OBJECTIVE

Direct factor Xa inhibitors rivaroxaban and apixaban are efficacious alternatives to warfarin and confer a lower risk of spontaneous intracranial hemorrhage (ICH); however, they lack a validated reversal strategy. We evaluated the efficacy and safety of 4-factor prothrombin complex concentrate (PCC) administration on rivaroxaban- and apixaban-mediated coagulopathy in patients with traumatic and spontaneous ICH.

METHODS

Retrospective review of patients presenting with traumatic and spontaneous ICH and concurrent use of rivaroxaban or apixaban. Demographic factors, reason for anticoagulation, hemorrhage type and location, Glasgow coma scale score, and when appropriate, ICH score, were included. Patient charts were reviewed for in-hospital mortality, thromboembolic events, pulmonary complications, worsening of hemorrhage, hemorrhagic complications after neurosurgical intervention, and 90-day modified Rankin scale score.

RESULTS

Eighteen patients met inclusion criteria; 16 used rivaroxaban and 2 used apixaban. Eight patients presented with traumatic ICH, 8 with hemorrhagic stroke, 1 with subarachnoid hemorrhage, and 1 patient with tumoral hemorrhage. Mean Glasgow coma scale score was 12.6 (range, 6-15) and mean ICH score was 2.3 (range, 0-4). After reversal with PCC, 1 patient (5.6%) demonstrated worsening of ICH on follow-up head computed tomography. PCCs were administered before emergent placement of an external ventricular drain in 1 individual, with no hemorrhagic complications. Six patients (33.3%) experienced in-hospital mortality: family withdrew care in 4 and 2 died due to pneumonia. There was 1 (5.6%) thromboembolic complication. Favorable outcomes at 90 days were seen in 6 patients (33.3%).

CONCLUSIONS

Despite no studies demonstrating the efficacy of 4-factor PCC administration for reversal of coagulopathy in patients on direct factor Xa inhibitors, our early experience demonstrates it to be safe, yet potentially reducing hemorrhagic complications and hematoma expansion in this critically ill population.