Abstract: Objective To explore the effects of the Hemopexin (Hx) on the generation of free radicals and brain edema after intracerebral hemorrhage (ICH). Methods One hundred SD rats were randomly and evenly divided into four groups (25 rats in each group) which named Sham group, ICH control group, Hx removal group and Hx intervention group respectively. There were five observation points (1 d, 3 d, 7 d, 14 d, 21 d) and which contain five rats for each. The stereotactic injection technique was used to make the ICH model, which adopted rat autologous whole blood that was removed or mixed with Hx and then injected to the right caudate nucleus of the brain. Sham group were only injected with 50 μL saline to the right caudate nucleus and ICH control group were injected with 50 μL autologous whole blood; Hx removal group were injected 50 μL autologous whole blood of removal Hx and Hx intervention group were injected 50 μL autologous whole blood which contain 0.25 mg (5 μg/μL) Hx. Bederson’s method was applied to evaluate whether the model was established successfully or not. Garcia’s method was used to estimate the neurological dysfunction scores by. Water contents of brain tissue around the hematoma was detected by dry-wet weigh method .The superoxide dismutase (SOD) activity were measured with the xanthine oxidase method. The content of the malonyldialdehyde (MDA) was measured by the thiobarbituric acid method. Pathological changes of brain tissue around the hematoma were detected by immunohistochemical method at each observation time points; and the immunohistochemical scores result was judged by the double semiquantitative evaluation method. Results Compared with Sham group, at 3-21 d, there were statistically significant differences ( P<0.05) in the neurological disorders and water content of the brain tissue and immunohistochemistry scores within ICH control group, Hx intervention group and Hx removal group. Compared with Sham group, at 1-21 d, there were statistically significant differences ( P<0.05) in SOD activity and the content of the MDA within ICH control group, Hx intervention group and Hx removal group. All the indexes above were superior in Hx intervention group to ICH control group ( P<0.05), and inferior in Hx removal group to ICH control group ( P<0.05). Conclusion The Hemopexinmay attenuate the generation of the free radicals and encephalaedema in the brain tissue around the hematoma after intracerebral hemorrhage.