Percutaneous left atrial appendage occlusion and risk of stroke, hospitalized bleeding and death in Medicare beneficiaries

Abstract

AbstractPurposeAmong patients with atrial fibrillation (AF), a nonpharmacologic option (e.g., percutaneous left atrial appendage occlusion [LAAO]) is needed for patients with oral anticoagulant (OAC) contraindications. Among beneficiaries in the Medicare fee‐for‐service coverage 20% sample databases (2015–18) who had AF and an elevated CHA2DS2‐VASc score, we assessed the association between percutaneous LAAO versus OAC use and risk of stroke, hospitalized bleeding, and death.MethodsPatients undergoing percutaneous LAAO were matched to up to five OAC users by sex, age, date of enrollment, index date, CHA2DS2‐VASc score, and HAS‐BLED score. Overall, 17 156 patients with AF (2905 with percutaneous LAAO) were matched (average ± SD 78 ± 6 years, 44% female). Cox proportional hazards model were used.ResultsMedian follow‐up was 10.3 months. After multivariable adjustments, no significant difference for risk of stroke or death was noted when patients with percutaneous LAAO were compared with OAC users (HRs [95% CIs]: 1.14 [0.86–1.52], 0.98 [0.86–1.10]). There was a 2.94‐fold (95% CI: 2.50–3.45) increased risk for hospitalized bleeding for percutaneous LAAO compared with OAC use. Among patients 65 to <78 years old, those undergoing percutaneous LAAO had higher risk of stroke compared with OAC users. No association was present in those ≥78 years.ConclusionIn this analysis of real‐world AF patients, percutaneous LAAO versus OAC use was associated with similar risk of death, nonsignificantly elevated risk of stroke, and an elevated risk of bleeding in the post‐procedural period. Overall, these results support results of randomized trials that percutaneous LAAO may be an alternative to OAC use for patients with contraindications.