JOURNAL OF NEUROTRAUMA Volume 9, Number 3, 1992 Mary Ann Lieben, Inc., Publishers
Protective Effect of Methylprednisolone on Vascular in Rat Spinal Cord Injury J.
Injury
XU,1 Z.X. QU,1 E.L. HOGAN,1 and P.L. PEROT, Jr.2
ABSTRACT
High-dose methylprednisolone (MP) given to patients within 8 h of traumatic spinal cord improved neural function at 6 and 12 months, suggesting a probable secondary injury process that may be amenable to therapeutic intervention. Vascular injury plays an important role in the secondary injury process of CNS trauma. We have examined the effect of MP on vascular changes, including tissue edema, vascular permeability, and polymorphonuclear (PMN) cell i n 111 trat ion in a rat model of spinal cord impact injury. MP significantly reduced extravasation of fluorescein isothiocyanate dextran (FITC-D), a macromolecular tracer, by 64.3% and 50.7% with trauma forces of 20 and 40 g-cm, respectively, when MP was administered IV immediately after trauma at a bolus of 165 mg/kg, with a subsequent continuous MP infusion
at 31.5 mg/kg/h for 23 h. MP reduced the water content in the 40 g-cm traumatic cord lesion to 73.0% compared to the traumatic control (74.3%, p < 0.001) at the same schedule of large dose 24-h infusion. The same doses of MP showed a trend to decrease the extent of neutrophil infiltration as determined by myeloperoxidase (MPO) activity, but the change was not significant. MP had a little effect in decreasing FITC-D extravasation and cord edema when given at a lower dose (bolus of 30 mg/kg with continued infusion of 1.3 mg/kg/h for 23 h). MP did not reduce extravasation of FITC-D and edema when administered IV as one bolus injection at high (165 mg/kg) or low (30 mg/kg) doses. These results indicate that appropriate dosing of high-dose MP has a protective effect on the vascular damage following spinal cord
injury.
INTRODUCTION extensive literature indicates that corticosteroids exert beneficial effects on spinal cord injury in animal models (Green et al., 1980; Means et al., 1981; Young and Flamm, 1982; Hall and Braughler, 1982). The available reports are not unanimous, however, in regard to the effects of steroid, and several
An
Departments of 'Neurology and 2Neurosurgery,
Medical
University of South Carolina, Charleston, South Carolina
245
XU ET AL.
have been noted (Faden et al., 1981; Hoerlein et al., 1983; for review see Young et al., 1990, and Anderson, 1992). In the first National Acute Spinal Cord Injury Study (NASCIS), high-dose methylprednisolone (MP) (1000 mg bolus and daily for 10 days) administered to patients with spinal cord injury did not improve recovery of motor or sensory function at 6 weeks and 6 months after injury (Bracken et al., 1984). A more recent study (NASCIS II), however, has shown that MP significantly improves
negative findings
neurologic recovery at 6 and 12 months after IV treatment if a very high dose is administered within 8 h of injury (Bracken et al., 1990, 1992). These exciting results have aroused our interest in studying the mechanism of action of MP in protecting the spinal cord against traumatic injury in a rat model. Vascular injury involving certain features of acute inflammation, including exudation of macromolecules (Hsu et al., 1985b; Qu et al., 1991), edema formation (Hsu et al., 1985a), polymorphonuclear neutrophil (PMN) infiltration (Means and Anderson, 1983; Xu et al., 1990), secretion of inflammatory mediators (Demediuketal., 1985; Hsu etal., 1985a; Xu etal., 1990, 1991), and platelet aggregation (Balentine, 1978; Goodman et al., 1979), is believed to have an important role in the secondary process initiated by spinal cord injury. Methylprednisolone, like other glucocorticoids, is a potent anti-inflammatory agent (Hsu and Dimitrijevic, 1990). The purpose of this study is to investigate the possible mechanism of actions of MP that may involve its anti-inflammatory action. We studied the effect of MP on vascular injury, edema formation, and PMN infiltration, three major features of acute inflammation that have been noted to develop in animal models of spinal cord injury (Means and Anderson, 1983; Hsu et al., 1985b; Xu et al., 1990). We also sought to determine the optimal dose and schedule of administration of MP in a rat spinal cord injury model. MATERIALS AND METHODS All of the chemicals from
were
obtained from
Upjohn (Kalamazoo, MI).
Sigma (St. Louis, MO) except for MP and vehicle,
which
were
Spinal Cord Injury The method for producing spinal cord injury in rat (Sprague-Dawley, 200-300 g) has been described in detail (Daniel et al., 1975 ; Hsu et al., 1985b). Briefly, rats were anesthetized with ketamine ( 100 mg/kg), and a laminectomy was made at T-12. Spinal cord injury was inflicted by dropping a 5 g cylindrical brass weight from a height of 4 (20 g-cm) or 8 (40 g-cm) cm onto an impounder that had been carefully placed on the dura overlying the dorsal surface of the spinal cord. Sham-operated rats (laminectomy) served as noninjury control.
Determination
of Vascular Permeability
The extent of vascular permeability change was determined by extravasation of fluorescein isothiocyanateconjugated dextran (FITC-D) (Qu et al., 1991). FITC-D was administered by IV infusion (20 mg/kg) 2 h before killing. The T-12 spinal cord segment was removed after intracardiac perfusion with 200 ml of saline, homogenized in 1.0 ml of 5% trichloroacetic acid (TCA), and centrifuged at 27,000g for 20 min. A 0.8 ml aliquot of supernatant was mixed with 1.8 ml of 0.2 M Tris (pH 8.2) for the measurement of FITC-D fluorescence intensity using an Aminco-Bowman spectrophotofluorometer (Silver Spring, MD) set at 490 nm for excitation and 521 nm for emission. A sample of blood was obtained from the heart before perfusion. For assay of the FITC-D level, 10 pJ of plasma was mixed with 1.0 ml of 5% TCA. After centrifugation, 0.8 ml of supernatant was added to 1.8 ml of 0.2 M Tris (pH 8.2), and FITC-D fluorescence was determined using the same procedure as for the spinal cord. The extent of vascular injury was estimated by a vascular injury index (VII) derived from the following formula (Chan et al., 1983).
_
Concentration of FITC-D/mg protein of spinal cord Concentration of FITC-D in p,l of plasma 246
METHYLPREDNISOLONE IN SPINAL CORD INJURY
Measurement
of Edema
Spinal Cord
in
T-12 cord segment was removed in a humidity chamber, and the wet weight was measured immediately. The cord segment was dried at 100CC to constant weight for determination of dry weight. Tissue edema was determined by calculating tissue water content, with % tissue water content (1 dry weight/wet weight) x 100%. =
—
Measurement
of Myeloperoxidase (MPO)
in
Spinal Cord
MPO activity in spinal cord was measured spectrophotometrically (Xu et al., 1990). Briefly, the T-12 cord segment was dissected, removed, and homogenized in 0.05 M phosphate buffer (pH 6.0) containing 0.5% hexadecyltrimethylammonium bromide using a sonicator. The supernatant after centrifugation at 27,000g for 15 min was assayed for MPO activity. MPO activity was determined in 2.9 ml of 0.05 M phosphate buffer (pH 6.0) containing 0.53 nM dianisidine dihydrochloride and 0.15 mM H202 (Bradley et al., 1982). The assay was started by adding 0.1 ml of sample. Absorbance at 460 nm was monitored using a Gilford spectrophotometer (model 240, Ciba-Corning, Oberlin, OH). One unit of MPO activity was defined as that degrading 1 p.mol of H202 in 1 min.
Protocol for MP Treatment 1. MP was administered IV as a bolus of 165 mg/kg over a period of 15 min immediately after and 45 min later a continuous infusion, 31.5 mg/kg/h, was begun and continued for 23 h before trauma, The total dose was 889 mg/kg. The animals were killed 24 h after trauma. killing. 2. MP was given IV as a bolus of 30 mg/kg over a period of 15 min immediately after trauma. Group minutes later, MP infusion was begun at 1.3 mg/kg/h for 23 h. The total dose was 60 mg/kg. The Forty-five animals were killed 24 h after trauma. Group 3. MP 165 mg/kg was given as a bolus over a period of 15 min 2 h before trauma. The animals were killed 4 h after trauma. Group 4. MP 30 mg/kg was given as a bolus over a period of 15 min 2 h before trauma. The animals were killed 4 h after trauma. Control animals for each treatment group were subjected to the same magnitude of injury but received the same volume of vehicle instead of MP in an identical fashion.
Group
RESULTS
Effect of MP
on
Vascular
Permeability
in Rat
Spinal Cord Injury
MP at a dose of 889 mg/kg including one bolus injection ( 165 mg/kg) and 31.5 mg/kg/h infusion over 23 h decreased the vascular injury index (VII) by 65.3% when the cord was subjected to 20 g-cm force trauma. The value of VII was 0.18 ± 0.07 u.1 plasma/mg protein in the MP-treated rats compared to 0.53 ± 0.27 \i\ plasma/mg protein in the vehicle-treated control (p < 0.01). MP administered in the same dose and manner decreased VII by 50.7% when the cord was subjected to 40 g-cm trauma force. The value of VII in the MP-treated group was 0.70 ± 0.42 u,l/mg protein compared to 1.42 ± 0.44 \iA plasma/mg protein in the vehicle control (p < 0.01)(Fig. 1). When MP was given at a dose of 30 mg/kg IV in a bolus over 15 min with subsequent infusion of 1.3 mg/kg/h for 23 h, MP decreased VII from 0.91 ± 0.51 to 0.39 ± 0.22. This was significant (p < 0.05) but not as effective as high-dose MP. When MP was also given as bolus only at high (165 mg/kg) or low (30 mg/kg) dose, MP decreased VII by 38.5% and 26.9% respectively, but the changes were
not
significant (Table 1).
247
XU ET AL.
Sham
Treatment Group Effect of MP on vascular injury index (VII). Treatment groups. Sham was subjected to laminectomy, group 1 subjected to 20 g-cm trauma, group 2 was subjected to 40 g-cm trauma. H indicates trauma control treated with vehicle.0 indicates MP treatment. MP 165 mg/kg IV bolus was injected with continuous infusion for 23 h (see protocol 1 ). The spinal cord was removed 24 h after 40 g-cm trauma. The value of VII is expressed as mean EUSD. n 3= 8 in each group. **Compared with trauma control p < 0.01.
FIG. 1. was
Effect of MP
on
Edema in
Spinal Cord Injury
MP at a dose of 889 mg/kg (a bolus dose of 165 mg/kg and continuing infusion at a rate of 31.5 mg/kg/h for 23 h) significantly reduced the traumatic edema. The water content was 73.03 ± 0.34% in the MP-treated group compared to 74.26 ± 0.36 in the untreated control (Fig. 2) (p < 0.001). MP did not have a significant effect on edema, employing either a low-dose MP infusion for 24 h (a bolus dose of 30 mg/kg followed by infusion of 1.3 mg/kg/h for 23 h) or a 165 mg/kg bolus injection 2 h before 40 g-cm trauma (p > 0.05) (Fig. 2). A 30 mg/kg bolus dose was not tested because of lack of effect at higher doses.
Effect of MP
on
PMN
Infiltration Reflected by MPO Activity
in
Spinal Cord Injury
MP at doses 889 mg/kg including a bolus injection (165 mg/kg) and continuing a 31.5 mg/kg/h infusion over 23 h reduced MPO activity by 23.9%, but the difference was not significant. MP did not affect MPO activity when MP was administered as a bolus only at high (165 mg/kg) or low (30 mg/kg) doses given before trauma
(Table 2).
Table 1. Effect
of
MP on Vascular Injury Index
(VII)a
VII
(mg/kg)
(ui plasma/ mg protein)
165 30
1.04 ± 0.52 0.64 ± 0.40 0.76 ± 0.60
Dose
Control MP
Decrease 15 6
(%)
p value
38.5 26.9
>0.05 >0.05
aThe value is expressed as mean ± SD. MP was given IV 2 h before 20 g-cm Spinal cord was removed 4 h after trauma.
trauma.
248
METHYLPREDNISOLONE IN SPINAL CORD INJURY
76 -i
DC LU
I-
Protective Effect of Methylprednisolone on Vascular in Rat Spinal Cord Injury J.
Injury
XU,1 Z.X. QU,1 E.L. HOGAN,1 and P.L. PEROT, Jr.2
ABSTRACT
High-dose methylprednisolone (MP) given to patients within 8 h of traumatic spinal cord improved neural function at 6 and 12 months, suggesting a probable secondary injury process that may be amenable to therapeutic intervention. Vascular injury plays an important role in the secondary injury process of CNS trauma. We have examined the effect of MP on vascular changes, including tissue edema, vascular permeability, and polymorphonuclear (PMN) cell i n 111 trat ion in a rat model of spinal cord impact injury. MP significantly reduced extravasation of fluorescein isothiocyanate dextran (FITC-D), a macromolecular tracer, by 64.3% and 50.7% with trauma forces of 20 and 40 g-cm, respectively, when MP was administered IV immediately after trauma at a bolus of 165 mg/kg, with a subsequent continuous MP infusion
at 31.5 mg/kg/h for 23 h. MP reduced the water content in the 40 g-cm traumatic cord lesion to 73.0% compared to the traumatic control (74.3%, p < 0.001) at the same schedule of large dose 24-h infusion. The same doses of MP showed a trend to decrease the extent of neutrophil infiltration as determined by myeloperoxidase (MPO) activity, but the change was not significant. MP had a little effect in decreasing FITC-D extravasation and cord edema when given at a lower dose (bolus of 30 mg/kg with continued infusion of 1.3 mg/kg/h for 23 h). MP did not reduce extravasation of FITC-D and edema when administered IV as one bolus injection at high (165 mg/kg) or low (30 mg/kg) doses. These results indicate that appropriate dosing of high-dose MP has a protective effect on the vascular damage following spinal cord
injury.
INTRODUCTION extensive literature indicates that corticosteroids exert beneficial effects on spinal cord injury in animal models (Green et al., 1980; Means et al., 1981; Young and Flamm, 1982; Hall and Braughler, 1982). The available reports are not unanimous, however, in regard to the effects of steroid, and several
An
Departments of 'Neurology and 2Neurosurgery,
Medical
University of South Carolina, Charleston, South Carolina
245
XU ET AL.
have been noted (Faden et al., 1981; Hoerlein et al., 1983; for review see Young et al., 1990, and Anderson, 1992). In the first National Acute Spinal Cord Injury Study (NASCIS), high-dose methylprednisolone (MP) (1000 mg bolus and daily for 10 days) administered to patients with spinal cord injury did not improve recovery of motor or sensory function at 6 weeks and 6 months after injury (Bracken et al., 1984). A more recent study (NASCIS II), however, has shown that MP significantly improves
negative findings
neurologic recovery at 6 and 12 months after IV treatment if a very high dose is administered within 8 h of injury (Bracken et al., 1990, 1992). These exciting results have aroused our interest in studying the mechanism of action of MP in protecting the spinal cord against traumatic injury in a rat model. Vascular injury involving certain features of acute inflammation, including exudation of macromolecules (Hsu et al., 1985b; Qu et al., 1991), edema formation (Hsu et al., 1985a), polymorphonuclear neutrophil (PMN) infiltration (Means and Anderson, 1983; Xu et al., 1990), secretion of inflammatory mediators (Demediuketal., 1985; Hsu etal., 1985a; Xu etal., 1990, 1991), and platelet aggregation (Balentine, 1978; Goodman et al., 1979), is believed to have an important role in the secondary process initiated by spinal cord injury. Methylprednisolone, like other glucocorticoids, is a potent anti-inflammatory agent (Hsu and Dimitrijevic, 1990). The purpose of this study is to investigate the possible mechanism of actions of MP that may involve its anti-inflammatory action. We studied the effect of MP on vascular injury, edema formation, and PMN infiltration, three major features of acute inflammation that have been noted to develop in animal models of spinal cord injury (Means and Anderson, 1983; Hsu et al., 1985b; Xu et al., 1990). We also sought to determine the optimal dose and schedule of administration of MP in a rat spinal cord injury model. MATERIALS AND METHODS All of the chemicals from
were
obtained from
Upjohn (Kalamazoo, MI).
Sigma (St. Louis, MO) except for MP and vehicle,
which
were
Spinal Cord Injury The method for producing spinal cord injury in rat (Sprague-Dawley, 200-300 g) has been described in detail (Daniel et al., 1975 ; Hsu et al., 1985b). Briefly, rats were anesthetized with ketamine ( 100 mg/kg), and a laminectomy was made at T-12. Spinal cord injury was inflicted by dropping a 5 g cylindrical brass weight from a height of 4 (20 g-cm) or 8 (40 g-cm) cm onto an impounder that had been carefully placed on the dura overlying the dorsal surface of the spinal cord. Sham-operated rats (laminectomy) served as noninjury control.
Determination
of Vascular Permeability
The extent of vascular permeability change was determined by extravasation of fluorescein isothiocyanateconjugated dextran (FITC-D) (Qu et al., 1991). FITC-D was administered by IV infusion (20 mg/kg) 2 h before killing. The T-12 spinal cord segment was removed after intracardiac perfusion with 200 ml of saline, homogenized in 1.0 ml of 5% trichloroacetic acid (TCA), and centrifuged at 27,000g for 20 min. A 0.8 ml aliquot of supernatant was mixed with 1.8 ml of 0.2 M Tris (pH 8.2) for the measurement of FITC-D fluorescence intensity using an Aminco-Bowman spectrophotofluorometer (Silver Spring, MD) set at 490 nm for excitation and 521 nm for emission. A sample of blood was obtained from the heart before perfusion. For assay of the FITC-D level, 10 pJ of plasma was mixed with 1.0 ml of 5% TCA. After centrifugation, 0.8 ml of supernatant was added to 1.8 ml of 0.2 M Tris (pH 8.2), and FITC-D fluorescence was determined using the same procedure as for the spinal cord. The extent of vascular injury was estimated by a vascular injury index (VII) derived from the following formula (Chan et al., 1983).
_
Concentration of FITC-D/mg protein of spinal cord Concentration of FITC-D in p,l of plasma 246
METHYLPREDNISOLONE IN SPINAL CORD INJURY
Measurement
of Edema
Spinal Cord
in
T-12 cord segment was removed in a humidity chamber, and the wet weight was measured immediately. The cord segment was dried at 100CC to constant weight for determination of dry weight. Tissue edema was determined by calculating tissue water content, with % tissue water content (1 dry weight/wet weight) x 100%. =
—
Measurement
of Myeloperoxidase (MPO)
in
Spinal Cord
MPO activity in spinal cord was measured spectrophotometrically (Xu et al., 1990). Briefly, the T-12 cord segment was dissected, removed, and homogenized in 0.05 M phosphate buffer (pH 6.0) containing 0.5% hexadecyltrimethylammonium bromide using a sonicator. The supernatant after centrifugation at 27,000g for 15 min was assayed for MPO activity. MPO activity was determined in 2.9 ml of 0.05 M phosphate buffer (pH 6.0) containing 0.53 nM dianisidine dihydrochloride and 0.15 mM H202 (Bradley et al., 1982). The assay was started by adding 0.1 ml of sample. Absorbance at 460 nm was monitored using a Gilford spectrophotometer (model 240, Ciba-Corning, Oberlin, OH). One unit of MPO activity was defined as that degrading 1 p.mol of H202 in 1 min.
Protocol for MP Treatment 1. MP was administered IV as a bolus of 165 mg/kg over a period of 15 min immediately after and 45 min later a continuous infusion, 31.5 mg/kg/h, was begun and continued for 23 h before trauma, The total dose was 889 mg/kg. The animals were killed 24 h after trauma. killing. 2. MP was given IV as a bolus of 30 mg/kg over a period of 15 min immediately after trauma. Group minutes later, MP infusion was begun at 1.3 mg/kg/h for 23 h. The total dose was 60 mg/kg. The Forty-five animals were killed 24 h after trauma. Group 3. MP 165 mg/kg was given as a bolus over a period of 15 min 2 h before trauma. The animals were killed 4 h after trauma. Group 4. MP 30 mg/kg was given as a bolus over a period of 15 min 2 h before trauma. The animals were killed 4 h after trauma. Control animals for each treatment group were subjected to the same magnitude of injury but received the same volume of vehicle instead of MP in an identical fashion.
Group
RESULTS
Effect of MP
on
Vascular
Permeability
in Rat
Spinal Cord Injury
MP at a dose of 889 mg/kg including one bolus injection ( 165 mg/kg) and 31.5 mg/kg/h infusion over 23 h decreased the vascular injury index (VII) by 65.3% when the cord was subjected to 20 g-cm force trauma. The value of VII was 0.18 ± 0.07 u.1 plasma/mg protein in the MP-treated rats compared to 0.53 ± 0.27 \i\ plasma/mg protein in the vehicle-treated control (p < 0.01). MP administered in the same dose and manner decreased VII by 50.7% when the cord was subjected to 40 g-cm trauma force. The value of VII in the MP-treated group was 0.70 ± 0.42 u,l/mg protein compared to 1.42 ± 0.44 \iA plasma/mg protein in the vehicle control (p < 0.01)(Fig. 1). When MP was given at a dose of 30 mg/kg IV in a bolus over 15 min with subsequent infusion of 1.3 mg/kg/h for 23 h, MP decreased VII from 0.91 ± 0.51 to 0.39 ± 0.22. This was significant (p < 0.05) but not as effective as high-dose MP. When MP was also given as bolus only at high (165 mg/kg) or low (30 mg/kg) dose, MP decreased VII by 38.5% and 26.9% respectively, but the changes were
not
significant (Table 1).
247
XU ET AL.
Sham
Treatment Group Effect of MP on vascular injury index (VII). Treatment groups. Sham was subjected to laminectomy, group 1 subjected to 20 g-cm trauma, group 2 was subjected to 40 g-cm trauma. H indicates trauma control treated with vehicle.0 indicates MP treatment. MP 165 mg/kg IV bolus was injected with continuous infusion for 23 h (see protocol 1 ). The spinal cord was removed 24 h after 40 g-cm trauma. The value of VII is expressed as mean EUSD. n 3= 8 in each group. **Compared with trauma control p < 0.01.
FIG. 1. was
Effect of MP
on
Edema in
Spinal Cord Injury
MP at a dose of 889 mg/kg (a bolus dose of 165 mg/kg and continuing infusion at a rate of 31.5 mg/kg/h for 23 h) significantly reduced the traumatic edema. The water content was 73.03 ± 0.34% in the MP-treated group compared to 74.26 ± 0.36 in the untreated control (Fig. 2) (p < 0.001). MP did not have a significant effect on edema, employing either a low-dose MP infusion for 24 h (a bolus dose of 30 mg/kg followed by infusion of 1.3 mg/kg/h for 23 h) or a 165 mg/kg bolus injection 2 h before 40 g-cm trauma (p > 0.05) (Fig. 2). A 30 mg/kg bolus dose was not tested because of lack of effect at higher doses.
Effect of MP
on
PMN
Infiltration Reflected by MPO Activity
in
Spinal Cord Injury
MP at doses 889 mg/kg including a bolus injection (165 mg/kg) and continuing a 31.5 mg/kg/h infusion over 23 h reduced MPO activity by 23.9%, but the difference was not significant. MP did not affect MPO activity when MP was administered as a bolus only at high (165 mg/kg) or low (30 mg/kg) doses given before trauma
(Table 2).
Table 1. Effect
of
MP on Vascular Injury Index
(VII)a
VII
(mg/kg)
(ui plasma/ mg protein)
165 30
1.04 ± 0.52 0.64 ± 0.40 0.76 ± 0.60
Dose
Control MP
Decrease 15 6
(%)
p value
38.5 26.9
>0.05 >0.05
aThe value is expressed as mean ± SD. MP was given IV 2 h before 20 g-cm Spinal cord was removed 4 h after trauma.
trauma.
248
METHYLPREDNISOLONE IN SPINAL CORD INJURY
76 -i
DC LU
I-
