The spinal cord is the most important part of the human body; it serves as a connection between the brain and the body. When the spinal cord is injured, the neural signals cannot travel between the brain and the body. This causes a complete or partial loss of sensory and motor function. The majority of the spinal cord injuries (SCI) are caused by apparent injury to the spinal canal including burst fractures, fracture dislocation, distraction injuries. Clinically, a burst fractures or dislocation are most common types of spinal column injuries associated with the SCI. Vertebral burst fractures result in spinal cord contusion from retropulsed bone fragments of the vertebral body are propelled into the spinal canal at high impact velocities, dislocation shears the spinal cord between adjacent vertebral levels, and distraction stretches the spinal cord. These distinct injury mechanisms produce varying patterns of primary spinal cord damage in experimental models and increasing injury severity results in a graded increase in the extent of spinal cord damage.
The SCIs can also result from the degenerative disease of the spine such as cervical spondylosis or spinal canal stenosis by ossified ligaments. Ossification of the posterior longitudinal ligament (OPLL) and ossification of the ligamentum flavum (OLF) have been recognized as causes of myelopathy due to thickening of the ligaments causing spinal cord compression. OPLL occurs primarily in the cervical spine while OLF is more common in the thoracic and thoracolumbar region. Computed tomography (CT) myelography and magnetic resonance (MR) imaging are used for diagnosis and pre-operative planning for the treatment of OPLL and OLF. OLF can be classified as fused and non-fused in the axial configuration, and round and beak types in the sagittal configuration. Based on plain lateral radiography, cervical OPLL can be classified into four types: segmental, continuous, mixed or circumscribed type, where the segmental and continuous types of OPLL were mostly recognized. Also, cervical OPLL was classified into two groups according to the axial ossified pattern of the OPLL on CT: central and lateral deviated type. The relationship between types of OPLL or occupancy rate and pathology in cervical compression myelopathy has been reported. Patients with OPLL usually require decompression surgery performed from either an anterior or posterior approach. Posterior approaches, which achieve decompression via posterior shifting of the cord, are feasible options for OPLL that extends more than two or three levels (e.g. continuous-type OPLL), and this approach is less technically challenging, more effective, and safer, especially for old patients. However, the SCI can be caused by surgical complications (for example, postoperative C5 palsy). The postoperative C5 palsies are most common complications seen after cervical surgery for OPLL. Although C5 palsy is a well-known complication of cervical spine surgery, the pathogenesis of this complication poorly understood and depends on many other factors, there are controversies still exist.
The SCI is a devastating and common neurological disorder that has profound influences on modern society from physical, psychosocial and socioeconomic perspectives. Currently, treatment options are limited, but experimental SCI models are being performed to understand pathophysiology neurological injuries. In addition, there are several widely used devices for spinal cord contusion injury, including the Ohio State University (OSU) device, the University of British Columbia multi-mechanisms injury device, the New York University (NYU) impactor, and the Infinite Horizon (IH) impactor. Although various devices and protocols have been employed to produce consistent injury severities, further investigation of the relationship between the key parameters of different contusion devices will improve our understanding of SCI mechanisms.
Due to the limitations associated with experimental SCI studies, the finite element (FE) analysis has been used to investigate mechanical parameters such as stress and strain, which are related to the neurological deficits. In this research, the biomechanical changes of the spinal cord due to degenerative spinal diseases or injuries and surgical treatments, and the relationships between mechanical parameters and neurological deficits were investigated to understand mechanisms of injury to the spinal cord using FE models of the human and animal spinal cord.
The degenerative diseases simulation models of OPLL and OLF showed that substantial increases in maximum stresses in the cord resulting in the manifestation of spinal cord symptoms occurred when the cross-sectional area (CSA) of the cord was reduced by 30% to 40% at 60% compression of the anterior-posterior diameter of the cord in OPLL and 4 mm compression in OLF. The influence of axial OPLL type was less than that of sagittal type, even though central type showed higher maximum stresses in the cord, especially for the continuous type. In the 60% of OPLL occupancy rate, the maximum stress was significantly high and CSA was reduced by more than 30% of intact area regardless of sagittal and axial types. In addition, a higher level of sagittal extension would increase the peak cord tissue stress which would be related to the neurological dysfunction and tissue damage.
The developed cervical spine and spinal cord models were used to investigate the effect of the extent of posterior decompression on the spinal cord in three types of surgical models for the continuous type of OPLL. As posterior decompression extended, stress and strain in the spinal cord decreased and posterior shifting of the cord increased. The location of the decompression extent also influenced shifting. Laminectomy and laminoplasty were very similar in terms of decompression results, and both were superior to hemilaminectomy in all parameters tested. Decompression to the extents of C3-C6 and C3-C7 of laminectomy and laminoplasty could be considered sufficient with respect to decompression itself.
The simulation of the flexion-extension of the cervical spine showed that static and dynamic factors are important for both preoperative planning, and understanding of postoperative complications. In the preoperative model, stress and strain in the cord increased and CSA of the cord decreased. The highest stress was observed during flexion while the greater reduction in CSA was observed during extension. After laminectomy, the stresses and strains were markedly reduced, but flexion led to increases stress in the cord because of residual dynamic compression occurred at segments with OPLL.
The simulations of SCI mechanisms (contusion, dislocation, and distraction) was developed based on clinical studies and analyzed the von-Mises stress and maximum principal strain in the spinal cord as well as the cord CSA for various injury severities. Dislocation and contusion showed greater stress and strain values in the cord than a distraction. The substantial increases in cord stress, as well as CSA reduction similar to or more than 30%, were produced at a 60% contusion and a 60% dislocation while the cord strain was gradually increased as injury severity elevated. These results indicated that the injury severity above a certain threshold may cause neurologic symptoms. In addition, the strong correlations between stress/strain and CSA reduction were quantified, which might be fundamental information in elucidating the relationship between radiographic and mechanical parameters related to SCI.
The developed cervical spine and spinal cord-nerve root complex model was used to investigate the cause of the postoperative C5 palsy. Our results suggest that high stress concentrated on the nerve roots after laminectomy could be the main cause of postoperative C5 palsy because of ossified ligament increased the spinal cord shift as well as root displacement. As the occupying ratio of OPLL increased, the stress and posterior shift of the spinal cord, as well as elongation and displacement of the nerve roots increased, but it has not significant effect on the stress in the cord and elongation of intradural nerve rootlets after laminectomy. The type of sagittal alignment also no influence on the change in the stress in the cord after laminectomy, but kyphosis cases with a high degree of occupying ratio resulted in greater increases in stress in the nerve root after laminectomy. Therefore, kyphosis with the high degree of the occupying ratio of OPLL could be a risk factor for poor surgical outcomes or postoperative complications, thus needs to be carefully considered regarding the surgical treatment.
The impact simulation models showed that stress in the cord was substantially elevated when the initial impact velocity of the pellet exceeded 4.5 m/s. Cord compression, reduction in CSA, and obliteration of the CSF increased gradually as the velocity of the pellet increased, regardless of the pellet size. The impact area of the pellet had less influence on biomechanical parameters than impact velocity of the pellet. Lastly, the quantitative relationship between NYU and IH impactor was established as a simple equation (Force = 28.2 ± 3.2?Height0.83±0.07) from the biomechanical parameters of the spinal cord. Thus, the key biomechanical parameter for a SCI contusion device can be converted or translated to that of another device in order to analyze experimental results from multiple contusion devices.
The research in this thesis covers a wide range of SCIs and surgical simulation studies from degenerative spinal diseases such as a various types of OPLL and OLF, surgical simulations for cervical OPLL, different types of SCI mechanisms including injury mechanisms related to postoperative C5 palsy, and modeling of the contusion devices for SCI research. We apply the FE modeling technique to accomplish the complex SCI studies, and also investigated the relationships between mechanical parameters such as stress or strain and neurological dysfunction and tissue damage to understand risk factors and causes associated with SCIs. The contribution of our works to this field is to improve our understanding of the biomechanics of SCI in the human and animal model of contusion SCI, as well as to provide essential information regarding the mechanisms and factors involved in SCI situation and surgical outcomes that can be incorporated into treatment and preventative strategies.