Document Type


Date of Degree

Spring 2017

Access Restrictions


Degree Name

PhD (Doctor of Philosophy)

Degree In

Biomedical Engineering

First Advisor

Nicole M. Grosland

First Committee Member

Matthew A Howard

Second Committee Member

Nicole A Kallemeyn

Third Committee Member

David G Wilder

Fourth Committee Member

Vincent Magnotta


Cervical myelopathy is the most common form of spinal cord injury in North America with roughly 19,000 new cases in the US every year. It results from chronic compression of the spinal cord by osteophytes, intervertebral disc herniation, and ossified ligaments. It commonly affects adults over the age of 50 years and causes upper extremity numbness, loss of hand dexterity, gait disturbances, and decreased proprioception. Recent studies imaging studies have shown this injury is highly dependent on the dynamic motion of the spine, often worsening in extreme flexion and extension. Surgical intervention is the accepted mode of treatment with the aim of decompressing the spinal canal and stabilizing the spine. However, 25% of patients have reoccurrence of symptoms indicating that surgical treatments may not be adequately addressing the injury. A main reason for this is little data has been reported on the spinal cord mechanics during cervical spinal motion in either healthy or cervical myelopathy subjects. To address this, we utilized MR imaging and finite element modeling to investigate spinal cord mechanics. As far as we know, we are the first group to obtain in vivo 3 dimensional spinal cord displacement and strain data from human subjects and the first to develop a C2 to T1 FE model of the healthy and cervical myelopathic spine and spinal cord.

Utilizing high resolution 3T MR imaging in neutral, flexion, and extension positions we were able to obtain spinal cord displacement and strain fields from both healthy subjects and cervical myelopathy subjects before and after surgical intervention. In healthy subjects, flexion motion of the spine causes the spinal cord to move superiorly and in extension the spinal cord moves inferiorly. During extension, localizations of high principal strain can be seen in healthy subjects at areas of bony impingement and dural buckling. In both flexion and extension, cervical myelopathy subjects exhibited very little spinal cord displacement due to spinal cord compression. Principal strains during flexion and extension were greater in cervical myelopathy patients than healthy patients, specifically at the C4-6 vertebral levels. Surgical treatments for cervical myelopathy did restore spinal cord motion however, not in the same pattern or direction as healthy subjects. Additionally principal strains of the spinal cord were not reduced after surgical intervention. This indicates that surgical interventions are not adequately addressing the altered mechanics of the spinal cord during cervical myelopathy.

To determine the how common surgical techniques for cervical myelopathy affect spinal cord mechanics, a FE model of the cervical spine and spinal cord was developed. The spinal cord motion was validated against MR imaging data obtained from normal subjects. Once validated, the model was used to develop a FE model of cervical myelopathy and surgical interventions. The native FE model predicted spinal cord motion well and replicated bony spinal cord impingement and dural buckling seen in healthy subjects. The FE model of cervical myelopathy also replicated spinal cord motion well as compared to MR imaging data of cervical myelopathy. Principal strains obtained from the healthy and cervical myelopathy FE models were similar in flexion however in extension, principal strains were higher at the C3, C6 and C7 levels. This is different than the patterns exhibited in the MR imaging and is most likely due to the percent of spinal cord compression induced in the FE model.

Three, C4 to C7 surgical interventions were introduced to the model: anterior discectomy and fusion, anterior discectomy and fusion with laminectomy, and double door laminoplasty. In flexion, all surgical treatments doubled spinal cord principal strains at the C3 level and minimally reduced tensile strain at C4. The majority of strain reduction occurred at C5-7. In extension, all surgical techniques increased principal strains at the C3 and C4 levels. Little or no reduction in principal strains was seen at the C5 and C7 levels. All surgical techniques reduced principal strains at the C6 level. Of the surgical techniques, ACDF tended to reduce spinal cord principal strains the least in both flexion and extension and tended to induce the highest von Mises stresses.

Combining the data obtained from MR imaging and FE modeling we can see that cervical myelopathy alters spinal cord mechanics by limiting spinal cord motion and increasing spinal cord strain. Additionally, current surgical techniques are not addressing the change in spinal cord mechanics effectively. Specifically after surgery, and especially with ACDF, spinal cord displacements and strains are being increased and transferred to different sections of the spinal cord. This indicates not only the need and importance of further research in spinal cord mechanics but also the need to improve treatments for cervical myelopathy which adequately restore the spinal cord mechanics.

Public Abstract

Cervical myelopathy is the most common form of spinal cord injury in North America with roughly 19,000 new cases in the US annually. It is caused by bony or soft tissue compression of the spinal cord which can get worse as a patient moves his or her head. Cervical myelopathy is very debilitating; these patients first loose feeling in their hands and as the injury worsens they can lose their ability to walk. The only way to stop this injury from progressing is to have a surgery that decompresses the cord. However, there are many different methods of doing this and no one knows exactly how each affects spinal cord compression. This means that sometimes, the first surgery doesn’t relieve the compression and secondary operations are needed.

What if we could know how each surgical method affects a patient’s spinal cord? This study focuses on using magnetic resonance imaging and computer models to answer this. Magnetic resonance imaging allows us to noninvasively “see” the spinal cord motion of cervical myelopathy patients. We can use this information to make computer models of cervical myelopathy, replicating the spinal cord compression during daily motion. Then we can try out different surgeries on this computer model and see which is best at decompressing the spinal cord. This information will enable surgeons to make more educated decisions when choosing surgical treatment methods for cervical myelopathy.


cervical myelopathy, finite element modeling, magnetic resonance imaging, spinal cord, spine


xvii, 114 pages


Includes bibliographical references (pages 105-114).


Copyright © 2017 Kirsten Elizabeth Stoner