Intracranial hemorrhage in patients with atrial fibrillation receiving anticoagulation with warfarin or edoxaban: An in-depth analysis from the ENGAGE AF-TIMI 48 randomized trial.

Abstract Intracranial hemorrhage (ICH) is a known risk of oral anticoagulation; delineating ICH attributes may provide nuanced guidance regarding atrial fibrillation management. We evaluated ICH characteristics and outcomes from Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation-Thrombolysis in Myocardial Infarction 48 (ENGAGE AF-TIMI 48), a randomized trial that compared two edoxaban regimens (higher-dose edoxaban regimen 60/30 mg (HDER), lower-dose edoxaban regimen 30/15 mg (LDER)) with warfarin in patients with atrial fibrillation. Patients who suffered ICH vs those who did not were compared and independent predictors of ICH were calculated. We also assessed ICH subtype and etiology. Of 21,105 randomized patients, 322 (1.53%) had ≥ 1 ICH for a total of 368 events. Intraparenchymal hemorrhage (HDER: HR 0.52 [95% CI 0.35–0.77], LDER: HR 0.22 [0.13–0.38]) and subdural hematoma (HDER: HR 0.29 [0.15–0.55], LDER: HR 0.26 [0.13–0.50]) were lower with both HDER and LDER vs warfarin. Subarachnoid hemorrhage frequency was similar in the HDER vs warfarin groups but lower in LDER. Compared to warfarin, edoxaban was associated with lower risk of spontaneous ICH (HDER: HR 0.47 [0.31–0.69], LDER: HR 0.34 [0.22–0.53]) and traumatic ICH (HDER: HR 0.32 [0.17–0.61], LDER: HR 0.31 [0.16–0.59]). In multivariable analysis, randomization to warfarin, increased age, and risk of falling remained independent predictors of ICH. In ENGAGE AF-TIMI 48, ICH was decreased in edoxaban-treated patients compared to warfarin-treated patients, including ICH of both spontaneous and traumatic causes. Both edoxaban regimens lowered intraparenchymal and subdural hemorrhages compared to warfarin. Patient characteristics and medical history may help guide anticoagulation management.