Effectiveness of high- versus low-dose four-factor prothrombin complex concentrate on hemostasis and thromboembolism in patients with oral factor Xa inhibitor-associated intracranial hemorrhage

Abstract

Abstract Background Four-factor prothrombin complex concentrate (4F-PCC) is frequently used off-label for the reversal of oral factor Xa inhibitor-associated bleeding. Dosing of 4F-PCC in routine practice remains inconsistent as published observational studies and clinical guidelines have provided conflicting results. Purpose We evaluated the incidence of hemostatic effectiveness and thromboembolism with high- versus low-dose 4F-PCC in a cohort of patients with oral factor Xa inhibitor-associated intracranial hemorrhage (ICH). Methods We identified patients treated with 4F-PCC during a hospitalization for apixaban- or rivaroxaban-associated ICH between December 2016 and August 2020. Patients were included if they were ≥18 years old, had a radiographically confirmed ICH, and received their last dose of apixaban or rivaroxaban <24 hours prior. Patients with a hematoma volume >60 mL, a Glasgow Coma Scale score <7, or planned surgery within 12 hours were excluded. Patients were divided into cohorts based on 4F-PCC dose administered: high- (≥35 IU/kg) or low-dose (<35 IU/kg). Our primary outcome was hemostatic effectiveness, defined as ≤35% increase in hematoma size from index to repeat scan at 12±5 hours after administration of 4F-PCC. If no repeat scan was available within that timeframe, the worst scan within 24 hours was used. If no repeat scan was available within 24 hours, hemostatic effectiveness was adjudicated based upon clinical findings by two trained investigators. Our secondary outcome was development of thromboembolism within 5 days of receiving 4F-PCC. To adjust for potential differences in baseline characteristics between 4F-PCC dosing groups, propensity score-overlap weighted logistic regression was used. Results are reported as adjusted odds ratios (ORs) with 95% confidence intervals (CIs). Results Ninety-five patients, 24 in the high- and 71 in the low-dose group, were included. Mean ± standard deviation (SD) age of the total study cohort was 80.3 ± 9.1 years, time from initial imaging to start of reversal was 2.4 ± 2.1 hours, and time from end of reversal to repeat scan was 6.8 ± 4.6 hours. Intracerebral ± intraventricular bleeds were most common (46.6%), with 7.4% of patients having infratentorial involvement, and 13.5% having bleeds ≥10 mL/≥10 mm in size. Most ICHs were single compartment (91.8%) and were of traumatic origin (68.7%). Following propensity score-overlap weighted logistic regression, no difference in hemostatic effectiveness was observed between patients receiving high- (80.3%) or low-dose (77.2%) 4F-PCC (OR 1.20; 95%CI 0.28 to 5.15). No cases of thromboembolism were observed in either 4F-PCC dosing arm. Conclusion High-dose (≥35 IU/kg) 4F-PCC was not associated with increased odds of achieving hemostatic effectiveness compared to the low-dose (<35 IU/kg) in patients experiencing apixaban- or rivaroxaban-associated ICH, with no incidence of thrombotic events.