LAAOS III

Published

Whitlock RP, et al. "Left atrial appendage occlusion during cardiac surgery to prevent stroke". The New England Journal of Medicine. 2021. 384(22):2081-2091.
PubMed • Full text • PDF • ClinicalTrials.gov

Clinical Question

In patients with atrial fibrillation undergoing cardiac surgery, does left atrial appendage occlusion reduce the risk of stroke compared to no procedure?

Bottom Line

Among patients with AF undergoing cardiac surgery, concomitant left atrial appendage occlusion at the time of cardiac surgery was found to reduce the risk of stroke compared to no left atrial appendage occlusion at the time of cardiac surgery.

Major Points

Atrial fibrillation places patients at risk of significant morbidity and mortality from stroke, with a key component of the care of patients with atrial fibrillation being stroke risk reduction. This is predominantly accomplished using anticoagulation to prevent thrombus formation. Studies in patients with non-rheumatic atrial fibrillation have found that the left atrial appendage accounts for approximately 90% of atrial thrombus location (it accounts for 60% among those with valvular AF, however). Therefore there has been interest in occlusion of the left atrial appendage to prevent thrombi formation and stroke.

The Left Atrial Appendage Occlusion Study (LAAOS III) was a blinded, randomized trial of surgical occlusion of the left atrial appendage at the time of another indicated cardiac surgery. The study found that left atrial appendage closure reduced the risk of ischemic stroke or systemic embolism by 2.2% absolute risk reduction during the almost four year follow-up period. There was no significant increase in operating time. Notably most patients remained on anticoagulation during the trial.

Overall, the findings support concomitant surgical left atrial appendage closure during another indicated cardiac surgery to reduce the incidence of ischemic stroke in patients with atrial fibrillation. Given that most patients remained on anticoagulation, the trial does not support left atrial appendage occlusion as an alternative to anticoagulation, unlike the PROTECT AF trial (2009) and the OPTION trial (2025), which studied the WATCHMAN device in comparison to anticoagulation.

Guidelines

AHA/ACC-led AF guidelines (2023, adapted)

• If AF and elevated CHA2DS2-VASc and undergoing cardiac surgery, surgical LAA exclusion with ongoing anticoagulation will lower stroke and systemic embolism risk (COR 1, LOE A)
  • Among above, the benefit of LAA exclusion without ongoing anticoagulation is of unclear benefit for stroke and systemic embolism risk (COR 2b, LOE A)

Design

• Multicenter, randomized, controlled trial
• N=4811
  • LAAOS (n=2400)
  • No LAAO surgery (n=2411)
• Setting: 105 centers in 27 countries
• Enrollment: 2012-2018
• Mean follow-up: 3.8 years
• Analysis: Intention-to-treat
• Primary Outcome: First occurrence of ischemic stroke or non cerebral systemic embolism during follow-up.

Population

Inclusion Criteria

• Age 18 years or older
• Planned for surgery with cardiopulmonary bypass
• History of atrial fibrillation and CHA2DS2-VASc of at least 2

Exclusion Criteria

• Mechanical-valve implantation
• Heart transplantation
• Surgery for complex congenital heart disease
• Isolated implantation of LVAD
• Prior surgery requiring opening the pericardium
• Previous implantation of left atrial appendage closure device

Baseline Characteristics

From the LAAO arm.

• Demographics: Age 71 years, 68% male
• AF type: Permanent 29%, persistent 24%, paroxysmal 47%
• Medical problems: Prior MI 24%, prior stroke 9%, rheumatic heart disease 7%, PAD 10%, HF 57%, DM 32%, aortic plaque 10%, current or former smoker 47%, HTN 82%
• CHA2DS2-VASc score: 4
• LVEF <50%: 31%
• Surgery: Isolated CABG 20%, isolated valve replacement 23%, any aorta procedure 6% other 56%
  • Valve replacement type: Mitral 36%, aortic 35%, tricuspid 17%, pulmonic <1%
• Concomitant surgical ablation of AF: 34%
• Occlusion attempted: 90%
  • Approach: Cut and sew 56%, stapler 11%, closure device 15%, closure from within 14%, other 4%

Interventions

• Randomized to undergo or not undergo left atrial appendage occlusion at the time of cardiac surgery.

  • Preferred technique was amputation and closure, but several techniques were allowed.
• Stratified by trial site
• Surgeons performing procedure aware of trial assignment, but participants, clinicians, and other personnel were not.

Outcomes

Comparisons are LAAO vs No LAAO.

Primary Outcomes

Components of the primary outcome are listed here for simplicity'

Ischemic stroke or systemic embolism

4.8% vs 7.0% (HR 0.67; 95% CI 0.53-0.85; P=0.001)

Ischemic stroke:4.6% vs 6.9% (HR 0.66; 95% CI 0.52-0.84)

Systemic embolism: 0.3 vs 0.3% (HR 0.86; 95% CI 0.29-2.55)

Secondary Outcomes

Stroke or systemic embolism

5.3% vs. 7.8% (HR 0.67; 95% CI 0.54-0.84)

Stroke: 4.7% vs 7.4% (HR 0.63; 95% CI 0.50-0.80)

All-cause mortality

22.6% vs 22.5% (HR 1.00; 95% CI 0.89-1.13)

HF hospitalization

7.7% vs. 6.8% (HR 1.13; 0.92-1.40)

Major bleed

10.4% vs. 11.2% (HR 0.93; 95% CI 0.78-1.11)

MI

2.1% vs. 2.3% (HR 0.87; 95% CI 0.59-1.28)

Ischemic stroke or systemic embolism

≤30 days after surgery: 2.2% vs. 2.7% (HR 0.82; 95% CI 0.57-1.18)

>30 days after surgery: 2.7% vs. 4.6% (HR 0.58; 95% CI 0.42-0.80)

Subgroup Analysis

The primary outcome was similar by age, sex, surgery type, or prior atrial fibrillation ablationsubgroups.

Additional Analyses

Bypass time (min)

119 +/- 48 vs 113 +/- 47 (difference 5; 95% CI 3-8)

Cross-clamp time (min)

86 +/- 37 vs 82 +/- 37 (difference 4; 95% CI 1-6)

Criticisms

• Unclear how this intervention might fare without long-term anticoagulation.
• Surgeons were not blinded, but it's not feasible to do so.
• Limited duration of follow-up.

Funding

• Canadian Institutes of Health Research
• Canadian Stroke Prevention Intervention Network
• Hamilton Health Sciences Research Institute
• Heart and Stroke Foundation of Canada
• Request for Applications Program-Research Strategic Initiatives
• Canadian Network and Centre for Trials Internationally
• McMaster University Surgical Associates