Aim To set up a novel animal model of intra-abdominal adhesion and to determine whether the tissue plasminogen activator activity (PAA) in exudate can be taken as an indicator to judge the formation of the adhesion. Methods　Rats were randomly divided into 2 groups. Each animal in both groups was opened the abdominal cavity via midline laparotomy without any disinfectant measures. 2-cm section from the cecal end was clamped and ligated, 1-cm cecum of the section was cut, and another 1-cm end from the ligated site was kept. After the content in the end was extruded, the cecum was put back without using any antibacterial agent. Before the skin closure, an effective combination AMD (allantoin, metronidazole and dexamethasone in combination), was given (ip) according to 1.5 ml per 100 g body weight (60.6 mg·kg-1). The control group was injected (ip) the same volume of normal saline. After 6 h, the exudate was extracted from drainage-tube, with the rats varying posture, and after 1 kw, the rats were killed for examining the intra-abdominal adhesion. The values of PAA of exudate in both groups were analyzed by the relative operating characteristic curve (ROC). Results　In the control group, all 20 rats formed the adhesions, the amount of exudate=(1.25±0.09) ml, the number of WBC(×103)=(23.1±6.6) mm3 and PAA=(0.9±0.4) IU·ml-1 in the exudate of abdominal cavity. In AMD group, however, there was not the adhesion formations (0/20), the amount of exuade was (0.52±0.04) ml (P<0.01), the number of WBC (×103) (10.6±4.2) mm3 (P<0.01), and PAA (23.1±6.6) IU·ml-1(P<0.01) in exuade.ROC analysis indicated that if PAA >1.24 IU·ml-1 in the exuade, the adhesion was difficult to form, and vice versa. Conclusion　This animal model can be taken as an effective tool to evaluate the human adhesion related to multi-links on the pathogenesis, and the PAA in exudate an indicator to judge intra-abdominal adhesion formation.