The cervical spine is an essential part of the human body responsible for the mobility and support of the head as well as the protection of the spinal cord. Flexion-distraction injuries are characterized by posterior disco-ligamentous injuries. They can lead to subluxation or dislocation of the articular facets and are among the most frequent injuries at the cervical spine. They also cause important neurological impairment. However, the injury mechanism leading to flexion-distraction injuries is still debated in the literature. Few experimental studies have been done specifically to reproduce this type of injuries. While disco-ligamentous injuries are recognized as unstable, there is little information on the contribution of the posterior structures to the post-traumatic spinal stability and the severity of neurological impairment following flexion-distraction injuries. Therefore, the objective of this thesis was to analyse the influence of the injury mechanism (loading conditions, vertebrae kinematics, injury patterns) on the instability (range of motion, relative intervertebral displacement) and on the spinal cord injury (spinal cord compression, von Mises stress, principal strains) in the context of flexiondistraction injuries. Thus, this work is divided into two parts: the traumatic behavior of the cervical spine leading to flexion-distraction injuries and the post-traumatic behavior of the cervical spine and the spinal cord.
To reach the first objective (O1), to characterize the head and cervical spine kinematics during a dynamic rear-head impact and the subsequent injuries, a test bench, composed of an adjustable seat and a horizontal impactor of 40 kg, was developed. This test bench was used in the context of an experimental campaign including six male post-mortem human subjects (PMHS) (82 to 96 years old). The PMHS were impacted at the rear of the head at 3,5 to 5,5 m/s with a posterior inclination of 0 to 25 degrees. The impactor was equipped with a triaxial load cell. Accelerometers were places at the sternum, forehead, and mouth of the subjects. Screws were inserted into the cervical spine vertebrae and equipped with two markers to allow 3D tracking by stereography. Markers were also placed on the subjects’ head and shoulder. The injuries due to the impact were assessed by computed tomography imaging and by dissection. The acceleration, impact force and head and cervical spine displacements curves during the impact were acquired. The six subjects sustained cervical spine injury. This study shows that flexion-compression loading applied to the head does lead to flexion-distraction injuries since four subjects sustained this type of injuries. Two of these subjects also had an articular facet fracture. The other two subjects suffered from a C2 fracture combined with a C1-C2 subluxation in one case. C6-C7 was the most often injured level and two subjects suffered from non-contiguous injuries. The presence of anterior osteophytes seemed to protect from upper cervical spine injuries and to protect the functional spinal units with osteophytic bridges. However, injuries were prone to occur at the spinal levels adjacent to the anterior osteophytes. Therefore, the presence of osteophytes influences the type and localisation of the injuries.
The second objective (O2), to characterize the injury mechanism in hyper-flexion and hyperflexion combined with compression loadings, was achieved using a finite element model (FEM) of the C4-C5 functional spinal unit extracted from SM2S (Spine model for safety and surgery) which was developed by the research group iLab Spine. The ligaments material properties were defined by non-linear stress and strain curves taken from the literature. The curves toe-region were calibrated against functional spinal units’ quasi-static flexion-extension curves. Maximal strain failure models were defined for the ligaments and the intervertebral disc (IVD). The model was validated against values of force and moment at failure. Then, the FEM was submitted to two types of loading up to failure: 1) flexion at 500 °/s and 2) flexion at 500 °/s with a compression of 0,2 m/s. Under flexion only, the ligamentum flavum (LF) and interspinous ligament (ISL) were ruptured first, followed by the posterior portion of the annulus, the supraspinous ligament (SSL) and the posterior longitudinal ligament (PLL). Under flexion-compression, the anterior part of the annulus was the first disrupted structure followed by the posterior part of the annulus. Then, the LF, ISL and SSL were ruptured. In both cases, the capsular ligaments (CL) were not ruptured. This suggests that different types of loading are necessary to cause their rupture and to cause the subluxation or dislocation of the articular facets. The results also show that the PLL is resistant to flexion and flexioncompression loading.
The achievement of the third objective (O3), to measure the pre and post-traumatic range of motion on PMHS following a rear-head dynamic impact, was attained during the experimental campaign on PMHS (O1). The head and intervertebral flexion-extension range of motion was measured before and after the impact on four subjects. The PMHS were filmed while motioned manually from a cervical spine neutral position to maximal flexion and then maximal extension. Following the impact, the head range of motion increased by 35 to 75 %. The intervertebral range of motion increased by 44 to 151 %. The global or intervertebral range of motion was lower for subjects with anterior osteophytes. Articular facets fracture created the most important increases in intervertebral range of motion highlighting the instability of this type of injury.
The fourth (O4) objective was to quantify the contribution of the posterior disco-ligamentous structures to the spinal stability of the cervical spine in flexion. First, the cervical spine FEM SM2S (C2-T1) was modified and calibrated. The material properties of the ligaments and IVD were adapted for the cervical spine and quasi-static loading conditions. The properties were calibrated against the intradiscal pressure and the intervertebral rotation in flexion-extension under pure flexion-extension moment loading (± 2 Nm). Then, injury to the posterior ligaments (LF, ISL and SSL) was modeled followed by a progressive IVD transversal rupture (1/3 of the antero-posterior length, 2/3 and finally complete rupture). The injuries were modeled at three levels: C2-C3, C4-C5 and C6-C7. The results showed that the ligaments disruptions had little impact at C2-C3 but increased the range of motion by 77 and 191 % at C4-C5 and C6-C7 respectively. On the contrary, the IVD complete rupture had an important impact at C2-C3 increasing the range of motion by 181 % and leading to facets subluxation at only 51 degrees of C2-T1 flexion.
To reach the fifth objective (O5), to characterize the spinal cord injury following flexiondistraction injuries, the SM2S spinal cord FEM was remeshed by brick elements and its geometry and material properties were improved. The link between the spinal cord and the medullary canal was modeled by spring elements. The FEM was verified against results of relative displacements between the spinal cord and the canal and against spinal cord principal strains at flexion and extension. Four combinations of injuries were tested subsequently at C4- C5, C5-C6 and C6-C7. These combinations aimed to quantify the contribution of the posterior disco-ligamentous structures to the protection of the spinal cord and to represent the diversity of injuries possible for flexion-distraction injuries. First, the LF, ISL and SSL were systematically ruptured. The injuries tested included the complete or partial transversal IVD rupture, PLL rupture and CL rupture. After injury modeling, the FEM was submitted to a pure moment in the sagittal plane (± 2 Nm). The principal strains and the von Mises stresses in the white and grey matters were measured as well as the lateral and antero-posterior compression of the spinal cord. In extension, the tested injuries had little impact on the spinal cord. In flexion, the complete rupture of the IVD combined with the rupture of all the posterior ligaments, excepted the PLL, was the situation leading to the highest von Mises stresses (47 to 66 kPa), the most extreme principal strains p1 (0.32 to 0.41 in the white matter) and p3 (-0.78 to -0.96 in the white matter) and the most important spinal cord compression (65 to 48 % in antero-posterior). The principal strains patterns in the spinal cord showed that the injury mechanism for these type of injuries during post-traumatic flexion, was an important compression of the anterior part of the white matter at the injured level combined with tension of the posterior part of the spinal cord. The CL followed by the IVD were identified as the most important structures for the protection of the spinal cord in post-traumatic flexion.
This research project has shown the relation between disco-ligamentous injury patterns and clinical instability. The hypothesis that flexion-distraction injuries are caused by flexioncompression loads applied to the head has also been demonstrated. New kinematic data for the head and the cervical spine during a rear-head impact were collected. The results from this thesis lead to clinical recommendations for the diagnostic and medical care of patients suffering from cervical spine injuries. The information acquired on the injury mechanisms will be useful for the development and evaluation of protective devices for the cervical spine. Finally, the realization of this thesis lead to improvement and development of numerical and experimental tools which will be used in the future to deepen our knowledge on spine trauma and spinal cord injury.
| Date | 31 Aug 2021 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Yvan Petit (Supervisor), Pierre Jean Arnoux (Supervisor) & Éric Wagnac (Co-supervisor) |
|---|