ESC 2012: Left atrial appendage closure is NOT the therapy of choice for many patients with AF?

Dr John Mandrola


August 27, 2012

One of the more interesting types of sessions at large meetings is the pro-contra debate. This one on left atrial appendage (LAA) occlusion devices did not disappoint.

As these devices near consideration for approval, I felt like it is a relevant topic.

Protagonist case:

German professor Dr Thomas Lewalter presented the protagonist view. The concept of LAA occlusion centers on the fact that most embolic strokes in AF originate from the left atrial appendage. The idea then is simple: take the appendage out of the picture. Two devices, the Watchman and Amplatzer Cardiac Plug , are closest to approval. After a period of weeks to months the device endothelializes, and anticoagulation is stopped. It's an appealing concept, because it trades a mechanical solution for lifelong anticoagulation.

The Watchman device was studied in a prospective randomized trial called PROTECT-AF , which was published in the Lancet in 2009. In this 700-patient trial, the efficacy of percutaneous closure of the LAA was noninferior to that of warfarin therapy. Device-treated patients had a higher rate of adverse safety events, which were mainly a result of periprocedural complications.

Dr Lewalter also presented recent data on the Amplatzer device. He pointed out that complications were lower and correlated with the learning curve. This update included zero strokes in the device arm. He told us that LAA occlusion is going on in 62 German centers and that he expected this device to evolve in the same way that PCI did decades ago.

Of course, the strongest of his arguments was not so much the strength of the procedure but rather the downsides of anticoagulation. That long-term anticoagulation has significant risks requires no further words.

In short, the protagonist case distills to the noninferiority of a risky procedure studied in a very small number of patients combined with the known downsides of chronic anticoagulation.

Antagonist case:

Noted anticoagulation expert Dr Stuart Connolly presented the much stronger case against current-day LAA occlusion devices.

He began on a positive note—saying that he thought the idea of closing the left atrial appendage is a worthy one. This would be the end of his concessions.

His strong and impressively presented case involved four focus points:

1. The left atrial appendage is only part of the stroke story

There are many other potential trouble sources of emboli, including the atrial septum, left-sided valves, carotids, and aorta. What's more, most of the risk factors for stroke in patients with AF are systemic—age, hypertension, and diabetes, for example. He made this interesting point: left atrial enlargement has never been shown to predict stroke. In short, since stroke risk in AF is a systemic—not anatomic—problem, occlusion of the LAA would likely always be an incomplete strategy.

2. Critique of the Watchman trial

He highlighted five major problems with the trial:

  • The number of patients studied was tenfold less than the anticoagulant trials.

  • Noninferiority trials are suspect because the researchers can set the noninferiority threshold anywhere they wish.

  • Patients treated with the device still had to be on warfarin in the periprocedure period. This limits the procedure to patients who can tolerate warfarin at least transiently.

  • LAA closure is supposed to work by preventing ischemic stroke, yet the benefit in the intervention arm was driven by a reduction in hemorrhagic strokes.

  • There were only 21 events in the trial. This led to wide confidence intervals.


3. Safety issues

This is a delicate and tricky procedure. The Watchman trial confirmed that adverse events were twice as likely in the device arm. These are not mild complications—stroke, significant pericardial tamponade, and device embolization. Proponents argue that adverse events are learning curve dependent. Dr Connolly reminded us that recent follow-up data showing lesser complications might be related to the fact that less-experienced centers were kicked out of the trial. Obviously, this is the opposite of what will happen if the device is approved.

(A personal sidebar: At Boston AF symposium this year, I watched a live case of LAA appendage closure from Italy. The audience gasped when a large air embolism appeared in the left atrium. Thankfully for the Italian patient on the table, the operators were able to aspirate it. And these were expert operators.)

4. A better way to reduce the risk of bleeding already exists

Dr Connolly points to the novel anticoagulants, which are backed by randomized clinical trials in more than 50 000 patients. Dabigatran, rivaroxaban, and apixaban all have favorable efficacy and safety profiles compared with warfarin. A universal finding of these trials was that novel anticoagulants reduced the risk of the most serious kind of bleeding—intracranial hemorrhage. The data with these drugs represent orders-of-magnitude greater evidence than that with LAA closure.

My take-home:

There are many factors driving the enthusiasm for LAA occlusion. These include the attractiveness of the concept, the dislike of anticoagulation, and that proceduralists like doing procedures. But yet, it's clear to me that this procedure is not ready for prime time. At this moment, it's too dangerous, too skill dependent, and the evidence for benefit is weak. As one who makes a living moving catheters in the left atrium, I can vouch for the fragility of the left atrial appendage. It's the part of the LA that worries me most.

Ending on a positive note, I was struck by Dr Connolly's concluding proposal: He suggested the possibility of using these devices in combination with anticoagulation. This strategy would get around the issue that stroke risk in AF parallels systemic factors more than anatomic ones.

Put me down in Dr Connolly's camp.



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.