Objective This study intended to establish rat venous replacement models through the “ cannula method ” and “ anastomosismethod ” and to identify a safe and effective venous replacement model for studies into the pathological mechanisms related to venous replacement. Methods SPF BN rats were used. The venous replacement models were constructed using the cannula or anastomosis method. The total operation time, blood flow blockage time and the 48 h survival rates after surgery were measured and analyzed. Abdominal vascular ultrasound was performed to observe the patency and blood flow velocity changes. The rats were killed to obtain the graft at different timepoint. Pathological changes were evaluated by histopathology, and TGF-β1 expression was detected by q-PCR. Results 36 rats with venous replacement were established via the cannula method or anastomosis method. There was no significant difference in the 48- hour survival rates [100. 0% (18 / 18) vs 94. 4% (17 / 18), P= 0. 310]. The total operation time and blood flow blocking time of the cannula method were shorter than that of the anastomosis method [(35. 8 ± 3. 6) min vs (56. 8 ± 4. 9) min, P< 0. 05;(14. 6 ± 2. 9) min vs ( 35. 1 ± 4. 5) min, P< 0. 05]. The vein blood flow velocity in the transplantation section increased in the early stage and gradually decreased with postoperative time. At 4 weeks, blood flow velocity in the cannula group was significantly lower than that in the anastomosis group (P< 0. 05). At 4 weeks after the operation, the formation of venous collateral circulation was seen around the vein of the transplanted section of models in the cannula group, but the transplanted vein was still unobstructed. There was no obvious formation of venous collateral circulation in the anastomotic group within 4 weeks. With the advancement of postoperative time, the wall of the transplanted vein thickened and the diameter reduced gradually in the two groups. The vein diameter was smaller in the cannula group than in the anastomosis group at each timepoint after the operation (P< 0. 05). In the cannula group, cellulose deposition was observed on the outer layer of the vein wall, forming fiber cladding in the muscle layer and intima hyperplasia and lumen stenosis. The degree of infiltration of mononuclear cells and CD4+ T lymphocytes in the transplanted vein wall in the anastomotic group was significantly higher than that in the cannula group (P< 0. 05 ). With the advancement of postoperative time, the expression of TGF-β1 in the vein wall of the transplanted section of the two groups increased gradually. At each timepoint, TGF-β1 expression in the cannula group was higher than that in the anastomosis group (P< 0. 05). Conclusions Both the cannula and anastomosis method can be used to safely and effectively establish a rat inferior vena cava replacement model. Compared with the cannula method , the vein replacement model established by the anastomosis method has lighter non-specific verification and wall remodeling, and has a higher long-term patency rate, which is more suitable for experimental studies related to vein replacement.