Eight Things to Know About the Surgical Left Atrial Appendage Occlusion (LAAOS III) Trial

John Mandrola, MD


May 15, 2021

In the Left Atrial Appendage Occlusion (LAAOS III) trial, slightly more than 4800 patients with atrial fibrillation (AF) scheduled for open-heart surgery were randomly assigned to add-on surgical closure of the left atrial appendage or heart surgery alone.

Both groups remained on standard medical therapy, including anticoagulation. Concomitant left atrial occlusion resulted in a statistically significant reduction in ischemic stroke or systemic embolism. No safety issues were observed.

Here are eight things to know about this practice-changing trial.

1. LAAOS III Had a Simple, Elegant Design

Many trials in the cardiovascular space feature complex designs and bloated composite endpoints. Not LAAOS III; the patients were randomly assigned to undergo left atrial appendage occlusion or not. It measured one important primary outcome: stroke or systemic embolism.

2. The Question Asked Is Crucial

The LAAOS III investigators asked whether closure of the left atrial appendage at the time of heart surgery reduced stroke in addition to baseline therapy with anticoagulation. That last phrase is key to avoid misinterpreting these data. Approximately 75% of patients in both groups remained on anticoagulation at 3 years.

Some may be disappointed that the investigators did not test appendage occlusion and removal of anticoagulation, but I find it reassuring that the authors studied a more fundamental question.

A mistake in thinking about stroke prevention in patients with AF is forgetting that stroke is a systemic disease. The Virchow triad holds that stasis combines with vessel wall injury and hypercoagulability to cause thrombus. Appendage occlusion may reduce stasis, but it likely does little to reduce the endothelial dysfunction and the hypercoagulable state seen in AF.

3. The Trial Has Internal Validity

Internal validity relates to trial conduct, and it affects our confidence in the results.

LAAOS III featured concealed allocation assignment, robust blinding (the surgeons were not involved after the procedure), minimal loss to follow-up, and little chance for bias in measurement of the primary outcome. The results were also consistent across subgroups—including in patients on or off anticoagulation.  Few patients crossed over from their assigned treatment, and the concordant results of the intention-to-treat and as-treated analyses add reassurance. Its lack of industry funding removes another potential bias.

4. The Trial Has External Validity

There is a tension in clinical trials between showing that a treatment works under controlled and ideal circumstances (efficacy) and whether a treatment works under “real-world” settings (effectiveness).

Trials that select for and enroll ideal patients produce results that are harder to translate to real-world practice, where patients are often older and burdened by multiple comorbidities.

LAAOS III enrolled a very typical group of patients having heart surgery. The mean age was 71 years, and the mean CHA₂DS₂-VASc  score was 4.2. About a third of patients had permanent AF, and two-thirds of patients underwent mitral and/or aortic valve surgery.

Only a few types of heart surgery were excluded: notably, off-pump surgery, mechanical valves, and reoperations. Another important caveat is that LAAOS III did not study stand-alone surgical appendage closure, and its results should not necessarily be extended to that procedure.

That said, LAAOS III has robust external validity.

5. The Results Are Statistically Strong

Surgical left atrial appendage occlusion resulted in a 33% reduction of stroke or systemic embolism. The 95% CIs ranged from as large as a 47% reduction to as small as a 15% reduction. The P value was .001.

The fact that the upper bound of the 95% CIs include a 15% reduction in stroke is reassuring.

Also reassuring is the low P value, which means that if surgical closure of the appendage had no stroke-reducing effect, the chance of observing these results  or an even greater risk reduction would be very low. The LAAOS III result, therefore, is unlikely to be due to chance.

6. The Effect Is Clinically Important

Findings that are statistically significant do not necessarily equate to a clinically important effect. I consider the 2.2–percentage point absolute risk reduction in stroke seen in LAAOS III clinically relevant for three reasons.

First, the Kaplan-Meier curves continue to separate over follow-up, indicating greater effect size over time.

Second, the risk reduction in LAAOS III compares favorably to that seen with many other well-accepted therapeutics in cardiovascular medicine: in JUPITER, rosuvastatin reduced  major adverse cardiac outcomes by about 1.2% vs placebo; in ARISTOTLE, apixaban reduced stroke/systemic embolism by about 0.6% compared with warfarin; and in CABANA, AF ablation was not associated with a statistically significant reduction in its composite primary endpoint of death, stroke, major bleeding, or cardiac arrest compared with medical therapy.

The most compelling reason that LAAOS III is clinically important is that surgical appendage occlusion came with little burden. It added only 6 minutes to bypass time and had no significant effect on postoperative bleeding, reoperation, or incidence of heart failure hospitalization.

7. Technique Was Vital

Although appendage occlusion in LAAOS III was highly successful, achieving complete surgical occlusion requires careful attention to technique. Simply closing the appendage with a double layer of sutures or use of a purse string may leave residual leaks, which could promote thrombus.

LAAOS III strongly encouraged amputation and primary closure of the appendage. Stapler and closure devices were allowed, but neither percutaneous closure nor purse-string closure was permitted. Intraoperative transesophageal echocardiography was recommended to demonstrate successful closure.

Generalizability of procedural and surgical trials is always a concern because operators selected for inclusion in the trial are usually experienced and skilled.

Cardiologists referring patients for this procedure will want to have frank discussions with their surgical colleagues to ensure replication of the techniques used in this trial.

8. The Results of LAAOS III Do Not Apply to Percutaneous Closure

LAAOS III studied appendage closure “in addition to” standard care, including anticoagulation. Percutaneous closure trials studied an entirely different strategy: that is, after a period of time, presumably with endothelialization of the device, the goal is to stop anticoagulation. In studies comparing the Watchman implantable device against warfarin, ischemic strokes were higher in the Watchman arm; therefore, the benefit (if any) from percutaneous closure must come from reducing the burden (major bleeding) of long-term use of anticoagulation.

Another reason LAAOS III does not inform the percutaneous device–based strategy is that direct surgical closure will surely be more complete than that achieved by endocardial devices, given the highly diverse appendage anatomy. Incomplete percutaneous closure, reported in nearly a third of patients implanted with the Watchman device in the PROTECT AF trial, is a major problem because it reduces the probability of stopping anticoagulation and may increase the risk for ischemic stroke.

Perhaps the most important difference, though, is that surgical closure is epicardial and leaves no foreign body exposed to the arterial blood pool in the left atrium.

Future Question and My Conclusion

The next major question is whether surgical closure of the appendage will allow discontinuation of anticoagulation. Answering that, of course, will require another trial.

When my patients with AF undergo heart surgery, I will ask my surgical colleagues to close the appendage using one of the approaches in LAAOS III.

It will not change the anticoagulation decision, but it will reduce the risk for future stroke. And that is a good thing.

Kudos to lead investigator, Richard Whitlock, MD, and his colleagues.

John Mandrola 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

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


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.