New Spinal Cord Injury Model Produced by Spinal Cord Compression in the Rat
TADATOSHI HASHIMOTO AND NAOHISA FUKUDA
Graded spinal cord injuries were produced in rats by compressing the spinal cord at the level of the Tll vertebra for 5, 15, 30, 60, or 180 min with stainless steel screws of 2-mm diameter and 2.8-mm length, or for 60 min with screws of the same diameter and various lengths (2.5,2.8,3.1, or 3.4 mm). The main neurologic symptoms caused by spinal cord compression were motor deficits, sensory deficits, and urinary incontinence. Neurologic scores, based on both motor and sensory deficits, correlated significantly with both the screw length and the duration of compression at every observation point from 4 hr to 21 days after removal of the screw. The incidence of urinary incontinence (from 24 hr to 21 days) and the percentage of rats surviving (from 14 days to 21 days) also correlated closely with the two factors (screw length and duration of compression). These results suggest that the present procedure could be a useful and simple model for studying traumatic spinal cord injury in rats. Key Words:
Spinal cord injury;
Compression;
Screw;
Neurologic
impairments;
Rat
INTRODUCTION Experimental spinal cord injuries in experimental animals, including rats, rabbits, cats, dogs, etc., have been widely investigated as models of central nervous system (CNS) injury. Among the methods reported, the weight-drop method, developed by Allen (1911), is a commonly used model. However, laminectomy or laparotomy, either of which is very surgically invasive, is needed to produce the spinal cord injury. Although the model is very valuable for estimating drug effects, it has been difficult to screen a large number of therapeutic agents for treatment of spinal cord injury (Faden, 1987) because many surgical procedures are required, and keeping treated animals in good condition is very labor-intensive. Therefore, it is very important to make a standardized animal model that is easy induction of neurologic symptoms. Also, it would be useful to produce groups of animals having spinal injuries with different degrees of severity that correspond to patients with mild, moderate, or severe functional deficits. In the present report, we describe a new, and simple standardized method for producing graded spinal cord injury models using rats. From the Research and Development Division, Takeda Chemical Industries, Osaka, Japan. Address reprint requests to: Tadatoshi Hashimoto, Biology Research Laboratories, Research and Development Divison, Takeda Chemical Industries, Ltd., 2-17-85, Juso-honmachi, Yodogawa-ku, Osaka 532 Japan. Received July 7, 1989; revised and accepted October 27, 1989. 203 Journal of Pharmacological Methods Q 1990 Elsevier Science Publishing
23, 203-212 (1990) Co., Inc., 655 Avenue of the Americas, New York, NY 10010
204
T. Hashimoto and N. Fukuda
MATERIALS AND METHODS Animals Male Wistat rats weighing about 300 g were used. They were kept under constant environmental conditions (24 ? 1°C and a regular 12-hr light/dark cycle), with food and water provided ad libitum, and were housed four rats to cage. Surgical Procedures and Postoperative
Care
A midline longitudinal incision was made in the back skin under pentobarbitalNa (30 mg/kg, ip) anesthesia, the fascia was cut in the midline, and the dorsal surface of the spinal column was exposed from TIO to T12 by paramedian incisions into the paraspinal muscles. The spinous process at the 11th thoracic (Tll) vertebra was dissected, and a small burr hole was made at the center of the dissected surface of the vertebra. A stainless steel screw of 2-mm diameter and 2.5-, 2.8-, 3.1-, or 3.4mm length was implanted into the burr hole made just above the dura mater, and then the incision was sutured. On the following day, rats showing no neurologic dysfunction were anesthetized by ether inhalation and the sutures were cut to expose the head of the screw. The spinal cord was compressed by inserting the screw as far as possible (Figure I). The screw was left in place for 60 min. In another group of rats, the spinal cord was compressed for 5,15, 30, or 180 min with screws of the same diameter and 2.8-mm length. Each experimental group comprised 20-24 rats. Following the removal of the screw, the incision was resutured. Rats having urinary incontinence were housed individually, and the urine was washed away with warm water, and they were wiped with paper towels daily.
FIGURE 1. Photograph showing the extent to which a stainless steel screw (S) of 2.8mm length (diameter: 2mm) is embedded in the vertebral foramen (Vf) of the eleventh thoracic vertebra. Sh, head of the screw; Va, vertebral arch; Vb, vertebral body.
Spinal Cord Injury by Compression
Neurologic
Symptoms and Scoring
The main neurologic symptoms caused by traumatic spinal cord compression were motor deficits, sensory deficits, and urinary incontinence. Neurologic scores (NS) from O-5 based on both motor and sensory deficits, were designated as follows: 0, no spontaneous movement of the hind limbs and no withdrawal response to tail pinching, including avoidance movements by forelimbs, biting of clamps, and vocalization; 1, no spontaneous movement of the hind limbs but showing a withdrawal response to tail pinching, including avoidance movement of the fore limbs, biting of clamps, and vocalization; 2, barely perceptible coordinated movement of the hind limbs and fore limbs; 3, well coordinated movement of the hind limbs and forelimbs but no weight bearing by four limbs; 4, able to walk with weight bearing by all four limbs but with an ataxic gait; 5, normal walking. Urinary incontinence was shown by wetness around the external genitalia because of urine. Observations of the neurologic symptoms were conducted in a blind manner 4 and 24 hr, and 2, 3, 5, 7, 14, and 21 days after removal of the screws. For animals that died before day 21, the NS and the occurrence or not of urinary incontinence determined at the final measurement point before death were used as the data for all measuring points after death. Analysis of Data The relationships between the neurologic score and the screw length, and between the neurologic score and the duration of compression, were determined by Spearman’s rank correlation coefficient (4), and the other relationships were determined by Pearson’s product moment correlation coefficient (r). RESULTS The main results of spinal cord injury in rats produced by compression at the level of the Tll vertebra using stainless steel screws of various lengths were motor deficits of the hind limbs, sensory deficits of the hind limbs and tail, and urinary incontinence. There was an apparent correlation between the motor and sensory deficits. All the rats showing complete sensory paralysis also showed complete loss of motor function in the hind limbs (NS, 0); however, many animals that showed no movement of the hind limbs did show withdrawal responses to tail pinching (NS, I). Relationship Between Screw length Urinary Incontinence
and NS, Survival Rate, and Incidence
of
Differences in the NS of rats with spinal cord injuries induced by compression of the spinal cord for 60 min at the Tll vertebra using screws of various lengths are shown in Figure 2. The severities of the neurologic symptoms shown by the rats whose spinal cords were compressed with screws of 2.5-mm length were relatively mild, and the median NSs at 4 hr, 3 days, and 10 days after removal of the screws were 2.5, 4, and 5, respectively. However, the neurologic symptoms of the rats
205
206
T. Hashimoto and N. Fukuda Length of Screu:
2.5mm
E l-l11 1161
0 ,r(
8 A
Length
of
Screu:
3.1m
Length
2 3
of
8
I
I
I
I
I
5
7
10
14
21
Screu:
3.4mm
3
5r
Days
after
Spinal
Cord
InJury,
FIGURE 2. Differences in the neurologic scores of rats with thoracic spinal cord (Tll) injury. Circles and lines represent individual and median values, respectively.
whose spinal cords were compressed with screws of 3.1 or 3.4 mm length were very severe, and the median NSs for both groups were 0 at 4 hr after removal of the screw, and even 21 days after spinal injury were 1 and 0, respectively. With the use of screws of 2.8 mm length, the median NSs of the rats were 1 from 4 hr to 3 days after screw removal, 2 on day 5, 2.5 on day 7, 3 on days IO and 14, and 3.5 on day 21 postinjury. As shown in Figure 3, the length of the screw used correlated significantly with the resulting NS at every observation point. For example, the median NSs 24 hr after removal of the screws in rats whose spinal cords were compressed for 60 min with screws of length 2.5, 2.8, 3.1, and 3.4 mm were 3, 1, 1, and 0, respectively, and the correlation coefficient (rs) was 0.743 (p < 0.0001). As shown in Figure 4A, the survival rate of rats with spinal injury caused by compression of the spinal cord with stainless steel screws of 2.5, 2.8, 3.1, and 3.4 mm length were 100, 91.6, 65, and 70% on day 14 postinjury (r = -0.89), respectively, and 100, 75, 45, and 55% on day 21 postinjury (r = -0.88), respectively. As shown in Figure 4B, the incidence of rats showing urinary incontinence were 10, 79.2, 100, and 100% 24 hr after removal of screws (r = 0.88), respectively, and 0, 45.8, 65, and 80% on day 21 postinjury (r = 0.96; p < 0.05), respectively.
Spinal Cord Injury by Compression Dar
1
2:s Dar
5
14
0.743
is
r=
2.8
2.5
Dar
r=
r=
Dar
MO. mol
3:1
0.806
3.4 -
2.5
‘Dar
r(0.ooo1
3.1
0.799
3
3.4
LB
7
r=
2.5
c
TADATOSHI HASHIMOTO AND NAOHISA FUKUDA
Graded spinal cord injuries were produced in rats by compressing the spinal cord at the level of the Tll vertebra for 5, 15, 30, 60, or 180 min with stainless steel screws of 2-mm diameter and 2.8-mm length, or for 60 min with screws of the same diameter and various lengths (2.5,2.8,3.1, or 3.4 mm). The main neurologic symptoms caused by spinal cord compression were motor deficits, sensory deficits, and urinary incontinence. Neurologic scores, based on both motor and sensory deficits, correlated significantly with both the screw length and the duration of compression at every observation point from 4 hr to 21 days after removal of the screw. The incidence of urinary incontinence (from 24 hr to 21 days) and the percentage of rats surviving (from 14 days to 21 days) also correlated closely with the two factors (screw length and duration of compression). These results suggest that the present procedure could be a useful and simple model for studying traumatic spinal cord injury in rats. Key Words:
Spinal cord injury;
Compression;
Screw;
Neurologic
impairments;
Rat
INTRODUCTION Experimental spinal cord injuries in experimental animals, including rats, rabbits, cats, dogs, etc., have been widely investigated as models of central nervous system (CNS) injury. Among the methods reported, the weight-drop method, developed by Allen (1911), is a commonly used model. However, laminectomy or laparotomy, either of which is very surgically invasive, is needed to produce the spinal cord injury. Although the model is very valuable for estimating drug effects, it has been difficult to screen a large number of therapeutic agents for treatment of spinal cord injury (Faden, 1987) because many surgical procedures are required, and keeping treated animals in good condition is very labor-intensive. Therefore, it is very important to make a standardized animal model that is easy induction of neurologic symptoms. Also, it would be useful to produce groups of animals having spinal injuries with different degrees of severity that correspond to patients with mild, moderate, or severe functional deficits. In the present report, we describe a new, and simple standardized method for producing graded spinal cord injury models using rats. From the Research and Development Division, Takeda Chemical Industries, Osaka, Japan. Address reprint requests to: Tadatoshi Hashimoto, Biology Research Laboratories, Research and Development Divison, Takeda Chemical Industries, Ltd., 2-17-85, Juso-honmachi, Yodogawa-ku, Osaka 532 Japan. Received July 7, 1989; revised and accepted October 27, 1989. 203 Journal of Pharmacological Methods Q 1990 Elsevier Science Publishing
23, 203-212 (1990) Co., Inc., 655 Avenue of the Americas, New York, NY 10010
204
T. Hashimoto and N. Fukuda
MATERIALS AND METHODS Animals Male Wistat rats weighing about 300 g were used. They were kept under constant environmental conditions (24 ? 1°C and a regular 12-hr light/dark cycle), with food and water provided ad libitum, and were housed four rats to cage. Surgical Procedures and Postoperative
Care
A midline longitudinal incision was made in the back skin under pentobarbitalNa (30 mg/kg, ip) anesthesia, the fascia was cut in the midline, and the dorsal surface of the spinal column was exposed from TIO to T12 by paramedian incisions into the paraspinal muscles. The spinous process at the 11th thoracic (Tll) vertebra was dissected, and a small burr hole was made at the center of the dissected surface of the vertebra. A stainless steel screw of 2-mm diameter and 2.5-, 2.8-, 3.1-, or 3.4mm length was implanted into the burr hole made just above the dura mater, and then the incision was sutured. On the following day, rats showing no neurologic dysfunction were anesthetized by ether inhalation and the sutures were cut to expose the head of the screw. The spinal cord was compressed by inserting the screw as far as possible (Figure I). The screw was left in place for 60 min. In another group of rats, the spinal cord was compressed for 5,15, 30, or 180 min with screws of the same diameter and 2.8-mm length. Each experimental group comprised 20-24 rats. Following the removal of the screw, the incision was resutured. Rats having urinary incontinence were housed individually, and the urine was washed away with warm water, and they were wiped with paper towels daily.
FIGURE 1. Photograph showing the extent to which a stainless steel screw (S) of 2.8mm length (diameter: 2mm) is embedded in the vertebral foramen (Vf) of the eleventh thoracic vertebra. Sh, head of the screw; Va, vertebral arch; Vb, vertebral body.
Spinal Cord Injury by Compression
Neurologic
Symptoms and Scoring
The main neurologic symptoms caused by traumatic spinal cord compression were motor deficits, sensory deficits, and urinary incontinence. Neurologic scores (NS) from O-5 based on both motor and sensory deficits, were designated as follows: 0, no spontaneous movement of the hind limbs and no withdrawal response to tail pinching, including avoidance movements by forelimbs, biting of clamps, and vocalization; 1, no spontaneous movement of the hind limbs but showing a withdrawal response to tail pinching, including avoidance movement of the fore limbs, biting of clamps, and vocalization; 2, barely perceptible coordinated movement of the hind limbs and fore limbs; 3, well coordinated movement of the hind limbs and forelimbs but no weight bearing by four limbs; 4, able to walk with weight bearing by all four limbs but with an ataxic gait; 5, normal walking. Urinary incontinence was shown by wetness around the external genitalia because of urine. Observations of the neurologic symptoms were conducted in a blind manner 4 and 24 hr, and 2, 3, 5, 7, 14, and 21 days after removal of the screws. For animals that died before day 21, the NS and the occurrence or not of urinary incontinence determined at the final measurement point before death were used as the data for all measuring points after death. Analysis of Data The relationships between the neurologic score and the screw length, and between the neurologic score and the duration of compression, were determined by Spearman’s rank correlation coefficient (4), and the other relationships were determined by Pearson’s product moment correlation coefficient (r). RESULTS The main results of spinal cord injury in rats produced by compression at the level of the Tll vertebra using stainless steel screws of various lengths were motor deficits of the hind limbs, sensory deficits of the hind limbs and tail, and urinary incontinence. There was an apparent correlation between the motor and sensory deficits. All the rats showing complete sensory paralysis also showed complete loss of motor function in the hind limbs (NS, 0); however, many animals that showed no movement of the hind limbs did show withdrawal responses to tail pinching (NS, I). Relationship Between Screw length Urinary Incontinence
and NS, Survival Rate, and Incidence
of
Differences in the NS of rats with spinal cord injuries induced by compression of the spinal cord for 60 min at the Tll vertebra using screws of various lengths are shown in Figure 2. The severities of the neurologic symptoms shown by the rats whose spinal cords were compressed with screws of 2.5-mm length were relatively mild, and the median NSs at 4 hr, 3 days, and 10 days after removal of the screws were 2.5, 4, and 5, respectively. However, the neurologic symptoms of the rats
205
206
T. Hashimoto and N. Fukuda Length of Screu:
2.5mm
E l-l11 1161
0 ,r(
8 A
Length
of
Screu:
3.1m
Length
2 3
of
8
I
I
I
I
I
5
7
10
14
21
Screu:
3.4mm
3
5r
Days
after
Spinal
Cord
InJury,
FIGURE 2. Differences in the neurologic scores of rats with thoracic spinal cord (Tll) injury. Circles and lines represent individual and median values, respectively.
whose spinal cords were compressed with screws of 3.1 or 3.4 mm length were very severe, and the median NSs for both groups were 0 at 4 hr after removal of the screw, and even 21 days after spinal injury were 1 and 0, respectively. With the use of screws of 2.8 mm length, the median NSs of the rats were 1 from 4 hr to 3 days after screw removal, 2 on day 5, 2.5 on day 7, 3 on days IO and 14, and 3.5 on day 21 postinjury. As shown in Figure 3, the length of the screw used correlated significantly with the resulting NS at every observation point. For example, the median NSs 24 hr after removal of the screws in rats whose spinal cords were compressed for 60 min with screws of length 2.5, 2.8, 3.1, and 3.4 mm were 3, 1, 1, and 0, respectively, and the correlation coefficient (rs) was 0.743 (p < 0.0001). As shown in Figure 4A, the survival rate of rats with spinal injury caused by compression of the spinal cord with stainless steel screws of 2.5, 2.8, 3.1, and 3.4 mm length were 100, 91.6, 65, and 70% on day 14 postinjury (r = -0.89), respectively, and 100, 75, 45, and 55% on day 21 postinjury (r = -0.88), respectively. As shown in Figure 4B, the incidence of rats showing urinary incontinence were 10, 79.2, 100, and 100% 24 hr after removal of screws (r = 0.88), respectively, and 0, 45.8, 65, and 80% on day 21 postinjury (r = 0.96; p < 0.05), respectively.
Spinal Cord Injury by Compression Dar
1
2:s Dar
5
14
0.743
is
r=
2.8
2.5
Dar
r=
r=
Dar
MO. mol
3:1
0.806
3.4 -
2.5
‘Dar
r(0.ooo1
3.1
0.799
3
3.4
LB
7
r=
2.5
c
