This study was performed using 18 thoracolumbar vertebrae T14 (T15) -L2 harvested from mature pigs (weight 110–130 kg). The porcine thoracic vertebrae have 14–15 segments and the lumbar vertebrae have 6–7 segments. Prior to their use in the study, the pigs were healthy and not exposed to any environmental factors that could affect their bone quality. All specimens were rigorously inspected to ensure that there were no defects. CT scan in the imaging department of Tianjin Hospital confirmed that the structure of them was intact and no deficiency (Figure 2). The paravertebral muscles and other soft tissues on both sides of the specimen were removed, and the intervertebral disc, spinous process, interspinous ligament, posterior longitudinal ligament, facet joint, ligamentum flavum and transverse process were preserved. Rinse with water and soak in 10% formaldehyde solution for 24 h. The study design and subjects are presented in Figure 3.

The 18 triple vertebrae specimens were completely immersed in 0.5000 mmol/L EDTA-Na₂ decalcification solution (pH 7.3 ± 0.1) (Lee et al., 2011; Hsieh et al., 2020; Liu JB. et al., 2022). Bone mineral density (BMD) of all vertebral bodies was measured weekly and the solution was changed until the percentage loss of BMD value reached 30%. After all specimens met the criteria of osteoporosis, a conical vertebral bone defect with a diameter of 16 mm and a depth of 30 mm was created in L1 of each specimen using a grinding drill. The animal model of KD was prepared by smearing the cancellous bone surface in the cone-shaped defect with bone wax to simulate the hardened bone at the defect site. L1 of each specimen was compressed to 1/2 the height of the original vertebral body by biomechanical testing machine (ElectroForce 3510, Bose) (Figure 4).

18 triple vertebrae specimens were randomly divided into 3 groups, which were recorded as group A, B, and C (n = 6/group). Group A underwent pure PKP; Group B underwent unilateral pedicle cement screw fixation combined with PKP; Group C underwent bilateral pedicle cement screw fixation combined with PKP.

The details of PKP surgical were performed based on standard procedure (Lei et al., 2020). The puncture position was adjusted according to the IVC to ensure that the tip cone was located in the cleft. The tip cone was withdrawn, and the guide drill was inserted in sequence to a distance of approximately 3–5 mm from the posterior edge of the vertebral body. The collapsed vertebral tissue was opened with a balloon to create a cemented cavity. Apply the PKP working sleeve to push the prepared bone cement along the vertebral arch in slow and staged injections, stopping the procedure immediately if high resistance is encountered or if the bone cement is close to the posterior wall of the vertebral body. The amount of bone cement was selected as 8 mL according to the size of the porcine vertebrae, and the distribution of the bone cement in the vertebral bone was observed. The curing time of the bone cement was 15 min. After waiting for sufficient time and confirming the curing of the bone cement, the pusher and working cannula were rotated and then withdrawn.

The following were details of the surgical procedure for novel pedicle bone cement screw fixation combined with PKP. The surface projections of the bilateral pedicle margin of the diseased vertebra were used as the pedicle screw entry point. Firstly, 1.5 mL bone cement was pushed in with the bone cement pusher, and appropriately sized pedicle bone cement screw were placed without penetration from the anterior cortical aspect. Bone cement was injected along this pedicle screw so that so that the cement flowing out of the anterior and lateral holes of the pedicle screw fused with the previously injected cement and became a single unit. After the cement had set, and the tail of the pedicle screw was broken.

CT was performed on all specimens immediately after surgery. Observation of cement filling, screw placement, vertebral body injury and spinal cord compression (Figure 5).

The upper and lower vertebrae of each specimen were embedded with denture base resin without crossing the intervertebral disc at depth. And in the procedure, try to make the part fixed to the MTS multi-degree of freedom simulation test system (MTS, Bionix 370.02, Figure 6) as regular in shape as possible, so that the specimens were sufficiently fixed during movement and did not become skewed. The upper and lower bases of each specimen were tightly attached to respective bases of the test machine with the T12/L1 and L1/L2 discs in a horizontal neutral position. The specimens were preloaded at a rate of 1 mm/min and stopped when the force reaches 100 N. The purpose was to eliminate creep movement and to make the vertebrae fully fixed in the mold so that they would not slip off and affect the determination of the value. The test force was set to zero after adequate fixation was completed. An axial load of 500 N was applied to simulate the mass of the upper body. The specimens were kept moist by spraying physiological saline every 5 min during the test to reduce the error caused by evaporative water loss.

SPSS 26.0 software (IBM, United States) was used for statistical analysis. The means and SD ( X ¯ ±s) were calculated for BMD during specimen decalcification. Medians and interquartile range [M(IQR)] were calculated for the data on the height of the vertebral body, ROM and bone cement displacement in each of these groups. The Kruskal–Wallis test was used for the comparative analysis of experimental results among different groups. The 5% significance threshold indicated a difference.

The changes in regional bone mineral density (BMD) from normal to osteoporotic status were shown in Figure 7. The initial BMD was recorded as 1.34 ± 0.15 g/cm², and after being treated with EDTA for 7 weeks, the BMD decreased to 0.89 ± 0.16 g/cm². After EDTA treatment for 7 weeks, the BMD decreased to below 30% of the normal vertebral value (p < 0.05), indicating the onset of osteoporosis in the experiment.

The vertebral heights at 500 N of axial compression in each group were shown in Table 1. There was no statistical difference in vertebral body height before treatment (p > 0.05) and after treatment (p > 0.05). After different treatments, all three groups had higher vertebral heights than before surgery (p < 0.05).

ROM measured at 5 Nm in six directions for intact vertebral specimens (N), KD specimens (KD), postoperative group A (A), postoperative group B (B), and postoperative group C (C) were shown in Figure 8; Table 2. Statistical analysis showed that the ROM of the KD specimens was greater (p < 0.05) compared to the N control group, except for right lateral bending. In the left/right axial rotation, the specimens were still significantly different (p < 0.05)from the intact specimens in terms of ROM after pure PKP.

All specimens underwent 6 times of activity with different deflection angles in 6 directions of flexion, extension, left/right lateral bending, left/right axial rotation. After repeated movement with 5° as the maximum activity angle, no significant displacement of bone cement was seen in all vertebrae on the CT images. After repeated movement with 10° as the maximum activity angle, the relative displacement of the bone cement in each group is shown in Table 3, and the partial post-activation CT of the bone cement is shown in Figure 9. Compared with pure PKP, pedicle cement screw fixation combined with PKP can significantly reduce the relative displacement of the bone cement (p < 0.05).

Biomechanical testing of new spinal implants in vitro is essential for their safety and efficacy (Lee et al., 2011). However, the limited availability of human cadavers leads to inconsistent specimen quality, making it difficult to obtain enough specimens for controlled experiments to explore the effects of these implants (Lee et al., 2011; Hsieh et al., 2023). The vertebrae of other large mammals are morphologically and biomechanically different from the human vertebrae, and their experimental parameters are not directly transferable to the human spine. However, due to the similarity of the porcine spine to the human in terms of size, nutritional make-up, bone structure, and mineral metabolism, it has been frequently used in some studies to verify the effectiveness of spinal fusion and internal fixation, and to a certain extent to draw clinically relevant conclusions (Lee et al., 2011; Hsieh et al., 2020; Laznovsky et al., 2022).

To model osteoporosis, demineralization can be performed using reagents such as hydrochloric acid and hydrogen peroxide, but the models prepared using these methods differ from the pathological process of human osteoporosis, which is more like osteomalacia (Elfar et al., 2014; Stewart et al., 2020). EDTA is a chelating agent that binds to the metal ion Ca²⁺ and acts more slowly, preserving the natural biological structure of collagen better than the stronger and faster acting hydrochloric or nitric acids (Lee et al., 2011). After 7 weeks, the difference in BMD values of the vertebrae before and after decalcification was more than 30%. Since there is currently no investigation data on peak BMD in porcine spine, the percentage of BMD loss was measured using the human osteoporosis standard, where 25% loss is considered an indicator of osteoporosis (Lee et al., 2011; Siris et al., 2014).

In the preparation of in vitro osteoporotic specimens, treatment with 10% formaldehyde solution for 24 h was a common approach in previous biomechanical experimental studies because of the prolonged exposure to ambient temperature (Lee et al., 2011; Hsieh et al., 2022). This preservative measure inevitably causes changes in the biomechanical properties, focusing particularly on Young’s modulus and impact energy in short time applications (Currey et al., 1995; Holewijn et al., 2017). This alteration makes it necessary to use smaller moments and slower angular velocities in the experiments to avoid damaging the specimens, which can also affect the experimental results to some extent.

In clinical practice, it has been found that KD occurs most commonly at the thoracolumbar junction, which is also the common location for osteoporotic vertebral fractures. From a biomechanical perspective, one important reason for the development of KD from vertebral microfractures is that the shear stress in the thoracolumbar segment changes after a large range of motion (Li et al., 2011). Following vertebral fracture, the vertebral morphology changes, mainly characterized by loss of vertebral height and spinal kyphosis (Röllinghoff et al., 2013). In this study, we found that satisfactory vertebral height was achieved regardless of the use of pedicle bone cement screw or not. This result is similar to the clinical treatment outcomes reported by Noriega et al. (2019) (Wang et al., 2021). It can be concluded that using the novel type of pedicle screw combined with PKP for the treatment of KD can achieve good effectiveness in static standing position. The restoration of vertebral height can prevent spinal center of gravity shifting, which in turn restores the internal biomechanical environment of the vertebrae and reduces the risk of nonunion (Cao et al., 2020; Najjar et al., 2023).

Both PVP and PKP are commonly used for vertebral compression fractures. Patients with severe OVCF can be treated with PVP in a prone position with the chest and hips resting on soft pillows to maintain spinal hyperextension to help restore vertebral body height (Chin et al., 2006; Shin et al., 2009; Yu et al., 2020). Since this experiment used an in vitro specimen, we had to use a balloon to expand the vertebral body when repairing the vertebral height.

After making the specimens into IVC-like vertebrae, ROM was significantly increased compared to intact vertebrae under 5 Nm torque. In the experiment, we observed that the increased instability mainly occurred at the site of the discontinuous cortical bone. Restoring the stability of the vertebral body is also a clear need in the treatment of KD (Yokoyama et al., 2013). The ROM of the vertebral body is an important parameter of vertebral stability. According to Denis’ three-column theory, recompression of the spine may be influenced by spinal stability (Denis, 1983). Low spinal stability may lead to an increased risk of recompression. In this study, filling with bone cement reduced the maximum displacement of the KD specimens and approached the intact vertebrae, which was also demonstrated by some related studies (Dai et al., 2022). However, in the PKP group, the specimens still differed from the intact vertebrae in terms of stability during rotation. This result suggests that the stability of the vertebral body can be further increased by pedicle screw fixation, thus reducing the incidence of recompression.

An important purpose of PKP and PVP is to return the patient to normal living conditions as soon as possible, which requires that the patient be able to perform the angle required after surgery. It is also required to reduce the probability of bone cement displacement during this period (Qi HR. et al., 2022). The gold standard in biomechanical testing is to apply moments to the specimens (Wilke et al., 1998). However, due to the in vitro osteoporotic specimens we used, which was more flexible than undecalcified specimens, this torque may result in a significant movement of the lumbar spine that should not have been accomplished. In previous studies, maximum rotational value was similar, with a range of 4.05° to 7.10° for lying and 9° to 14° for standing (Yeager et al., 2014; Breen et al., 2019; Daniel et al., 2023). In consideration of the maximum ROM of L12-L2 and avoiding excessive damage to the experimental specimen during the experiment, 5° was used to simulate ROM for lying and 10° to simulate ROM of standing (Yeager et al., 2014; Breen et al., 2019; Daniel et al., 2023).

Previous clinical case reports have also shown displacement of bone cement, but our experimental results generally have a higher displacement rate (Korovessis et al., 2013), perhaps due to the extreme smoothness of the cortical bone defect and the complete coverage of the cancellous bone surface by bone wax in our experimental models to better control the number of variables. The KD model designed in the experiment meets the clinical and radiographic characteristics of intravertebral cleft (Formica et al., 2016; Qi J. et al., 2022). This makes it possible that the measured displacement of the bone cement in the experiment indicates the trend and magnitude of displacement. The screw can release bone cement through the lateral hole at the front end of the screw, thoroughly filling the cracks in the vertebral body, so that the bone cement and the screw are connected as a whole. At the same time, the screw connects the strongest part of the bone at the pedicle, and ultimately reinforces the interaction between the bone cement and the surrounding bone tissue as a “bridge”. This achieves the function of preventing the bone cement from loosening and displacing.

The experiment has convincingly demonstrated that the novel bone cement screws provide excellent fixation of bone cement masses, providing them with less mobility and more stability. However, the experiment has certain limitations. First, the specimens used were porcine spines prepared by formaldehyde solution and EDTA, which inevitably led to altered the soft tissue properties and thus the overall biomechanical properties and therefore cannot be identical to human specimens. Second, the number of specimens was small, and more data are needed to support the results. Third, during the experiments, shear forces were still unavoidable, with a maximum value of 0.3 Nm. To obtain more reliable test results, these forces had to be reduced further, even to pure moments. Finally, the bone cement relative displacement was linked to the ROM, which is not correspond to the common recommendations in previous studies.