Preparation of pelvic specimens
Five adult embalmed cadaveric pelvic specimens (provided by the Department of Anatomy, Shandong First Medical University) were selected, including three male and two female, aged 42 – 58 years old, with an average of 48.3 years old. The L3 vertebral body was reserved to 10 cm of the proximal femur of both sides (Figure 1). Pelvic fracture, tumor, forced spondylitis, sacroiliac sclerosis, rheumatoid arthritis and other diseases were excluded by examination, and osteoporosis was excluded by bone density test. Specimens were wrapped in double plastic bags and stored at -20°C. The specimens were thawed at room temperature. Skin, muscle, fat and other tissues were removed. The complete pelvic bone, ligament structure and hip joint were retained. The ligament structure mainly included suprapubic ligament, pubic arch ligament, sacroiliac posterior ligament, sacroiliac anterior ligament, sacroiliac interosseous ligament, hip joint accessory ligament. The upper and lower ends of the specimens were embedded with methyl methacrylate-polymer resin to enable fixation to the mechanical testing machine.
Establishment and fixation of pelvic Tile C1 injury model
After the creep was eliminated, experimental results from biomechanics tester of complete pelvis specimens were used as control group. Then, the pubic symphysis was cut with electric saw, and the left sacral Denis II fracture was established (Figure 1). A series of surgical procedures were subsequently performed by Professor Baisheng Fu. Anatomical reduction of pelvic fracture was performed. Five-hole reconstruction plate was used to fix the separated pubic symphysis and lumbo-iliac fixation (the L4, L5 pedicle screws (6.5-mm diameter, 45-mm long), and iliac screws (7.5-mm diameter, 80-mm long), Medtronic-WeiGao Inc., WeiHai, China) was used for the treatment of unstable posterior pelvic ring injury: three channels of bilateral single iliac screw, including channel A from PSIS to AIIS, channel B from 1cm medial and 1cm caudal of PSIS to AIIS, channel C from 2cm below PSIS to AIIS (Figure 2). The pedicle screws were laterally straight and parallel to the vertebral endplate. The iliac screws were entering from three channels and a 7-mm ball tip feeler was inserted into the channel to ensure its completion. Subsequently, the 7.5-mm diameter iliac screws were placed. Finally, L4-L5 pedicle screws and iliac screws were connected by a curved rod and a cross-link was fixed between the L5 pedicle screws and iliac screws.
Biomechanical tests
The pelvic specimens were fixed on the special fixture of the American E10000 material mechanics testing machine (Provided by Institute of Orthopedics, Soochow University) (Figure 2). The L3 vertebral body was kept in a horizontal state at all times. Bilateral anterior superior iliac spine and pubic symphysis were placed on the same coronal plane in order to simulate the force on the pelvis when standing. The compression load of L3 vertebral segment was 0-500N [15] and the stress load speed was 3 mm/min through the upper loading connector. The analysis software Bluehill 2.0 provided by the mechanical testing machine automatically recorded the load-displacement curve and calculated the compressive stiffness (N/mm). Each model was tested three times and averaged. After each test cycle was completed, the pelvic specimen was carefully checked to be complete and the internal fixators did not loosen or break, the next channel test was randomly carried out. In the torsional load experiment, 6N·m torsional load was applied to the specimen through the rotational axis. The torque-torsional Angle curve was automatically recorded by WaveMatrix software and calculated the torsional stiffness (N·m /°). The same, each specimen was tested three times, and the average rotation angle was calculated. During the experiment, normal saline was sprayed regularly to keep the specimen moist.
Finite element injured models and finite element analyses
One normal adult male volunteer (48 years old, 175cm, 70Kg) was recruited. The study protocol was approved by the Ethics Committee of Shandong Provincial Hospital. The volunteer agreed with written informed consent. The radiographic data of lumbar spine, pelvis and femur were obtained by CT scan (Siemens Spiral CT, Germany, 0.625mm slice thickness), and imported into Mimics 21.0 (Materialise, Belgium) in Dicom format. And, then, these files were processed by Geomgic studio 12.0 (Geomgic, USA). Pro/Engineer 5.0 (PTC, USA) and Hypermesh 2017 (Altair, USA) were used to draw iliac screw, pedicle screw, longitudinal rod, connector, transverse connecting rod and reconstruction plate, screw, etc. The material properties and characteristics, Young's modulus, and the structure of the model ligament were set. The three-dimensional finite element model of L4-pelvic-proximal femur, pedicle screw and iliac screw were imported into Ansys 19.0 (SASI, USA) for finite element analyses (Figure 3). The finite element model included lumbar (L4 and L5), pelvis, and proximal femur. The full pelvis was composed of the left ilium, sacrum, right ilium, and symphysis pubis, and these bones consisted of the cortical bone and cancellous bone. The anterior sacroiliac, interosseous sacroiliac, posterior sacroiliac, sacrotuberous, and sacrospinous ligaments were also created to simulate normal condition. The linear elastic isotropic material properties were used, and the properties of the bones and ligaments were shown in Table 1. The same, the pelvic Tile C1 injury model (pubic symphysis separation, left sacral Denis II fracture) was established. Simulate 5-hole reconstruction plate fixation of the pubic symphysis and lumbo-iliac fixation for the treatment of posterior pelvic ring surgery. Finite element analyses were used to explore the biomechanical characteristics of bilateral single iliac screw, divided into three channels: channel A from PSIS to AIIS, channel B from 1cm medial and 1cm caudal of PSIS to AIIS, channel C from 2cm below PSIS to AIIS. The number of elements for implants was 1,713,729 for channel A, 1,715,997 for channel B, and 1,713,492 for channel C, respectively. The number of nodes for implants was 2,794,487 for channel A, 2,798,784 for channel B, and 2,795,149 for channel C, respectively. The standing state of the human body was simulated and the boundary conditions were set at the bilateral femoral ends. The 500 N vertical downward load was applied on the upper surface of L4 vertebral body, and the torque in different directions of 10 N·m was applied to simulate the working conditions of flexion, extension, lateral bending and rotation. The stress nephogram, displacement nephogram and deformation nephogram of the internal fixation, vertebral body and iliac bone were obtained by finite element software. The maximum von Mises stress of internal fixation and the maximum von Mises stress of vertebral body and ilium in different channels were compared.
Statistical methods
The maximal compressive displacement and torsional angle between the peak and trough points of curves were obtained. The compressive and torsional stiffness of fixation construct were calculated by the following formulas:
Compressive stiffness = 500 (N) / maximum compressive displacement (mm)
Torsional stiffness = 6 (N·m) / maximum torsional angle (°)
SPSS software (version 20.0 Chicago, IL, USA) was used for statistical description and analysis of the experimental results. The measurement data were expressed as mean ± standard deviation (SD). The experimental data were tested to meet the normal distribution and the homogeneity of variance. One-way analysis of variance was used for comparison between groups. LSD method was used for pairwise comparison. The difference was statistically significant when P < 0.05.