Abstract: [Objective] The purpose of this article aimed to evaluate the biomechanical difference of three cortical support concepts (positive, neutral, negative) for intertrochanteric fractures. [Methods] Synthetic femur bone models (SYNBONE) were used to simulate three types of femoral intertrochanteric fractures including type-A1.1, type-A2.1 and type-A3.2. According to different types of cortical support, each kind of fracture was fixed with PFNA and then divided into three groups including positive, neutral and negative support group. Static axial compression test was used to measure the proximal femur subsidence under different loads and the axial stiffnessof the models. [Results] For type-A1.1, the proximal femur subsidence of positive and neutral groups were lower than negative group, and the axial stiffness was greater than negative group (P<0.05). To compare the proximal femur subsidence in type-A2.1, the positive group was the smallest, followed by neutral group, and negative group was the largest. The axial stiffness was the largest in positive group, followed by neutral group and the smallest in negative group, with statistically significant difference between the groups (P<0.05). The proximal femur subsidence of neutral group was less than both the positive and negative group in type-A3.2, and the axial stiffness was greater than the latter two groups (P<0.05). [Conclusions] Positive medial cortical support can be considered as the optimal reduction state for type-A1/A2 intertrochanteric fractures, while type-A3 reverse obliquity trochanteric fractures should obtain maximum anatomic reduction and neutral support.