J Neurosurg 85:131–137, 1996
Effect of magnesium given 1 hour after head trauma on
brain edema and neurological outcome
ZEEV FELDMAN, M.D., BORIS GUREVITCH, M.D., ALAN A. ARTRU, M.D.,
ARIEH OPPENHEIM, M.D., ESTHER SHOHAMI, PH.D., ELI REICHENTHAL, M.D.,
AND YORAM SHAPIRA, M.D., PH.D.
Departments of Neurosurgery and Anesthesiology, Soroka Medical Center and the Faculty of Health
Science, Ben-Gurion University, Beer-Sheva, Israel; Department of Anesthesiology, University of
Washington School of Medicine, Seattle, Washington; and Departments of Anesthesiology, Hadassah
Medical Center, and Pharmacology, School of Pharmacy, Hebrew University, Jerusalem, Israel
U Excitatory amino acids (EAA), mainly glutamate and aspartate, are released in excessive amounts from terminals
of ischemic or traumatically injured neurons. These excessive levels of EAAs initiate a cascade of events believed to
lead to secondary delayed damage to the surrounding brain. The N-methyl-D-aspartate receptor antagonists MK-801
and ketamine are reported to suppress excessive EAA release and to attenuate the development of focal brain edema
following neuronal injury. Magnesium is also reported to work at the postsynaptic receptor to reduce the neurotoxic
effect of glutamate. The present study was undertaken to examine the effect of postinjury treatment with Mg++ on brain
edema and neurological outcome after traumatic brain injury.
Sixty-nine rats that survived halothane anesthesia and closed head trauma (CHT) were randomly assigned to one of
seven experimental groups: sham, CHT, and CHT with administration of Mg++ 1 hour postinjury. At 48 hours, brain
tissue Mg++ concentration (calculated from optical density using a standard curve) was significantly increased compared to baseline levels (10.06 6 2.44 mg/g vs. 6.83 6 0.81 mg/g, p , 0.01 calculated by one-way analysis of variance). Also at 48 hours postinjury, brain tissue specific gravity in the contused hemisphere of Mg++-treated rats was
significantly greater than that in the contused hemisphere of untreated rats, indicating attenuation of brain edema formation by Mg++. The neurological severity score (NSS) of rats treated with Mg++ improved significantly at both 18 and
48 hours, compared to baseline values obtained 1 hour after CHT but prior to administration of Mg++ (11.2 6 2.5 vs.
15.2 6 4.1, p = 0.03; and 12.3 6 6.1 vs. 17.3 6 3.6, p = 0.004, respectively). In the untreated groups, the NSS at 18
and 48 hours was not significantly different from baseline values (that is, no neurological improvement). The present
study indicates that postinjury treatment with Mg++ attenuates brain edema formation and improves neurological outcome after experimental CHT.
KEY WORDS • head injury • brain edema • neurological outcome •
magnesium • rat
amino acids (EAA), mainly glutamate
and aspartate, are released into the extracellular space in excessive amounts from terminals of
ischemic or traumatically injured neurons.9,21 These excessive levels of EAAs initiate a cascade of events believed
to lead to secondary delayed damage to the surrounding
brain. Rothman36 and Choi and colleagues3,5 have suggested two possible mechanisms for the induction of cell
death by EAA. The first is the induction of Cl2 and Na+
influx leading to neuronal swelling, and the second is
the induction of Ca++ influx, leading to delayed damage.
Normally, when energy is available, the postsynaptic
response is terminated by reuptake of the EAAs, by extrusion of Ca++ and Na+ from the intracellular space, and
restoration of K+ and Cl2 gradients across the membranes.
However, if energy is lacking due to ischemia or trauma,
postsynaptic activation will be prolonged and enhanced.62
E
XCITATORY
J. Neurosurg. / Volume 85 / July, 1996
Three major EAA receptors have been described, selectively activated by N-methyl-D-aspartate (NMDA), by
kainate, and by quisqualate.62 Two separate postsynaptic
channels have been proposed to be gated by these receptors. A monovalent cation channel activated by the kainate
and quisqualate receptors allows influx of Na+ and efflux
of K+, whereas the other, which is gated by the NMDA
receptor, allows the influx of Ca++ and Na+ and the efflux
of K+.14 A variety of competitive and noncompetitive
antagonists block the NMDA-gated channel. Several of
these NMDA receptor antagonists have been investigated
for their potential as treatments to decrease the effect of
EAAs and reduce toxicity after brain injury.9,14,27,49,55,57
Glutamate release in brain tissue has been reported in
association with brain injury in rats.7,24,25,33 The noncompetitive agent phencyclidine attenuated long-term behavioral deficits after traumatic brain injury (TBI) in the
131
Z. Feldman, et al.
rat.18,57 Ketamine and dextrorphan blocked NMDA excitation, improving the neurological outcome after TBI in rats
and reducing the increase in brain tissue Ca++ and the decrease in brain tissue Mg++.9,44,48 Shapira, et al.,49 reported
that the beneficial effects of ketamine administration following TBI were dose and time related. Ketamine was
effective in ameliorating neurological dysfunction at a
dose of 120 mg/kg when administered 1 to 2 hours postinjury, but it was ineffective at lower doses and when
administered at 4 hours postinjury. Another noncompetitive NMDA receptor antagonist, MK-801, reduced the
increase in tissue Na+ and prevented the decline in total
Mg++ concentration seen after TBI in rats, attenuating the
development of focal brain edema and improving neurological outcome.27,55 An Mg++ deficiency significantly
exacerbated neurological dysfunction and increased mortality following brain injury in rats, whereas pre- and posttreatment intravenous administration of Mg++ prevented
the decline in brain tissue Mg++, and significantly improved posttraumatic neurological outcome.26,28
Posttraumatic brain edema is thought to be generated
in part by blood-brain barrier (BBB) breakdown and extravasation of protein, that is, vasogenic edema, and in
part by cytotoxic edema. The significant reduction in
brain edema formation observed after treatment with
the noncompetitive NMDA receptor antagonist MK-801
is attributed mostly to a decrease in the cytotoxic portion
of the posttraumatic edema.27,55 Magnesium also is an
NMDA receptor antagonist and has been reported to
decrease regional brain tissue water content and to decrease memory impairment after fluid-percussion injury
in rats.31,63
There are no reports on the effect of postinjury treatment with Mg++ on neurological outcome using a scoring system that provides data on motor deficits, reflexes,
mobility, and balance, data that are included in neurological outcome scores used in clinical practice. Accordingly,
the present study was undertaken to examine the effect of
postinjury treatment with Mg++ on neurological outcome
utilizing a scoring system similar to that used in numerous
previous reports on neurological outcome with and without “protective” treatments following brain injury in
rats.10,43–61 In addition, this study sought to determine if the
decrease in regional brain edema previously reported with
Mg++ therapy following fluid-percussion brain injury in
rats would also occur following Mg++ therapy in the present animal model of TBI.
Materials and Methods
Eighty-five Sprague–Dawley rats weighing 260 6 39 g (mean 6
standard deviation (SD)) were initially enrolled in the study. Experimental conditions for the 7 groups are summarized in Table 1.
Animal Model
The weight-drop device was designed to deliver a standard blow
to the cranium resulting in a controlled cerebral injury.52 Impact was
delivered by a silicone-coated 5-mm metal tip extruding from a
platform that falls down a frame. The impact imparted to the cranium of the rat is proportional to the momentum of the platform at the
end of its free fall. For rats weighing 250 to 300 g, a 7-cm height of
free fall was used. This has been shown to produce an impact energy of 0.5 J over the skull.52 The reproducibility of the impact was
determined previously by measuring the velocity developed during
132
the free fall of the platform and the change in the velocity of the tip
on collision with the target. This model of cranial injury has been
used in multiple previous studies.10,43–61 The hemisphere ipsilateral
to the impact displays hemorrhagic contusion that develops to an
area of hemorrhagic necrosis by 18 hours postinjury. Blood-brain
barrier disruption, measured by Evans blue dye extravasation, occurs as early as 15 minutes postimpact, reaches its maximum level
at 4 hours, and resolves thereafter, with BBB integrity returning to
normal at 7 days postinjury.52 Decreased brain tissue specific gravity (indicating cerebral edema) is observed at 15 minutes postinjury,
reaches maximum levels at 24 hours, and gradually resolves thereafter, with brain tissue specific gravity returning to normal values at
7 days postinjury.50
Surgical Procedure
Rats were prepared for surgery by anesthetization with halothane
and allowed to breathe spontaneously. Maintenance of adequate
anesthesia for the experimental procedure was confirmed by the
loss of corneal reflexes. Once the corneal reflexes were abolished, a
midline scalp incision was made and the scalp and underlying muscles were reflected laterally. In 61 rats, closed head trauma (CHT)
was delivered to the skull over the frontal portion of the left cerebral
hemisphere via a weight-drop device. After scalp incision with or
without CHT, anesthesia was discontinued, animals were returned
to their cages, and unlimited food and water were supplied.
Following CHT, 12 rats became apneic and died, and four others
died from massive bleeding. These rats were excluded from the
study. The 45 rats that survived CHT were randomly assigned to
experimental Groups 2, 4, 5, and 7, and the 24 rats receiving anesthesia and surgical preparation without CHT were randomly assigned to experimental Groups 1, 3, and 6.
Determination of Neurological Status
The neurological status of the rats was assessed using the neurological severity score (NSS) outlined in Table 2. The NSS determines the clinical condition of the rat following CHT, with a score
of 0 indicating no neurological deficit and a score of 25 indicating
the most severe impairment. The NSS was determined 1 hour after
the head injury to assess its severity and served as a baseline for
comparison with later evaluations at 18 (Groups 1–4) or 48 hours
(Groups 5–7).
Experimental Protocol
Rats were assigned randomly to one of seven experimental
groups (Table 1). In Group 1 no cranial impact and no Mg++ treatment were given. Rats were killed after 18 hours. In Groups 2, 4, 5,
and 7, cranial impact was delivered. Rats were killed at 18 (Groups
2 and 4) and 48 hours (Groups 5 and 7). In the three other groups
(1, 3, and 6), the scalp was incised and the skull was exposed, but
no cranial impact was delivered. Rats were killed at 18 (Group 3)
and 48 hours (Group 6). At 1 hour after CHT or scalp incision, and
after the NSS had been determined, 600 mg/kg11,35,69 of MgSO4 was
administered subcutaneously to rats in Groups 3, 4, 6, and 7. The
NSS assessed the effect of the experimental treatment and was
determined before the rats were killed. The examination was performed in a lighted room by an observer blinded to the experimental groups.
Determination of Specific Gravity
Rats were decapitated at 18 hours (Groups 1–4) and 48 hours
(Groups 5–7) after the CHT and their brains were rapidly removed
(44 6 6 seconds). For the traumatized hemisphere, tissue samples
were taken from an area adjacent to the traumatic lesion, and for the
nontraumatized hemisphere, samples were taken from a corresponding contralateral site. Samples of up to 100 mg containing
portions of the temporal and frontal cortexes from each hemisphere
(gray and subcortical white matter) were hand cut on an iced cooled
glass plate. Pieces were placed into gradient columns for specific
gravity determination. The specific gravity of brain tissue was
determined by the method of Marmarou, et al.,22 using linear gradient columns of kerosene and bromobenzene. A calibration curve
J. Neurosurg. / Volume 85 / July, 1996
Magnesium, brain edema, and neurological outcome
TABLE 1
Experimental protocol in seven groups of rats undergoing Mg11
treatment for closed head trauma*
Group
No. of
Rats
Closed Head
Trauma
Mg11
Treatment
Time of
Death (hrs)
1
2
3
4
5
6
7
8
13
8
10
13
8
9
no
yes
no
yes
yes
no
yes
no
no
yes
yes
no
yes
yes
18
18
18
18
48
48
48
* Eighty-five rats were initially enrolled, but 16 rats assigned to Groups
2, 4, 5, and 7 did not survive closed head trauma and were excluded from
the study.
was determined for each column using anhydrous KSO4 solutions of
known specific gravity (1.045, 1.040, 1.035, and 1.025).
Determination of Mg++ Content in Brain Tissue
Samples weighing 50 to 100 mg were taken from an area adjacent to the traumatic lesion for determination of Mg++ concentration
in brain tissue in the traumatized hemisphere and from a corresponding contralateral site in the nontraumatized hemisphere. The
tissue was prepared and optical density was determined by atomic
absorption using a densitometer at a wavelength of 572 nm (model
403; Perkin-Elmer Corp., Norwalk, CT). Tissue levels of Mg++ were
calculated from the measured optical density using a standard curve.
The Effect of Mg++on Serum Osmolality
To control for the effect Mg++ might have on serum osmolality
and thus on brain edema, 600 mg/kg of MgSO4 was administered
subcutaneousely to 10 rats. Serum osmolality was determined using
a Piske 1-10 osmometer at 0.5, 1, 24, and 48 hours postinjection.
Statistical Analysis
Brain tissue specific gravity and Mg++ concentrations were tabulated as the mean 6 SD. These data were compared among groups
receiving and not receiving Mg++ treatment using one way analysis
of variance (ANOVA), followed by the Student-Newman-Keuls
multiple comparisons test. The NSS was tabulated as median values
(range) and the data was compared among groups receiving and not
receiving Mg++ using the Mann–Whitney test for nonparametric
data. A probability value of less than 0.05 was considered significant.
Results
Mortality Rate
Ten (38%) of 26 rats in Groups 2 and 5 (CHT without
Mg++ treatment) and three (16%) of 19 in Groups 4 and 7
(CHT with Mg++ treatment) died. The mortality rate was
not statistically different between rats receiving and not
receiving Mg++ (Fisher’s exact test, p = 0.18). The median
NSS for rats that died was 20 (range 14–25), as compared
to 16 (range 9–23) for the surviving rats. Six rats died 3
hours postinjury, six were dead at 18 hours, and one at 48
hours postinjury.
Neurological Outcome
At 1 hour postinjury (baseline), the NSS of rats with
CHT scheduled to subsequently receive Mg++ (Groups 4
and 7) was 17 (range 10–23), not significantly different
J. Neurosurg. / Volume 85 / July, 1996
TABLE 2
Method used to measure neurological severity score for rats at
1, 18, and 48 hours after closed head trauma
Characteristic
mobility
inability to exit from a circle 50 cm in diameter when
placed in center
within 30 min
within 60 min
at .60 min
hemiplegia (inability to resist forced changes in position)
inability to walk straight when placed on the floor
inability to move
reflexes
flexion of hindlimb when raised by the tail
loss of startle reflex
loss of righting reflex
for 20 min
for 40 min
for 60 min
behavior
loss of seeking behavior
prostration
functional tests
failure in beam-walking task
8.5 cm wide
5.0 cm wide
2.5 cm wide
failure in beam-balancing task (1.5 cm wide)
for 20 sec
for 40 sec
for 60 sec
stability on beam balance (1.5 cm wide)
able to walk, normal gait
able to walk, impaired gait
unable to walk, steady balance on beam
unable to walk, steady balance, all limbs on beam
unable to walk, unsteady balance, unable to place all
limbs on beam
effort on beam balance (1.5 cm wide)
unable to stay on the board
unable to try to stay on the board
maximum score
Points
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
1
1
25
from the baseline NSS of rats with CHT assigned to the
untreated groups (16 (range 9–20), Groups 2 and 5).
However, at both 18 and 48 hours, the NSS of rats treated
with Mg++ improved significantly compared to baseline
values (10 (range 7–19) vs. 16 (range 10–18), p , 0.05,
and 8 (range 6–20) vs. 17 (range 13–23), p , 0.05, respectively). There was no significant difference between baseline and NSS at 18 or 48 hours in the untreated groups.
The difference between NSSs at 1, 18, or 48 hours in treated and untreated groups is shown in Fig. 1.
The NSS is a complex value, integrating the results of a
variety of tests (Table 2). The influence of Mg++ treatment
on individual motor functions was therefore examined.
The data of the NSSs were analyzed for the individual
motor tasks. The most pronounced effect was on beam
balance and beam walk at 48 hours postinjury, as summarized in Table 3.
Specific Gravity
The specific gravity of the contused hemisphere at 18
133
Z. Feldman, et al.
TABLE 3
Percentage of rats able to perform motor tasks at 18 and 48
hours after closed head trauma*
18 Hours
Task
48 Hours
CHT +
CHT +
CHT
Mg11
CHT
Mg11
(Group 2) (Group 4) p Value (Group 5) (Group 7) p Value
beam walk (width in cm)
8.5 cm
31.0
5 cm
15.4
2.5 cm
0
beam balance (seconds)
20
46.0
40
31.0
60
15.4
60
40
10
NS
NS
NS
15.4
0
0
66
44
22
,0.05
,0.05
NS
60
50
20
NS
NS
NS
23
0
0
88
66
66
,0.01
,0.01
,0.01
* CHT = closed head trauma; NS = not significant.
† According to Fisher’s exact test.
FIG. 1. Bar graph showing the neurological severity score at 1,
18, and 48 hours after closed head trauma (CHT). The neurological severity score of animals treated with Mg++ was significantly
lower at 18 and 48 hours than at baseline, indicating neurological
recovery (asterisk). The neurological severity score of animals that
were not treated with Mg++ was unchanged compared to the baseline score.
and 48 hours in rats sustaining CHT (Groups 2, 4, 5, and
7) was decreased compared to that in the corresponding
left hemisphere of sham-operated rats (p , 0.05), indicating brain edema in the groups with CHT (Table 4). There
was a significant difference between the specific gravities
of the contused hemispheres of treated and untreated rats
at 48 hours (Groups 5 and 7, p , 0.05). There was no significant difference between the specific gravities of the
contused hemispheres in treated and untreated animals at
18 hours. In both treated and untreated animals, specific
gravity values were significantly higher at 48 hours after
injury compared to the values at 18 hours (p , 0.01).
Concentrations of Mg++ in Brain Tissue
Concentrations of Mg++ in the brain tissue of the different experimental groups are summarized in Table 5.
Baseline Mg++ levels in brain tissue as measured in Group
1 (sham) were 6.83 6 0.81 mg/g brain tissue. Compared
to this baseline value, treatment with 600 mg/kg of
MgSO4 significantly increased Mg++ concentration in
brain tissue after 48 hours to 10.06 6 2.44 mg/g in Group
6 and to 9.35 6 1.75 mg/g in Group 7 (p , 0.01,
ANOVA). However, the Mg++ concentration in the brain
tissue of Groups 6 and 7 was not significantly different
compared to that at 48 hours in Group 5 (CHT without
Mg++ treatment). At 18 and 48 hours after CHT, Mg++ concentration in the brain tissue of the groups receiving CHT
and no Mg++ (Groups 2 and 5) was not significantly different from the baseline Mg++ concentration in brain tissue
(Group 1).
Serum Osmolality After Mg++ Administration
Baseline serum osmolality was 285 6 15 mOsm/kg.
At 0.5 hour postinjection of 600 mg/kg of MgSO4, se134
rum osmolality was 302 6 14 mOsm/kg, at 1 hour it was
290 6 16 mOsm/kg, at 24 hours it was 280 6 17
mOsm/kg, and at 48 hours it was 290 6 14 mOsm/kg. At
no time was the serum osmolality statistically different
from baseline values.
Discussion
The results of the present study demonstrate that postinjury treatment with Mg++ after experimental CHT attenuated brain edema formation and significantly improved
neurological outcome. One possible mechanism for its
beneficial effect following CHT is the positive role of
Mg++ in general cellular metabolism and function. Magnesium is essential for normal cell functions such as membrane integrity, cellular respiration, transcription by messenger RNA, protein synthesis, glucose and energy
metabolism, maintenance of normal Na+ and K+ gradients,
and regulation of Ca++ transport and accumulation.4,6,
8,13,14,39,66–68
A second possible mechanism for its beneficial
effect following CHT is the antagonistic action of Mg++ on
NMDA receptors. Magnesium ions are known to exert a
gating effect on the NMDA receptors; Mg++ blocks the
NMDA receptor–ion channel in the brain and in the spinal
cord.23,30 Depletion of Mg++ exposes the neurons to the
toxic effect of the EAAs, and impaired Na+, K+, and Ca++
gradients promote further damage to the injured brain.8 In
addition, intracellular Mg++ has also been shown to have a
voltage-gating role in NMDA receptors and to regulate
the release of EAAs.20,37,38 Although some noncompetitive NMDA receptor antagonists (MK-801, ketamine, and
phencyclidine) have been shown to have a more potent
protective effect after brain trauma, their severe side effects limit their usefulness in humans.14 By comparison, as
an NMDA receptor antagonist Mg++ is appealing as a safe
therapeutic agent for the injured brain.
The results of the present study demonstrated that postinjury treatment with Mg++ attenuated brain edema formation to some extent. The Mg++ presumably decreased
edema formation by a direct effect on regulation of normal intracellular Na+ and K+ gradients, the impairment of
which might enhance posttraumatic edema formation,8
J. Neurosurg. / Volume 85 / July, 1996
Magnesium, brain edema, and neurological outcome
TABLE 4
Specific gravity values of the left hemisphere in the different
experimental groups of rats receiving closed head trauma
Group
Time of
Death (hrs)
No. of
Samples
1
2
3
4
5
6
7
18
18
18
18
48
48
48
8
8
8
8
8
8
8
Lt Hemisphere
Specific Gravity*
1.0455 6 0.002
1.0317 6 0.002†
1.0458 6 0.008
1.0325 6 0.019†
1.0345 6 0.002†‡
1.0459 6 0.005
1.0361 6 0.002†‡§
* Values are expressed as means 6 standard deviation.
† Significantly different from Group 1, p , 0.05.
‡ Significantly different from corresponding treatment group at 18
hours, p , 0.05.
§ Significantly different from untreated Group 5, p , 0.05.
and/or an NMDA receptor antagonist effect that protected
neurons from the deleterious effect of EAAs and thus reduced cytotoxic brain edema. The fact that posttraumatic
edema is partially due to BBB disruption may account for
the lack of total reversal of brain edema formation. The
present study also demonstrated a significant improvement in neurological outcome after CHT in rats treated
with Mg++ at 1 hour postinjury, compared to animals without Mg++ treatment. These results are in concordance with
numerous studies. In vivo pretreatment with Mg++ attenuated its decline in the brain tissue, significantly improved
neurological outcome,26 and protected against irreversible damage after spinal ischemia,34,65 whereas preinjury
dietary depletion of Mg++ resulted in lower Mg++ concentrations in the brain and worsened neurological outcome.26,67 Postinjury treatment with Mg++ 30 minutes after
brain trauma has been reported to improve neurological
outcome compared to controls.28
The results of the present study are consistent with two
previous studies on the effect of Mg++ on brain edema and
memory after TBI.31,63 In one study, Sprague–Dawley rats
weighing 350 to 400 g received 125 mmol of MgCl2 intravenously at 15 minutes after fluid-percussion injury to the
parasagittal region of the brain. The rats had been trained
prior to injury to locate a platform in a Morris water maze.
At 42 hours after injury, decreased median memory scores
(indicating memory dysfunction) seen in untreated rats
were doubled (indicating improved memory) in rats treated with MgCl2.63 In the second study, Sprague–Dawley
rats weighing 350 to 400 g received 300 mmol/kg of
MgCl2 intravenously at 15 minutes after fluid-percussion
injury to the parietal cortex. At 48 hours postinjury, brain
tissue water content was decreased in the ipsilateral hippocampus of the rats receiving MgCl2 compared to untreated rats, but not at the injury site, adjacent to the injury
site, ipsilateral thalamus, or contralateral sites.31
Several studies have reported a decline of Mg++ concentration in brain tissue following TBI in untreated
rats.27,66,67 However, in the present study no significant difference was found between Mg++ concentrations at 18 and
48 hours postinjury. Several factors may contribute to the
finding of decreased Mg++ concentration in the brain tissue of the untreated rats in previous studies27,66,67 but not in
the present study. One possible factor is the difference in
J. Neurosurg. / Volume 85 / July, 1996
TABLE 5
Brain tissue Mg11 levels in the different experimental groups of
rats receiving closed head trauma
Group
Time of
Death (hrs)
No. of
Samples
Brain Tissue
Mg11 (mg/g)*
1
2
3
4
5
6
7
18
18
18
18
48
48
48
9
5
6
7
5
12
7
6.83 6 0.81
7.75 6 0.95
8.9 6 1.53
7.79 6 1.47
9.19 6 4.45
10.06 6 2.44†
9.35 6 1.75†
* Values are expressed as means 6 standard deviation.
† Significantly different from Group 1, p , 0.05.
method of producing cerebral injury. Previous studies
used a fluid-percussion model of cerebral injury. Unlike
the present model, which produces injury at the site of cranial impact, injury produced by some fluid-percussion
models is ipsilateral to, but remote from, the site of dural
impact.27 Although a decrease in Mg++ concentration in
brain tissue was reported in samples from the injury site at
48 hours after fluid-percussion injury, Mg++ concentration
in brain tissue samples from the impact site at 48 hours
was not significantly different from control.27 A second
possible factor is the difference in severity of injury. With
the fluid-percussion model, it was reported that the magnitude and duration of decrease in Mg++ concentration in
brain tissue following TBI was positively related to the
severity of injury.66,67 The fact that the Mg++ concentration
in brain tissue was decreased at 48 hours after fluid-percussion injury,27 but not in the present study at 48 hours
after CHT, may be related to the severity of injury as
reflected by the 25% to 36% mortality rate after fluid-percussion injury,27 as compared to the 19% mortality in the
present study. A third possible factor is the difference in
the anesthetics used during surgical preparation and delivery of TBI. Halothane, the anesthetic used in the present
study, exerts significantly different effects on movement
into and within brain tissue of water, electrolytes, small
hydrophilic amino acids, proteins, and glucose,1,2,12,15–17,
29,32,40–42,64
as compared to pentobarbital, the anesthetic
used in the previous studies.27,66,67
Not all previous studies have reported a decline of Mg++
concentration in brain tissue following brain injury in
untreated subjects. Helpern, et al.,19 used magnetic resonance spectroscopy to study patients with acute focal cerebral ischemia and found a significant elevation of free
intracellular Mg++ during the early days after onset. A significant association between brain tissue acidosis and high
Mg++ concentration was established, and the authors postulated that breakdown of high-energy phosphates or competitive displacement from adenosine triphosphate binding by H+ accounts for this rise. This may explain the trend
toward higher Mg++ concentration found at 48 hours in
animals sustaining CHT compared to sham-operated animals in the present study.
Conclusions
In summary, treatment with Mg++ at 1 hour after CHT
135
Z. Feldman, et al.
increased the concentration of Mg++ in brain tissue at 48
hours, increased brain tissue specific gravity (indicating
an attenuation of brain edema formation) at 48 hours, and
improved the NSS at 18 and 48 hours. These beneficial
effects of Mg++ in TBI are similar to those previously reported with other NMDA receptor antagonists. Magnesium may hold more promise for clinical use following
CHT, based on previous reports of minimal side effects
with Mg++ but severe side effects with other NMDA receptor antagonists.14
18.
19.
20.
References
1. Alexander SC, Helmer PH, Ramirez O, et al: Effects of general
anesthesia on canine blood-brain barrier glucose transport, in
Harper M, Jennett B, Miller D, et al (eds): Blood Flow and
Metabolism in the Brain. Edinburgh: Churchill Livingstone,
1975, pp 9.37–9.40
2. Artru AA: Reduction of cerebrospinal fluid pressure by hypocapnia: changes in cerebral blood volume, cerebrospinal fluid volume and brain tissue water and electrolytes. II. Effects of
anesthetics. J Cereb Blood Flow Metab 8:750–756, 1988
3. Choi DW: Calcium-mediated neurotoxicity: relationship to specific channel types and role in ischemic damage. Trends Neurosci 11:465–469, 1988
4. Choi DW: Ionic dependence of glutamate neurotoxicity. J Neurosci 7:369–379, 1987
5. Choi DW, Maulucci-Gedde MA, Kriegstein AR: Glutamate
neurotoxicity in cortical cell culture. J Neurosci 7:357–368,
1987
6. Clark AF, Roman IJ: Mg2+ inhibition of Na2+-stimulated Ca2+
release from brain mitochondria. J Biol Chem 255:6556–6558,
1980
7. Condorelli DF, Dell’Albani P, Kaczmarek L, et al: Glial fibrillary acidic protein messenger RNA and glutamine synthetase
activity after nervous system injury. J Neurosci Res 26:
251–257, 1990
8. Ebel H, Günther T: Magnesium metabolism: a review. J Clin
Chem Clin Biochem 18:257–270, 1980
9. Faden AI, Demediuk P, Panter SS: The role of excitatory amino
acids and NMDA receptors in traumatic head injury. Science
244:798–800, 1989
10. Feldman Z, Zachari S, Reichenthal E, et al: Brain edema and
neurological status with rapid infusion of lactated Ringer’s or
5% dextrose solution following head trauma. J Neurosurg 83:
1060–1066, 1995
11. Fischer S, Wolf G, Keilhoff G, et al: Mg2+ antagonizes aamino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-induced
brain damage and convulsions. Eur J Pharmacol 232:
121–124, 1993
12. Forster A, Van Horn K, Marshall LF, et al: Anesthetic effects
on blood-brain barrier function during acute arterial hypertension. Anesthesiology 49:26–30, 1978
13. Garfinkel L, Garfinkel D: Magnesium regulation of the glycolytic pathway and the enzymes involved. Magnesium 4:
60–72, 1985
14. Gentile NT, McIntosh TK: Antagonists of excitatory amino
acids and endogenous opioid peptides in the treatment of experimental central nervous system injury. Ann Emerg Med 22:
1028–1034, 1993
15. Gjedde A, Rasmussen M: Pentobarbital anesthesia reduces
blood-brain glucose transfer in the rat. J Neurochem 35:
1382–1387, 1980
16. Gumerlock MK, Neuwelt EA: The effects of anesthesia on
osmotic blood-brain barrier disruption. Neurosurgery 26:
268–277, 1990
17. Hannan CJ Jr, Kettler TM, Artru AA, et al: Blood-brain barrier
permeability during hypocapnia in halothane-anesthetized
136
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
monkeys, in Strandl FL (ed): Fourth Colloquium in Biological Sciences: Blood-Brain Transfer. Annals of the New
York Academy of Sciences, Vol 529. New York: The New
York Academy of Sciences, 1988, pp 172–174
Hayes RL, Jenkins LW, Lyeth BG, et al: Pretreatment with
phencyclidine, an N-methyl-D-aspartate antagonist, attenuates
long-term behavioral deficits in the rat produced by traumatic
brain injury. J Neurotrauma 5:259–274, 1988
Helpern JA, Vande Linde AMQ, Welch KMA, et al: Acute elevation and recovery of intracellular [Mg2+] following human
focal cerebral ischemia. Neurology 43:1577–1581, 1993
Johnson JW, Ascher P: Voltage-dependent block by intracellular Mg2+ of N-methyl-D-aspartate-activated channels. Biophys
J 57:1085–1090, 1990
Kohmura E, Yamada K, Hayakawa T, et al: Hippocampal neurons become more vulnerable to glutamate after subcritical
hypoxia: an in vitro study. J Cereb Blood Flow Metab 10:
877–884, 1990
Marmarou A, Poll W, Shulman K, et al: A simple gravimetric
technique for measurement of cerebral edema. J Neurosurg
49:530–537, 1978
Mayer ML, Westbrook GL, Guthrie PB: Voltage-dependent
block of Mg2+ of NMDA responses in spinal cord neurones.
Nature 309:261–263, 1984 (Letter)
McDonald JW, Fix AS, Tizzano JP, et al: Seizures and brain
injury in neonatal rats induced by 1S,3R-ACPD, a metabotropic glutamate receptor agonist. J Neurosci 13:4445–4455,
1993
McDonald JW, Johnston MV: Pharmacology of N-methyl-Daspartate-induced brain injury in an in vivo perinatal rat model.
Synapse 6:179–188, 1990
McIntosh TK, Faden AI, Yamakami I, et al: Magnesium deficiency exacerbates and pretreatment improves outcome following traumatic brain injury in rats: 31P magnetic resonance spectroscopy and behavioral studies. J Neurotrauma 5:17–31,
1988
McIntosh TK, Vink R, Soares H, et al: Effect of noncompetitive
blockade of N-methyl-D-aspartate receptors on the neurochemical sequelae of experimental brain injury. J Neurochem 55:
1170–1179, 1990
McIntosh TK, Vink R, Yamakami I, et al: Magnesium protects
against neurological deficit after brain injury. Brain Res 482:
252–260, 1989
Nemoto EM, Stezoski SW, MacMurdo D: Glucose transport
across the rat blood-brain barrier during anesthesia. Anesthesiology 49:170–176, 1978
Nowak L, Bregstovski P, Archer P, et al: Magnesium gates glutamate-activated channels in mouse central neurones. Nature
307:462–465, 1984 (Letter)
Okiyama K, Smith DH, Gennarelli TA, et al: The sodium channel blocker and glutamate release inhibitor BW1003C87 and
magnesium attenuate regional cerebral edema following experimental brain injury in the rat. J Neurochem 64:802–809, 1995
Pashayan AG, Mickle JP, Vetter TR, et al: Blood-CSF barrier
function during general anesthesia in children undergoing ventriculoperitoneal shunt placement. Anesth Rev 15:30–31, 1988
Patel AJ, Hunt A, Seaton P: The mechanism of cytosine arabinoside toxicity on quiescent astrocytes in vitro appears to be
analogous to in vivo brain injury. Brain Res 450:378–381,
1988
Robertson CS, Foltz R, Grossman RG, et al: Protection against
experimental ischemic spinal cord injury. J Neurosurg 64:
633–642, 1984
Rothe F, Wolf G, Fischer S, et al: Quinolinate and kainate facilitate magnesium penetration into brain tissue. Neuroreport 4:
205–207, 1993
Rothman SM: The neurotoxicity of excitatory amino acids is
produced by passive chloride in flux. J Neurosci 5:1483–1489,
1985
J. Neurosurg. / Volume 85 / July, 1996
Magnesium, brain edema, and neurological outcome
37. Rothman SM: Synaptic activity mediates death of hypoxic neurons. Science 220:536–537, 1983
38. Rothman SM, Olney JW: Glutamate and the pathophysiology
of hypoxic-ischemic brain damage. Ann Neurol 19:105–111,
1986
39. Rubin H: Magnesium deprivation reproduces the coordinate
effects of serum removal or cortisol addition on transport and
metabolism in chick embryo fibroblasts. J Cell Physiol 89:
613–626, 1976
40. Sage JI, Duffy TE: Pentobarbital anesthesia: influence on amino
acid transport across the blood-brain barrier. J Neurochem 33:
963–965, 1979
41. Saija A, Princi P, De Pasquale R, et al: Modifications of the permeability of the blood-brain barrier and local cerebral metabolism in pentobarbital- and ketamine-anaesthetized rats. Neuropharmacology 28:997–1002, 1989
42. Schettini A, Furniss WW: Brain water and electrolyte distribution during the inhalation of halothane. Br J Anaesth 51:
1117–1124, 1979
43. Shapira Y, Artru AA, Cotev S, et al: Brain edema and neurologic status following head trauma in the rat. No effect from
large volumes of isotonic or hypertonic intravenous fluids, with
or without glucose. Anesthesiology 77:79–85, 1992
44. Shapira Y, Artru AA, Lam AM: Ketamine decreases cerebral
infarct volume and improves neurological outcome following
experimental head trauma in rats. J Neurosurg Anesthesiol
4:231–240, 1992
45. Shapira Y, Artru AA, Qassam N, et al: Brain edema and neurologic status with rapid infusion of 0.9% saline or 5% dextrose
after head trauma. J Neurosurg Anesthesiol 7:17–25, 1995
46. Shapira Y, Artru AA, Yadid G, et al: Methylprednisolone does
not decrease eicosanoid concentrations or edema in brain tissue
or improve neurologic outcome after head trauma in rats.
Anesth Analg 75:238–244, 1992
47. Shapira Y, Davidson E, Weidenfeld Y, et al: Dexamethasone
and indomethacin do not affect brain edema following head
injury in rats. J Cereb Blood Flow Metab 8:395–402, 1988
48. Shapira Y, Lam AM, Artru AA, et al: Ketamine alters calcium
and magnesium in brain tissue following experimental head
trauma in rats. J Cereb Blood Flow Metab 13:962–968, 1993
49. Shapira Y, Lam AM, Eng CC, et al: Therapeutic time window
and dose response of the beneficial effects of ketamine in
experimental head injury. Stroke 25:1637–1643, 1994
50. Shapira Y, Setton D, Artru AA, et al: Blood-brain barrier permeability, cerebral edema, and neurologic function after closed
head injury in rats. Anesth Analg 77:141–148, 1993
51. Shapira Y, Shohami E: Experimental studies on brain oedema
after blunt head injury: experimental approaches from animal
experimentation to actual or possible clinical application. Eur
J Anaesthesiol 10:155–173, 1993
52. Shapira Y, Shohami E, Sidi A, et al: Experimental closed head
injury in rats: mechanical, pathophysiologic, and neurologic
properties. Crit Care Med 16:258–265, 1988
53. Shapira Y, Yadid G, Cotev S, et al: Accumulation of calcium in
the brain following head trauma. Neurol Res 11:169–172,
1989
54. Shapira Y, Yadid G, Cotev S, et al: OKY-046 inhibits throm-
J. Neurosurg. / Volume 85 / July, 1996
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
boxane synthesis with no effect on brain edema and neurological status in head traumatized rats. Prostagl Leukot Essent
Fatty Acids 36:49–55, 1989
Shapira Y, Yadid G, Cotev S, et al: Protective effect of MK801
in experimental brain injury. J Neurotrauma 7:131–139, 1990
Shohami E, Nates JL, Glantz L, et al: Changes in brain polyamine levels following head injury. Exp Neurol 117:189–195,
1992
Shohami E, Novikov M, Bass R: Long-term effect of HU-211,
a novel non-competitive NMDA antagonist, on motor and
memory functions after closed head injury in the rat. Brain Res
674:55–62, 1995
Shohami E, Shapira Y, Cotev S: Experimental closed head injury in rats: prostaglandin production in a noninjured zone.
Neurosurgery 22:859–863, 1988
Shohami E, Shapira Y, Rosenthal J, et al: Superoxide dismutase
activity is not affected by closed head injury in rats. J Basic
Clin Physiol Pharmacol 2:103–109, 1991
Shohami E, Shapira Y, Sidi A, et al: Head injury induces increased prostaglandin synthesis in rat brain. J Cereb Blood
Flow Metab 7:58–63, 1987
Shohami E, Shapira Y, Yadid G, et al: Brain phospholipase A2
is activated after experimental closed head injury in the rat. J
Neurochem 53:1541–1546, 1989
Siesjo BK, Bengtsson F: Calcium fluxes, calcium antagonists,
and calcium-related pathology in brain ischemia, hypoglycemia, and spreading depression: a unifying hypothesis. J
Cereb Blood Flow Metab 9:127–140, 1989
Smith DH, Okiyama K, Gennarelli TA, et al: Magnesium and
ketamine attenuate cognitive dysfunction following experimental brain injury. Neurosci Lett 157:211–214, 1993
Stewart PA, Hayakawa EM, Carlen PL: Ethanol and pentobarbital in combination increase blood-brain barrier permeability
to horseradish peroxidase. Brain Res 443:12–20, 1988
Vacanti FX, Ames A III: Mild hypothermia and Mg++ protect
against irreversible damage during CNS ischemia. Stroke 15:
695–698, 1984
Vink R, McIntosh TK, Demediuk P, et al: Decline in intracellular free Mg2+ is associated with irreversible tissue injury after
brain trauma. J Biol Chem 263:757–761, 1988
Vink R, McIntosh TK, Demediuk P, et al: Decrease in total and
free magnesium concentration following traumatic brain injury
in rats. Biochem Biophys Res Commun 149:594–599, 1987
Vink R, McIntosh TK, Weiner MW, et al: Effects of traumatic
brain injury on cerebral high-energy phosphates and pH: a 31P
magnetic resonance spectroscopy study. J Cereb Blood Flow
Metab 7:563–571, 1987
Wolf G, Fischer S, Hass P, et al: Magnesium sulphate subcutaneously injected protects against kainate-induced convulsions
and neurodegeneration: in vivo study on the rat hippocampus.
Neuroscience 43:31–34, 1991
Manuscript received September 27, 1995.
Accepted in final form February 14, 1996.
Address reprint requests to: Zeev Feldman, M.D., Department of
Neurosurgery, Baylor College of Medicine, 6560 Fannin, Suite 900,
Houston, Texas 77030.
137