were associated with a twofold increase in ischemic stroke/systemic embolism.

"The risk is higher. We have two different datasets that are consistent, so I tend to believe it," said Reddy, Icahn School of Medicine at Mount Sinai, New York City, who served as the formal discussant for the study.

He noted that his study found a nonsignificant annual increase in thromboembolic events of about 0.5% in patients with a leak, similar to what was observed in the NCDR registry. Leaks that persist out to 1 year are also known to double the subsequent risk.

"On the other hand, the absolute risk is only about 15%, and some of the leaks seen at 45 days will close by 1 year," Reddy said. "We want to prevent leaks, but what do we do with our patients who have leaks in an early time point?"

Alkhouli replied that these leaks are the "nagging" clinical question, but highlighted that the present data reflect previous practice with a first-generation device. Operators today have more experience and tools, like CT planning and steerable sheaths to facilitate implantation, all of which reduce leaks. The 1-year residual leak rate, for example, is only 10% with the newer Watchman Flex device.

"This data are a mix: reassuring and concerning a bit," he added. "The absolute risk is small, yet it is a preventative procedure, so we would like to get it as good as we can."

The results, Alkhouli cautioned, should not be used to support closing small leaks in all patients using coiling because there are limited data to suggest it works in the long-term.

"I think we should study this more with the current practice, with the newer devices, with the current techniques…, and then randomize these patients with small leaks to maybe a longer-term anticoagulation and then examine them at a later time to see if those leaks close or are still associated with events," he said.

"I think that’s a very important point," agreed Reddy. "Coiling or whatever has its own risk."

In addition to the use of a first-generation device, the study was limited by variations in imaging and peri-device leak size measurements, a lack of data on interventional leak management, and follow-up limited to a year, Alkhouli acknowledged.

The study was funded by Boston Scientific, which had no role in the study design or interpretation of the data. Alkhouli is on the advisory board of Boston Scientific. Reddy reports relationships with numerous companies, including consultant fees/honoraria from Boston Scientific.

J Am Coll Cardiol EP. Published online April 3, 2022. Abstract

Follow Patrice Wendling on Twitter: @pwendl For more from | Medscape Cardiology, follow us on Twitter and Facebook.


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.