Abstract： Objective To establish a big animal model of secondary infection of acute necrotizing pancreatitis (ANP). Methods Thirty young pigs were allocated to experiment group ( n = 20 ) or control group (n = 10). The ANP model was induced by retrograde injection of a mixture solution of 5% sodium taurocholate and 5% trypsin (0. 5 ml/kg body weight) into the main pancreatic duct and ligation of the proximal end of the main pancreatic duct, and then the second step was injecting 3 ～ 4 ml living Escherichia coli (E coli) suspension (108/ml) to the necrotic area of the pancreas by fine needle aspiration technique under CT guidance in the experiment group, and by injecting 3 ～ 4 ml inactivated E coli in the control group using the same method. Multi-slice spiral CT dynamic enhanced scan was performed in both groups 1 day and 2 or 3 days after ANP modeling and 5 days after bacterial injection to calculate the CTSI score. Serum amylase, blood WBC count and blood bacterial culture was performed in both groups. 5 days later, the animals were scarified to observe the infected or necrosis foci, and perform smear, bacterial culture and pathologic examinations of the tissue around the infected or necrosis foci. Results The ANP secondary infection model was successfully established in 16 of the 20 animals in the study group, with a success rate of the 80.0% (16/20). There were 17 foci where the positive rate of bacterial culture was 100% (17/17 foci), and the success rate of blood bacterial culture was 68.8%(11/16). In the control group, the ANP model was established successfully in 7 of 10 animals (70%), except for one case of contamination, only one foci was identified;the positive rate of bacterial culture and the success rate of b|ood bacterial culture was 14.3% (1/7). Serum amylase and white blood WBC count increased with similar trends, WBC count in the study group was significantly higher than that in the control group (P<0.01). The mean CT severity index(CTSI) was all ≥4 in beth groups, indicating the severity was moderate to severe. Conclusions A stable and reliable model of secondary infection of ANP in big could be established satisfactorily by injecting active E. coli into the pancreatic necrosis tissue under CT guidance, which helps further pathogenic mechanism studies and clinical studies, especially imaging studies.