A Finite Element Analysis And Clinical Study On Lumbar Interspinous Process Distraction Device

Background:Spinal fusion surgery is looked as gold standard to treat spinal diseases. With the improvement of recognition of adjacent segment degeneration (ASD) after spinal fusion, non-fusion technique emerged. Interspinous process distraction device (IPD) belongs to techniques of lumbar posterior non-fusion. IPD can restrict some extension of lumbar spine, and the distraction of IPD can reduce the ligamentum flava invading into spinal canal, increase size of intervertebral foramen, so the symptoms of spinal canal stenosis and decompression of nerve roots can be relieved accordingly. Preserving spinous process, most vertebral lamina and superspinous ligament can keep the stability of posterior column and protect dural sac, meanwhile be helpful to reduce cyclic compression from soft tissue and re-stenosis of spinal canal. Dynamic fixation with IPD is easily accepted by doctors and patients due to simple operation and minimal trauma. But the long-term effect of IPD is uncertain since the period of application is short and overall basic research of all kinds of IPDs is absent. Therefore, general biomechanic evaluation, especially effect on the adjacent segments, is essential to predict the long-term clinical effect and develop non-fusion technique. This study is to choose Coflex system as typical instrument of IPDs, and to assess IPD's effect on implanted and adjacent segments by the biomechanical and clinical outcome of Coflex system.Objectives:1. To develop and validate an anatomic detailed finite element model of the lumbosacral vertebral spine.2. With three-dimensional finite element method, to analyse the effect of Coflex system on segmental range of motion (ROM), intervertebral disc stress (IDS) and facet joint pressure (FJP) by comparing with decompression model and rigid fixation model with pedicle screw system.3. To treat and follow up lumbar disc herniation (LDH) with Coflex, and to evaluate the clinical and radiological effect.Methods:1. The lumbosacral spine geometries were determined from CT images of a healthy male adult. The finite element model (L3-S) was constructed by the combination of software package Geomagic Studio 9.0, HyperMesh 10.0 and Abaqus6.9-1. ROMs of L4/5 were calculated after being subjected to moments 7.5Nm and 500N preload and generating flexion, extension, lateral bending and axial rotation. For validation of the model, the results were compared with literatures.2. The model of Coflex system was established with software Rhino 4.0 and was compared with material object.3. Based on the intact finite element model, new models were generated by simulating decompression by fenestration, dynamic fixation with Coflex system and rigid fixation with pedicle screw system. With a preload of 500N and 7.5Nm moment in flexion, extension, lateral bending and axial rotation was applied on all models, ROMs, IDSs and FJPs of all segments were calculated.4. From 2007 to 2008,16 LDH patients were treated with dynamic fixation with Coflex, and the other 16 patients with posterior lumbar interbody fusion (PLIF). The clinical effect was evaluated by Visual Analogue Scale (VAS) and Oswestry disability index (ODI); ROM of lumbar segments was measured by X-ray film.Results:1.The finite element model concluded 120,570 nodes and 31,4487 elements. ROM in flexion (FLE), extension (EXT), lateral bending (LB) and axial rotation (AR) of L4/5 was consistant with literatures.2. The model of Coflex device conclude 30,757 nodes and 16,047 elements. The appearance and performance of the model was the same as the material object and could be used for experiment.3. The effect on the surgical segment: ROM, IDS and FJP of decompression model (MDec) increased; those of dynamic fixation with Coflex model (MCof) decreased, especially in EXT; those of rigid fixation with pedicle screw model (MPS) decreased significantly. The effect on the adjacent segments: ROM of MDec decreased, IDS increased; ROM, IDS and FJP of MCof increased slightly; those of MPS increased significantly, and the increasing degree of L3/4 was more than that of L5S1.4. The stresses of Coflex device and spinous process increased, especially while EXT.5. All cases were followed up. In Group Coflex, VAS reduced from 7.8±1.2 before surgery to 2.1±0.6 at final follow-up; ODI from 30.8±3.2 to 4.6±1.2; and ROM of lumbar spinal segments were improved. ADH decreased, PDH and FH increased after operation (P < 0.01). In Group fusion, VAS reduced from 7.2±1.1 to 2.0±0.6; ODI from 29.9±3.0 to 4.5±0.9,and there was a significant increase in ROM of L3/4 at final follow-up (P< 0.01). ADH, PDH and FH increased after operation (P<0.01). There was no significant difference on clinical effect between two groups.Conclusions:1. The finite element model was validated and could be used on biomechanics test.2. ROM, IDS and FJP of implanted segments dynamicly fixed with Coflex were reduced partly, but ROM and IDS of the adjacent segments weren't influnced significantly, and FJP of adjacent segments were increased.3. After surgery fixed with Coflex, the stresses of Coflex device and spinous processes were increased, and the risk of fracture of spinous process and fatigue break of Coflex implant is increased accordingly.4. The short-term clinical effect of dynamic fixation with Coflex on LDH was satisfying, and now fixation with Coflex could be one of available treatments.