Abstract: Objective To investigate the curative effects of various infusion volumes on liver-related metabolic mechanism in the treatment of hemorrhagic shock. Methods A severe hemorrhagic shock rabbit model was established in 30 rabbits. The rabbits were randomly divided into three groups： non-infusion group （A）, conventional infusion group （B）, and excessive infusion group （C） （n=10 in each group）. Taking group B as the control, groups A and C were observed for the damage of non-infusion and excessive infusion, respectively. The outcomes in the three groups and their relations with liver tissue metabolism changes were analyzed with gas chromatograph-mass spectrometer （GC-MS）. Results The mortality in groups A, B, and C group were 80%, 0%, and 70%, respectively. The liver tissue metabolic profile in group B showed statistically significant difference compared with that in groups A and B. In group C, the levels of 21 metabolites were lower than those in group B, and the levels of 8 metabolites were lower than those in group A. The relative contents of various metabolites were correlated with infusion volumes, and the succinic acid content was associated with death events （P<0.05）. Conclusion The conventional infusion has significant curative effect on hemorrhagic shock. The metabolites of liver tissues with excessive infusion are generally decompensated and have longer survival time than those in non-infusion group, which may caused by the excessive infusion-induced blood volume increase after hemorrhagic shock. Tissue fluid dilution is an important cause of death.

Key words: forensic pathology, hemorrhagic shock, liver, metabolism, infusion quantity, rabbits