Arrhythmias & electrophysiology
Causal Machine Learning for Left Atrial Appendage Occlusion in Patients With Atrial Fibrillation
Authors' conclusion
Novel machine learning algorithms were developed to identify patients who are more likely to benefit from LAAO vs DOACs. This information can support clinical decision-making to determine which patients should be referred to subspecialists for further examination and discussion of LAAO.
Study context
- Population
- Adult patients with atrial fibrillation treated with either left atrial appendage occlusion (LAAO) or direct oral anticoagulant (DOAC) between March 13, 2015, and December 31, 2019identified, matched cohort of 28,930 patients · matched cohort 28,930
- Participants
- 744,190 patients
- Comparison
- LAAO vs a direct oral anticoagulant
- Outcome
- Primary composite outcome of:
- Ischemic stroke
- Systemic embolism
- Major bleeding
- All-cause mortality
- Follow-up
- 1 year and 2 years
Methodological detailsOne-to-one propensity score matching, causal forest model, OptumLabs Data Warehouse
- One-to-one propensity score matching
- Causal forest model
- OptumLabs Data Warehouse
Main finding
As reported in the abstract
Average treatment effect
−0.68%
CI −1.4% to 0.06%
at 1 year; no difference
Average treatment effect
−2.9%
CI −3.7% to −2.0%
at the end of 2 years; lower risk with LAAO
Further findings
Baseline characteristics of the matched cohort
- Mean age
- 76.8 ± 6.3 years
- Female
- 5,818 patients (40%)
- Mean CHA2DS2-VASc score
- 5.8
Classification of treatment effect of LAAO · At the end of 2 years
- Potentially benefiting from LAAO
- 30.1%
- Neutral
- 69.7%
- Potentially harmed by LAAO
- 1.4%
- Of the overall cohort
Original publication
Causal Machine Learning for Left Atrial Appendage Occlusion in Patients With Atrial Fibrillation. JACC Clin Electrophysiol. Published 12 February 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →