Abstract: Objective This study is aimed at establishment of emodin-induced drug-induced liver injury in rats. Methods According to dose of emodin daily intake，100 male Wistar rats were randomly divided into five groups，e.g. high-dose group （16mg.kg^-1?d^-1），a larger dose （8 mg?kg^-1?d^-1），the middle dose group （4mg?kg^-1?d^-1），low-dose group （2 mg?kg^-1?d^-1） and the normal control group （fed an equal volume of distilled water），and each group had 20 rats. 5 rats were killed in each group at the end of 4th and 8th week. Other rats were all killed in the end of 12th week. The body weight of rats was measured. With HE and VG staining，the pathology of liver tissues were observed under microscope. The degree of fibrosis in liver tissues were compared by Radit analysis. Results Rats lived good in each group，and there were no deaths in each group or ascites formation;The body weight in high dose group were significantly lighter，and the body weight in normal group，low dose group，and middle dose group at 12th week were（335±10.56） g，（350.4±7.23） g and（338.6±7.54） g，respectively，much heavier than in high dose group[（300.4±8.91）g，P<0.01]；The inflammation activity and degree of hepatic fibrosis in liver tissues were both significantly more obvious in high dose group at 12th week（x²=14.75，P<0.05）. Conclusion The liver injury model in rats could be induced by emodin perfusion，which needs to be explored further in the future.

Key words: Drug-induced liver injury, Emodin, Pathology, Model