Coagulation abnormalities following primary intracerebral hemorrhage.

Systemic hemostatic activation following primary intracerebral hemorrhage (PICH) has been described, particularly with intraventricular or subarachnoid extension. Our objective was to study the occurrence of abnormalities of coagulation as measured by partial thromboplastin time, international normalized ratio, and platelet count in patients with PICH and no obvious cause for a pre-existing coagulopathy. Charts of PICH patients admitted between November 1991 and December 2001 were reviewed. We excluded patients with an underlying lesion, cranial trauma, anticoagulation, liver failure or sepsis. All patients had partial thromboplastin time, international normalized ratio, and platelet count measured on admission. An international normalized ratio > 1.4, partial thromboplastin time > 35, and platelet count < 100,000 were considered abnormal based on standardized values for our laboratory. All patients underwent a computed tomography (CT) scan on admission. Repeat CT was obtained for evidence of neurological deterioration. One hundred ninety-two patients with intracerebral hemorrhage were studied. Thirty-seven were excluded because of a possible underlying cause for a pre-existing coagulopathy. Thirteen of one hundred and fifty-five (8.4%) patients were found to have a coagulopathy based on our criteria. Three of thirteen (23%) patients with coagulopathy versus 3/142 (2%) without suffered neurological deterioration with evidence of hematoma enlargement (P = .008). Eleven of sixty-seven (17%) patients with intraventricular/subarachnoid extension versus 2/88 (2%) without had a coagulopathy (P = .002). Eight of thirteen (61%) patients with coagulopathy versus 29/142 (20%) without were dead at 30 days (P = .003). Coagulation abnormalities without an obvious etiology that may be consistent with low grade disseminated intravascular coagulation are seen in 8.4% of patients with PICH and are associated with extension into the subarachnoid and intraventricular compartments, neurological deterioration with hematoma expansion, and mortality at 30 days. This may represent a target for therapeutic intervention.