Weekly Journal Scan: Is left atrial appendage closure really non-inferior to anticoagulation in atrial fibrillation?

This comment refers to ‘Left Atrial Appendage Closure or Anticoagulation for Atrial Fibrillation’ which was published in the New England Journal of Medicine, https://doi.org/10.1056/NEJMoa2517213. Atrial fibrillation remains a major contributor to stroke, with an untreated annual risk estimated at approximately 3%–5%.2 Oral anticoagulation has substantially reduced this burden, initially with warfarin and subsequently with NOACs, which have lowered the annual risk of stroke to approximately 1.1%–1.5% with a more favourable safety profile. Nevertheless, bleeding remains a persistent and clinically meaningful limitation of lifelong anticoagulation, with major bleeding rates of approximately 2%–3% per year. In addition, treatment discontinuation is not uncommon, reaching up to 20%–30% at 2 years.3,4 These limitations have sustained interest in nonpharmacological strategies targeting the left atrial appendage, the primary site of thrombus formation in AF.5,6 The evidence supporting LAAC has evolved progressively. Early randomized trials such as PROTECT-AF and PREVAIL demonstrated non-inferiority of device-based closure compared with warfarin, supporting regulatory approval and guideline adoption as an alternative embolic protection strategy in patients unsuitable for long-term anticoagulation.7,8 Observational registries subsequently confirmed improving procedural safety and feasibility in routine clinical practice.9,10 With the advent of NOACs, the benchmark for comparison shifted. In this contemporary setting, PRAGUE-17 demonstrated that LAAC could achieve outcomes comparable to NOACs in selected high-risk patients, with a potential reduction in bleeding over time, while OPTION extended this concept to patients undergoing AF ablation.11,12 More recently, in the older and frailer population enrolled in CLOSURE-AF (mean age, 78 years; CHA2DS2-VASc, 5.2; HAS-BLED, 3.0), LAAC failed to demonstrate non-inferiority compared with best medical therapy and was associated with a higher incidence of the primary composite end point—stroke, systemic embolism, major bleeding, or cardiovascular or unexplained death (16.8% vs 13.3%)—at a median follow-up of 3 years.13