Degenerativecervical myelopathy (DCM) is a major cause of adult spinal cord dysfunction that develops due to osteoligamentous degeneration of the spine and associated cervical spinal cord compression.1除了静态颈脊髓受压sion, neck motion in DCM patients increases spinal cord stress and strain, which further contributes to spinal cord dysfunction.2–4During neck flexion, the spinal cord has been shown to lengthen longitudinally and narrow in the anteroposterior direction, and this spinal cord stretch (strain) is further accentuated by ventral spinal cord compression.2In neck extension, anteroposterior spinal cord compression (stress) can be exacerbated by shingling of the laminae and infolding of the ligamentum flavum,5especially in cases with coexisting ventral compression. Presurgical neck mobility is correlated with clinical function in DCM.6The effect of stress/strain on spinal cord dysfunction in DCM is further supported by the description of cellular pathways that mediate stretch-associated axonal7and myelin8injuries.
Posterior cervical spinal cord decompression with or without spinal fusion is a common treatment for DCM, and the three major posterior cervical approaches include laminectomy, laminoplasty, and laminectomy with fusion. All three surgical procedures decompress the spinal cord dorsally by removing laminae and the ligamentum flavum; however, they have different effects on neck motion. While laminectomy and laminoplasty are considered motion-preserving surgeries, laminectomy with fusion considerably reduces neck motion.9Prior studies have shown that restriction of cervical range of motion (ROM) after posterior cervical surgery is associated with improved neurological outcomes,10and spinal cord biomechanics may contribute to inconsistent neurological recovery after surgery for DCM. Given the unique effects of laminectomy, laminoplasty, and laminectomy with fusion on neck motion, it is expected that each surgical approach will have a distinct impact on spinal cord biomechanics that can affect neurological recovery.
Direct measurement of human spinal cord stress and strain after surgery is not currently feasible. Finite element models (FEMs) of the spinal cord can assess spinal cord biomechanics and predict postsurgical spinal cord stress and strain during simulated neck motion.11–14Finite element modeling is an established technique to understand the biomechanics of complex systems such as the human spine and spinal cord. Individual elements with specific material properties recreate complex anatomical geometries in an FEM. Computational analyses are then conducted with simulated loads to evaluate biomechanical responses of the tissues. FEM studies overcome the limitations of cadaver models, which have intrinsic biological variability. FEMs can be used to measure tissue biomechanics where clinical testing is not feasible, especially when testing the efficacy of multiple potential surgical options for a given patient’s spinal disorder or anatomy. Since spinal alignment, geometry, and spinal cord morphology can impact spinal cord biomechanics, these data can be incorporated into FEMs to predict patient-specific responses prior to surgery and boost clinical translation.15但是,没有事先研究脊髓b相比iomechanics between the different posterior surgical approaches for DCM. Our hypothesis was that laminectomy with fusion will show greater reductions in spinal cord stress/strain compared with motion-preserving surgeries such as laminectomy and laminoplasty. The aim of this study was to compare spinal cord biomechanics after laminectomy with fusion, laminectomy, and laminoplasty using a patient-specific FEM for DCM.
开云体育世界杯赔率
Patient-Specific Cervical Column FEM Development
Presurgical MRI (Fig. 1) data from a volunteer with mild DCM (modified Japanese Orthopaedic Association score 16) were used to develop a 3D patient-specific FEM of the human cervical spine and spinal cord. The patient was diagnosed on the basis of motor dysfunction of the upper and lower limbs and mild sensory loss in the upper limbs. A T2-weighted fast spin echo image of the cervical spine was acquired using a 3T GE Premiere scanner with the following parameters: TR 2500 msec, TE 122 msec, and 3D isotropic resolution of 0.8 × 0.8 × 0.8 mm. The research MRI and procedures of this study were approved by the institutional review board, and informed consent was obtained from the patient.
A–C:Presurgical sagittal T2-weighted MR image (A), presurgical patient-specific FEM (B), and postsurgical sagittal FEM for laminectomy (C).D–F:Coronal views of the FEMs for laminectomy and fusion (D), laminectomy (E), and laminoplasty (F). Figure is available in color online only.
Osteoligamentous spinal geometries were measured on MRI using the RadiAnt DICOM viewer (Medixant) as described previously.15Disc degeneration at each spinal level was graded based on T2-weighted MRI.16The vertebral width and lateral mass width were measured on axial MR images. Using these dimensions, the previously validated generic spine FEM17–19was morphed into a patient-specific FEM. Grading of disc degeneration, measured on MRI,16was incorporated into the model. With the severity of degeneration, the segmental stiffness at the most degenerated level (C5–6) increased, and intervertebral disc bulging and annulus stress and strain decreased.20A total of 44,799 hexahedral elements and 35,679 quadrilateral elements were used in the cervical spine model.
Generic FEM
The creation of our generic cervical spine FEM has been described previously.21,22For the spinal cord FEM, we used normative MRI-derived cervical spinal canal measurements,23as well as material properties of the human spinal cord and adjacent structures. The generic model did not exhibit spinal canal stenosis or spinal cord compression. The morphable generic spine FEM was validated from human cadaver cervical columns, with the model-predicted ROMs at all segmental levels within a mean ± 1 standard deviation.24,25
Patient-Specific Cervical Spinal Cord FEM Development
Development of the patient-specific cervical spinal cord FEM has been described previously.15Human material properties used in the spinal cord components were obtained from previous studies.15The presurgical patient-specific FEM was created by inserting the patient-specific spinal cord FEM into the patient-specific cervical column FEM. The spinal cord was unconstrained within the spinal canal.
Surgical Simulation
Different posterior surgeries were performed to decompress the spinal cord at index levels. The spine surgeon (A.V.) supervised the simulation of the operation, which was based on the method used in the actual in vivo situation. Titanium alloy hardware was modeled using previously published material properties: elastic modulus of 110 GPa and a Poisson ratio of 0.3.26The spinal cord was manually moved dorsally (average of 2 mm) at the level of decompression to reflect the predicted spinal cord shift, which occurs after posterior decompression for DCM.27To determine the effect of spinal cord shift after decompression, we compared outputs for 1-, 2-, and 3-mm dorsal spinal cord shift at the level of decompression for the laminoplasty model during flexion and extension.
Open-Door Laminoplasty
An open-door laminoplasty was simulated by creating a full-thickness trough on one side of the C4, C5, and C6 laminae at the junction of the lamina and the lateral mass of the patient-specific vertebral mesh. On the contralateral side, a partial-thickness trough was simulated to produce a hinge along the lamina-facet junction. The interspinous ligaments and ligamentum flavum were disconnected at the C3–4 and C6–7 levels. The C4, C5, and C6 laminae were turned 13°–16° toward the hinge, and the laminoplasty was kept open using a titanium plate. Hexahedral elements were used to mesh the titanium plate using ANSA software (BETA CAE Systems). The 1-mm-thick double-bend plate (ARCH laminoplasty system, DePuy Synthes) was bent to establish the best anatomical fit. Tight contact was formulated at the interface between the lateral mass/lamina and the laminoplasty plates.
Laminectomy
This surgical technique was simulated by removing the spinous process, lamina, ligamentum flavum, and interspinous ligaments from the preoperative patient-specific model at the concerned levels (C4–6). The ligamentum flavum and interspinous ligaments were also removed at adjacent levels (C3–4 and C6–7) to complete the C4–6 laminectomy.
Laminectomy and Fusion
The laminectomy model (C4–6) described above was used as a baseline for the creation of the laminectomy and fusion (C3–7) model. Using CAD software (CATIA V6, Dassault Systèmes), a 3.5-mm-diameter titanium longitudinal connecting rod bending from 0° to 20°, set screw, and screw head were modeled. The rod was bent to provide proper contact between the rod and lateral mass. The interaction between the screw sets and bone interface were simulated as a tight constraint to simulate rigid connection between the screw and rod.
Loading and Boundary Conditions
The FEMs were exercised under flexion-extension (sagittal bending), lateral bending, and axial rotation. The bending moments of 2 Nm applied in the sagittal plane combined with a follower load of 75 N were applied to pre- and postoperative FEMs.28Follower cables were used to apply the follower load as described by Patwardhan et al.29No follower load was applied for lateral bending and axial rotation modes as previously recommended.25The inferior surface of the T1 vertebra was completely constrained in all 6 degrees of freedom.
Spinal cord stress (von Mises stress) and strain (maximum principal strain) were recorded during simulated neck motion. Segmental ROM, capsular ligament strain, and intradiscal pressure were also measured for each surgical simulation and compared with the preoperative patient-specific FEM. The average stress/strain values were recorded after eliminating values below the 5th percentile and values above the 95th percentile. Nonparametric tests were used to compare outputs between surgical FEMs using IBM SPSS version 23 (IBM Corp.).
Results
Comparison Between the Generic FEM and Preoperative Patient-Specific Model
Spinal cord stress/strain was increased in the presence of spinal cord compression, and this effect was seen even at multiple spinal levels and most prominently during neck extension (Fig. 2).
Spinal cord von Mises stress and maximum principal strain during simulated neck flexion and extension in a generic FEM compared with a patient-specific FEM in DCM. Figure is available in color online only.
Validation of Surgical Models
Laminectomy and laminoplasty FEMs were validated for segmental ROM using data from a prior cadaver study by Kode et al.30Segmental ROMs (Fig. 3椎板切除术和板成形术《w)ithin 1 standard deviation for all loading modes (flexion-extension, lateral bending, and axial rotation). Similarly, segmental ROM for the fused and unfused segments for the laminectomy with fusion FEM were within 1 standard deviation of a prior cadaver study by Kretzer et al.31
Bar graphs showing that the ROMs of our surgical FEMs are within 1 standard deviation of previously published cadaver studies. Figure is available in color online only.
Spinal Cord Stress
Spinal cord stress at the decompressed levels was reduced in neck extension for both laminectomy with fusion and laminoplasty but increased for laminectomy (Fig. 4). For the surgical levels, the average changes in stress were −30.4%, +37.8%, and −12.6% for laminectomy with fusion, laminectomy, and laminoplasty, respectively, in neck extension. In neck flexion, the average changes in stress across the surgical levels were +2.9%, +50.0%, and +57.4% for laminectomy with fusion, laminectomy, and laminoplasty, respectively. Spinal cord stress at the surgical levels for lateral bending was reduced for all three models (laminectomy with fusion: −28%; laminectomy: −17%; and laminoplasty: −12%). Axial rotation at the surgical levels was associated with reduced spinal cord stress after laminectomy with fusion (−38%) and laminoplasty (−3%) but increased after laminectomy (+23%).
Distribution of spinal cord stress and strain across the cervical spinal cord during neck motion in laminectomy with fusion (LF), laminectomy (LN), and laminoplasty (LP) FEMs. Figure is available in color online only.
Spinal Cord Strain
Spinal cord strain was increased in neck extension for all three surgeries but decreased in flexion (Fig. 4). The increase in strain with neck extension was maximum for the laminectomy model. For the surgical levels, the average changes in strain were +21.9%, +118.4%, and +24.6% in neck extension for laminectomy with fusion, laminectomy, and laminoplasty, respectively. In neck flexion, the average changes in strain across the surgical levels were −34.4%, −7.9%, and 2.1% for laminectomy with fusion, laminectomy, and laminoplasty, respectively. Spinal cord strain was increased across the surgical levels for lateral bending (laminectomy: +26.1%; and laminoplasty: +56.7%) and axial rotation (laminectomy: +75.1%; and laminoplasty: +20.9%) in the laminectomy and laminoplasty models. Spinal cord strain was reduced for both lateral bending (−33.8%) and axial rotation (−32.1%) in the laminectomy with fusion model.
Segmental ROM
Segmental ROM was reduced for laminectomy with fusion and laminoplasty most prominently at the decompressed levels (Fig. 5). The mean decrease in ROM at the surgical levels (C4–5, C5–6, and C6–7) was higher for laminectomy with fusion (flexion: −76.5% vs −43.1%; and extension: −89.3% vs −55.4%) compared with laminoplasty. At the levels of decompression, segmental ROM increased after laminectomy in neck extension (+5.7%) but not in flexion (−7.2%). In the sagittal bending modes, laminoplasty showed increased off-axis motion for axial rotation (flexion: +70%; and extension: +117%) and lateral bending (flexion: +446%; and extension: +327%) compared with laminectomy with fusion and laminectomy. For lateral bending, ROM at the surgical levels was reduced for laminectomy with fusion (−40%) and increased for laminectomy (+36%) and laminoplasty (+21%). In axial rotation, ROM at the surgical levels was reduced by 74% for laminectomy with fusion and 18% for laminoplasty. For laminectomy, the ROM was increased by 23%.
Bar graphs showing segmental spinal cord stress (A), spinal cord strain (B), and ROM (C) for flexion-extension, lateral bending, and axial rotation. The decompressed levels are outlined by thedotted lines. Figure is available in color online only.
Capsular Ligament Strain
Capsular ligament strain in the preoperative model showed asymmetrical changes: higher on the left at C5–6 in extension, but higher on the right at C4–5 and C6–7 (flexion and extension) and C5–6 in flexion (Table 1). Mean capsular ligament strain at the operated levels (C4–5, C5–6, and C6–7) was reduced for laminectomy with fusion and laminoplasty bilaterally. In the laminoplasty model, there was a decrease in mean capsular ligament strain bilaterally at the operated levels. For laminectomy, there was an increase in the mean capsular ligament strain at the operated levels in extension (left: 18.6%; and right: 7.5%) compared with flexion as well as in axial rotation (left: 3.9%; and right: 3.2%).
Intradiscal pressure and capsular ligament strain during simulated neck motion for patient-specific laminectomy with fusion, laminectomy, and laminoplasty FEMs
| Laminectomy (C4–6) & Fusion (C3–7) | Laminectomy (C4–6) | Laminoplasty (C4–6) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Flexion | Extension | Lat Bending | Axial Rotation | Flexion | Extension | Lat Bending | Axial Rotation | Flexion | Extension | Lat Bending | Axial Rotation | |
| Intradiscal pressure | ||||||||||||
| C2–3 | 4.65 | −83.46 | −32.13 | −8.37 | −1.04 | 85.31 | 88.24 | 3.72 | 0.00 | −71.36 | 145.59 | 26.51 |
| C3–4 | 8.19 | −97.02 | −66.02 | −89.08 | 6.21 | 1.95 | 5.51 | 6.22 | 7.56 | −17.04 | 2.97 | 2.16 |
| C4–5 | −12.82 | 92.87 | −70.06 | −68.32 | −2.10 | −34.78 | −41.01 | −10.07 | −16.36 | 64.21 | −81.12 | −74.50 |
| C5–6 | −16.36 | −59.23 | −34.26 | −31.23 | −4.60 | −26.54 | −18.82 | −28.23 | −18.17 | 14.62 | −24.61 | −82.28 |
| C6-7 | −2.83 | −92.52 | 79.75 | −28.66 | −1.54 | 11.63 | −63.77 | 122.93 | −2.29 | 14.29 | −74.02 | 12.99 |
| C7–T1 | 4.63 | −4.03 | −1.96 | 6.14 | 1.39 | −1.10 | 4.21 | 8.68 | 6.03 | −3.55 | 13.39 | 27.95 |
| Capsular ligament strain (max principal strain) | ||||||||||||
| Lt side | ||||||||||||
| C2-3 | 30.54 | 7.95 | −5.47 | 3.74 | 4.84 | 2.30 | −4.00 | 10.21 | 13.55 | 9.83 | −10.95 | 0.08 |
| C3–4 | −31.25 | −76.18 | −76.53 | −98.36 | 29.58 | 21.41 | 11.47 | 18.06 | 40.90 | 30.71 | 12.04 | 7.22 |
| C4–5 | −77.89 | −87.56 | −92.13 | −97.65 | −0.98 | 37.31 | 1.26 | 2.41 | −86.30 | −57.71 | −87.59 | −91.89 |
| C5–6 | −54.09 | −48.78 | −28.27 | −89.42 | −19.50 | 9.06 | 9.88 | 8.89 | −71.07 | −45.64 | −66.10 | −93.46 |
| C6–7 | −85.52 | −62.15 | −60.73 | −76.00 | −18.61 | 9.38 | 4.18 | 0.27 | −3.84 | 36.81 | 140.30 | 10.92 |
| C7–T1 | −6.09 | 6.59 | 2.37 | −3.24 | −16.67 | 2.56 | 4.84 | −3.55 | −21.74 | 24.93 | 34.09 | 7.38 |
| Rt side | ||||||||||||
| C2-3 | 48.10 | 34.92 | 9.86 | 7.32 | 13.61 | 17.09 | 5.69 | 6.19 | 11.24 | 0.00 | 6.42 | 6.31 |
| C3–4 | −78.15 | −84.73 | −90.91 | −98.21 | 48.35 | −6.72 | 6.59 | 18.62 | 50.60 | −34.61 | 4.05 | 18.81 |
| C4–5 | −69.26 | 25.78 | −94.60 | −84.65 | −9.72 | 10.16 | −3.89 | −1.98 | −84.81 | −66.41 | −71.32 | −90.30 |
| C5–6 | −82.45 | −73.61 | −97.15 | −90.26 | −17.55 | 5.56 | −1.84 | 6.16 | −82.45 | −74.07 | −73.57 | −92.04 |
| C6–7 | −93.42 | −38.58 | −82.09 | −86.48 | −14.55 | 6.84 | −0.53 | 5.50 | 7.27 | −28.26 | 1.95 | 23.07 |
| C7–T1 | 13.84 | 14.58 | 1.72 | 5.75 | −1.55 | 10.49 | −2.74 | 1.13 | 13.02 | −4.82 | −1.31 | 10.15 |
All values are percentage changes with respect to the presurgical FEM.
Intradiscal Pressure
Intradiscal pressure at the surgical levels was lower in all bending modes for the laminectomy with fusion model. For the laminectomy model, intradiscal pressure decreased in all modes except for axial rotation, which was associated with an increase by 28.2%. In the laminoplasty model, lower disc pressures were noted in all modes except for neck extension, where the disc pressure increased by 31% (Table 1).
Comparison of Spinal Cord Biomechanics for Laminectomy and Laminoplasty Versus Laminectomy With Fusion
Compared with laminectomy and laminoplasty, spinal cord biomechanics for laminectomy with fusion revealed significantly reduced median extension stress (13.7 kPa vs 9.7 kPa, p = 0.03), lateral bending strain (0.01 vs 0.007, p = 0.007), axial rotation stress (3.7 kPa vs 2.1 kPa, p = 0.04), and axial rotation strain (0.017 vs 0.009, p = 0.04). Stress and strain were elevated in the ventral and lateral funiculi in the preoperative model. The spatial distribution of high stress/strain persisted for laminoplasty and laminectomy models, most prominently during neck extension (Fig. 6). In the laminectomy with fusion model, ventral and ventrolateral stress and strain were reduced during neck extension.
Spatial distribution of spinal cord von Mises stress and maximum principal strain at the C4–5 level comparing preoperative and postoperative FEMs. Figure is available in color online only.
Effect of Spinal Cord Shift
Stress and strain were consistently higher for 2- and 3-mm dorsal shift compared with 1-mm shift at all spinal levels including the levels of decompression (Fig. 7).
Comparison of spinal cord stress/strain for 1-, 2-, and 3-mm dorsal spinal cord shift after laminoplasty showing greater spinal cord stress/strain for 2- and 3-mm shift compared with 1-mm shift. Figure is available in color online only.
Discussion
本研究使用一个特定的有限元法来测量the unique effect of posterior cervical surgery on spinal cord biomechanics in DCM. All surgical approaches were associated with increased spinal cord stress in flexion and increased spinal cord strain in extension, although laminectomy with fusion was associated with the smallest increase. Laminoplasty was associated with reduced segmental ROM in flexion and extension but higher segmental ROM for lateral bending. Laminectomy was associated with persistent elevations in spinal cord stress, which was associated with an increase in segmental ROM. Compared with motion-preserving surgeries (i.e., laminectomy and laminoplasty), laminectomy with fusion was associated with significantly lower spinal cord stress/strain in extension as well as lateral bending and axial rotation modes.
Laminectomy, laminoplasty, and laminectomy with fusion have distinct effects on segmental ROM in the cervical spine. Complete resection of the posterior lamina, interspinous ligaments, and ligamentum flavum in laminectomy contributes to an increase in segmental ROM, most prominently with neck extension. Although open-door laminoplasty is considered a motion-preserving surgery, this surgery results in a decrease in segmental ROM from 12% to 46%.32–34Progressive ankylosis of the joints is known to further reduce the ROM in laminoplasty at long-term follow-up.35Comparatively, laminectomy with fusion is associated with substantial reduction in segmental ROM across the fused segments early after surgery. Our FEMs also demonstrated that open-door laminoplasty induces off-axis motion in sagittal bending modes. This is likely a result of the unilateral full-thickness laminotomy on one side and resection of the ligamentum flavum at the top and bottom of the laminoplasty construct as well as asymmetrical facet arthropathy seen in patients. Although the clinical impact of changes in ROM and spinal cord biomechanics after laminoplasty need to be explored further, Morio et al.10showed that reduced mobility at C4–5 after laminoplasty is associated with improved symptoms of myelopathy.
The changes in disc pressure in our model were impacted by the change in material properties induced by disc degeneration,16which were incorporated into the model. Prior studies have shown that multilevel laminectomy is associated with increased sagittal ROM, while laminoplasty causes moderate reductions in ROM in sagittal bending but increased ROM in lateral bending.30Higher capsular strain in our laminectomy model in extension is likely due to higher segmental ROM induced by resection of osteoligamentous structures. Asymmetrical changes in capsular ligament strain in our FEM further emphasize the value of patient-specific FEMs where geometry of facet joints is not identical bilaterally. Increased capsular strains also create laxity in the joint over time and may contribute to an increase in mobility after laminectomy. These changes may further exacerbate presurgical facet arthropathy. Further studies are necessary to determine if capsular strain and disc pressure measurements can better optimize surgical decision-making.
Posterior cervical surgery for DCM removes dorsal compression of the spinal cord. This is associated with a dorsal shift of the spinal cord and reduction in spinal cord compression. A number of factors impact postoperative spinal cord biomechanics, including levels of decompression, sagittal alignment, neck motion, and magnitude of dorsal spinal cord shift. Although dorsal shift of the spinal cord allows for indirect ventral decompression after posterior cervical surgery, we found that the magnitude of shift after laminoplasty is associated with distinct changes in spinal cord biomechanics. While the magnitude of shift on MRI is greater for laminectomy with fusion than for laminoplasty (possibly because of complete removal of the ligamentum flavum and spinal lamina in laminectomy with fusion), it has not been shown to correlate with clinical outcomes.27Our FEM showed that this shift was associated with decreased spinal cord stress in extension. Ventral compression, due to disc material or osteophytes, which are not addressed with posterior cervical surgery, can further exacerbate spinal cord stress during neck flexion.2Preserved ROM after laminectomy contributes to persistent adverse spinal cord tension, and this may explain late deterioration or plateauing of recovery that can be seen after laminectomy for DCM.36Prior FEM studies have shown higher spinal cord stress and strain in flexion after laminectomy, especially with more severe cord compression.37Increased spinal cord stress/strain was also noted at unoperated levels in our FEMs. Spinal cord stress/strain at unoperated levels may also be due to adjacent-segment effect, mild stenosis, ventral disc protrusions, and sagittal alignment. Changes in spinal cord stress/strain at C7–T1 are likely because the T1 vertebra was constrained in our model. Constraining the spinal column at T1 has been performed in other cervical spine FEMs28with the rationale that the motion inferior to this level is small compared with the motions cephalad because of the presence of the rib cage. Compared with the preoperative model, spinal cord stress was elevated in flexion, while strain was elevated in extension at all adjacent segments for the laminectomy with fusion model. However, the magnitude of adjacent-segment changes in spinal cord biomechanics was lower for laminectomy with fusion than for laminoplasty and laminectomy. This effect may be due to some of the limitations of the FEM approach. Neck ROM, which is a major determinant of spinal cord stress/strain, was not standardized between the three models. Only the applied load was standardized, and since the laminoplasty and laminectomy had higher ROMs than laminectomy with fusion (Fig. 4), these models also showed greater spinal cord stress and strain in the adjacent segments. The FEM approach does not include time-dependent adjacent-segment disc degeneration and spondylosis that would impact spinal cord stress/strain. Therefore, our FEMs only reflect an early postoperative simulation. It is likely that progressive adjacent-segment disc degeneration would contribute to increased adjacent-segment cord compression and higher spinal cord stress and strain after laminectomy with fusion over time; however, this is not reflected in the current FEM approach.
The use of a patient-specific FEM in this study further enhances clinical translation to assist with surgical decision-making. Prior spinal cord FEMs for DCM use generic simulations of spinal cord compression to quantify intrinsic spinal cord stress.11,13Our patient-specific model showed a substantial increase in spinal cord stress/strain across all spinal levels, indicating a global effect on spinal cord biomechanics likely related to the alignment and osteoligamentous geometries. These results show that estimation of clinically relevant spinal cord biomechanics necessitates the incorporation of patient-specific geometries for FEM studies in DCM. Our model incorporates spinal alignment and geometries specific to the patient, and our modeling method reduces time needed to create a patient-specific cervical spine and spinal cord FEM. Our FEM includes all anatomical structures around the spinal cord such as dura mater, pia mater, denticulate ligaments, and CSF. The increased anatomical detail compared with prior generic FEMs further improves accuracy of spinal cord biomechanical measurements.11–13The original intact spine FEM was validated with segmental motions at all levels of the spine under the same loading modes.24The postsurgical FEMs were validated for segmental ROMs using data from prior cadaver studies30,31for laminoplasty, laminectomy, and laminectomy with fusion, and our responses were within 1 standard deviation for all loading modes, which is within the range described in other FEM studies.24,38From a true validation perspective, we did not validate this patient-specific model, since segmental motions after surgery were not measured for this patient. In addition, we simulated multiple surgeries with this model, and it is not possible to obtain validation data as only one surgery was performed in the patient. A future study will be conducted to obtain pre- and postsurgical segmental ROM from a DCM patient and use it to validate the DCM model. Since it is not feasible to directly measure spinal cord stress/strain in humans, we could not validate the spinal cord biomechanics outputs.
未来的工作将集中在临床correl定义ates of spinal cord biomechanics measured using a patient-specific FEM. Our FEM technique can help with surgical decision-making for the individual patient since it incorporates patient-specific geometry. Surgeons can use simulated spinal cord biomechanics to determine whether a motion-sparing approach will yield a reduction in postsurgical spinal cord stress/strain. In patients with mild to moderate stenosis at levels adjacent to a planned fusion, patient-specific FEMs can predict adjacent-segment spinal cord stress/strain that may warrant inclusion of these levels in the initial construct. Since inadequate correction of sagittal balance is associated with poorer outcomes after surgery for DCM,39predicted spinal cord stress/strain can assist with planning the extent of surgical correction of kyphosis. Although not performed in this study, future work will compare anterior and posterior approaches for DCM to assist with surgical choice in cases with equipoise. Patient-specific FEMs yield opportunities for a personalized medicine approach for spinal cord biomechanics in DCM. The patient-specific approach was based on the actual 3D pathological anatomy of a DCM patient. This preliminary numerical study with a single case shows a proof of concept for a personalized medicine approach to surgical treatment of DCM. Although spinal cord biomechanics contributes to spinal cord pathology in DCM, the inability to estimate these responses for the individual patient limits the incorporation of these data into surgical decision-making. Using our method, it is possible to apply computational biomechanics to the individual DCM patient and determine the effects of different surgical procedures on spinal cord stress/strain. The results cannot be generalized to all DCM patients, and a future study with a larger cohort is necessary to create more generalizable results. Clinical correlates of spinal cord biomechanics need to be defined to better utilize these parameters in surgical decision-making. Dorsal shift of the spinal cord was made based on prior imaging studies;27however, this shift may not accurately reflect the true shift after surgery for this patient. Patient-specific tissue properties of the spinal cord based on age and spinal cord degeneration are currently not available, but this information will further improve accuracy of spinal cord stress/strain measurements. The open-door laminoplasty technique is commonly performed at our center, but our FEM can also be used to simulate other laminoplasty techniques. To reduce confounders, we restricted the surgical approach to posterior cervical spine surgery. A future study will compare spinal cord biomechanics between anterior and posterior fusion surgeries for DCM.
Conclusions
Using a patient-specific FEM, we show that laminectomy, laminectomy with fusion, and laminoplasty are associated with distinct changes in spinal cord stress and strain. Compared with motion-preserving approaches such as laminectomy and laminoplasty, laminectomy with fusion was associated with lower spinal cord stress and strain.
Acknowledgments
Funding for this work was provided by AO Spine North America and North American Spine Society (to A.V.). The research was also supported by the Office of the Assistant Secretary of Defense for Health Affairs through the Broad Agency Announcement under award no. W81XWH-16-1-0010. It was also supported by the Department of Veterans Affairs Medical Research.
Disclosures
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
Author Contributions
Conception and design: Vedantam, Harinathan, Yoganandan. Acquisition of data: Harinathan, Warraich, Budde. Analysis and interpretation of data: Vedantam, Harinathan, Budde, Yoganandan. Drafting the article: Vedantam, Harinathan. Critically revising the article: Vedantam, Yoganandan. Reviewed submitted version of manuscript: Vedantam, Yoganandan. Approved the final version of the manuscript on behalf of all authors: Vedantam. Administrative/technical/material support: Vedantam, Harinathan. Study supervision: Vedantam, Yoganandan.
References
-
1 ↑
BadhiwalaJH,AhujaCS,AkbarMA,et al.Degenerative cervical myelopathy—update and future directions.Nat Rev Neurol.2020;16(2):108–124.
-
2 ↑
BreigA,TurnbullI,HasslerO.Effects of mechanical stresses on the spinal cord in cervical spondylosis. A study on fresh cadaver material.J Neurosurg.1966;25(1):45–56.
-
3
FehlingsMG,SkafG.A review of the pathophysiology of cervical spondylotic myelopathy with insights for potential novel mechanisms drawn from traumatic spinal cord injury.Spine (Phila Pa 1976).1998;23(24):2730–2737.
-
4 ↑
亨德森FC,GeddesJF,VaccaroAR,WoodardE,BerryKJ,BenzelEC.Stretch-associated injury in cervical spondylotic myelopathy: new concept and review.开云体育app官方网站下载入口.2005;56(5):1101–1113.
-
5 ↑
ChenIH,VasavadaA,PanjabiMM.Kinematics of the cervical spine canal: changes with sagittal plane loads.J Spinal Disord.1994;7(2):93–101.
-
6 ↑
LiuS,LafageR,SmithJS,et al.Impact of dynamic alignment, motion, and center of rotation on myelopathy grade and regional disability in cervical spondylotic myelopathy.J Neurosurg Spine.2015;23(6):690–700.
-
7 ↑
MaxwellWL,McCreathBJ,GrahamDI,GennarelliTA.Cytochemical evidence for redistribution of membrane pump calcium-ATPase and ecto-Ca-ATPase activity, and calcium influx in myelinated nerve fibres of the optic nerve after stretch injury.J Neurocytol.1995;24(12):925–942.
-
8 ↑
MaxwellWL,KosanlavitR,McCreathBJ,ReidO,GrahamDI.Freeze-fracture and cytochemical evidence for structural and functional alteration in the axolemma and myelin sheath of adult guinea pig optic nerve fibers after stretch injury.J Neurotrauma.1999;16(4):273–284.
-
9 ↑
RheeJM,BasraS.Posterior surgery for cervical myelopathy: laminectomy, laminectomy with fusion, and laminoplasty.Asian Spine J.2008;2(2):114–126.
-
10 ↑
MorioY,YamamotoK,TeshimaR,NagashimaH,HaginoH.Clinicoradiologic study of cervical laminoplasty with posterolateral fusion or bone graft.Spine (Phila Pa 1976).2000;25(2):190–196.
-
11 ↑
LévyS,BaucherG,RochePH,EvinM,CallotV,ArnouxPJ.Biomechanical comparison of spinal cord compression types occurring in degenerative cervical myelopathy.Clin Biomech (Bristol, Avon).2021;81:105174.
-
12
NishidaN,KanchikuT,ImajoY,et al.Stress analysis of the cervical spinal cord: impact of the morphology of spinal cord segments on stress.J Spinal Cord Med.2016;39(3):327–334.
-
13 ↑
StonerKE,Abode-IyamahKO,FredericksDC,ViljoenS,HowardMA,GroslandNM.A comprehensive finite element model of surgical treatment for cervical myelopathy.Clin Biomech (Bristol, Avon).2020;74:79–86.
-
14 ↑
YangS,QuL,YuanL,et al.Finite element analysis of spinal cord stress in a single segment cervical spondylotic myelopathy.Front Surg.2022;9:849096.
-
15 ↑
VedantamA,PurushothamanY,HarinathanB,ScrippS,BuddeMD,YoganandanN.Spinal cord stress after anterior cervical diskectomy and fusion: results from a patient-specific finite element model.Ann Biomed Eng.Published online December 20, 2022. doi:10.1007/s10439-022-03118
-
16 ↑
SuzukiA,DaubsMD,HayashiT,et al.Magnetic resonance classification system of cervical intervertebral disk degeneration: its validity and meaning.Clin Spine Surg.2017;30(5):E547–E553.
-
17 ↑
ChoiH,PurushothamanY,BaisdenJ,YoganandanN.Unique biomechanical signatures of Bryan, Prodisc C, and Prestige LP cervical disc replacements: a finite element modelling study.Eur Spine J.2020;29(11):2631–2639.
-
18
YoganandanN,PurushothamanY,ChoiH,JebaseelanD,BaisdenJ.Biomechanical effects of uncinate process excision in cervical disc arthroplasty.Clin Biomech (Bristol, Avon).2021;89:105451.
-
19 ↑
JohnJD,ArunMWJ,YoganandanN,KurpadSN,GurunathanSK.Mapping block-based morphing for subject-specific spine finite element models.Biomed Sci Instrum.2017;53(1):1–6.
-
20 ↑
KumaresanS,YoganandanN,PintarFA,MaimanDJ,GoelVK.椎间盘变性,骨赘fo的贡献rmation in the cervical spine: a biomechanical investigation.J Orthop Res.2001;19(5):977–984.
-
21 ↑
JohnJD,ArunMWJ,GurunathanSK,YoganandanN.颈椎有限元模型with anatomically accurate asymmetric intervertebral discs.Paper presented at: Summer Biomechanics, Bioengineering and Biotransport Conference;June 21-24, 2017;Tucson, AZ.
-
22 ↑
JohnJD,Saravana KumarG,YoganandanN.Cervical spine morphology and ligament property variations: a finite element study of their influence on sagittal bending characteristics.J Biomech.2019;85:18–26.
-
23 ↑
UlbrichEJ,SchranerC,BoeschC,et al.Normative MR cervical spinal canal dimensions.Radiology.2014;271(1):172–182.
-
24 ↑
WheeldonJA,StemperBD,YoganandanN,PintarFA.Validation of a finite element model of the young normal lower cervical spine.Ann Biomed Eng.2008;36(9):1458–1469.
-
25 ↑
BarreyC,RousseauMA,PersohnS,CampanaS,PerrinG,SkalliW.Relevance of using a compressive preload in the cervical spine: an experimental and numerical simulating investigation.Eur J Orthop Surg Traumatol.2015;25(suppl 1):S155–S165.
-
26 ↑
TanLA,YoganandanN,ChoiH,PurushothamanY,JebaseelanD,BoscoA.Biomechanical analysis of 3-level anterior cervical discectomy and fusion under physiologic loads using a finite element model.Neurospine.2022;19(2):385–392.
-
27 ↑
AshanaAO,AjiboyeRM,SheppardWL,et al.Spinal cord drift following laminoplasty versus laminectomy and fusion for cervical spondylotic myelopathy.Int J Spine Surg.2021;15(2):205–212.
-
28 ↑
KodeS,KallemeynNA,SmuckerJD,FredericksDC,GroslandNM.The effect of multi-level laminoplasty and laminectomy on the biomechanics of the cervical spine: a finite element study.Iowa Orthop J.2014;34:150–157.
-
29 ↑
PatwardhanAG,HaveyRM,GhanayemAJ,et al.Load-carrying capacity of the human cervical spine in compression is increased under a follower load.Spine (Phila Pa 1976).2000;25(12):1548–1554.
-
30 ↑
KodeS,GandhiAA,FredericksDC,GroslandNM,SmuckerJD.Effect of multilevel open-door laminoplasty and laminectomy on flexibility of the cervical spine: an experimental investigation.Spine (Phila Pa 1976).2012;37(19):E1165–E1170.
-
31 ↑
KretzerRM,HsuW,HuN,et al.Adjacent-level range of motion and intradiscal pressure after posterior cervical decompression and fixation: an in vitro human cadaveric model.Spine (Phila Pa 1976).2012;37(13):E778–E785.
-
32 ↑
ManzanoGR,CasellaG,WangMY,VanniS,LeviAD.A prospective, randomized trial comparing expansile cervical laminoplasty and cervical laminectomy and fusion for multilevel cervical myelopathy.开云体育app官方网站下载入口.2012;70(2):264–277.
-
33
MachinoM,YukawaY,HidaT,et al.Cervical alignment and range of motion after laminoplasty: radiographical data from more than 500 cases with cervical spondylotic myelopathy and a review of the literature.Spine (Phila Pa 1976).2012;37(20):E1243–E1250.
-
34 ↑
NagamotoY,IwasakiM,SugiuraT,et al.In vivo 3D kinematic changes in the cervical spine after laminoplasty for cervical spondylotic myelopathy.J Neurosurg Spine.2014;21(3):417–424.
-
35 ↑
SeichiA,TakeshitaK,OhishiI,et al.Long-term results of double-door laminoplasty for cervical stenotic myelopathy.Spine (Phila Pa 1976).2001;26(5):479–487.
-
36 ↑
KatoY,IwasakiM,FujiT,YonenobuK,OchiT.Long-term follow-up results of laminectomy for cervical myelopathy caused by ossification of the posterior longitudinal ligament.J Neurosurg.1998;89(2):217–223.
-
37 ↑
KhuyagbaatarB,KimK,ParkWM,LeeS,KimYH.Increased stress and strain on the spinal cord due to ossification of the posterior longitudinal ligament in the cervical spine under flexion after laminectomy.Proc Inst Mech Eng H.2017;231(9):898–906.
-
38 ↑
PanzerMB,CroninDS.C4-C5 segment finite element model development, validation, and load-sharing investigation.J Biomech.2009;42(4):480–490.
-
39 ↑
RoguskiM,BenzelEC,CurranJN,et al.Postoperative cervical sagittal imbalance negatively affects outcomes after surgery for cervical spondylotic myelopathy.Spine (Phila Pa 1976).2014;39(25):2070–2077.
