*2.2. Mechanical Loading*

A load was applied to the head of the femur using a custom mold. Each potted femur was placed in a material testing machine (Electroplus E10000, Instron, Norwood, MA, USA) and held with custom fixation devices at 6◦ valgus to simulate anatomical positioning with weight bearing. The specimens were supported in the testing machine by a ball bearing to avoid uncontrolled torque or bending, as previously described by Cordey et al. [10]. Each distal femur was firmly held in a pre-shaped auto-polymerized acrylic resin (Vertex Dental, Zeist, Netherlands) until the lateral and medial condyles were in contact with the mold as it hardened. The mold aimed to evenly distribute the axial force applied during the testing process (Figure 2).

**Figure 1.** Two different constructs of the osteoporotic femur model with IMN and radiographs: (**A**) IMN with two standard proximal locking screws (SN group) and (**B**) IMN with two reconstruction proximal locking screws (RN group). IMN, intramedullary nailing.

**Figure 2.** Experimental setup for measuring the stiffness and failure load after creating the osteoporotic femur models with different proximal interlocking screw constructs.

The experiments were designed to measure the structural stiffness, failure load, and failure mode of each proximal interlocking screw construct by applying axial compression. Before conducting the experiments, an axial preload of 100 N was applied to the servohydraulic testing machine and the femoral constructs to obtain stable results. The test was then performed by loading a 1500 N weight (twice the force applied to the femoral head of a 75 kg adult) at a velocity of 10 N/s in the direction of axial compression, as described by Grisell et al. [11]. This process was repeated five times, and the weight was loaded at 10 N/s on each femur until compressive failure. All constructs were ramped to failure by increasing the force to 10 N/s, and the load, displacement, and time data were collected at a sampling rate of 20 Hz. The axial displacements from the initial position to the preload and from the preload to the maximum load were continuously recorded

using the crosshead motion sensor of the servohydraulic testing machine. The degree of displacement corresponding to the increase in axial load was determined for each femur. The stiffness of each femur was calculated as the slope of the elastic portion of the force versus displacement curve, and the mean value was considered. Failure was defined as screw breakage or fracture of a part of the construct, and the force applied to the femoral head at the time of failure was measured. If none of the aforementioned failure criteria were satisfied through direct observation, a sudden reduction in force, as revealed by the force versus displacement graph, was considered a failure [12].
