Abstract： Objective To investigate the effect and mechanism of sodium ferulate (SF) on injury of gastric mucosa after traumatic hemorrhage shock resuscitation in rabbits.Methods One hundred healthy rabbits were randomly ( random number) divided into 4 groups ( n =25 in each group):control group ( A),model group ( B),pre-resuscitation SF group (C) and post-resuscitation SF group (D).The gastric mucosa injury model was established by using a method of comminuted fracture of femur and blood depletion.SF 30 mg/kg was injected into vein of rabbits' ear 20 min before resuscitation in group C and 30 min after resuscitation in group D,while rabbits of remaining groups received equal volume of normal saline instead.The gastric mucosa was obtained 90 min after resuscitation.The damage index (DI) of gastric mucosa was observed with method of Guth and the ultra-structure of parietal cell of stomach was observed under electronic microscope and the contents of TXB2 and 6-Keto-PGF1α in gastric tissue homogenate were determined with radio-immunity methods,and the ratios of TXB2/6-Keto-PGF1α were calculated.Data were analyzed by ANOVA ( LSD-t test ),and P ＜ 0.05 was considered as statistical significance. Results Under the electronic microscope,the secreting tubules were observed to be closed tightly in the parietal cells of stomach in the group A,showing a static status.However,in the group B,the number of normal secreting tubules was increased and the lumens were enlarged obviously.Compared with the group B,the number of normal secreting tubules was decreased and the enlargement of secreting tubules was not obvious in group C.The degree of changes in secreting tubules in group D was that between group C and group B.Compared with group A,the DI,the content of TXB2 and ratio of TXB2 to 6-Keto-PGF1α in other three groups were higher [DI: (81.5+13.6), (61.3+18.2), (70.5+17.2) vs.(4.2+2.7); the contents of TXB2:(4.95 +0.51),(3.75+0.64),(4.39±0.69) vs.(2.76±0.44); and the ratios of TXB2 to 6-KetoPGF1α:(0.064±0.002),(0.037±0.005), (0.049±0.002) vs.(0.027±0.002)] (P＜0.01),but the contents of 6-Keto-PGF1α in other 3 groups were lower [ (77.9±8.9),(96.4±11.2),(89.2+11.4) vs. (109.3±7.6)] (P＜0.05orP＜0.01).Compared with group B,theDI [ (61.3±18.2),(70.5±17.2) vs.(81.5±13.6)] and the contents of TXB2 [ (3.75±0.64), (4.39±0.69) vs.(4.95±0.51)] and the ratios ofTXB2 to6-Keto-PGF1α [ (0.037±0.005), (0.049±0.002) vs.(0.064 ±0.002)] in groups C and D were lower (P ＜ 0.05 or P ＜ 0.01 ),but the contents of 6-KetoPGF1α in groups C and D [ (96.4 ± 11.2),( 89.2 ± 11.4) vs.(77.9 ± 8.9) ] were higher ( P ＜ 0.05 or P ＜ 0.01 ).Compared with group C,the DI [ ( 70.5 ± 17.2) vs.61.3 ± 18.2) ] and the contents of TXB2 [ (4.39 ± 0.69) vs.(3.75 ± 0.64) ] and the ratios of TXB2 to 6-Keto-PGF1α [ (0.049 ± 0.002 ) vs.(0.037 +0.005) ] in group D were higher ( P ＜ 0.05 or P ＜ 0.01 ),but the content of 6-Keto-PGF1α in group D [ ( 89.2 ± 11.4 ) vs.(96.4 ± 11.2) ] was lower ( P ＜ 0.05 ).Conclusions SF can attenuate the injury of gastric mucosa after traumatic hemorrhage shock resuscitation in rabbits,and its therapeutic effects is better when it is administered before resuscitation than those as it is administered after resuscitation.The possible mechanism is associated with the effects of improving balance between TXB2 and 6-Keto-PGF1α and inhibiting the secreting function of parietal cell of stomach.