Objective: To compare the effectiveness and safety of different methods to construct animal models of aortic arch dissection (AAD), and explore safe and effective methods for constructing AAD animal models. Methods: Twenty-four healthy mongrel dogs were divided into 4 groups by random number table (n=6). Group A: Venous incision needle high pressure water flow impact method; Group B: Venous incision needle non-high pressure water flow impact method; Group C: Transarterial sheath non-high pressure water flow impact method; Group D: Two-way balloon expansion combined with elastase perfusion method. Imaging examinations were performed immediately and 7 days after operation, aortic tissue biopsy and pathological staining were performed 15 days after operation to observe the formation of AAD. The operation time, aortic blood flow block time, model construction success rate, dissection tear length, postoperative survival rate and survival time of four groups of experimental dogs were collected to compare the effectiveness and safety of different construction methods. Results: There were no significant difference of the gender, age and weight between four groups of experimental dogs (all P>0.05). The operation time of four groups of experimental dogs were (111.6±8.0), (168.0±17.4), (164.4±13.9), (202.8±21.5)min, and the difference was statistically significant (F=39.973, P<0.001). The operation time of group A was significantly lower than group B, C and D (all P<0.001). The aortic blood flow block time of four groups of experimental dogs were (5.2±1.8), (19.6±3.8), (20.6±3.9), and (18.6±3.0) min, and the difference was statistically significant (all P<0.001). The aortic blood flow block time of group A was significantly lower than group B, C and D (F=27.598, P<0.001). The four groups of experimental dogs had 5, 5, 4, and 1 model were successfully constructed, respectively, and the difference was statistically significant (P=0.008). The successful rate of model construction in group A was significantly higher than that in group D (P=0.040). The dissection tear length of four groups were (14.4±3.0), (11.3±4.2), (7.0±2.3), (4.7±0.6) cm,and the difference was statistically significant (F=8.103, P=0.003). The dissection tear length of group A was significantly longer than group C, D (all P<0.05). The postoperative survival time were 15.0(10.0, 15.0), 5.0(3.0, 10.0), 3.5(1.5, 4.8), 10.0(2.8, 15.0) days, and the difference was statistically significant (χ2=7.825,P=0.036). The postoperative survival time of group A was significantly higher than group B, C (all P<0.05). There was no significant difference in the survival rate of the four groups (P=1.000). The pathological staining results showed that the elastic fiber at the tearing point of AAD was destroyed, and the elastic fiber on the outer wall of the false cavity was over-stretched, which was consistent with the pathological changes of aortic dissection. Conclusion: Transvenous incision needle high-pressure water flow impact modeling method is easy to operate. The aortic blood flow block time is short, the dissection tear length is wide, and the postoperative survival time is long, can be used as the preferred method of animal AAD model construction.