1. Experimental materials
The following materials were used in this study: 24 artificial standard femora (third-generation compound femora, medium 3304; Sawbones, a department of the Pacific Research Laboratory in Vashon, Washington); a 12-hole Cindy steel plate and a corresponding attached steel plate (DePuy Synthes, Solothurn, Switzerland); and the Ortho-Bridge System, comprised of a 6 mm diameter connecting rod, a fixing block, and a fixing screw (Tianjin Weiman Biomaterials Co., Ltd.).
2. Experimental grouping
This experiment comprised the four following groups: (i) Group A, a LCP + LAP fixed group; (ii) Group B, a bridging double-rod and double-cortex fixation group; (iii) Group C, a bridging double-rod and single-cortex fixation group; and (iv) Group D, a bridging single-bar cross-fixed group. Six experimental models were constructed for each group. The number of experimental models was determined in accordance with similar experiments undertaken internationally [21], and also met requirements for statistical analysis.
3. Model construction
This experiment was completed in the Newton Laboratory of Tianjin Weiman Biomaterials Co., Ltd. First, orthopaedic specialists performed a standard femoral neck osteotomy using Sawbones with a swing saw at a height of 10 mm near to the lesser trochanter. A hole was cut and the medullary cavity was opened using a medullary cavity file. The medullary cavity was enlarged to match the diameter of the medullary cavity expander. After placing a cement restrictor at the distal end of the medullary cavity, an appropriate amount of Heraeus bone cement was manually applied to fill the cortex, and the prepared femoral stem prosthesis was implanted. Twenty-four Sawbone femora were made. First, we used a 12-hole Cindyʼs plate, a corresponding attached plate, and the OBS for internal fixation. The bones were then sawed, and four groups of models were made to treat the B1 fractures. The length of the OBS connecting rod was consistent with a standard 12-hole locking steel plate, which was placed at the planned position under the femoral trochanter. In the bridge double-cortex fixation group, the bicortical screw channel at the upper end of the fracture was located under fluoroscopic observation (Fig. 2).The specific fixing requirements of each group were as follows:
(i) Group A, the LCP + LAP fixation group
We selected a Synthes locking (12 holes) wide dynamic compression plate, an attachment plate, and corresponding locking screws. Four 5-mm locking screws at the proximal end of the fracture, including two double-cortex locking screws were attached to the plate, two other screws were monocortical, and there were three 5-mm double-cortex locking screws at the distal end of the fracture.
(ii) Group B, the bridging double-rod and double-cortex fixation group
Two connecting rods (length, 22 cm; diameter, 6 mm), and three single-rod and single-hole fixation blocks and double-cortex locking screws were used to fix the proximal end of the fracture from top to bottom. Near the fracture end, a double-rod single-hole fixation block and a single-cortex locking screw were used for fixation. The distal end of the fracture was fixed with three double-rod single-hole fixation blocks and three double-cortex locking screws.
(iii) Group C, the bridging double-rod single cortex fixation group
Two connecting rods (diameter, 6 mm) were selected, along with three single-rod single-hole fixation blocks, one double-rod single-hole fixation block, and four screws, which were all single cortex locking screws from top to bottom at the proximal end of the fracture. We used three double-rod single-hole fixation blocks and three double-cortex locking screws for fixation at the distal end of the fracture.
(iv) Group D, the bridging single-rod cross-fixation group
One connecting rod (length, 22 cm; diameter, 6 mm) and one connecting rod with a shorter length were selected, and the two rods were separated at a certain distance. There were four single-rod and single-hole fixed blocks and four single-cortex locking screws at the proximal end of the fracture. From top to bottom, the first and third screws were located on the long connecting rod, and the second and fourth screws were located on the short connecting rod. Spacing of these two sets of screws was staggered at approximately 90°, resulting in fracture, and four double-cortex locking screws were used. From top to bottom, the first and second screws were located on the short connecting rod, and the third and fourth screws were located on the long connecting rod. Spacing of these two sets of screws was staggered at approximately 90° (Fig. 3).
4. Biomechanical test
Dental tray powder (polymethyl methacrylate formaldehyde) was poured on the distal femur, and an appropriate iron cup was used to fix the distal femur in the experiment. Outside the bone cement and 2.5 cm away from the distal end of the femoral stem prosthesis, an industrial wire saw was then used to cut a 45° oblique fracture line from the upper lateral side to the lower medial side (simulated anatomical reduction and Vancouver B1 simple fracture fixation), so that the supporting effect of internal fixation could be prevented during the axial compression test [21]. All samples were subjected to axial compression and torsion tests, and the stiffness and torsion angle values of the different system groups were obtained when treating simple fractures. An industrial milling machine was then used to form a 5-mm bone gap at the sawing bone site to simulate the comminuted fracture model, and two groups of axial compression stiffness and the torsion angle data were obtained when testing and treating complex fractures again. Finally, all samples were subjected to axial compression failure, and failure compression force values were obtained. In this study, simple fracture was defined as the initial saw bone, comminuted fracture was defined as the 5-mm bone gap, and the axial compression failure experiment was defined as the continuous application of load during axial compression until irreversible failure of the implant or the femur occurred.
5. Experimental tests
5.1 Axial compression test
The experiment was performed using a microcomputer-controlled electronic universal testing machine (Equipment model E45.105, Fig. 4). The cast iron cup at the distal femur was matched and fixed to the testing machine, the femoral head prosthesis at the proximal femur was in contact with the white polyethylene cylinder of the testing machine, and the experiment was performed under an initial vertical load of<100N, amaximum vertical load of 1000N, and a displacement loading rate of 8 mm/min, and the load-displacement curve was obtained. The slope of the curve was obtained using computer software(TW-Elite)connected to the testing machine.
In the crushing fracture test, changes in the stiffness of the fracture end after contact were observed, and the load was approximately 500N. The stiffness values without contact were compared to eliminate test-related influencing factors. The load level was lower than the physiological level in many daily life activities; however, to prevent permanent damage to the sample during the stiffness test, a lower load level was selected [21].
5.2 Torsion test
This experiment was performed using a torsion testing machine (Equipment model ND-200; Fig. 6). The femur was horizontally installed on the equipment, the cast iron cup at the distal end of the femur was matched and fixed to the testing machine, and the femoral head was fixed with clamps at both ends of the testing machine. The experimental conditions comprised a maximum of 10 N.m and a loading rate of 90°/min. P-main computer software was connected to the testing machine, which was used to obtain the torque-rotation curve, and a lower load was selected to avoid permanent damage to the sample during testing.
5.3 Axial compression failure test
The axial compression failure test was the final failure test, which was conducted using a microcomputer-controlled electronic universal testing machine (equipment model: E45.105). The cast iron cup at the distal femur was matched and fixed to the testing machine, the femoral head prosthesis at the proximal femur was in contact with the white polyethylene cylinder of the testing machine, and the vertical initial load was applied to the femoral head within 100 N, and the displacement loading rate was 8 mm/min, until the implant or femur was irreversibly damaged, and the failure mode and verticality during failure were recorded.
6. Experimental evaluation
The stiffness value indicates the capacity of an implant to resist deformation in the elastic stage, and it is used as a standard evaluation of axial compression. Under the same axial pressure, the greater the stiffness value, the smaller the deformation of the plant and the firmer it is. The torsion experiment was used to compare the difference in torsion angle between different implanted structures under the same torque load, that is, torsion stiffness. Under the same torsion force, the torsion angle of the endophyte decreases, which means that the endophyte is stronger. The aim of the axial compression failure test was to detect the maximum load that the structure of the inner plants could bear. The greater the maximum load, the stronger the resistance of the inner plants to destructive force and the better the overall strength.
7. Statistical analysis
To analyse the differences among all of the structures, after ensuring the normal distribution of the test data, the data were analysed using one-way ANOVA, and the significance level was set at P < 0.05. When the variance between the control groups maintained homogeneity, the Bonferroni method was used for multiple comparisons in pairs, while the Tamhane method was used for multiple comparisons when the variance between the control groups did not maintain homogeneity. When analysing the comparison between the simulated simple fracture group and the simulated comminuted fracture group, a t-test was performed, and the Bonferroni significance level was adjusted to P < 0.0125. The adjustment value was calculated through dividing the P-value of the 95% confidence interval by the number of constructions compared, that is, Bonferroni = the P-value of the 95% confidence interval/number of constructions = 0.05/4 = 0.0125.