Development of a Chronic Cervical Cord Compression Model in Rats: Changes in the Neurological Behaviors and Radiological and Pathological Findings

Abstract

Cervical myelopathy is caused by chronic segmental compression of the spinal cord because of degenerative changes of the spine. However, the exact mechanisms of chronic cervical cord compression are not fully understood. The purpose of this study was to validate a new animal model of chronic cervical cord compression capable of reproducing the clinical course without laminectomy in rats. A polyethylene line attached to a plastic plate was fastened with three turns around the vertebral body of C4 in 1-month-old rats. After surgery, the polyethylene line grows deeper into the dorsal wall of the spinal canal along with the growth of the spinal canal and vertebral body, producing a gradual compression of the spinal cord. The results show that this cervical canal stenosis (CCS) model in rats caused motor deficits and sensory disturbances 9 months after initiating CCS; however, no clinical manifestations took place until 6 months. The intramedullary high-intensity area on T2-weighted images was observed in 70% of the CCS model rats at 12 months after initiating CCS. In histological sections, the spinal cord was compressed along the entire circumference at 12 months after initiating CCS. The number of ventral neurons was decreased, and the white matter showed wallerian degeneration. This model might reproduce characteristic features of clinical chronic cervical cord compression, including progressive motor and sensory disturbances after a latency period and insidious neuronal loss, and represents chronic compression of the cervical spinal cord in humans.

Introduction

I t is generally considered that the genesis of chronic compression myelopathy associated with pathological conditions of the cervical spine, such as spondylosis, ligamentum ossification, spinal canal stenosis, and disc herniation, may result from both mechanical compression and vascular problems. Chronic compression myelopathy impairs motor and sensory functions insidiously and progressively. Many patients remain asymptomatic for years, despite the presence of compression and deformity of the cord during this period, and once the disease becomes symptomatic, neurological disturbances progress with few changes in the impingement. A survey of cervical spine magnetic resonance imaging (MRI) in a randomly chosen asymptomatic cohort showed cord impingement in 26% of subjects older than 64 years, and in 16% of subjects aged 45 to 64 years (Teresi et al., 1987). Clinically, even in cases with severe compression confirmed by neuroimaging, examination often shows unexpectedly insignificant neurological findings, and considerable neurological improvement may be expected after surgical decompression. Thus one of the characteristics of chronic compression of the spinal cord not present in acute spinal cord injury is spinal plasticity (Fukushima et al., 1991). To date, clinically post-operative results of cervical chronic myelopathy keep improving (Baba et al., 1997). However, there are still many unsatisfactory outcomes, even after sufficient decompression surgery. This means that the cervical cord has been injured so severely that there is little reversibility in the process. The exact pathological mechanisms of the injury process remain to be elucidated, but circulatory insufficiency and venous congestion have been postulated as possible contributors (Gooding, 1974; Taylor et al., 1964; Wilson et al., 1969). Histological examination of autopsy specimens of the spinal cord showed loss of motor neurons and vascular degeneration in the gray matter, as well as demyelination and swelling of the axons (Barnes and Saunders, 1984; Hogan and Romanul, 1966; Mair and Druckman, 1953). An appropriate model for the pathophysiology of this condition should exhibit a latency period after induction of compression, with insidious onset of neurological dysfunction, followed by a phase of progressive disturbance (Kim et al., 2004; Xu et al., 2008). Furthermore, it is of great importance to make a model with no injury to the spinal cord. However, the pathogenesis of chronic compression myelopathy has remained unclear because of the absence of representative animal models for the studies. The purpose of this study was to produce a new animal model of chronic cervical cord compression satisfying these clinical conditions, with the assessment of motor and sensory functions using MRI and histopathological examination.

Methods

Animals

The experiment was carried out under the guidance of the local animal ethics committee in accordance with the guidelines on animal experiments in our university, Japanese government animal protection and management law, and Japanese government notifications on feeding and safe keeping of animals. Experiments were conducted on 27 male Sprague-Dawley rats (Clea, Tokyo, Japan), aged 3 weeks with a mean body weight of 51.0±1.1 g (mean±standard deviation [SD]). The animals were housed in conditions with controlled temperature (23±1°C), humidity (55.0±5 %), and lighting (on from 07:00 h to 20:00 h); while being allowed free access to water and food.

Production of the cervical canal stenosis model

Before surgery, the animals were kept in their cages for 1 week for adaptation to the environment. The animals were anesthetized with an intraperitoneal injection of 35 mg/kg sodium pentobarbital (Nembutal, 50 mg/mL; Abbott Laboratories, North Chicago, IL). Each animal was placed in the prone position on a heated plate (HM-300; Yamato Co., Ltd., Tokyo, Japan), and surgery was performed in a strictly aseptic manner. Body temperature was maintained at 37°C, and blood pressure in a tail artery was monitored by a non-invasive monitor (MK-1030; Muromachi Kikai Co., Ltd., Tokyo, Japan), to maintain constant physiologic levels during surgery. A longitudinal skin incision was made over the spinous process of C4, and bilateral paravertebral muscles were dissected bluntly from the lamina. Both laminae were exposed, and bilateral C4 and C5 roots were identified and dissected under a microscope. Then the spinous process of C4 was resected, and decortication of the laminae was performed. The left C4 root was retracted laterally and a polyethylene line (Tole Co., Tokyo, Japan) was passed ventral to the vertebral body of C4 and fixed to a plastic plate (1×2×0.5 mm) against the ventral aspect of the vertebral body of C4, and was then fastened with three turns (Fig. 1A and B). In this model, the polyethylene line grows deeper into the dorsal wall of the spinal canal due to the growth of the spinal canal and vertebral body, thus gradually compressing the spinal cord (Fig. 1C and D). The control animals were only subjected to dissection and exposure of the laminae of C4 and C5, but were not fastened. The control (n=10) and model rats (n=10) were kept for 12 months in the laboratory conditions described above.

FIG. 1.

(A) A diagram of the cervical spine fastened with polyethylene line at the C4 level. (B) The vertebral body and spinal canal expand during development; however, the spinal canal at the fastened site (C4) cannot expand locally, and the spinal cord will thus be gradually compressed. (C) An intraoperative photograph of a 3-week-old rat showing the cervical spine after a polyethylene line was fastened.

Neurological assessment

Basso, Beattie, and Bresnahan rating scale

Neurological motor function in each group were assessed at 1, 3, 6, and 9 days, and 12 months after surgery using the Basso, Beattie, and Bresnahan (BBB) locomotor rating scale (Basso et al., 1995). The BBB is based on locomotor ability following experimental spinal cord injury. Briefly, the BBB is a 21-point ordinal scale ranging from 0 (no discernible hindlimb movement) to 21 (consistent and coordinated gait with parallel paw placement of the hindlimbs and consistent trunk stability). Scores from 0–7 rank the early phase of recovery, with return of isolated movements of three joints (hip, knee, and ankle), scores from 8–13 describe the intermediate recovery phase, with return of paw placement, stepping, and forelimb-hindlimb coordination, and scores 14–21 rank the late phase of recovery, with return of toe clearance during the step phase, predominant paw position, trunk stability, and tail position. Two examiners observed and evaluated the locomotion of each rat in 4-min periods. The neurological examinations took place following a double-blind technique.

Measurement of maximum treadmill speed

Maximum treadmill speed was measured using a treadmill device (MK-680S; Muromachi, Tokyo, Japan). Treadmill speed was started at 5 m/min for 1 min, and after that it was increased by 1 m/min every 5 sec. Measurement was terminated and treadmill speed was recorded when the rat dropped out of the treadmill. A single investigator who was blinded to study group performed the maximum treadmill speed tests.

Sensory assessment

For mechanical sensory assessment, the rats were placed on a flat surface with the trunk lightly restrained by an examiner, and a von Frey filament (100 g) was applied to the dorsal surface of the forepaw and hindpaw. The filament was applied for 10 trials of 3 sec per trial at approximately 3-min intervals. The occurrence of paw withdrawal was expressed as the response frequency (i.e., the number of trials accompanied by paw withdrawal/10×100).

Magnetic resonance imaging study

The MR imaging studies were performed at 3, 6, and 12 months after surgery on a 0.4-T permanent magnet (APERTO Inspire; Hitachi, Tokyo, Japan), with a knee surface coil (9D-knee coil; Hitachi). Sequences included T1-weighted [350/25 (TR/TE)] and T2-weighted spin-echo images [5000/119 (TR/TE)], using a 150-mm field of view (FOV), 5-mm slice thickness, 256×256 matrix, and four excitations. The scan time was 5–6 min. All animals were scanned in the prone position, and the cervical portion was placed in the center of the coil, which was set in the isocenter of the static magnetic field.

The cross-sectional areas and compression ratio of the spinal cord were obtained in T1-weighted axial views passing through the area that received the maximum compression level, using a digitizer with MR apparatus. The compression ratio was obtained as the ratio of the anteroposterior diameter to the transverse diameter of spinal cord (Tarlov et al., 1953). The compression ratio and cross-sectional area of the spinal cord were compared in the control and CCS model rats.

The presence or absence of intramedullary high signal intensity area was examined using T2-weighted sagittal images. The quantitative assessment was performed by determining the signal intensities in the cervical cord using a region of interest (ROI) technique. Signal-difference-to-noise ratios (SDNRs) of all sequences were evaluated for the cervical spinal cord to assess the contrast between the spinal cord and the surrounding tissue of the spinal canal. SDNRs were assessed by separately measuring average signal intensities (mean in the ROI) of the spinal cord and outside the neck in the phase-encoding direction, including ghosting artifacts. The difference between the two values obtained was then divided by the standard deviation of background noise in the same image (Crooks and Ortendahl, 1989; Wolff and Balaban, 1997).

Preparation of histopathological examination

After MR imaging studies, all animals were sacrificed for histopathological examinations. The animals were fixed by intra-aortal perfusion with 4% buffered paraformaldehyde (Nakalai Tecque, Kyoto, Japan). The cervical vertebral column was removed and fixed in the same solution for 1 week. The specimens were then decalcified in 0.5 M ethylenediaminetetraacetic acid (EDTA-2Na; Wako Pure Chemical, Osaka, Japan), and embedded in paraffin using standard procedures. The light microscopy specimens were 600 serial 5-μm-thick transverse sections sliced from the rostral to the caudal end of the C2–C7 segment. Every 10th section was stained with Nissl stain, and photomicrographs were taken covering the whole ventral horn in each of the stained sections.

The motoneurons were identified with large nuclei and well-developed, densely-stained Nissl bodies in the cytoplasm. The nucleus, normally centrally placed, contains a well-demarcated round nucleolus. To obtain a precise count of the neurons without redundancy or discharge, we selected every second 5-μm-thick slice, leaving a 5-μm gap between each slice, based on the following stereological considerations: the characteristic large nucleoli have diameters of approximately 5 μm, and the size is quite uniform (Kitamura et al., 1986); and choosing the slice thickness and interval identical to the diameter (5 μm) of the spherical nucleolus means each section will contain tangential contours of the nucleoli with their centers located within 2.5 μm outside the slice. Thus counting the nucleoli of the motor neurons that appear on a 5-μm slice yields the number of those with the center located within 5 μm on each side from the middle of the slice, thus leaving 5-μm intervals between the sections, meaning that the stereological count avoids omission or redundancy in the number of motoneurons.

Statistical analysis

The data are expressed as average±standard error of the mean (SEM). Statistical analysis of BBB scale scores, maximum treadmill speed, sensory assessment, MR imaging studies, and numbers of motoneurons were performed using a nonparametric test (Wilcoxon's rank sum test) with SPSS statistical software, version 11.0J (SPSS Inc., Chicago, IL). Statistically significant differences were defined as p<0.05.

Results

Postoperatively, the animals were in good, healthy condition and did not develop any infections. The survival rate of control and CCS rats was 76.9% (10/13 animals), and 71.4% (10/14 animals), at 12 months after surgery, respectively. Their body weight increased rapidly, from 51.0±1.1 g (mean±SEM) at surgery, to 481.7±14.7 g (control rats), or to 465.0±19.8 g (CCS model rat), at 3 months after surgery. Afterward, body weight gradually increased in both groups, and the body weight of control and CCS model animals reached 706.7±43.2 g and 679.1±43.9 g at 12 months after surgery, respectively. There were no differences in the course of growth between the two groups.

Neurological outcome

Neurological functions were examined at 3, 6, 9, and 12 months after surgery. At 3 and 6 months after initiating CCS, no rats had signs of myelopathy, and BBB scale scores as well as maximum treadmill speeds were similar for the two groups. Two of 10 CCS model rats had decreased BBB scale scores and maximum treadmill speed at 9 months after initiating CCS, and all animals in the CCS model group developed signs of myelopathy at 12 months after initiating CCS. BBB scores of CCS model rats (17.3±1.6) was lower than those of control rats (20.8±0.4; Fig. 2A and Table 1), and the maximum treadmill speed of CCS model rats (21.7±4.5 m/min), was lower than that of control rats (33.3±3.5 m/min) 12 months after initiating CCS (Fig. 2B and Table 1). In the von Frey filament test, the response frequency of CCS model rats had a similar course to that of control rats until 6 months after initiating CCS, when 3 of 10 CCS model rats developed mild hypesthesia of the forepaws and hindpaws after 9 months. Afterward, all CCS model rats had decreased response frequency of the forepaws (26.0±12.6%), and the hindpaws (21.0±23.8%), at 12 months (Fig. 2C and Table 1).

FIG. 2.

The time course of behavioral assessment from production of the cervical canal stenosis (CCS) model to 12 months later. (A) Basso, Beattie, and Bresnahan (BBB) scale scores. (B) Maximum treadmill test speed. (C) Von Frey filament test results (*p<0.05).

Table 1.

The Results of Behavioral Assessment, Magnetic Resonance Imaging, and Histological Study after 12 Months

	Behavioral assessment					MR imaging					Histological study
	Animal no.	BBB score	Maximum treadmill speed (m/min)	Von Frey test of forepaw (%)	Von Frey test of hindpaw (%)	Cross-sectional area (mm ² )	Anteroposterior diameter (mm)	Transverse diameter (mm)	Compression ratio (%)	SDNR	Numbar of motoneurons (cells/60 slices)
Control rats	1	21	36	80	80	10.1	3.1	6.0	50.7	27.9	951
	2	21	37	90	80	11.2	3.2	5.1	63.3	13.4	840
	3	21	33	100	100	11.6	3.3	5.9	54.9	19.0	893
	4	21	39	90	90	14.7	3.5	4.8	72.9	18.8	840
	5	21	31	100	90	14.5	3.2	6.3	51.0	15.5	861
	6	20	35	90	100	13.9	3.0	6.1	49.1	21.1	899
	7	21	30	100	100	12.3	3.2	6.5	48.8	20.3	880
	8	21	42	100	80	11.9	3.2	5.3	59.7	17.8	794
	9	20	32	90	100	13.0	3.2	6.9	46.5	24.6	921
	10	21	33	80	90	12.7	2.9	6.9	42.7	27.3	880
	Average	20.8	34.8	92.0	91.0	12.6	3.2	6.0	54.0	20.6	875.8
	SEM	0.1	1.2	2.5	2.8	0.5	0.0	0.2	2.8	1.5	14.1
CCS model rats	1	18	23	50	30	11.1	1.7	5.4	31.8	27.4	682
	2	19	25	60	60	10.9	1.5	5.6	38.0	23.5	713
	3	17	18	20	30	7.3	1.0	3.8	27.5	38.6	507
	4	19	29	40	60	11.1	1.8	5.3	34.5	27.3	775
	5	16	15	20	30	8.8	1.1	4.2	26.8	29.9	413
	6	17	20	20	10	7.7	1.3	5.0	25.3	23.9	629
	7	15	18	30	10	7.5	1.2	4.7	26.5	32.8	453
	8	15	17	30	20	8.4	2.0	5.2	25.8	36.7	465
	9	19	27	30	40	11.0	2.2	4.7	46.8	25.3	650
	10	17	20	10	30	9.3	1.1	3.8	28.9	26.1	587
	Average	17.2	21.2	31.0	32.0	9.3	1.5	4.8	31.2	29.2	587.4
	SEM	0.5	1.5	4.8	5.5	0.5	0.1	0.2	2.2	1.7	38.7

BBB, Basso, Beattie, and Bresnahan locomotor rating scale; SDNR, signal-difference-to-noise ratio; SEM, standard error of the mean.

Magnetic resonance imaging

The MR imaging studies were performed at 3, 6, and 12 months after initiating CCS. On MR imaging, the spinal cords of CCS model animals gradually become flattened from 6 months, worsening up to 12 months after initiating CCS (Fig. 3). The cross-sectional area of the spinal cord obtained from T1-weighted axial imaging also decreased from 6 months after initiating CCS, and it was 9.3±1.6 mm² after 12 months, which was 74% of the size in control animals (Fig. 4A). The anteroposterior diameter of the spinal cord of CCS model animals was 1.5±0.4 mm, which was about 47% narrower than that of control rats. The compression ratio of CCS model rats was also decreased, to 47.7±6.5% at 6 months after initiating CCS, decreasing to 31.2±7.1% at 12 months, which was significantly different from control rats. The intramedullary high-intensity area on T2-weighted imaging was not observed at 6 months after initiating CCS; however, after 12 months, 7 of 10 CCS model rats presented with it, and the SDNRs of CCS model rats were significantly higher than those of normal control rats after 12 months (Table 1).

FIG. 3.

Magnetic resonance imaging of before (A), and after surgery, of control (B, C, and D), and cervical canal stenosis (CCS) model rats (E, F, and G). (A) Before surgery (1 month after birth), (B and E) 3 months after surgery, (C and F) 6 months after surgery, and (D and G) 12 months after surgery. Flattening of the spinal cord was observed from 6 months forward (F), and gradually worsened until 12 months after surgery in CCS model animals (G). The intramedullary high-intensity area on T2-weighted imaging was observed at 12 months after surgery in CCS model rats (arrow in G; scale bars=10 mm).

FIG. 4.

Changes in cross-sectional area (A) and compression ratio (B) of spinal cord at the maximum compression level.

Histopathological observations

All animals were sacrificed at 12 months after initiating CCS for histopathology. There was no epidural scar tissue such as that seen post-laminectomy in any animal. In the control rats, the spinal cord shape was well preserved, and motoneurons in the anterior columns were identified by their large nuclei and well-developed, densely-stained Nissl bodies in the cytoplasm (Fig. 5A, C, and E). However, in the CCS model rat, the spinal cord was compressed along the entire circumference, where in the gray matter it was associated with flattening of the ventral horn. The ventral neurons were flattened, small, and decreased in numbers of motoneurons (Table 1 and Fig. 5B, D, and F). The white matter showed myelin destruction and spongy axonal degeneration in all CCS model rats. These pathological changes occurred predominantly under the area of compression.

FIG. 5.

Transverse sections of the spinal cord at the C4–C5 level of control rats (A, C, and E) and cervical canal stenosis (CCS) model rats (B, D, and F). Flatter, smaller, and fewer neurons were seen in the gray matter, while myelin destruction and wallerian degeneration were seen in the white matter at the compression site (Nissl staining; magnification 1.25×for A and B, 20×for C and D, and 40×for E and F).

Discussion

Cervical myelopathy is the most serious consequence of compressive spinal cord disease such as spondylosis, ligamentum ossification, spinal canal stenosis and disc herniation, but its pathophysiology is not understood in detail. One reason is that a satisfactory model of chronic cervical cord compression has not been produced in experimental animals yet. So far, various methods have been undertaken to induce spinal cord compression; however, most of the models available are only suitable for acute or subacute compression and involve laminectomy to bring about the compression, so epidural tissues such as blood vessels are injured and scar tissue forms. There has been no appropriate model for analysis of the effects of chronic cervical cord compression. The other models include transplantation of tumor cells (Delattre et al., 1989; Ikeda et al., 1980; Manabe et al., 1989; Ushio et al., 1977), placement of screws with gradual tightening (Al-Mefty et al., 1993; Gooding et al., 1975; Hukuda and Wilson, 1972; Schramm et al., 1978; Shinomiya et al., 1992), implantation of an expanding sheet (Kasahara et al., 2006; Kim et al., 2004), and the genetic mouse making ossification of spinal ligaments (Hosoda et al., 1981). These models have several inconveniences, such as the time course of the tumor models being too rapid, epidural tissues being injured during the direct implantation of the sheet or screw placement, and a lack of options of where the compression site in genetically modified mice will take place other than the C1–C2 vertebral level. By comparison in our model, the polyethylene line suppressed the enlargement of the spinal canal, taking advantage of spinal growth itself, which gradually compresses the spinal cord. The cervical cord compression was produced without performing surgery on the spinal canal, therefore epidural tissues were not injured, and scar formation after laminectomy did not occur.

The bone maturity of rats and humans was around 3–4 months of age and 13–16 years old, respectively (Adler et al., 1983; Moskowitz et al., 1990). The spinal cord growth of rats and humans is completed around 3 months and 18 years after birth, respectively (Moskowitz et al., 1999). Therefore, rats aged 3 weeks and 12 months correspond to adolescence and advanced age in humans, respectively. Some studies have been made of the association between morphological changes of the compressed spinal cord on CT myelography or MR imaging, and clinical symptoms in humans. The spinal cord was reported to lose its functional tolerance if the transverse area measured by CT myelography and MR imaging was less than 55–75% of the normal value (Fujiwara et al., 1988, 1989; Fukushima et al., 1991; Hukuda and Wilson, 1972; Mizuno et al., 1992; Okada et al., 1994; Saito et al., 1992). In addition, a characteristic feature of human cervical cord compression is the delayed and insidious appearance of clinical symptoms after induction of spinal cord compression. In this study, compression of the cervical cord did not result in acute persistent neurological deficits. All animals returned to a normal neurological status after surgery. On MR imaging, the spinal cord of CCS model animals gradually became flattened from 6 months on after surgery. Neurological deficits developed in a delayed fashion; however, the earliest abnormalities emerged 9 months after the initial compressive procedure. The neurological deficits were progressive in nature, gradually becoming worse, much like the human condition. Thus this model is unique because of the slow course of the disease, and the lack of surgical damage to the spinal epidural tissues as well as to the spinal cord. Additionally, behavioral assessments were helpful and meaningful to accurately evaluate specific functions of the spinal cord. In our model, the conditions seen in chronic cervical cord compression in humans were simulated much more accurately than by other models.

At present, MR imaging is the preferred radiological diagnostic tool, because it can anatomically depict how the spinal cord is compressed, and it can also reflect the pathological changes in the spinal cord. Many authors have reported intramedullary high signal intensity on T2-weighted MR imaging in patients with compression lesions of the cervical spinal cord (Matsuda et al., 1991; Matsumoto et al., 2000; Takahashi et al., 1989). This imaging modality has been reported to have low sensitivity for the detection of cervical myelopathy, with published estimates ranging from 15–65%. Furthermore, the high signal intensity related to cervical myelopathy appears only on T2-weighted images of patients in late clinical stages. However, the pathological process underlying the high signal intensity seen on T2-weighted images remains imprecise. The increased intensity may be due to a water increase in myelomalacia, chronic ischemia, inflammatory edema, or cavitation (Al-Mafty et al.,1988; Faiss et al., 1990; Ito et al., 1996; Matsuda et al., 1991; Matsumoto et al., 2000; Takahashi et al., 1989). In animal experiments, there have been only a few MR imaging studies of chronic spinal cord compression, because of the difficulty in establishing an accurate model. Al-Mefty and associates (1993) demonstrated a chronic cervical cord compression model in dogs, obtained T2-weighted MR imaging, and reported intramedullary high-intensity areas in a snake-eye pattern at the compression site, which were indicative of cystic necrosis of gray matter found at the time of histological study. Kanchiku and associates (2001) also demonstrated a chronic cervical cord compression model in the rabbit, and reported high-intensity areas at the compression site that correlated to necrotic changes and gliosis in the gray matter, with demyelination and axonal degeneration in the white matter. In our study, the intensity quantified as the SDNR was significantly high at the compression area. Histological sections showed disappearance of ventral neurons in the gray matter and wallerian degeneration in the white matter that correlated with the T2 high-intensity areas seen at the compression site. Kobayashi and colleagues (2008) demonstrated that the evidence of active wallerian degeneration in the gray and white matter seen in the area of intramedullary edema was due to breakdown of the blood–spinal cord barrier. Therefore, intramedullary increased signal intensity changes seen on T2-weighted MR imaging are probably signs of edema and an inflammatory reaction in the cervical cord. Further studies will be needed to confirm this hypothesis using this model.

In studies of cadaveric cervical spinal cords, Hatayama and co-workers (1992), and Kameyama and associates (1995), observed a decreased number of anterior horn cells in patients who had suffered from motor paresis due to cervical ossification of the posterior longitudinal ligament, which preceded findings of gliosis, demyelination, and atrophy of anterior gray matter. They also reported a close relationship between the severity of external compression and atrophy of the anterior horn in human cadavers. Neuronal loss at the site of cervical cord compression has been reported in other animal models (Xu et al., 2008). If the disturbance of neurons caused by compression and the resulting central chromatolysis are mild, the neurons can recover fully after compression is relieved (Kobayashi et al., 2007). However, it seems likely that sustained mechanical compression of the cervical cord could result in irreversible damage to the neurons of the ventral horn, such as apoptosis. Chromatolysis does not necessarily foreshadow neuronal cell death, although specific signals seen during chromatolysis may be required to initiate apoptosis (Kerr and Harmon, 1991; Price and Porter, 1972; Sen, 1992). In our study we saw evidence that the motoneurons in the CCS rat model were decreased by 65% compared to controls at 1 year post-surgery. This finding indicated that the decrease seen in the number of motoneurons preceded the onset of the disturbances of motor and sensory function.

This animal model might reproduce characteristic features of clinical chronic cervical cord compression, including progressive motor and sensory disturbances after a latent period, and insidious neuronal loss. The MR imaging and histological findings found in this animal model were similar to findings in human subjects, so rats prepared in this way should be useful for the study of chronic compression of the human cervical cord.

Conclusion

The present study validates a new animal model of chronic cervical cord compression without direct injury to the spinal cord of the rat. In this model, rats whose C4 vertebral bodies were constricted using a polyethylene line presented with motor deficits and sensory disturbances 9 months after initiating CCS; however, no clinical manifestations took place until 6 months post-surgery. This insidious and delayed onset of symptoms is one of the most typical characteristics of chronic compression of the spinal cord. Therefore, this model may reproduce the behavior of chronic cervical spinal cord compression in humans.

Footnotes

Author Disclosure Statement

We thank Dr. Katsuhiko Hayakawa and Mr. Takashi Nakane of the Radiology Department of Aikou Orthopaedic Hospital for help with the MR imaging procedures. This work was supported by a grant-in aid from the Ministry of Education, Science and Culture of Japan (grant no. 21791388).

References

Adler

J.H.

, Schoenbaum

, Silberberg

1983. Early onset of disk degeneration and spondylosis in sand rats (Psammomys obesus) Vet. Pathol., 20:13–22.

Al-Mefty

, Harkey

H.L.

, Marawi

, Haines

D.E.

, Peeler

D.F.

, Wilner

H.I.

, Smith

R.R.

, Holaday

H.R.

, Haining

J.L.

, Russell

W.F.

, Harrison

, Middleton

T.H.

1993. Experimental chronic compressive cervical myelopathy. J. Neurosurg., 79:550–561.

Al-Mefty

, Harkey

L.H.

, Middleton

T.H.

, Smith

R.R.

, Fox

J.L.

1988. Myelopathic cervical spondylotic lesions demonstrated by magnetic resonance imaging. J. Neurosurg., 68:217–222.

Baba

, Maezawa

, Uchida

, Furusawa

, Wada

, Imura

1997. Plasticity of the spinal cord contributes to neurological improvement after treatment by cervical decompression. A magnetic resonance imaging study. J. Neurol., 244:455–460.

Barnes

M.P.

, Saunders

1984. The effects of cervical mobility on the natural history of cervical spondylotic myelopathy. J. Neurol. Neurosurg. Psychiatry, 47:17–20.

Basso

, Beattie

M.S.

, Bresnahan

J.C.

1995. A sensitive and reliable locomotor rating scale for open field testing in rats. J. Neurotrauma, 12:1–21.

Crooks

L.E.

, Ortendahl

D.A.

1989. Measuring signal-to-noise ratios in MR imaging. Radiology, 173:265–267.

Delattre

J.Y.

, Arbit

, Thaler

H.T.

, Rosenblum

M.K.

, Posner

J.B.

1989. A dose-response study of dexamethasone in a model of spinal cord compression caused by epidural tumor. J. Neurosurg., 70:920–925.

Faiss

J.H.

, Schroth

, Grodd

, Koenig

, Will

, Thron

1990. Central spinal cord lesions in stenosis of the cervical canal. Neuroradiology, 32:117–123.

10.

Fujiwara

, Yonenobu

, Ebara

, Yamashita

, Ono

1989. The prognosis of surgery for cervical compression myelopathy. An analysis of the factors involved. J. Bone Joint Surg., 71-B:393–398.

11.

Fujiwara

, Yonenobu

, Hiroshima

, Ebara

, Yamashita

, Ono

1988. Morphometry of the cervical spinal cord and its relation to pathology in cases with compression myelopathy. Spine, 13:1212–1216.

12.

Fukushima

, Ikata

, Taoka

, Tanaka

1991. Magnetic resonance imaging study on spinal cord plasticity in patients with cervical compressive myelopathy. Spine, 16:S534–S538.

13.

Gooding

M.R.

1974. Pathogenesis of myelopathy in cervical spondylosis. Lancet, 2:2280–2282.

14.

Gooding

M.R.

, Wilson

C.B.

, Hoff

J.T.

1975. Experimental cervical myelopathy. Effects of ischemia and compression of the canine cervical spinal cord. Adv. Neurosurg., 43:9–17.

15.

Hatayama

, Kaneda

, Sato

, Abumi

, Oshio

, Nagashima

, Oguma

1992. Clinical review of poor results for myelopathy caused by ossification of the posterior longitudinal ligament. Investigation Committee Report on the Ossification of the Spinal Ligaments. Kurokawa

Japanese Ministry of Public Health and Welfare: Tokyo, 107–111.

16.

Hogan

E.L.

, Romanul

F.C.

1966. Spinal cord infarction occurring during insertion of aortic graft. Neurology, 16:67–74.

17.

Hosoda

, Yoshimura

, Higaki

1981. A new breed of mouse showing multiple osteochondral lesions-twy mouse. Ryumachi, 21:157–164.

18.

Hukuda

, Wilson

C.B.

1972. Experimental cervical myelopathy: Effects of compression and ischemia on the canine cervical cord. J. Neurosurg., 37:631–652.

19.

Ikeda

, Ushio

, Hayakawa

, Mogami

1980. Edema and circulatory disturbance in the spinal cord compressed by epidural neoplasms in rabbits. J. Neurosurg., 52:203–209.

20.

Ito

, Oyanagi

, Takahashi

H.E.

, Ikuta

1996. Cervical spondylotic myelopathy. Spine, 21:827–833.

21.

Kameyama

, Hashizume

, Ando

, Takahashi

, Yanagi

, Mizuno

1995. Spinal cord morphology and pathology in ossification of the posterior longitudinal ligament. Brain, 118:263–278.

22.

Kanchiku

, Taguchi

, Kaneko

, Yonemura

, Kawai

, Gondo

2001. A new rabbit model for the study on cervical compressive myelopathy. J. Orthop. Res., 19:605–613.

23.

Kasahara

, Nakagawa

, Kubota

2006. Neuronal loss and expression of neurotrophic factors in a model of rat chronic compressive spinal cord injury. Spine, 31:2059–2066.

24.

Kerr

J.F.R.

, Harmon

B.V.

1991. Definition and incidence of apoptosis: an historical perspective. Apoptosis: The Molecular Basis of Cell Death. Tomei

L.D.

, Cope

F.O.

Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 5–29.

25.

Kim

, Haisa

, Kawamoto

, Kirino

, Wakai

2004. Delayed myelopathy induced by chronic compression in the rat spinal cord. Ann. Neurol., 55:503–511.

26.

Kitamura

, Nishiguchi

, Okubo

, Chen

K.L.

, Sakai

1986. An HRP study of the motoneurons supplying the rat hypobranchial muscles: central localization, peripheral axon course and soma size. Anat. Rec., 216:73–81.

27.

Kobayashi

, Baba

, Uchida

, Kokubo

, Takeno

, Yayama

, Miyazaki

, Nakajima

, Nomura

, Yoshizawa

2008. Synapse involvement of dorsal horn in experimental lumbar nerve root compression. A light and electron microscopic study. Spine, 33:716–723.

28.

Kobayashi

, Uchida

, Yayama

, Takeno

, Shimada

, Kubota

, Nomura

, Miyazaki

, Baba

2007. Motor neuron involvement in experimental lumbar nerve root compression. A light and electron microscopic study. Spine, 32:627–634.

29.

Mair

W.G.

, Druckman

1953. The pathology of spinal cord lesions and their relation to the clinical features in protrusion of cervical intervertebral discs: a report of four cases. Brain, 76:70–79.

30.

Manabe

, Tanaka

, Higo

, Park

, Ohno

, Tateishi

1989. Experimental analysis of spinal cord compressed by spinal metastasis. Spine, 14:1308–1315.

31.

Matsuda

, Miyasaki

, Tada

, Yasuda

, Nakayama

, Murakami

, Matsuo

1991. Increased MR signal intensity due to cervical myelopathy. J. Neurosurg., 74:887–892.

32.

Matsumoto

, Toyama

, Ishikawa

, Chiba

, Suzuki

, Fujimura

2000. Increased signal intensity of the spinal cord on magnetic resonance images in cervical compressive myelopathy: does it predict the outcome of conservative treatment? Spine, 25:677–682.

33.

Mizuno

, Nakagawa

, Iwata

, Hashizume

1992. Pathology of spinal cord lesions caused by ossification of the posterior longitudinal ligament, with special reference to reversibility of the spinal cord lesion. Neurol. Res., 14:312–314.

34.

Moskowitz

, Weinberger

, Shurtleff

D.B.

1999. Normal growth of the spinal cord. Eur. J. Pediatr. Surg. Suppl., 1:48–49.

35.

Moskowitz

R.W.

, Ziv

, Denko

C.W.

, Boja

, Jones

P.K.

, Adler

J.H.

1990. Spondylosis in sand rats: a model of intervertebral disc degeneration and hyperostosis. J. Orthop. Res., 8:401–411.

36.

Okada

, Ikata

, Katoh

, Yamada

1994. Morphologic analysis of the cervical spinal cord, dural tube and spinal canal by magnetic resonance imaging in normal adults and patients with cervical spondylotic myelopathy. Spine, 19:2331–2335.

37.

Price

D.L.

, Porter

K.R.

1972. The response of ventral horn neurons to axonal transection. J. Cell Biol., 53:24–37.

38.

Saito

, Mimatsu

, Sato

, Hashizume

1992. Histopathologic and morphometric study of spinal cord lesion in a chronic cord compression model using bone morphogenetic protein in rabbits. Spine, 17:1368–1374.

39.

Schramm

, Hashizume

, Takahashi

1978. Histological findings in graded experimental spinal cord compression in the cat. Adv. Neurosurg., 5:368–372.

40.

Sen

1992. Programmed cell death: concept, mechanism and control. Biol. Rev., 67:287–319.

41.

Shinomiya

, Mutoh

, Furuya

1992. Study of experimental cervical spondylotic myelopathy. Spine, 17:S383–S387.

42.

Takahashi

, Yamashita

, Sakamoto

, Kojima

1989. Chronic cervical cord compression: Clinical significance of increased signal intensity on MR images. Radiology, 173:219–224.

43.

Tarlov

I.M.

, Klinger

, Vitale

1953. Spinal cord compression studies. I. Experimental techniques to produce acute and gradual compression. Arch. Neurol. Psychiat., 70:813–819.

44.

Taylor

A.R.

1964. Vascular factors in the myelopathy associated with cervical spondylosis. Neurology, 14:62–68.

45.

Teresi

L.M.

, Lufkin

R.B.

, Reicher

M.A.

, Moffit

B.J.

, Vinuela

F.V.

, Wilson

G.M.

, Bentson

J.R.

, Hanafee

W.N.

1987. Asymptomatic degenerative disk disease and spondylosis of the cervical spine: MR imaging. Radiology, 164:83–88.

46.

Wilson

C.B.

, Bertan

, Norrell

H.A.

, Hukuda

1969. Experimental cervical myelopathy II. Acute ischemic myelopathy. Arch. Neurol., 21:571–589.

47.

Wolff

S.D.

, Balaban

R.S.

1997. Assessing contrast on MR imagings. Radiology, 202:25–29.

48.

Ushio

, Posner

J.B.

, Shapiro

W.R.

1977. Experimental spinal cord compression by epidural neoplasm. Neurology, 27:422–429.

49.

, Gong

W.M.

, Li

, Zhang

, Yin

D.Z.

, Jia

T.H.

2008. Destructive pathological changes in the rat spinal cord due to chronic mechanical compression. Laboratory investigation. J. Neurosurg. Spine, 8:279–285.