Background and purpose: intracerebral hemorrhage associated with oral anticoagulants has a poor prognosis. Current treatment guidelines are based on case series and plausibility only, and a common consensus on effective hemostatic therapy is missing. We compared the effectiveness of diverse hemostatic approaches in a mouse model of warfarin-associated intracerebral hemorrhage.
Methods: male C57BL/6 mice received anticoagulant treatment with warfarin (0.4 mg/kg for 3 days). Intracerebral hemorrhage was induced by striatal injection of collagenase, and 30 minutes later, mice received an intravenous injection of saline (200 μL n=15), prothrombin complex concentrate (100 U/kg, n=10), fresh-frozen plasma (200 μL, n=13), recombinant human Factor VII activated (3.5 mg/kg, n=8 and 10 mg/kg, n=8), or tranhexamic acid (400 mg/kg, n=12). Intracerebral hemorrhage volume was quantified on T2-weighted images after 24 hours.
Results: mean hematoma volumes were 7.4 ± 1.8 mm(3) in the nonwarfarin controls and 21.9 ± 5.0 mm(3) in the warfarin group receiving saline. Prothrombin complex concentrate (7.5 ± 2.3 mm(3)) and fresh-frozen plasma (8.7 ± 2.1) treatment resulted in significantly smaller hematoma volume compared with saline. Recombinant human Factor VII activated (10 mg/kg: 14.7 ± 3.4; 3.5 mg/kg: 15.0 ± 6.8 mm(3)) and tranexamic acid (16.2 ± 4.1 mm(3)) were less effective. Water content in the hemorrhagic hemisphere was similar in all groups except for tranexamic acid in which it was significantly increased.
Conclusions: prothrombin complex concentrate and fresh-frozen plasma effectively prevent hematoma growth in murine warfarin-associated intracerebral hemorrhage, whereas Factor VIIa was less effective. Tranexamic acid exacerbates perihematoma edema in this mouse warfarin-associated intracerebral hemorrhage model.