Asymptomatic hemorrhagic transformation of cerebral infarction does not worsen long-term outcome.

Hemorrhagic transformation of ischemic stroke occurs relatively frequently. Many patients with hemorrhagic infarction on brain imaging do not show obvious signs of clinical deterioration, and such intracerebral hemorrhage is, therefore, designated asymptomatic. As a post hoc analysis, we explored whether such asymptomatic intracerebral hemorrhage may exert possible subtle adverse effects that might lead to a worse long-term outcome. Patients had an ischemic stroke and were enrolled in the National Institute of Neurological Disorders and Stroke tissue plasminogen activator (t-PA) stroke trial (n = 624). Computed tomography brain scan was performed 24 hours, 7 to 10 days, and 3 months after stroke, and at any time when clinical deterioration was observed. A neuroradiologist blinded to clinical information evaluated each scan for the presence of hemorrhage. Symptomatic intracerebral hemorrhage was defined as a computed tomographically documented hemorrhage that was temporally and causally related to deterioration of the patient's clinical condition in the judgment of the clinical investigator. Asymptomatic intracerebral hemorrhage was defined as computed tomographically documented hemorrhage that was not associated with deterioration in the patient's neurologic condition in the judgment of the clinical investigator. Three-month favorable outcome was assessed using the National Institutes of Health Stroke Scale, Barthel Index, and modified Rankin Scale. Wilcoxon and Fisher's exact tests were performed to determine whether asymptomatic intracerebral hemorrhage had an effect on outcome at 3 months. In all, 21 (3.4%) patients had asymptomatic intracerebral hemorrhage (13 in the t-PA group and 8 in the placebo group, P = .16). There was no statistical interaction between treatment and asymptomatic intracerebral hemorrhage on 3-month favorable outcome (P = .31). Initial analysis suggested that asymptomatic intracerebral hemorrhage decreased the odds of a favorable outcome. When controlling for computed tomographic lesion volume, however, a difference in the odds of a favorable outcome could no longer be detected for those with asymptomatic intracerebral hemorrhage compared with those without asymptomatic intracerebral hemorrhage. Patients enrolled in the National Institute of Neurological Disorders and Stroke t-PA Stroke Study had better long-term outcome regardless of whether asymptomatic intracerebral hemorrhage occurred. Asymptomatic intracerebral hemorrhage does not appear to influence long-term outcome after ischemic stroke.