Does impaired renal function and other high bleeding risk factors increase the risk of adverse clinical outcomes in Asian patients with atrial fibrillation treated with edoxaban?
In Asian patients with atrial fibrillation treated with edoxaban, impaired renal function and additional risk factors such as low BMI significantly amplify the risk of both thromboembolic and bleeding events.
Abstract Introduction In patients with atrial fibrillation (AF) and impaired renal function (RF), the risks of ischaemic stroke (IS) and major bleeding (MB) are elevated. While these associations are well established, the combined impact of additional high bleeding risk factors—such as low body mass index (BMI) or advanced renal impairment among patients receiving direct oral anticoagulants (DOACs) remains insufficiently characterised. Purpose To characterise clinical event rates and outcomes at 2 years in patients with AF on edoxaban with impaired RF, including the influence of additional bleeding risk factors. Methods ETNA-AF-ASCA is a prospective, observational, non-interventional study of patients with AF from South Korea/Taiwan, Thailand, and Hong Kong receiving edoxaban at baseline (BL). RF was indexed by creatinine clearance (CrCl). Unadjusted Cox regression models were used to evaluate the association between BL bleeding risk factors and clinical outcomes at 2 years: (1) impaired vs normal RF (CrCl 60 vs ≥60 mL/min), and in patients with impaired RF, (2) moderate/low vs high bleeding risk (DOAC score), (3) underweight (BMI ≤18.5 kg/m²) vs normal weight (BMI 18.5–23.0 kg/m²), and (4) tiered CrCl (30 vs 30–60 mL/min). Hazard ratios (HRs), 95% confidence intervals (CIs), and p-values are reported. Results Of 3299 patients with AF on edoxaban at BL, 1509 (49.2%) had impaired RF (CrCl 60 ml/min) and 1561 (50.8%) had normal RF (CrCl ≥60 ml/min). Patients with impaired RF were older, more often female, had lower BMI, higher CHA2DS2-VASc scores, more comorbidities, and more likely to receive edoxaban 30 mg at BL (Table 1). Compared with normal RF, impaired RF was associated with higher risks of IS/transient ischaemic attack (TIA)/systemic embolic event (SEE; HR 95% CI; 1.79 1.09–2.93; P=0.0217), MB (1.71 1.02–2.88; P=0.0429), major gastrointestinal (GI) bleeding (2.73 1.14 - 6.54, P=0.0241), all-cause death (ACD) (3.07 [1.97–4.80; P0.0001), and worse net clinical outcome (NCO; 2.31 1.71–3.13; P0.0001; Table 2). Among patients with impaired RF, those with high DOAC score–indexed bleeding risk had significantly worse NCOs compared with moderate/low risk (3.59 1.88–6.85; P=0.0001). BMI and RF tiers were analysed in patients with impaired RF. Patients who were underweight (BMI ≤18.5 kg/m²) vs normal weight (BMI 18.5–23.0 kg/m²) were at higher risk for ACD (2.42 1.08–5.41; P=0.0312) and NCO (2.00 1.11–3.62; P=0.0217). Patients with low CrCl (30 mL/min) vs those with moderate CrCl (30–60 mL/min) had markedly higher risks of MB (2.62 1.27–5.42; P=0.0092), major GI bleeding (5.62 2.22–14.24; P=0.0003), ACD (3.96 2.47–6.34; P0.0001), cardiovascular mortality (3.50 1.20–10.23; P=0.0223), and NCO (2.78 1.90–4.06; P0.0001). Conclusion Impaired RF remains a critical risk factor for bleeding and thromboembolic events; additional bleeding risk factors amplify vulnerability.
Choi et al. (Mon,) studied this question.