Biomechanical analysis of a novel height-adjustable nano-hydroxyapatite/polyamide-66 vertebral body: a finite element study

Corpectomy is generally accepted as an effective surgical procedure for spinal cancer metastasis, infection, deformity, and traumatic injuries [1, 2]. However, restoration of the spinal column during surgery remains a technical challenge in clinical practice [3, 4]. To create a biological environment for fusion, rigid stabilization with an ideal vertebral body graft is highly recommended, and a number of interbody graft types have been constructed including allografts, cement, metal, and synthetic materials [5‐9]. Among them, titanium mesh cages (TMC) and artificial vertebral body (AVB) such as the VLIFT cage made of metal alloy material have been used widely for their good mechanical properties [10‐13]. However, an increasing number of studies have demonstrated that these types of implants are often associated with some troublesome implant-associated complications such as stress shielding, high subsidence rate, and fatigue failure [1, 14, 15].

Although some in vivo and in vitro studies have compared the biomechanical properties of the TMC and AVB systems and found no significant difference between them [16, 17], few studies have addressed the mechanisms underlying these implant-associated complications, and there is little knowledge about the stress acting inside these prostheses. In addition, the complications related to these two prostheses are not all the same; for instance, AVB has a higher subsidence and revision surgery rate than TMC, suggesting that there may exist different mechanical mechanisms in these two prostheses [1]. To the best of our knowledge, no study has reported the use of FEM analysis to investigate the biomechanical properties of these two prostheses.

Various types of bioceramics have been used for treating bone defects [7, 18‐20], but they have some drawbacks such as weak mechanical properties and chemical stability, which limit their clinical applications [21]. To address these issues, many composite systems have been explored as bone substitute materials, including HA reinforced polyethylene, polylactide, collagen, and Polyactive™ [22‐25]. To enhance HA bioceramic toughness, Wei and Li [26] employed a novel method to make biomaterial n-HA/PA66 composite and found that the composite with 64.25 wt% n-HA had excellent mechanical properties close to the natural bone.

Nano-hydroxyapatite/polyamide66 (n-HA/PA66) as a biomimetic biomaterial has been approved for clinical application for more than 10 years, and many products made of this material have been used for spinal reconstruction [26‐28]. Additionally, some previous investigations have proved that n-HA/PA66 has a good clinical application with good mechanical performance in bony fusion [28, 29]. To reduce the incidence of implant-related complications, we have developed a novel height-adjustable vertebral body (HAVB) made of n-HA/PA66, which was reported in our previous study [30]. However, its biomechanical properties were not fully discussed in that paper because of limited data at that time. The objective of this study was to use FEM analysis to compare the biomechanical properties of this novel prosthesis with TMC and AVB, and evaluate their biomechanical efficacy in spinal stability reconstruction.

Corpectomy has become a common surgical procedure for spinal tumors, deformity, infection, and trauma [1, 2]. Accordingly, various types of interbody cages have been developed and applied clinically. Among them, TMC and AVB have become the most commonly used prostheses due to their good biomechanical properties and a high fusion rate. However, with the increased number of patients and prolonged follow-up periods, more implant-associated complications have been reported [33]. To overcome the disadvantages of these spinal interbody prostheses, we have developed a novel HAVB prosthesis made of n-HA/PA66. The details of the design process were reported in our previous study [30]. However, we did not fully discuss its biomechanical efficacy in that paper because of the lack of sufficient evidence. In this study, we used the FE method to investigate the performance of HAVB in spinal stability reconstruction and compared its biomechanical properties with those of the TMC and AVB systems.

As TMC and AVB have been widely used in clinical practice, many in vitro cadaveric studies have been performed to verify the efficacy of TMC and AVB in spinal stability reconstruction [13]. To compare the in vitro biomechanical properties of three different expandable cages with a nonexpendable cage, Rober et al. [13] conducted a cadaveric study and reported that no significant difference could be determined. In contrast, Knop et al. [21] reported that Synex was associated with significantly higher stiffness and lower ROM for rotation and bending as compared with TMC. It was found in our study that AVB was associated with the greatest ROM reduction in all loading conditions but no significant difference in ROM reduction was observed as compared with HAVB and TMC, indicating that spinal stability of these three FE models is similar.

Cappuccino et al. [34] reported that additional bilateral pedicle screw and rod fixation could provide the maximum ROM reduction for single-level lumbar fusion. Many other studies have also emphasized the importance of additional posterior stabilization [17, 35]. In our study, a noticeable ROM reduction was observed in all instrumented FE models. The reduction of flexion ROM was relatively higher than that of extension ROM in all models, with ROM reduction being the greatest at L2–4 (86% for HAVB and 91% for TMC and AVB). Additionally, we observed that the stress distribution of flexion on the prosthesis was higher than that of extension, indicating that both the posterior fixation system and the anterior prosthesis played an important role in ROM reduction. As the spikes at both ends of the TMC and AVB systems are embedded into the endplate, it greatly restricts ROM of extension. Otherwise, AVB exhibited greater ROM rotation reduction than TMC and HAVB in overall model and all intersegments except in L1–2, indicating that the type of prosthesis plays a critical role in ROM reduction, which is consistent with the finding of Knop et al. [36]. All these results illustrate that both the prosthesis and posterior stabilization play an essential role in spinal stability and HAVB has the similar biomechanical efficacy in spinal stability reconstruction as compared with TMC and AVB.

Although there is no significant difference in biomechanical properties between TMC and AVB, inconsistent clinical complications have been reported. Mark et al. [1] reported that AVB had a higher subsidence and revision surgery rate compared with TMC. It was found in our study that there was a remarkable difference in stress distribution between these prostheses and the adjacent endplates. In TMC and AVB, the peak stress on the L2 caudal and L4 cranial endplates occurred at left or right rotation, which was much higher than that in HAVB model, while the HAVB model and the INTACT model showed the similar stress value and distribution in all motions. The stress on HAVB also demonstrated the similar trend that the maximum stress occurred at flexion and rotation, indicating that the implants and the interfaced endplate bear more stress in flexion or rotation than that in any other motions. Therefore, more attention should be paid to the protection of flexion and rotation during postoperative rehabilitation training.

A tremendous amount of force concentration was observed at one point of the spikes in AVB, and this may be useful in explaining the incidence of adjacent vertebral fracture and low back pain after surgery. While an astonishing force concentration was detected in TMC and AVB, the peak stress value on HAVB was much lower, probably due to two main reasons. One is that the material of n-HA/PA66 has a similar elasticity modulus with our human cortical bone, and this can effectively reduce the stress shielding effect. The other is the morphological design of HAVB that enlarges the contact surface with the endplate, which helps disperse the stress loaded on the prosthesis. Given some advantageous biomechanical properties over TMC and AVB, HAVB can be used as a viable option for spinal stability reconstruction.

The result of biomechanical analysis in this study demonstrates that this novel prosthesis of HAVB made of n-HA/PA66 has the similar stress value and distribution compared with INTACT model in all motions, indicating that the material of n-HA/PA66 is a viable material for bone tissue implantations. In addition, some other new bioceramics have been studied recently and enhanced with different nanocomposties [9, 21, 37]. For example, Khandan et al. [7, 8] studied the mechanical and biological properties of the bredigite-magnetite nanocomposite with various amounts of magnetite and found that the bredigite-30 wt% magnetite was an optimal sample with a fracture toughness of 2.69 MPa m1/2 and a Young’s modulus of 29 GPa. Its excellent biomechanical properties make it a suitable candidate for bone implantations. It is therefore warranted to pay more attention to these newly developed biocomposites.

There are several limitations in our study. First, as the two components of the HAVB system are rigidly fixed with no relative sliding in all loading conditions, it may not reflect the real clinical situation. In addition, we failed to consider the effect of bone grafting or bone cement filling in the prosthesis, knowing that they may also affect spinal stability. Finally, as the material properties applied to the element of the FE model do not exactly reflect the real behavior of the human lumbar spine, the result of FE analysis should be interpreted as a trend only, and further in vitro and vivo studies are required.

The present study has demonstrated that HAVB has the similar biomechanical efficacy in spinal stability reconstruction as compared with TMC and AVB. The difference in clinical complications between TMC and AVB can be explained by the differential stress values and stress distributions on these prostheses. More attention should be paid to the protection of flexion and rotation during postoperative rehabilitation training. The material and configuration design of the prosthesis is closely related to the complications associated with the implant. Given some advantageous biomechanical properties over TMC and AVB, HAVB may prove to be a promising implant for spinal column reconstruction. Further in vivo and vitro studies are still required to validate our findings and conclusions.