Time course of energy perturbation after compression trauma to the spinal cord: An experimental study in the rat using microdialysis

Surg Neurol

297

1993 ;39 :297-304

Time Course of Energy Perturbation After Compression Trauma to the Spinal Cord : An Experimental Study in the Rat Using Microdialysis Y. Zhang, M .D., L. Hillered, M .D., Y. Olsson, M.D., and A. Holtz, M .D . Laboratory of Neuropathology, and Departments of Neurosurgery, and Clinical Chemistry, University Hospital, Uppsala, Sweden

Zhang Y, Hillered L, Olsson Y, Holtz A . Time Course of Energy Perturbation After Compression Trauma to the Spinal Cord : An Experimental Study in the Rat Using Microdialysis . Surg Neurol 1993 ;39 :297-304 .

Changes occurring in the extracellular fluid (ECF) concentration of energy-related metabolites were investigated in a well-characterized model of compression trauma to the spinal cord . Microdialysis probes were inserted into exposed grey matter of the dorsal horn at the level of Th 7-8, and perfused with mock cerebrospinal fluid . The trauma was produced 2 hours later by compression of the cord with a 9-, 35-, or 50-g load for 5 min . Microdialysis samples (10-minute fractions) were collected for another 2 hours following decompression . The trauma was associated with an accumulation of lactate, inosine, and hypoxanthine, and an increase in the lactate/pyruvate ratio in the ECF, indicating a profound disturbance in energy metabolism . These changes were related to the severity of spinal cord injury as well as to the spinal cord blood flow (SCBF) reductions and neurological deteriorations previously determined . Following decompression, all ECF metabolites normalized within 20-40 min after mild (9 g) to moderate (35 g) trauma . After severe trauma (50 g), resulting in complete ischemia during compression, followed by irreversible paraplegia, there was a partial recovery of ECF inosine and hypoxanthine, whereas the increase in lactate and the lactate/pyruvate ratio persisted . The results suggest that penumbra conditions prevail during the early posttraumatic period when the degree of trauma results in severe neurological deterioration and that ECF lactate levels in the spinal cord is a sensitive indicator of secondary ischemia after compression injury . Energy metabolism ; Lactate ; Microdialysis ; Rats ; Spinal cord injury KEY WORDS :

Addren reprint requests to.• lays Hillered, M .D ., Departments of Clinical Chemistry and Neurosurgery, Uppsala University Hospital, S-751 85 Uppsala, Sweden . Received July 22, 1992 ; accepted September 4, L992 . 1993 by

Elsevier Science

Publishing Co ., Inc .

Trauma to the spinal cord may result in a broad spectrum of neurological disabilities ranging from transient symptoms in mild cases to complete and permanent disability in severe ones [1, 9] . Primary mechanical forces may result in damage to neurons, glial cells, and myelin . In addition, secondary pathophysiological processes may occur, mainly due to spinal cord hypoperfusion [6] . Among the secondary pathophysiological processes, special interest has been focused on lipid peroxidation [7], release of endogenous opioid-like peptides [10], ionic and metabolic disturbances [30], and release of excitatory amino acids [21, 25] . Previous studies on blood flow changes after graded spinal cord trauma have indicated that a posttraumatic hypoperfusion occurs, which is related to the severity of impact (19, 27] . This hypoperfusion may lead to secondary ischemia and energy perturbation, aggravating the neurologic sequelae [11] . Results from brain ischemia and trauma research suggest that disturbances in energy metabolism can be accurately estimated by measuring changes in the extracellular fluid (ECF) concentration of energy-related metabolites (i .e ., lactate, pyruvate, inosine, and hypoxanthine) using microdialysis [13, 16-18, 23] . The main advantage with this approach is that it allows repeated measurements in the same individual so that the time course of the pathophysiological process can be monitored [15, 26] . Previously, microdialysis has been used only in a few studies on the spinal cord : after continuous posterior epidural stimulation [28], after administration of depolarizing agents [29], and after impact injury to the spinal cord measuring serotonin, norepinephrine [20], and excitatory amino acids [21, 25] . The aim of the present study was to monitor disturbances in energy metabolism during the first 2 hours after a graded compression trauma to the spinal cord and correlate the findings to the time course of spinal cord blood flow (SCBF) changes and neurological deterioration known from previous studies in the same animal model . 0090-3019193/$6 .00



2 98

Surg Neurol 1993 ;39:297-304

Materials and Methods Animals and Spinal Cord Injury Sprague-Dawley rats weighing between 350 and 450 g were used, and they had free access to food and water before the experiments . They were anesthetized with a solution containing fluanisone 2 .5 mg/ml and midazolam 1 .25 mg/ml, subcutaneously, totally 1 .5-2 .0 ml . Polyethylene catheters (PE 50) were inserted into the left carotid artery and jugular vein for blood sampling, continuous monitoring of mean arterial blood pressure (MABP) and infusions, respectively . The rats were then placed prone, and by using a rectal thermistor, we were able to maintain body temperature at -37'C . The spinal cord was exposed by a dorsal laminectomy of Th 7-8 under microscopic guidance leaving the dura intact. The animals were then tracheotomized . Muscular paralysis was achieved by suxamethonium, and the animals were connected to a small-animal respirator (BraunMelsungen, Melsungen AG, ERG) and artificially ventilated with a 7Wo : 30% N,O/O, mixture at a frequency of 65 per minute and a tidal volume of 3 ml. Before implantation of the microdialysis probe the ventilation was adjusted to pH 7 .35-7 .45 ; Pco2 4 .5-5 .5 kPa ; Pot 9 .5-18 kPa . Animals not fulfilling these criteria with a base excess more than ±3 mmol/L, or showing microscopical signs of spinal cord injury after laminecromy were excluded .

Experimental Groups The animals were divided into four groups prior to spinal cord injury .

• Group 1 . Laminectomized controls . • Group 2 . A compression of 9 g was applied for 5 min . This compression does not result in any neurological deficit but corresponds to a load resulting in a myelographic block and a slight weakening in sensory-evoked potential (SEP) . SCBF is reduced to 25% during compression and recovers to 60% of control 60 min after decompression [ 19] . • Group 3 . A load of 35 g was applied for 5 min resulting in a severe but reversible paraparesis with a 90% recovery over 10-14 days 124] . The load produces a marked impact on SEP, and SCBF is reduced to 10% of control during compression and recovers to 30% at 60 min after injury [19] .

Zhang et al

•

Group 4 . A compression of 50 g was applied for 5 minutes leading to paraplegia with only minor recovery over the following 10-14 days [241 . SEP is extinguished, and SCBF is close to zero during compression and 10% of control at 60 minutes after injury [191.

Implantation of Dialysis Probe and Spinal Cord Injury The animals were placed in a specially designed stereotaxic frame allowing implantation of the microdialysis probe and production of spinal cord injury independently, without moving the animal (Figure 1) . The microdialysis probe (CMA/11 ; diameter : 0 .24 mm ; membrane length : 1 mm ; CMA/Microdialysis, Stockholm, Sweden) was implanted through a dural incision into the grey matter of the dorsal horn . The dialysis system was perfused with artificial cerebrospinal fluid (CSF) (NaCl, 148 mM ; CaC1 2 , 1 .2 MM ; MgC1 2 , 0 .9 mm ; KCI, 2 .7 mm) at a flow rate of 2 ltl/min, and the dialysate was sampled in 10-minute fractions during 2 hours before and 2 hours after the spinal cord injury . At 2 hours after probe implantation (steady-state period), the spinal cord injury was produced using a compression device whereby a slightly curved rectangular plate (2 .2 x 5 .5 mm) with a predetermined load was applied on the exposed dura for 5 min [19, 241 . The microdialysis probe was removed immediately before and reinserted within 1 min after trauma into exactly the same position of the spinal cord using the stereotaxic device . This procedure has been successfully used in studies on experimental brain injury [23] .

Chemical Analyses The microdialysis samples were divided into two aliquots and analyzed on two different high-performance liquid chromatography (HPLC) systems . Lactate and pyruvate were detected by ultraviolet (UV) light at 214 nm [141, and the purine metabolites, inosine, and hypoxanthine by UV-light at 254 nm [12] . The basal level of the metabolites were defined as the mean dialysate concentration in the last six fractions of the steady-state period .

Statistics For comparison of multiple means for each metabolite (within groups), statistical analysis was performed using one-way analysis of variance and Tukey's multiple-range test (Statgraphics 2 .6, STSC Inc ., Rockville, Maryland) . Values in tables and figures are given as the mean ±



Surg Neurol 299 1993 ;39 : 2 9 7 -304

Spinal Cord Injury and Energy Metabolism

Figure 1 . Photograph illustrating the stereotaxs frame . The spine was .ctabilized in a fixed position by clamping the spinous process of Th 6 and 9, respectively . with the forceps mounted on the framework, The figure illustrates the compression device ith the predetermined rectangularload on top (~!per, !are. white arrow) and the microdialysis probe illustrated by a needle (lower,,fine, white arrow) .

SD . Differences with a p value <0 .05 were considered statistically significant . Results Physiological Variables

The values of the physiological variables, recorded after laminectomy but before implantation of the microdialysis probe and up to 2 hours after trauma, are given in Table 1 . The mean arterial blood pressure was lowest in the 50-g compression groups as expected [ 191 . There was no significant difference in body weight and duration of operation between the groups .

In group 4, lactate rose tenfold with the peak value in the second fraction after impact . In this group there was only a partial recovery with a persistent lactate increase throughout the 2-hour observation period . There was no significant change in the dialysate concentration ofpyruvate in response to trauma . Therefore, the increase in the lactate/pyruvate ratio roughly corresponded to the change in lactate . The basal level in the lactate/pyruvate ratio in group 1 was between 15 and 20, in agreement with previous studies [26} . In group 4, the lactate/pyruvate ratio rose from 15 to 95 after trauma and then remained at an increased level of 50-60 throughout the experiment .

Lactate and Pyruvate

Figure 2 gives the dialysate concentration of lactate in a group of laminectomized control animals (group 1) subjected to removal and reinsertion of the microdialysis probe but no trauma . There was a stable basal lactate level of about 30 µM throughout the experiment . As expected from previous work on brain trauma [231, there was no effect of the reimplantation procedure on the lactate concentration (black arrow) . The basal level was 30-35 µM in all four experimental groups . Figure 2 further summarizes the effect of the graded spinal cord impact on the dialysate lactate content . In groups 2 and 3, lactate increased six- to sevenfold above the basal level in the first fraction following the trauma . Lactate normalized promptly (within about 30 min) in group 2, whereas recovery in group 3 was more sluggish .

Purine Metabolites

In group 1, the inosine level of the dialysates remained at a stable basal level close to zero throughout the experiment (Figure 3) . As noted for lactate, the reimplantation procedure produced no change in the inosine level . Figure 3 further illustrates the effect of spinal cord trauma on the dialysate concentrations of inosine . From a control level of close to zero, inosine rose to 0 .7-0 .9 gM in the first fraction after impact in groups 2 and 3 . In group 4, inosine increased to 1 .4 µM in the corresponding fraction . Recovery of inosine was directly related to the degree of impact, restitution being progressively delayed with increasing severity of the trauma . Recovery was complete at about 40 min postinjury in group 4 . I nosine .



300

Surg Neurol 1993 ;39 : 2 97-304

Zhang et al

Group 3

Figure 2 . Effect of graded spinal cord compression trauma on the extracellular lactate levels as reflected by microdialysate concentrations from rats subjected to sham operation (group I) . mild (group 2), moderate (group 3). or severe (group 4) injury. Values mere the means offive to six experiments ±SD . A rrows indicate production of trauma. 'Denotes statistically significant differenre from the basal preinjury level . Forfurther explanation, see text.

Group 4

100

400

300

300

200

200

100

0

0 0

30

60

90

120 150 180

Minutes

0

30

60

90

120

150

180

Minutes

Hypoxanthine. The dialysate level of hypoxanthine remained at a basal level close to zero throughout the experiment in group 1 (Figure 4) . Figure 4 also summarizes the results of the hypoxanthine response to trauma . As noted for lactate and inosine, there was no effect of the reimplantation procedure . In groups 2 and 3, hypoxanthine increased to 1 .3 and 1 .6 µM, respectively, in the first fraction after injury. The increase of hypoxanthine was larger in group 4, with a peak value of 2 .8 µM occurring in the second fraction after trauma . Recovery of hypoxanthine was progressively delayed with increasing severity of the impact . Complete restitution occurred at about 80 min after injury. Thus, both the accumulation and the recovery of hypoxanthine was delayed in group 4 compared to inosine, reflecting that inosine is a catabolic precursor of hypoxanthine .

Discussion In this investigation we studied the time course of changes in the extracellular concentration of energyrelated metabolites in relation to graded compression trauma to the spinal cord using microdialysis . The trauma was associated with an accumulation of lactate, inosine, hypoxanthine, and an increase in the lactate/pyruvate ratio in the ECF, indicating a profound disturbance of

energy metabolism [cf. 13, 16, 23) . The energy perturbation during compression and recovery was related to the severity of the impact, with the most pronounced changes in the group in which the trauma led to irreversible paraplegia (group 4) . The ECF increase in lactate and purine metabolites was also related to the degree of SCBF reduction previously determined (19] . The most pronounced changes occurred in group 4 with a tenfold increase in lactate corresponding to complete ischemia during compression . This is similar to the change in whole-brain lactate after 5 min of complete ischemia in fed (normoglycemic) animals [22]. In this situation, adenine nucleotides (ATP, ADP, AMP) are rapidly degraded with an accumulation of inosine and hypoxanthine [8] . Following decompression, SCBF recovers to 60% of control by 60 min postinjury in group 2 [19] . This level of injury results in a myelographic blockage and a slight weakening in SEP, but no neurological deterioration . All ECF metabolites consequently returned to their preinjury level within 20-30 minutes . In group 3 recovery of SCBF is only partial with a flow level of 30% of control reached 60 min after trauma [19) . The biochemical correlate was a slightly delayed recovery of the energy-related metabolites . Lactate tended to remain elevated after decompression

3

a 5 .1 ±0 .3 4 .9 ±0 .1 5 .2 ±0 .2 4 .8 ±0 .3

d

7 .44 *--0 .02 7 .43 ±0 .03 7 .45 ±0 .02 7 .45 `--0 .01

c

5 .0 ---0 .3 5 .1 ±0 .3 5 .1 ±0.1

h 5 .0 ±0 .3 4 .8 '-0 .3 4 .9 ±0 .4 5 .1 ±0.2

(kPa) c

118 -04 11_4 -06 ii A 0,7

b 11 .7 ±1 .0 12 .0 ±0 .5 12 .1 -08 12+0 +0 .6

a 117 ±1 .1 11 .9 ±0 .7 11 .3 ± .11 12 .0 ---0 .8

d 5 .1 ±0.2 4 .8 ±0 .2 5 .0 ---0 .3 4 .9 ±0 .2

Abbreviations Hct, hematoctit, Glucose, blood glucose ; other abbreviations as per text . a, After laminecromy . b, At the end of the steady-state period prior to spinal cord injury, c, At 5 minutes after compression. d, At 2 hours after compression .

7 .42 ±0 .04 7 .43 ±0 .02 7 .44 ±0 .01

7 .42 ±0 .02 7 .42 ±0 .04 7 .44 ±0 .04 7 .43 ±0 .02

7 .43 .-0 .02 7 .44 ±0.02 7 .43 ±0 .03 7 .44 ±0 .03

1

2

c

b

pH

Po (kPa)

± 12 1 2 .t

441 t

±1 . 1

112 ±14 104 -11 113

7 .6 +_1 .6 8 .0 ±-2 .1 +~ 8 .4 11 .2 ±2 .1 13 .1 -2 1 1~ 4 42 '-1 41 1--2 43

43 ±-1 43 ±2 44

11 .4 ±1 .3 118 '02 11 .3 '08 11 .4 +06

a d

b

d

b

Glucose (mmol/L) d

Hct (%)

Physiological Variables Prior to Implantation of Dialysis Probe and Up to 2 Hours After Spinal Cord Injury

a

.

Group

Table

99

+12 15 96 ±5 107

b

' 814

143 ' 16

109 17

d

101 ±5 104

c

117 '--16 128

MAPS (mmHg)



Surg Neurol 1993 ;39 :297-304

302

INOSINE

Group 2

Group 1

1 .5

1 .5 1 .0 0 .5

0.5 0 0

30

60

90

120 150 180

0

30

60

90

120 150 160

Group 4

Group 3

0 Minutes

60

Figure 3 . Effect of graded spinal cord compression trauma on the extracellular inosine levels as reflected by microdialysate concentrations from rats subjected to sham operation (group 1), mild (group 2), moderate (group 3), or severe (group 4) injury . Other conditions as per Figure 2.

90 120 150

Minutes

HYPOXANTHINE

Group 2

Group I 3 .0

3 .0 2 .5 2 .0 1 .5 1 .0

I

0 .5 0

Group 4

Group 3 3 .0

3 .0

2 .5

2 .5

2 .0

2 .0 1 .5

1 .0

Lo

0.5

0 .5 0

0 0

30

60

90

120 150 180

Minutes

0

30

60

90

120 150 180

Minutes

Figure 4 . Effect of graded spinal cord compression trauma on the extracelular hypoxanthinelevels as reflected by microdialysate concentrations from rats subjected to sham operation (group 7) . mild (group 2), moderate (group 3), or severe (group 4) injury . Other conditions as per Figure 2 .



Spinal Cord Injury and Energy Metabolism

Surg Neurol 303 1993 ;39 :297-304

without reaching statistical significance . This probably

jury . The important implication of the penumbra con-

reflected that the posttraumatic level of SCBF was close

cept is that the tissue may be rescued if blood flow is

to the flow threshold for electrical failure resulting in

improved or if the metabolic needs are reduced, within

increased lactate production according to results from

a certain window of opportunity . ECF lactate measured

graded cerebral ischemia experiments [4] . This level of

continuously by microdialysis may be a useful way of

injury, indeed, results in transient paraparesis with a

monitoring the efficacy of new therapies aiming at im-

marked impact on SEP [19], supporting this suggestion . Decompression in group 4 is accompanied by a persis-

proving SCBF and/or reducing the metabolic demand of the injured spinal cord under penumbra conditions .

tent hypoperfusion with an SCBF of 10% of control 60 min postinjury . Although with a considerable delay, this was associated with a complete recovery of ECF inosine and hypoxanthine, probably reflecting the combination of salvage of inosine and hypoxanthine for re-

The authors wish to acknowledge the skillful assistance provided by Ms . Ulla Karlsson and Mr . Patrik Jansson . Financial support was obtained from the Swedish Medical Research Council (project nos . 7888 and 03020), the Laerdal Foundation for Acute Medicine, and the Tore Nilsson s Foundation, Sbderbergs, Hedlunds and Thurings stiftelser .

synthesis of adenine nucleotides and some washout to the systemic circulation ; however, the lactate increase was irreversible (Figure 2) . How can this be explained? Based on results from work on cerebral ischemia, one would expect that the SCBF level of 10% of control is close to the flow threshold for energy and ion pump failure [4] . Thus, the spinal cord tissue may be under penumbra conditions following decompression, which is defined as a state in which the tissue is below the flow threshold for electrical failure and thus being electrically silent, but above the flow threshold for energy and ion pump failure [5] . The metabolic characteristics of this border line situation is that ATP levels are upheld by the oxygen supplied by the remaining perfusion, whereas lactate production is greatly increased (and phosphocreatine levels reduced), reflecting a partial disturbance of energy metabolism [5] . To our knowledge this is the first study applying microdialysis for sequential measurements of the posttraumatic energy perturbation in individual animals . Previous studies have used analysis of whole-tissue energy metabolite concentrations, where the results from one animal can be used only for a single time point . In a study by Anderson et al [23], employing a feline spinal cord compression (190-g load for 5 minutes) injury model resulting in irreversible paraplegia, it was found that lactate increased fourfold and ATP levels dropped . There was no sign of metabolic recovery during the 24hour observation period . The results are not, however,

References I . Austin GM . Trauma . In : Austin GM, ed . The spinal cord, third edition. New York : Igako-Shoin Medical Publishers, 1983 : 215-61 . 2 . Anderson DK, Means ED, Waters TR . Spinal cord energy metabolism in normal and postlaminectomy cats . J Neurosurg 1980 ; 52 :387-91 . 3 . Anderson DK, Means ED, Waters TR, Spears CJ . Spinal cord energy metabolism following compression trauma to the feline spinal cord . J Neurosurg 1980 ;53 :375-80 . 4 . Astrup J, Symon L, Branston NM, Lassen NA . Critical evoked potential and extracellular K' and H' at critical levels of brain ischemia . Stroke 1977 ;8 :51-7 . 5 . Astrup J, Siesjb BK, Symon L . Thresholds in cerebral ischemiathe ischemic penumbra . Stroke 1981 ;12 :723-56 . Balenrine JD . Hypothesis in spinal cord trauma research . In : Central nervous trauma status report . Becker DP, Povlishock JT, eds . Washington : NINCDS, National Institutes of Health . 1985 :455-61 . 7 . Braughler JM, Hall ED . Central system nervous trauma and stroke . Biochemical considerations for oxygen free radical formation and lipid peroxidation . J Free Radic Biol Med 1989 ;289-301 . S . Deuricke B, Gerlach E, Dierkesmann R . Abbau freier Nucleotide in Herz, Skeletmuskel, Gehirn and Leber der Ratte bei Sauersroffmangel . Pflugers Arch Ges Physiol 1966 ;292 :239-54 . 9 . DuckerTB, Walleck CA . Recovery from cord injury . In : Central nervous trauma status report . Becker DP, Povlishock JT, eds . Washington : NINCDS, National Institutes of Health, 1985 : 369-74 .

changes were not measured and because a different spe-

10 . Faden Al . Recent pharmacological advances in experimental spinal cord injury . Theoretical and methodological considerations . Trends Neurosci 1983 ;9 :375

cies was used . The former may explain why the lactate

11 . Faden At, Demediuk P, Panter SS, V ink R. The role of excitatory

directly comparable to ours as regional metabolic

increase was only fourfold, reflecting the average of grey and white matter responses to impact . Our results suggest that there is a partial recovery of energy metabolism early after trauma in the grey matter of the dorsal horn, even though the 50-g compression (group 4) results in an irreversible paraplegia and an extinguished signal conduction [191 . Furthermore, our findings indicate that ECF lactate is a sensitive indicator of secondary ischemia after spinal cord compression in-

amino acids and NMDA receptors in traumatic brain injury . Science 1989;244 :798-800 . 12 . Fredholm BE, Sollevi A . The release of adenosine and inosine from canine subcutaneous adipose tissue by nerve stimulation and noradrenaline . J Physiol (Load) 1981 ;313 :351-67 . 13 . Hagberg H, Andersson P, Lacarewicz J, Jacobson I, Butcher S, Sandberg M . Exrracellular adenosine, inosine, hypoxanthine, and xanthine in relation to tissue nucleotides and purines in rat striatum during transient ischemia . J Neurochem 1987 ;49 :227-31 . 14 . Hallstrbm A, Carlsson A, Hillered L, Ungerstedt U . Simultaneous determination of lactate, pyruvare and ascorbate in microdialysis



304

Surg Neurol 1993 ;39 ;297-304

samples from rat brain, blood, fat and muscle using high performance liquid chromatography . J Pharmacol Meth 1989 ; 22 :113-24 15 . Hillered L, Persson L . Microdialysis for metabolic monitoring in cerebral ischemia and trauma : Experimental and clinical studies . In : Microdialysis in the neurosciences. Robinson TE, Justice Jr JB, eds . Amsterdam : Elsevier, 1991 ;389-405 . 16 . Hillered L, Hallstrbm, Segersvard 5, Persson L, Ungerstedt U . Dynamics of extracellular metabolites in the striatum after middle cerebral artery occlusion in the rat monitored by intracerebral microdialysis . J Cereb Blood Flow Metab 1989 ;9 :607-16 . 17 . Hillered L, Persson L, Ponren U, Ungerstedt U . Neurometabolic monitoring of the ischemic human brain using microdialysis . Acta Neurochir 1990 ;102 :91-7 . 18 . Hillered L, Kotwica Z, Ungerstedt U . Interstitial and cerebrospinal fluid levels of energy-related metaholites after middle cerebral artery occlusion in rats . Res Exp Med 1991 ;191 :219-25 .

Zhang cr al

upon restitution of brain energy metabolism following total ischemia. Brain Res 1974 ;77 :173-86 . 23 . Nilsson P, Hillered L, Ungerstedt U, Ponren U . Changes in cortical extracellular levels of energy related metaholites and amino acids following concussive brain injury in rats . J Cereb Blood Flow Metab 1990 ;10 :631-7 . 24 . Nystrom B, Berglund J-E . Spinal cord restitution following compression injuries in rats . Acts. Neurol Scand 1988 ;78 :467-72 . 25 . Panter SS, Yum SW, Faden AT . Alterations in extracellular amino acids after traumatic spinal cord injury . Ann Neurol 1990 ; 27 :96-9. 26 . Persson L, Hillered L. Chemical monitoring of neurosurgical intensive care patients using intracerebral microdialysis . J Neurosurg 1992 ;76 :72-80 . 27 . Sandier AN, Tator CH . Effect of acute spinal cord compression injury on regional spinal cord blood flow in primates . J Neurosurg 1976 ;45 :660-76 .

19 . Holtz A, Nystrom B, Gerdin B . Spinal cord blood flow measured by 14-C iodoantipyrine auroradiography during and after graded spinal cord compression in rats . Surg Neurol 198931 :35060 .

28 . Simpson RK Jr, Robertson CS, Goodman JC . Release of segmental amino acid transmittors in response to peripheral afferent and motor cortex stimulation . Life Sci 1991 ;49(17) :113-18 .

20 . Liu DX, Valadez V, Sorkin LS, McAdoo DJ . Norepinephrine and serotonin release upon impact injury to rat spinal cord . J Neurorrauma 1990 ;7 :219-27 .

29 . Sorkin LS, Hughes MG, Liu D, Willis Jr WD, McAdoo DJ . Release and metabolism of 5-hydroxytryptamin in the cat spinal cord examined with microdialysis . J Pharmacol Exp Ther 1991 ; 257 :192-9 .

21 . Liu D, Thangnipon W, McAdoo DJ . Excitatory amino acids rise to toxic levels upon impact injury to the rat spinal cord . Brain Res 1991 ;547 :344-8. 22 . Ljunggren B, Norberg K, Siesjo BK . Influence of tissue acidosis

30 . Young W . Blood flow, metabolic and neurophysiological mechanisms in spinal cord injury . In : Central nervous trauma status report. Becker DP, Povlishock JT, eds . Washington: NINCDS, National Institutes of Health . 1985 ;463-73 .