OBJECTIVE：To explore the effect of different plasmapheresis supplement timing on therapeutic efficacy of toxic liver injury. METHODS：96 patients with toxic liver injury and divided into group A，B ，C and D with 32 cases in each group according to different plasmapheresis supplement timing. All patients received plasmapheresis supplement based on routine treatment. In group A，synchronized fluid replacement was 100％ fresh plasma；in group B，synchronized fluid replacement was 40％ normal saline firstly，and then 60％ fresh plasma；in group C，substitute liquid was given till the in vitro blood reached 12％ of circulation amount，supplement order as 40％ normal saline for the first supplement，and then add 60％ fresh plasma. The clinical symptoms and signs，liver function，prothrombin activity and blood biochemical indicators were observed in each group before and after treatment. RESULTS：Plasmapheresis supplement had good therapeutic efficacy on toxic liver injury；but the time of clinical symptoms and signs disappearance in group C was significantly shorter than that in group A and B，and the group B was significantly shorter the group A，with statistical significance（P＜0.05）. The plasma bilirubin，prothrombin activity，albumin levels of 3 groups after treatment were significantly better than before，with statistical significance（P＜0.05）；the group C was better than the group A and B，and the group B was better than the group A，with statistical significance（P＜0.05）. Compared with before treatment，the levels of AST and ALT in 3 groups decreased significantly after treatment，with statistical significance（P＜0.05）；the group B and C were better than the group A，with statistical significance（P＜0.05）；the group B and C was similar to each other，without statistical significance（P＞0.05）. CONCLUSIONS：Different plasmapheresis supplement timing have different effect on toxic liver injury. The supplement method that giving 40％ normal saline for the first supplement，and then add 60％ fresh plasma when in vitro blood reach 12％ of circulation amount has more significant effect.