Biomechanical Comparison Of Modified TARP Technique Versus Goel Technique For The Treatment Of Basila Invagination By Finite Element Analysis

BackgroundBasilar invagination (BI) is characterized by irreducible atlantoaxial dislocation and upward migration of the tip of the odontoid process of C2 vertebra. Basilar invagination is usually secondary to congenital regional malformation, or bone diseases such as rheumatoid arthritis, hyperparathyroidism, Paget disease, osteogenesis imperfect and rickets. It results in progressive compression on cervical spinal cord, leading to profound neurologic deficits and even death.BI can be divided into clivus type and odontoid type based on the presence or absence of clinical and radiological evidence of instability of the craniovertebral junction. The pathogenesis of BI appears to be different in the two types. The clivus type BI showed that the odontoid process and clivus remained anatomically aligned despite the presence of BI and other associated anomalies. In this clivus type, the radiological findings showed that the cerebellar was forced from the foramen magnum, causing brainstem compression and neurological symptoms. Therefore, foramen magnum bone decompression appears to be a rational surgical treatment for patients having clivus type BI. Unlike the clivus type BI, the radiological findings suggested that the odontoid type BI is characterized by upward the migration of odontoid process result in direct compression of the brainstem ventrally following atlantoaxial dislocation, and so the key treatment to this type is to achieve anatomic reduction and pull the den downward to decompress the ventral cord directly. Several authors of clinical studies have shown that some patients with BI can be reduced in traction before surgery or during surgery. However, clinical studies have demonstrated that the contraction of anterior muscles, ligaments, and capsules of atlantoaxial joint, especially the osteophytes and scar tissue inside the atlanto-dens interspace prevent the reduction in odontoid type BI.As for the cases of BI with irreducible atlantoaxial dislocation, which can not be reduced in traction before surgery, the main surgical techniques of treatment of BI are TARP technique and Goel technique.In 2004, Goel first reported the spacer or cage was inserted into a distracted atlantoaxial joint after removing the articular surface cartilage to reduce the BI and achieve atlantoaxial fusion through posterior approach. We treat the technique of the C1 lateral mass screw, C2 pedicle screw and bar (or plate) fixation combined with Cage (ClLS+C2PS+Cage)as "Goel technique". In 2013, Chandra et al. proposed that the Goel technique only reduction of odontoid process but not the atlas, and used the cage in the atlantoaxial joint as a pivot to reduce the atlas. They used the bicortical C2 laminar screw instead of the C2 pedicle screw to avoid injury of the vertebral artery. We treat the technique of the C1 lateral mass screw, bicortical C2 laminar screw and bar (or plate) fixation combined with Cage (C1LS+BC2PS+Cage) as "modified Goel technique". There have been no reports to determine the biomechanical differences between C1LS+C2PS+Cage and C1LS+C2PS+Cage in C1-2 fixation.In 2004, Yin et al. first reported transoral atlantoaxial reduction plate (TARP) fixation, and clinical studies have demonstrated that the transoral anterior atlantoaxial release followed by TARP fixation can achieve reduction, decompression, fixation and fusion of C1-C2 through a transoral-only approach. The atlantoaxial fusion cage with TARP device may increase stability and fusion rates, maintain or improve atlantoaxial fusion angle and prevent bone graft collapse, extrusion, resorption and micromotion. We treated the TARP+Cage technique as "modified TARP technique" (TARP+Cage). However, few studies have compared the biomechanical differences between TARP+Cage and C1LS+C2PS+Cage for treatment of BI.The FE analysis is well suited to parameter studies and allows to investigate the stress distribution of the instrumentation than vitro experiment. In 2000, Puttlitz et al first reported validated FEM of the upper cervical spine (C0-C2), which was used to study the pathology of rheumatoid arthritis in the craniovertebral junction. Since then, the FEM of the C0-C2 complex was developed to simulate the complex kinematics of the upper cervical spine and has already been used widely for analyzing biomechanical of various instrumentation system designed for use in the upper cervical spine.Part one Biomechanical comparison of two modified Goel techniques for the treatment of basilar invagination by finite element analysisObjectivesTo determine the biomechanical differences between C2 pedicle screw and bicortical C2 laminar screw with intra-articular Cage in C1-2 fixation by finite element analysis.MethodsTo build an FEM, computed tomography images were obtained at 0.5-mm intervals on C0-C2 from a 31-year-old healthy man with a height of 175 cm. The images were performed for boundary detection by custom-made software. The commercially available finite-element program Abaqus 6.9 (Dassault Systemes, USA) was used to model the C0-C2 complex and to evaluate differences between the C1 lateral mass screw+Cage+C2 pedicle screw (C1LS+Cage+C2PS) and C1 lateral mass screw+Cage+bicortical C2 laminar screw (C1LS+Cage+BC2LS) for C1-C2 fixation. The FEM of the C0-C2 complex included cortical bone, cancellous bone and cartilage, and 10 types of ligaments (anterior longitudinal ligament, anterior atlanto-occipital membrane, tectorial membrane, posterior atlanto-occipital membrane, posterior atlanto-axial membrane, joint capsules, alar ligament, apical ligament, transverse ligament and cruciate ligament-vertical portion). The transverse ligament is low elastic tissue and quite tough, so it was modeled with 4-node membrane elements. All other ligaments were modeled with linear contact elements applying only to the tension force. The C1-C2 vertebral were defined as an internal cancellous bone core surrounded by a 1.5-mm-thick cortical outer shell, whereas occiput (CO) were defined as entire cortical bone. The thickness of cartilage of C0-C2 complex was 3.0mm.An unstable model was also established after removing the transverse ligament, and then elements were added at C0-C1 joint to simulate assimilation of C1.Two different implanted models:ClLS+Cage+C2PS and C1LS+Cage+BC2LS were integrated at the C1-2 segment into the unstable model. To study the biomechanics, vertical load of 40 N was applied in the inferior direction on the occipital condyles, to simulate head weight and 1.5 Nm torque was applied to the occiput to simulate flexion, extension, lateral bending, and rotation.ResultsTo validate intact model, the ROM of the C0-C1 and C1-C2 segments was calculated and compared with the results of in vitro tests performed by Panjabi and Dvorak et al and Panjabi et al, as well as the results of in FE tests performed by Zhang et al. We found good agreement between our results and previously published data.Unstable model via removing all transverse ligament elements increased ROM of C1-C2 segment in flexion, extension, lateral bending, and axial rotation by 35.2%, 16.4%,4.0%, and 5.6%, respectively, as compared with the intact model, which were consistent with those from biomechanical study performed by Li et al. Based on these findings, we confirmed the validity of our FEM of the upper cervical spine.There was no significant difference in the range of motion between C1LS+Cage+BC2LS and ClLS+Cage+C2PS implanted models,<0.1° for all loading cases, and also there was no significant difference in stress distribution and maximum stress between the 2 implanted models. Bone graft stress of the 2 implanted models, especially the C1LS+Cage+C2PS fixation model, were minimum under extension loading condition.ConclusionsOur FE results indicate that in cases in which the anatomy limits placement of C2PS, a device using BC2LS is a viable option, and the C1LS+Cage+BC2LS fixation offer similar stability to C1LS+Cage+C2PS fixation for the treatment of Basilar invagination. Compared to C2PS technique, the BC2L is an easy and effective technique and it can avoid vertebral artery and spinal cord injury. To avoid the instrumentation failure and stress shelter, after surgery, neck extension movement should be restricted or banned.Part two Biomechanical comparison of modified TARP technique versus Goel technique for the treatment of basilar invagination by finite element analysisObjectivesTo determine the biomechanical differences between modified TARP technique and Goel technique for treatment of Basilar invagination (BI) by finite element analysis.MethodsA validated three-dimensional finite element model of the intact upper cervical spine was established, and an unstable model was also established after removing the transverse ligament. The transoral atlantoaxial reduction plate+Cage (TARP+Cage) fixation model was established according to the method below:the entry point of the unicortical C1 anterior lateral mass screw was located 5 mm away from the inner-lower edge of the C1 lateral mass and the entry angle of C1 anterior lateral mass screw was 5°-10° upward and 10°-15° outward along the longitudinal axis of the C1 lateral mass. The C2 anterior pedicle screw entry point was 5 mm below the vertex point of medial corner of superior lateral mass of C2, and the entry angle of C2 anterior pedicle screw was 9.3° to 28.3° laterally in the axial plane and 6.5° to 21.5° inferiorly relative to the C2 lateral mass in the saggital plane. In the same way, the C1 lateral mass screw+C2 pedicle screw+Cage (C1LS+C2PS+Cage) fixation model was established according to the method below:the starting point of C1 lateral mass screw was located at the center of the lateral mass, along the C1-C2 joint 5° to 10° medially and 10° to 15° upward. The C2 pedicle screw was placed at the medial border of the C2 pedicle. The starting point was typically "high and inside", and the entry angle of C2 pedicle screw was the C1-C2 joint 16.5° to 23.8° medially and 25.3° to 36.7° upward. Then an unstable model treated with TARP+Cage fixation, was compared to that with C1LS+C2PS+Cage fixation. To study the biomechanics, vertical load of 40 N was applied in the inferior direction on the occipital condyles, to simulate head weight and 1.5 Nm torque was applied to the occiput to simulate flexion, extension, lateral bending, and rotation.ResultsBoth devices significantly reduced ROM compared with the intact state. In comparison with the C1LS+C2PS+Cage model, the TARP+Cage model reduced the ROM by 44.7%、30.0% and 10.5% in extension, lateral bending, and axial rotation, while the TARP+Cage model increased the ROM by 30.0% in flexion. In flexion, lateral bending and axial rotation, the maximum stress of anterior C1LS and anterior C2PS in TARP+Cage model showed lower than that of C1LS and C2PS in ClLS+C2PS+Cage model, respectively, and the maximum Stress of TARP in TARP+Cage model showed lower than that of rod in C1LS+C2PS+Cage model in both flexion and extension.ConclusionsOur results indicate that the TARP+Cage fixation may offer higher stability to C1LS+C2PS+Cage in extension, lateral bending, and axial rotation but lower stability in flexion. Compared to Goel technique, the modified TARP technique not only have the capability of transferring the load and distributing the stress, but also provide neural decompression, stabilization and fusion, and restore C1-C2 normal fusion angle. Prospective, randomized and multicenter clinical trials of the two different surgical techniques are required to validate the advantages of modified TARP technique for treatment of BI.