LAACS
(2022)Objective
To examine if routine closure of the left atrial appendage in patients undergoing open-heart surgery provides long-term protection against cerebrovascular events independently of atrial fibrillation history, stroke risk, and oral anticoagulation use
Study Summary
• Protective effect was independent of pre-surgery atrial fibrillation status and stroke risk
• No increase in mortality or safety concerns with LAA closure
Intervention
Surgical left atrial appendage closure added to elective open-heart surgery
Inclusion Criteria
Patients undergoing elective first-time open-heart surgery
Study Design
Arms: LAA closure group vs standard care (open LAA)
Patients per Arm: 100 vs 86
Outcome
• Clinical stroke/TIA events: 10.0% vs 15.1% (P = 0.252)
• All-cause mortality: 23.0% vs 27.9% (P = 0.401)
Bottom Line
Adding surgical left atrial appendage closure to elective first-time open-heart surgery reduced the composite of ischemic stroke, TIA, and silent cerebral infarct over a mean 3.7-year follow-up (5.0% vs 16.3%; HR 0.3, 95% CI 0.1–0.8; p=0.02), with the protective effect independent of baseline AF status, CHA2DS2-VASc score, and OAC use, and no procedural bleeding events attributed to closure.
Major Points
- First randomized trial of routine LAACS added to elective first-time open-heart surgery, enrolling patients with and without prior AF
- ~70% relative reduction in the composite of ischemic stroke, TIA, and silent cerebral infarct (HR 0.3, 95% CI 0.1–0.8; p=0.02) in ITT analysis
- Protective effect not modified by baseline AF status, CHA2DS2-VASc score, or OAC use (all interaction p>0.4)
- Most primary events in the control arm occurred beyond the first post-operative year, supporting an AF-mediated late-stroke mechanism
- No procedural adverse events (e.g. bleeding) attributable to LAACS were recorded; no difference in all-cause mortality
- Study protocol recommended (but did not mandate) double closure with purse-string plus running suture
Study Design
- Study Type
- Prospective, randomized, open-label clinical trial (blinded MRI adjudication)
- Randomization
- Yes
- Blinding
- Open-label treatment assignment; radiologists reviewing brain MRIs were blinded to randomization
- Sample Size
- 187
- Follow-up
- Mean 3.7 ± 1.6 years (up to 6 years); total 684 patient-years
- Centers
- 2
- Countries
- Denmark
Primary Outcome
Definition: Composite of post-operative symptomatic ischemic stroke, transient ischemic attack, or imaging evidence of new silent cerebral infarct (per clinical CT/MRI or protocol MRI)
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 16.3% (14/86) | 5.0% (5/101) | 0.3 (0.1–0.8) | 0.02 |
Limitations & Criticisms
- Study halted before reaching planned randomization of 200 patients; final N=187
- Substantial cross-overs (two LAACS-randomized patients did not undergo closure due to technical issues; five control patients underwent intra-operative ablation with LAA closure)
- Mixed surgery types (CABG, valve, combined) with varying baseline risk
- Full paired MRI sets available in only 75 patients (~40%), potentially introducing selection bias
- Only 10 LAACS patients accepted post-operative TEE, so unrecognized incomplete closure cannot be excluded
- Composite primary endpoint that includes imaging-only silent cerebral infarcts; symptomatic-only sensitivity analysis lost significance (p=0.08)
- LAA closure technique not always documented in detail
- Study not powered to demonstrate a reduction in clinical stroke alone
Citation
J Cardiothorac Surg. 2018;13:53