Evaluating stroke-prevention strategies in patients with atrial fibrillation by age

The EuroPCR Course Directors have selected 3 major late-breaking trials that will be presented for the first time during EuroPCR 2026 and are set to impact practice worldwide. Among them is a subgroup analysis of the CHAMPION-AF trial investigating stroke-prevention strategies in patients with atrial fibrillation by age.

Background 

For patients with atrial fibrillation (AF), the use of oral anticoagulant therapy to prevent stroke is limited by the risk of bleeding. Left atrial appendage closure (LAAC) is considered for patients who are unsuitable candidates for long-term anticoagulation, but its role in patients who are eligible for anticoagulants has not been established. Earlier this year, results were published from the CHAMPION-AF trial comparing device-based LAAC with direct oral anticoagulant (DOAC) therapy among patients with AF who were candidates for anticoagulation. The primary efficacy endpoint of death from cardiovascular causes, stroke or systemic embolism at 3 years was noninferior for LAAC vs. DOAC therapy, while LAAC was superior to DOAC for the primary safety endpoint of non-procedure-related bleeding at 3 years. 

Jens Erik Nielsen-Kudsk presented results from a prespecified sub analysis of the CHAMPION-AF trial, which evaluated the efficacy and safety of LAAC vs. DOAC by age.

Results 

• The analysis included data from 1,915 patients aged <75 years and 1,085 patients aged ≥75 years. 
• The incidence of the primary efficacy endpoint of death from cardiovascular causes, stroke or systemic embolism at 3 years was similar for LAAC vs. DOAC in patients aged <75 years (hazard ratio [HR] 1.07; 95% confidence interval [CI] 0.67–1.71; log rank p=0.7789) and ≥75 years (HR 1.34; 95% CI 0.85–2.12; log rank p=0.2036; p interaction=0.5014). 
• Significantly lower incidence of non-procedural major and clinically relevant non-major bleeding was observed with LAAC vs. DOAC in patients aged <75 years (HR 0.64; 95% CI 0.50–0.82; log rank p<0.0001) and ≥75 years (HR 0.68; 95% CI 0.51–0.91; log rank p=0.0002; p interaction=0.7281). 
• In the intention-to-treat analysis, the incidence of ischaemic stroke was higher in LAAC vs. DOAC, but this difference was no longer significant among patients who received their assigned therapy as intended. 
• Rates of disabling ischaemic stroke were very low and similar across treatment arms and age groups. 

Key learnings 

• These results indicate that the efficacy and safety of LAAC vs. DOAC are not affected by age.

Conclusions and PCR recommendations 

These findings suggest that age alone should not preclude the use of LAAC in otherwise suitable candidates. It should be noted that caution is needed when interpreting subgroup analyses as they are associated with certain limitations including reduced power compared with the overall analysis. The choice between LAAC and DOAC should be made on an individualised basis in a shared decision-making process with patients.