2026 年 57 巻 4 号 p. 50-56
Background: Apixaban, a direct oral anticoagulant (DOAC), is widely used for stroke prevention in patients with nonvalvular atrial fibrillation (AF). Its dose is adjusted based on age, body weight and renal function; however, even when patients meet criteria for the standard dose, physicians may prescribe a reduced dose at their discretion (underdosing). Previous studies have reported comparable risk of stroke/systemic embolism between patients receiving the standard-dose (5 mg twice daily) and those receiving an under-dose (2.5 mg twice daily), whereas bleeding risk was lower in the latter. Conversely, another report suggested that under-dosing of DOACs increases thromboembolism risk without reducing bleeding risk. Therefore, this study aimed to evaluate whether thromboembolic risk differs between standard-dose and under-dose apixaban, and to explore the influence of concomitant medications and patient characteristics on these outcomes.
Methods: Medical records of 973 patients who received apixaban were retrospectively reviewed. Among the patients, 644 received standard-dose and 329 received under-dose apixaban. The follow-up period was defined as the interval from the date of apixaban initiation to occurrence of a thromboembolic event, such as cerebral infarction or systemic embolism. For patients without events, follow-up ended on the last day of apixaban administration. Thromboembolic events occurred in 85 patients (8.7%). Covariates included in multivariable Cox proportional hazards models were selected based on their clinical relevance. To avoid overlapping adjustment and improve interpretability, we constructed two models; Model 1 included the CHADS2 and HAS-BLED scores as composite variables, whereas Model 2 included the individual components of these scores instead of the composite scores.
Results: In multivariate analysis, patients who received β-blockers had a significantly lower risk of thromboembolic events in Models 1 and 2. In contrast, no difference in thromboembolic events was observed between the standard-dose and under-dose groups.
Conclusion: Concomitant β-blocker use was independently associated with lower thromboembolic risk in apixaban-treated patients with AF. No significant difference was observed between standard-dose and under-dose apixaban; however, larger multicenter studies are needed to address limited events and residual confounding.
