Cervical compressive myelopathy (CCM) is a common cause of neurological disability.
Chronic spinal cord compression is mainly caused by chronic degenerative changes in the cervical spine, disk herniation, ossification of the posterior longitudinal ligament, and ossification of the ligamentum flavum, resulting in slowly progressive impairment in motor and sensory functions of the limbs 1–3. Currently, surgical decompression can attenuate the progression of CCM, but most patients are still left with significant neurological impairment 4. No effective pharmacological treatment is available to improve neurological disability in patients with postoperative persistence of neurological deficits and patients who are unable to tolerate surgical treatment. The detailed pathophysiological mechanisms of CCM are poorly understood.
Previously described animal models of CCM are mostly a replacement of acute compression methods using impactors such as the Infinite Horizon impactor and the New York University impactor 5. Few reports are available on animal models of ‘chronic’ or ‘slow-progressive’ compressive myelopathy. Kim et al. 6 and Long et al. 7 reported a rat model of chronic compressive myelopathy that uses a water-absorbing polymer sheet inserted under a spinal lamina. Kim et al. 6 used a sheet measuring 0.7 mm×3 mm×5 mm that expands in volume by 230%. By contrast, Long et al. 7 used a sheet measuring 1 mm×3 mm×1 mm that expands in volume by 700%. The optimal size and coefficient of expansion of the polymer sheet have not yet been established. The aim of the present study was to optimize these properties in an ideal rat model of CCM using a water-absorbing polymer sheet.
Materials and methods
We used 30 eight-week-old female Sprague-Dawley rats in the present study. Rats were anesthetized with sevoflurane and fixed in a prone position. An incision was made in the middle of the cervical segments. The C4–C7 laminae were exposed and the C6 lamina was removed. A sheet of water-absorbing polymer was inserted into the C4–C5 sublamina space using a surgical microscope (Fig. 1). We sought to produce an experimental model of myelopathy by static spinal cord compression from the sheet expanding gradually and dynamic compression as a result of flexion and extension of the cervical spine.
We used an expandable sheet made of a water-absorbing polyurethane elastomer. We used Aquaprene Dx (Sanyo Chemical Industries, Kyoto, Japan), which expands in volume by 200%, and Sealer Tape BW (Azuma Gum Industries Company Limited, Gunma, Japan), which expands in volume by 350%. These polymer sheets expand after implantation, absorbing tissue water for 48–72 h to reach their maximum volume. These expandable materials did not show any soft tissue reaction.
Rats were divided randomly into five experimental groups (n=6 per group) by the expansion rate and thickness of the sheets. Sheets that expanded in volume by 200% (Aquaprene Dx; Sanyo Chemical Industries) were placed in rats in groups 1 (3 mm×5 mm×0.5 mm) and 2 (3 mm×5 mm×0.7 mm). Sheets that expanded in volume by 350% (Sealer Tape BW; Azuma Gum Industries Company Limited) were placed in rats in groups 3 (3 mm×5 mm×0.5 mm) and 4 (3 mm×5 mm×0.7 mm). A sheet (3 mm×5 mm×0.7 mm) was placed in the C4–C5 sublamina space momentarily in rats in the sham control group and then removed.
After the surgery, we evaluated the severity of paralysis because of spinal cord compression for 12 weeks. Locomotor function of the forepaws was assessed using a Forelimb Locomotor Scale (FLS) score (0–17 points) 8 and that of the hind paws using a Basso, Beattie, and Bresnahan locomotor rating scale (BBB) score (0–21 points) 9. Behavioral evaluations were performed before surgery, 1 and 4 days, and 1, 2, 3, 4, 6, 8, 10, and 12 weeks after surgery.
Twelve weeks after surgery, the rats were anesthetized with pentobarbital intraperitoneally. Transcardial perfusion with 4% paraformaldehyde in PBS was used to fix tissue. The cervical spine was removed and postfixed in the same fixative overnight and stored in 20% sucrose in PBS at 4°C. The cervical cord was embedded in optimal cutting temperature compound (Sakura Fine Technical, Tokyo, Japan). The cryoprotected samples were frozen at −80°C.
Transverse sections (10 µm, 10 sections) at the C4/C5 and the C7 segment of the spinal cord were used. These sections at the C4/C5 segment were stained with cresyl violet and sections at the C7 segment were stained with luxol fast blue (LFB). Micrographs were produced using a microscope fitted with a digital camera. The number of motor neurons in the anterior horn at the C4–C5 segment was counted. The LFB-positive area in the corticospinal tract at the C7 segment was calculated. To count motor neurons and calculate the LFB-positive area, we used image analysis software (Image J, version 1.6.0, Rasband WS; US National Institutes of Health, Bethesda, Maryland, USA).
We measured Young’s modulus of the water-absorbing sheets and spinal cord to determine their hardness using a mechanical testing device. Eight-week-old Sprague-Dawley rats were anesthetized with pentobarbital intraperitoneally and their spinal cords were harvested. C1–C7 laminae were removed. The spinal cord was removed and cut into 5 mm longitudinal samples. Aquaprene Dx (expansion rate: 200%) and Sealer Tape BW (expansion rate 350%) were absorbed with water for a week and maximally expanded. We used spinal cords and four kinds of samples: a sheet with 200% expansion before water absorption, a sheet with 200% expansion after water absorption, a sheet with 350% expansion before water absorption, and a sheet with 350% expansion after water absorption (n=3). The anteroposterior and transverse diameters of each sample were measured using a digital caliper (Digimatic Caliper, Mitutoyo, Japan). Samples were loaded in a mechanical compression testing device (DCS-2000 Universal Testing Machine; Shimadzu Corp., Kyoto, Japan). The actuator was driven at 2 mm/min until 70% strain. The stress–strain curve of each sample was plotted from the results of the compression test. Young’s modulus of each sample was identified as the slope of a line tangential to the stress–strain curve 10. Irreversible neurological deficits have occurred after 30–65% spinal cord compression 1,11,12. A stress–strain curve for the spinal cord was constructed using the compression test. Young’s modulus for the spinal cord at 50–60% strain was measured. Subsequently, we measured Young’s modulus for water-absorbing sheets at a stress range of 50–60% strain of the spinal cord.
Statistical differences in the FLS score, BBB score, number of motor neurons in the anterior horn, LFB-positive area, and Young’s modulus were tested using Turkey’s honest significant difference test. P value less than 0.05 was considered significant. Statistical analyses were carried out using the JMP 10 software package (SAS Institute Inc., Cary, North Carolina, USA).
The BBB score and the FLS score did not change significantly in the control group during 12 weeks postoperatively (Fig. 2). The BBB score and the FLS score decreased at several days postoperatively, but recovered in several weeks in all groups, except for the control group. ‘Slow-progressive’ paralysis appeared in group 3 (0.5-mm-thick sheet with 200% expansion) at 8 weeks and in group 4 (0.7-mm-thick sheet with 200% expansion) at 4 weeks. At 4 and 6 weeks postoperatively, the average locomotor scores in group 4 were significantly lower than the scores in group 3 (P<0.05). At 12 weeks postoperatively, the average BBB score in group 3 was 10.4±1.2 and the FLS score was 8.8±1.8. The average BBB score in group 4 was 9.0±0.5 and the FLS score was 7.2±0.7. The average locomotor scores in these groups were significantly lower than the scores in group 1 (0.5 mm-thick sheet with 350% expansion), group 2 (0.7 mm-thick sheet with 350% expansion), and the control group (P<0.05). By contrast, from 1 day to 2 weeks postoperatively, fast progressive paralysis was observed and after 3 weeks, recovery from paralysis was observed in groups 1 and 2. At 12 weeks postoperatively, there was no significant difference in the locomotor scores between groups 1, 2, and the control group.
Macroscopic findings at 12 weeks postoperatively showed large erosion of the lamina and mild spinal cord compression despite the considerable swelling of the sheet with 350% expansion (Fig. 3b). By contrast, there was small erosion of the lamina and relatively severe spinal cord compression despite the smaller swelling of the sheet with 200% expansion (Fig. 3c).
Cresyl violet staining
Cavitation and a loss of motor neurons in the anterior horn were observed in all groups, except for the control group (Fig. 4). The number of motor neurons in groups 1–4 was significantly lower than that in the control group (P<0.05).
Luxol fast blue staining
A loss of the LFB-positive area and demyelination in the corticospinal tract were observed in group 3 (0.5-mm-thick sheet with 200% expansion) and group 4 (0.7-mm-thick sheet with 200% expansion) (Fig. 5). By contrast, demyelination was not observed in group 1 (0.5-mm-thick sheet with 350% expansion) or group 2 (0.7-mm-thick sheet with 350% expansion). The LFB-positive area in groups 3 and 4 was significantly lower than that in groups 1, 2, and the control group (P<0.05).
A stress–strain curve of the spinal cord was obtained using the compression test (Fig. 6a). From the stress–strain curve of the spinal cord, the average stress required for 50% and 60% compression of the spinal cord was 55±9.4 and 172±18 kPa, respectively. The stress–strain curve of each water-absorbing sheet sample was obtained (Fig. 6b) and Young’s modulus in the stress range of 50–60% strain of the spinal cord (55–172 kPa) was calculated (Fig. 6c). Young’s modulus required for 50–60% compression of the spinal cord was 1130±125 kPa. Young’s modulus for the sheet with 350% expansion was 3786±268 kPa before and 1119±26 kPa after water absorption. Young’s modulus for the sheet with 200% expansion was 3445±20 kPa before and 2021±93 kPa after water absorption. Young’s modulus for the sheet with 200% expansion after water absorption was significantly higher than that for the spinal cord (P<0.05). By contrast, there was no significant difference between Young’s modulus for the sheet with 350% expansion after water absorption and that for the spinal cord.
This study shows that the polymer sheet with 200% expansion after water absorption was suitable for modeling chronic compressive myelopathy. By contrast, rat models using polymer sheets with 350% expansion reflected temporary and acute myelopathy, which was not suitable to model chronic compressive myelopathy.
In rat models using the sheet with 200% expansion, deterioration occurred temporarily because of surgical invasion, but progressive paralysis occurred after 4–8 weeks postoperatively. Previous reports of a rat model of chronic compressive myelopathy using 3 mm×5 mm×0.7 mm water-absorbing sheets with 230% expansion were considered 6,13,14. Kurokawa et al. 13 reported that the ability for treadmill exercise by the rats declined after 6 weeks postoperatively. Yamamoto et al. 14 reported that deterioration in grip strength occurred after surgery and recovered in several weeks, and progressive weakness was observed 7–8 weeks postoperatively. In the present study, the results of behavioral evaluations in the groups using sheets with 200% expansion were consistent with these previous reports.
We found loss of motor neurons in the anterior horn and demyelination in the corticospinal tract at 12 weeks postoperatively in rat models using sheets with 200% expansion. Histopathology of cervical spondylotic myelopathy at human autopsy and of rat models showed cavitation and loss of motor neurons in the anterior horn and demyelination in the corticospinal tract 15–17. Our findings were consistent with these previously reported findings. Therefore, rat models using sheets with 200% expansion reflected slow-progressive myelopathy and were considered to be appropriate as models of chronic compressive myelopathy.
By contrast, temporary paralysis and no demyelination of corticospinal tract at 12 weeks postoperatively were observed in rat models using sheets with 350% expansion after water absorption. These findings were considered to be acute onset and temporary myelopathy because of surgical invasion and excessive expansion of sheets. Possible reasons for the temporary paralysis and recovery in models using sheets with large expansion are as follows: first, Kim et al. 6 reported the reasons for recovery of paralysis in a model of chronic compressive myelopathy, which included a partial reduction of the compression to the spinal cord by erosion and remodeling of the lamina overlying the polymer sheet. Another reason might be related to the hardness of the water-absorbing sheet after expansion. Mechanical testing showed that the sheet with 200% expansion after water absorption was harder than the spinal cord, but the sheet with 350% expansion after water absorption was not. Therefore, paralysis might have been temporary and recovered in models using sheets with 350% expansion because the sheet with 350% expansion after water absorption was not harder than the spinal cord. Paralysis might be slowly progressive in models using sheets with 200% expansion because the sheets with 200% expansion are harder than the spinal cord.
The behavioral results for the models using sheets with 200% expansion were an earlier appearance of paralysis in the subgroup of models using 0.7-mm-thick sheets than in the subgroup using 0.5-mm-thick sheets. However, temporary paralysis after surgery because of surgical invasion was greater in the subgroup of models using 0.7-mm-thick sheets. If the thickness of the sheet is more than 0.7 mm, invasion is even greater and the risk of iatrogenic spinal cord injury increases. In the unpublished data from preliminary experiments, spinal cord injury frequently occurred after inserting 1.0-mm-thick sheets under the laminae. Kubota et al. 18 reported a mouse model of subclinical spinal canal stenosis. The mouse model had no paralysis for 6 weeks after a 300-µm-thick sheet was inserted under the laminae. If the thickness of the sheet was less than 0.3 mm, the probability of paralysis was considerably less. Therefore, a thickness of 0.5–0.7 mm was considered suitable for rat models of chronic compressive myelopathy.
An animal model of CCM using a chronic compression device 19, and flat plastic screw implantation through spinal lamina 20, showed acute paralysis or progressive recovery several weeks after surgery. Long et al. 7 used a water-absorbing sheet with 700% expansion. However, their observation period was only 1 week and insufficient for chronic compressive myelopathy. These models did not reflect ‘chronic’ or ‘slow-progressive’ myelopathy. We considered that delayed and slow-progressive paralysis was necessary to model chronic compressive myelopathy.
Spondylosis and the dynamic factor were not considered in the present study. To reflect the pathophysiology of CCM of humans, it is necessary to develop a compressive myelopathy model that takes these factors into consideration.
A polymer sheet that expands in volume by 200% after water absorption is an ideal material for rat models of CCM. The hardness of the material inserted under the spinal lamina is important to reflect chronic compressive myelopathy.
This work was supported by grants-in-aid for Scientific Research from the Ministry of Education, Science and Culture of Japan (15561169).
Conflicts of interest
There are no conflicts of interest.
1. Fehlings MG, Skaf G. A review of the pathophysiology of cervical spondylotic myelopathy with insights for potential novel mechanisms drawn from traumatic spinal cord injury. Spine 1998; 23:2730–2737.
2. Bohlman HH, Emery SE. The pathophysiology of cervical spondylosis and myelopathy. Spine 1988; 13:843–846.
3. Panjabi M, White A 3rd. Biomechanics of nonacute cervical spinal cord trauma. Spine 1988; 13:838–842.
4. Kalsi-Ryan S, Karadimas SK, Fehlings MG. Cervical spondylotic myelopathy: the clinical phenomenon and the current pathobiology of an increasingly prevalent and devastating disorder. Neuroscientist 2013; 19:409–421.
5. Khuyagbaatar B, Kim K, Kim YH. Conversion equation between the Drop Height in the New York University Impactor and the Impact Force in the Infinite Horizon Impactor in the Contusion Spinal Cord Injury Model. J Neurotrauma 2015; 32:1987–1993.
6. Kim P, Haisa T, Kawamoto T, Kirino T, Wakai S. Delayed myelopathy induced by chronic compression in the rat spinal cord. Ann Neurol 2004; 55:503–511.
7. Long HQ, Li GS, Lin EJ, Xie WH, Chen WL, Luk KD, Hu Y. Is the speed of chronic compression an important factor for chronic spinal cord injury rat model? Neurosci Lett 2013; 545:75–80.
8. Cao Y, Shumsky JS, Sabol MA, Kushner RA, Strittmatter S, Hamers FP, et al. Nogo-66 receptor antagonist peptide (NEP1-40) administration promotes functional recovery and axonal growth after lateral funiculus injury in the adult rat. Neurorehabil Neural Repair 2008; 22:262–278.
9. Basso DM, Beattie MS, Bresnahan JC. A sensitive and reliable locomotor rating scale for open field testing in rats. J Neurotrauma 1995; 12:1–21.
10. Bosisio MR, Talmant M, Skalli W, Laugier P, Mitton D. Apparent Young’s modulus of human radius using inverse finite-element method. J Biomech 2007; 40:2022–2028.
11. Ichihara K, Taguchi T, Shimada Y, Sakuramoto I, Kawano S, Kawai S. Gray matter of the bovine cervical spinal cord is mechanically more rigid and fragile than the white matter. J Neurotrauma 2001; 18:361–367.
12. Galle B, Ouyang H, Shi R, Nauman E. Correlations between tissue-level stresses and strains and cellular damage within the guinea pig spinal cord white matter. J Biomech 2007; 40:3029–3033.
13. Kurokawa R, Nagayama E, Murata H, Kim P. Limaprost alfadex, a prostaglandin E1 derivative, prevents deterioration of forced exercise capability in rats with chronic compression of the spinal cord. Spine 2011; 36:865–869.
14. Yamamoto S, Kurokawa R, Kim P. Cilostazol, a selective type III phosphodiesterase inhibitor: prevention of cervical myelopathy in a rat chronic compression model. J Neurosurg Spine 2014; 20:93–101.
15. Karadimas SK, Moon ES, Yu WR, Satkunendrarajah K, Kallitsis JK, Gatzounis G, Fehlings MG. A novel experimental model of cervical spondylotic myelopathy (CSM) to facilitate translational research. Neurobiol Dis 2013; 54:43–58.
16. Mizuno J, Nakagawa H, Chang HS, Hashizume Y. Postmortem study of the spinal cord showing snake-eyes appearance due to damage by ossification of the posterior longitudinal ligament and kyphotic deformity. Spinal Cord 2005; 43:503–507.
17. Kameyama T, Hashizume Y, Ando T, Takahashi A, Yanagi T, Mizuno J. Spinal cord morphology and pathology in ossification of the posterior longitudinal ligament. Brain 1995; 118 (Pt 1):263–278.
18. Kubota K, Saiwai H, Kumamaru H, Kobayakawa K, Maeda T, Matsumoto Y, et al. Neurological recovery is impaired by concurrent but not by asymptomatic pre-existing spinal cord compression after traumatic spinal cord injury. Spine, 37:1448–1455.
19. Lee J, Satkunendrarajah K, Fehlings MG. Development and characterization of a novel rat model of cervical spondylotic myelopathy: the impact of chronic cord compression on clinical, neuroanatomical, and neurophysiological outcomes. J Neurotrauma 2012; 29:1012–1027.
20. Sun Y, Zhang LH, Fu YM, Li ZR, Liu JH, Peng J, et al. Establishment of a rat model of chronic thoracolumbar cord compression with a flat plastic screw. Neural Regen Res 2016; 11:963–970.