New Data Increase Caution on Left Atrial Appendage Occlusion

I remain concerned about the irrational exuberance among some of my colleagues toward left atrial appendage occlusion devices for the prevention of stroke in patients with atrial fibrillation.

In short, these devices are plugs that doctors place into the left atrial appendage. The idea is to stop clots from forming or escaping from the left atrial appendage. It was a good theory because the appendage can be a place that harbors clots.

One of the barriers to overcome with these plugs is prevention of clots on the device. The plug sits in the left atrium exposed to the low flow of blood in that chamber.

Foreign bodies exposed to blood leads to clotting; it’s exactly why metal stents used in coronary arteries render the person dependent on potent clot-blocking drugs. The appendage plug, too, needs drugs to prevent clots from forming on it.

Here’s the problem: many of the patients getting these devices are getting them because they have had major bleeding events while taking anticoagulants like warfarin or NOACs.They are deemed ineligible to take drugs that block the body’s clotting system.

But anti-platelet drugs, e.g. aspirin, can cause bleeding–especially in the elderly. See the BAFTA trial.(Ref below).

Two studies chosen as featured presentations at the European Heart Rhythm Meeting did not deliver any reassurance of these devices.

Both studies looked at real-world data on how the device performs when used by regular doctors outside the clinical trial. Real-world observational studies are important because a therapy (drug or device) may not deliver the same outcomes when used outside the protective confines of a trial.

A French study took data from numerous centers in an effort to study the presence of clots on the device, and the influence of anti-clotting drugs to prevent these clots.

A German study, sponsored by industry, also looked at the presence of clots as well as overall complications from the procedure. This study took data from more than 61 centers across the world.

The details of the studies are included in my column: Real-World Data on Left Atrial Appendage Closure Does Not Reassure

The gist of these studies was that clots on the device are not rare; potent clot-protecting drugs are likely required to prevent clots (at least for a period of weeks-months); the presence of clots increase the risk of stroke, and finally, major procedural complications are in the range of 4%.

These findings bolster my already cautions approach to this procedure.

Remember, left atrial appendage closure is a preventive procedure. Its benefit is a probabilistic one in the future. If you start with a 3-4% major complication rate, future risk reductions have to be higher than that to deliver a net gain. Thus far, the data from the randomized controlled trials don’t show that it is.

Why my colleagues are speeding ahead with this risky procedure is hard for me to understand. The conclusion of my analysis of these two most recent studies is that we need a head-to-head comparison of the device versus no therapy in patients who cannot take clot-blocking drugs. My prediction is that the device will be no better or worse than no therapy.



Mant J, Hobbs FDR, Fletcher K, et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. The Lancet.370:493-503.

(Note, I do not use the term blood-thinner. Anticoagulants (warfarin and NOACs) and anti-platelet (aspirin and clopidogrel) drugs do not thin the blood; they inhibit parts of the normal clotting system.)


  1. Dr Mandrola ,
    Does this caution include the type of LAA occlusion procedure that is done during a typical CABG where a clip is placed on the outside rather than a device placed on the inside

    1. This is a different question. There is much less data on surgical closure. Surgical closure can be done in many different ways. One of the problems with surgical closure occurs when the surgeon gets partial closure; this can increase the chance for clots.

      One of the advantages of surgical closure is that there is no foreign body left in the body.

      Finally, it’s not clear that appendage closure in any way results in better outcomes, because the appendage is but one of only many means for stroke. The older and sicker a patient gets, the more likely are non-appendage causes of stroke–carotid disease, intra-cerebral vascular disease etc.

  2. Dr. John,

    First off, THANK YOU for all the empowering information. I am a 40yo who had my 1st episode of AF 4 days after a Marathon PR 3:02 and qualifying for Boston. I was cardioverted 7 days after onset because my vagal tone was good enough that my rate never increased above 104bpm. I have now been in NSR for 3 months and perform a ECG every week at work.

    TEE on cardioversion day showed no clots, no defects, no valve or structural problems.

    I want to continue to run although I don’t need to run fast again. Is continuing to do marathons a good idea? What are the chances it was a single event and will never happen again since I am young. I have anxiety about doing it again since it happened after a hard race. I know you don’t have a crystal ball but the recommendations of my local Cardiologist seemed skewed because the lack of knowledge on the subject and I know you are an athlete and have AF yourself.

Comments are closed.