Percutaneous Left Atrial Appendage Occlusion: ACC Data Challenge Promise vs Reality

John M. Mandrola, MD


April 04, 2022

The American College of Cardiology (ACC) 2022 Scientific Session featured a handful of studies on percutaneous left atrial appendage occlusion. The news was concerning.

One theme involved the challenge of actually occluding the appendage with these devices; the other theme was the types of patients who are getting this supposedly preventive procedure in the US.

The Challenge of Percutaneous Appendage Occlusion

Mohamad Alkhouli, MD, from the Mayo Clinic, Rochester, Minnesota, presented a study looking at the clinical impact of residual leaks following appendage closure with Watchman (Boston Scientific). JACC-EP simultaneously published the study.

One of the practical challenges with appendage occlusion is that human left atrial appendages vary so much in size and shape. A partial occlusion can create an even greater nidus for thrombus. Indeed, a recent report found that leaks were associated with a near doubling of risk of an embolic events at one-year.

Alkhouli and colleagues queried the mandated NCDR-LAAO US registry and identified more than 55,000 patients who had Watchman implanted over a 3-year period.

Based on transesophageal echocardiograms at 45 days, they identified three groups: those with no leak, small leaks (< 5 mm), or large leaks (≥ 5 mm).

The first and perhaps most remarkable finding was that 1 in 4 patients (n = 13,258) had a small leak. Large leaks were less common at 0.7%.

They then reported anti-thrombotic drug use. (Post-device regimens can include both antiplatelet and anticoagulant drugs.) Remarkably, even in patients with no leaks, 13% of patients remain on either an anticoagulant or P2Y12 inhibitor drug at 1 year. In patients with small leaks, 15% of patients were taking one of these agents.

The third part of their paper reported the association of embolic and bleeding events with leaks. Using the no leak group as the reference, patients with small leaks had a statistically significant 15% higher rate of stroke, transient ischemic attack, or systemic embolism, and an 11% higher rate of major bleeding.

Rates of ischemic and bleeding events were also higher in patients with large leaks but because of smaller numbers, the statistical threshold for significance was not met.

Who Is Getting This Procedure:

Two posters at ACC used Medicare data to shed light on which patients are getting this procedure in the US.

Allen Wang, MD, from Beth Israel Deaconess Medical Center in Boston presented a poster looking at frailty in patients undergoing left atrial appendage closure.

They searched a Medicare database and found more than 21,000 patients who had the procedure over a 3-year period. They deemed the patients low, intermediate, or high-risk based on a validated frailty score:

They reported three main findings:

  • Nearly half the patients who had appendage closure were considered frail, including 16% who met criteria for high-risk frailty.

  • Mortality rates were 16.1% for the low-risk group, 27.6% for the intermediate-risk group, and 41.1% for the high-risk group (P < .001).

  • After adjusting for comorbidities, high vs low frailty was associated with eightfold higher risk of a long hospital stay, fivefold higher 30-day mortality and threefold higher 1-year mortality.

A second poster from the University of Florida and Penn State groups (I was a co-author) compared baseline comorbidities of US patients who are having left atrial appendage closure with those who were enrolled in the regulatory trials, PROTECT-AF and PREVAIL.

We used a Medicare claims database to find patients who had the procedure between 2015 and 2018. We then compared this random sample of 2780 patients to those enrolled in the pivotal trials.

The main results were that Medicare-insured patients who have had appendage closure were significantly older, had more hypertension, congestive heart failure, diabetes, and a higher-rate of prior bleeding than those enrolled in the pivotal trials. The CHA2DS2-VASc score was also higher: 4.8 ± 1.4 vs 3.4 ± 1.5 and 3.8 ± 1.2.


The National Cardiovascular Data Registry (NCDR) data tells us that 1 in 4 patients implanted with a Watchman had a peri-device leak. This is an existential problem because patients with leaks are more likely to continue anti-thrombotic drugs and leaks associate with a higher rate of ischemic and bleeding events. The selling point of this 'preventive' procedure is to remove the need for anti-thrombotic drugs while also reducing ischemic and bleeding events.

Leaks related to percutaneous appendage closure have become prominent enough to have led to a new branch of cardiology: percutaneous closure of leaks with other devices, detachable coils and even radiofrequency energy. Think about that for a moment.

Proponents of percutaneous appendage closure will counter with the fact that with greater experience and newer iterations of the device, there will be fewer peri-device leaks. They may cite the PINNACLE FLX study that reported a 10% peri-device leak rate at 12 months.The problem is that this study was done in only 29 centers and included only 400 patients. Let's see what the registry shows when this next-generation Watchman device is unleashed on low-volume operators in many more patients.  

The two posters describing who is getting these devices frightens me more than the device leaks.

Here is why: older patients with more comorbid conditions and frailty will have greater competing causes of stroke. Occlusion of the appendage can only prevent one type of stroke — that from appendage thrombus. But in older, sicker patients, there are many more causes of stroke, such as large and small vessel atherosclerosis.

I don't believe that the pivotal trials show convincing data. But even if you took a rosy view of that data, it stretches credibility to say it justifies implanting these devices in older patients with many comorbidities.

The other problem with doing this procedure in such patients is the upfront risk.

Again, for the sake of argument, let's say there is a future benefit (stroke or bleeding decrease) from this procedure. US patients still have to accept a major complication rate of 5%-10%. Frail, older, and female patients are at even greater risk of complications — further decreasing the chance for a net benefit.


As I have said before, the answer to the percutaneous left atrial appendage debate is painfully simple: randomize.

Consider that patients with this much frailty and comorbidities will have high rates of events. That makes for a short study.

Instead, we will continue doing this invasive and quite risky procedure not knowing whether it helps people.

It's one thing to give people a vitamin that doesn't work; it's another to put a foreign body in them. I remain worried that our field is making a big mistake with this approach to stroke prevention.

John Mandrola, MD, practices cardiac electrophysiology in Louisville, Kentucky and is a writer and podcaster for Medscape. He espouses a conservative approach to medical practice. He participates in clinical research and writes often about the state of medical evidence. 

Follow John Mandrola on Twitter

Follow | Medscape Cardiology on Twitter


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.