The Pathophysiology of Concussion
Stefano Signoretti, MD, PhD, Giuseppe Lazzarino, PhD, Barbara Tavazzi, PhD,
Roberto Vagnozzi, MD, PhD
Abstract: Concussion is defined as a biomechanically induced brain injury characterized
by the absence of gross anatomic lesions. Early and late clinical symptoms, including
impairments of memory and attention, headache, and alteration of mental status, are the
result of neuronal dysfunction mostly caused by functional rather than structural abnor-
malities. The mechanical insult initiates a complex cascade of metabolic events leading to
perturbation of delicate neuronal homeostatic balances. Starting from neurotoxicity, ener-
getic metabolism disturbance caused by the initial mitochondrial dysfunction seems to be
the main biochemical explanation for most postconcussive signs and symptoms. Further-
more, concussed cells enter a peculiar state of vulnerability, and if a second concussion is
swelling. This condition of concussion-induced brain vulnerability is the basic pathophys-
iology of the second impact syndrome. N-acetylaspartate, a brain-specific compound
representative of neuronal metabolic wellness, is proving a valid surrogate marker of the
post-traumatic biochemical damage, and its utility in monitoring the recovery of the
aforementioned “functional” disturbance as a concussion marker is emerging, because it is
easily detectable through proton magnetic resonance spectroscopy.
PM R 2011;3:S359-S368
Concussion is the most common form of traumatic brain injury (TBI) worldwide [1,2]. In
European countries, approximately 235 people per 100,000 are admitted annually to the
hospital after TBI, 80% of which are classified as belonging in the mild TBI (mTBI) category
[3,4]. This phenomenon mirrors U.S. figures, in which approximately 1.5-8 million people
of mTBI is relatively high, death from this type of trauma appears to be very low (6-10 per
100,000/year), and only 0.2% of all patients with mTBI who visit emergency departments
(EDs) will die as a direct result of this injury .
broadly accepted view is that mTBI is indeed a very frequent entity but is not a very serious
injury, leading only to transient disturbances, and that no intervention other than observa-
tion typically is required [5-10]. However, according to a recent report revealing that the
diagnosis of an intracranial hematoma in such patients was made with a median delay of 18
hours , the quality of the “observation” that mildly injured patients receive while in the
hospital is of utmost concern. In the United States, it has been found that neurological
observations were documented in only 50% of patients admitted with a mild head injury
, and in Europe, patients with mTBI historically have been observed on nonspecialist
wards by nurses and doctors not experienced in neurological observation. The issue of
whether to perform imaging tests, observe, or discharge a patient with mTBI is one of the
be under-reported by patients and underestimated by physicians [6-11].
The label “mild” in mTBI does not reflect the severity of the underlying metabolic and
physiologic processes, if not even the potential clinical manifestations. The word “mild”
implies the general absence of overt structural brain damage. However, long beyond the
typically reported recovery interval of 1 week to 3 months, at least 15% of persons with a
S.S. Division of Neurosurgery, Department of
Neurosciences Head and Neck Surgery, S.
Camillo Hospital, Rome, Italy
Disclosure: nothing to disclose
G.L. Department of Biology, Geology and En-
vironmental Sciences, Division of Biochemis-
try and Molecular Biology, University of Cata-
nia, Catania, Italy
Disclosure: nothing to disclose
B.T. Institute of Biochemistry and Clinical Bio-
chemistry, Catholic University of Rome, Rome,
Disclosure: nothing to disclose
R.V. Department of Neurosciences, University
of Rome “Tor Vergata,” Via Montpellier 1,
Rome 00133, Italy Address correspondence
to R.V.; e-mail: firstname.lastname@example.org
Disclosure: nothing to disclose
Printed in U.S.A.
© 2011 by the American Academy of Physical Medicine and Rehabilitation
Vol. 3, S359-S368, October 2011
history of mTBI continue to see their primary care physician
because of persistent problems [12-16]. In addition, various
health care professionals frequently become involved in the
care of persons with mTBI, including family practice physi-
cians, behavioral psychologists, clinical psychologists, neu-
ropsychologists, neurologists, psychiatrists, neuro-ophthal-
mologists, neurosurgeons, physiatrists, nurses, occupational
therapists, and physical therapists.
Awareness of the potential of a high level of disability after
mTBI is increasing. The provision of comprehensive diagnostic
otherwise would spend prolonged periods off work or depen-
dent on others. Yet considerable confusion and inconsistency
authors’ effort to piece together the current concepts and the
and to emphasize the nuances involved in conducting research
in this area.
Although concussion certainly is blended into the vast world of
and distinct entity because not all cases of mTBI are truly
“concussive”; thus the 2 terms refer to different constructs and
should not be used interchangeably . That being said, the
authors understand the common synonymous acceptance of
mTBI and concussion. During the past decade, concussion has
been considered by 3 international consensus conferences in
which it has been defined and redefined by a panel of experts,
until finally and unanimously the following statement was
reached: “Concussion is a complex pathophysiological process
Given this general and propaedeutical definition, several
common features were added by the consensus panel to better
explain the nature of this peculiar brain injury . In brief,
concussion typically results in the rapid onset of short-lived
impairment of neurologic function, which resolves spontane-
ously. Postconcussive symptoms may be prolonged in a small
percentage of cases, but the acute clinical symptoms largely
reflect a functional disturbance rather than a structural injury,
which usually is confirmed by the absence of abnormalities on
not involve loss of consciousness.
Notwithstanding such a comprehensive, well-designed, and
multifunctional definition, a certain degree of confusion still
exists regarding the compelling pathomechanisms that are trig-
gered by the mechanical insult and that unfold thereafter. The
dominant theory that diffuse axonal injury (DAI) is the main
neuropathological process behind concussion is proving to be
weak or, at best, inconclusive, given the current literature and
the fact that neuronal injury inherent to mTBI improves with
few lasting clinical sequelae in the vast majority of patients.
of the spectrum continuum that is DAI. A large body of clinical
and experimental evidence suggests that such a distinctive
consequence of complex biochemical and neurochemical cas-
The distinction between DAI and concussion is not merely
theoretical, and from a biomechanical perspective, it is well
and 18,000 rad/s2are required, respectively. Although other
is capable of generating DAI , it recently has been reported
that much smaller head acceleration values ranging from 4500
to 5500 rad/s2are needed to provoke a concussion .
THE MECHANICAL INSULT AND THE
“IGNITION” OF THE NEUROCHEMICAL
Concussive head injury causes the brain to experience a me-
chanical “shake,” by virtue of the action of the acceleration and
ical and neurometabolic events.
The sudden stretching of the neuronal and axonal mem-
branes initiates an indiscriminate flux of ions through previ-
ously regulated ion channels and transient physical membrane
of a multitude of neurotransmitters, particularly excitatory
amino acids (EAAs) [27,28], resulting in further changes of
neuronal ionic homeostasis. Among the EAAs, glutamate plays
and D-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
ionic channels. N-methyl-d-aspartate receptor activation is re-
sponsible for a further depolarization, ultimately causing an
influx of calcium ions into the cells.
The essential point of this post-traumatic ionic cellular
derangement is mitochondrial calcium overloading [29-31],
which is responsible for inducing changes of inner mem-
brane permeability with consequent malfunctioning, uncou-
pling of oxidative phosphorylation, and finally, organelle
swelling [32,33]. As suggested by experiments in which the
mitochondrial capacity to catalyze the tetravalent reduction
of molecular oxygen through the electron transport chain
appears compromised, dysfunctional mitochondria become
the main intracellular source of reactive oxygen species
(ROS) [34-36], inducing a phenomenon known as oxidative
stress. The occurrence of an overproduction of ROS beyond
Signoretti et alPATHOPHYSIOLOGY OF CONCUSSION
sively leads to a decrease of antioxidant cell defenses with
consequent irreversible modification of biologically impor-
tant macromolecules. ROS-mediated damage, mainly char-
acterized by the onset of lipid peroxidation, is revealed by
measuring tissue malondialdehyde, a compound undetect-
able during normal conditions .
As clearly demonstrated in bench studies, this event occurs
very rapidly, starting 1 minute after trauma and persisting for
24-48 hours after injury . Once the lipid peroxidation
reaction chain is initiated, it spontaneously propagates and
causes significant ascorbate depletion, explained either by the
direct oxidizing action of ROS on ascorbate or by its use in the
redox cycling of ?-tocopherol (vitamin E), which represents
the only membrane-bound lipid soluble compound capable of
breaking lipid peroxidation reaction chain .
Although no full explanation has been found, a collateral
phenomenon occurring during the onset of oxidative stress is
a condition that jeopardizes all the oxidoreductive reactions,
anisms for this phenomenon are the hydroxyl radical–induced
hydrolysis of the N-glycosidic bond of reduced nicotinamide
adenine dinucleotide (phosphate) and the activation of the oxi-
dized form of the enzyme nicotinamide adenine dinucleotide
glycohydrolase . Both mechanisms cause the hydrolysis of
is, adenosine diphosphate (ADP)-ribose(P) and nicotinamide.
Experimental evidence showed that methylenetetrahydrofolate
reductase can be subject to direct ROS attack and subsequent
irreversible degradation of a consistent amount of the reduced
To re-establish pretrauma ionic balance, the Na1/K1 adeno-
sine triphosphate (ATP)-dependent pumps must work at their
maximal capacities, and a high level of glucose oxidation is
urgently required to satisfy this sudden increased energy de-
drial functioning, most of glucose consumption is coupled to
oxygen consumption, thus optimizing ATP generation. How-
ever, damaged by the calcium overloading and under multiple
attacks from ROS, most of these oxidoreductive reactions are
impaired, and the mitochondria cannot maintain the correct
phosphorylating capacity. This scenario results in a rapid net
such as high-energy phosphates (eg, ATP and guanosine
triphosphate). This phenomenon is mirrored by a proportional
increase of their dephosphorylated products (ie, ADP, adeno-
sine monophosphate, guanosine diphosphate, guanosine
monophosphate, nucleosides, and oxypurines). Particularly in-
teresting are the increases of xanthine (5?) and uric acid (7?),
which strongly suggests the activation of xanthine oxidase, a
tion via the catalytic mechanism of this enzyme .
Thus it happens that during the time of maximum energy
request, the concussion-induced transient mitochondrial
malfunctioning causes an imbalance between ATP consump-
tion and production, a condition that obligates neurons to
work overtime via the more rapid, but less efficient, oxygen-
glucose consumption and the yet unfulfilled energy require-
ment explain the paradoxical temporary increase in neuronal
glucose consumption, notwithstanding a period of general
metabolic depression. In fact, local cerebral metabolic rates
levels within the first 30 minutes after injury and may last
from 30 minutes to 4 hours [46-50].
The overall evidence from these studies demonstrates that
the traumatic insult is directly responsible for sudden bio-
chemical changes, beginning immediately after injury, and
leads to subsequent depression of brain energy metabolism.
Even if it is considered a “mild” form of TBI, concussion is
able to cause profound biochemical changes, with the only
difference being that the described modifications are fully
reversible . As recently reported, the metabolic derange-
ment and the post-mTBI “energy crisis” are considered
chiefly responsible for the compromised synaptic plasticity
and the subsequent cognitive deficits .
A SURROGATE MARKER OF
POSTCONCUSSIVE BRAIN DAMAGE:
When proton magnetic resonance spectroscopy (1H-MRS) was
applied to the human brain, it was evident that the most prom-
inent proton signal, detectable after having suppressed the pro-
ton signal of water, was that of a metabolite known as N-
acetylaspartate (NAA). Subsequently, NAA became the most
in brain1H-MRS studies. These findings captured the attention
of the general neurosciences, dramatically accelerating the pace
of research into the neurochemistry and neurobiology of a
molecule indeed definable as “unique” .
Although the exact biochemical role of this compound
remains to be fully established, brain NAA was found in
concentrations hundreds-fold higher than in non-nervous
system tissues and therefore was considered a brain-specific
metabolite and an in vivo marker of neuronal density
[53,54]. A decrease in NAA has been observed in many
neurological diseases that cause neuronal and axonal degen-
eration, such as tumors, epilepsy, dementia, stroke, hypoxia,
multiple sclerosis, and many leukoencephalopathies. Con-
versely, the only known pathologic state characterized by a
dramatic increase in cerebral NAA is an autosomal-recessive
genetic leukodystrophy (Canavan disease) caused by the
synthesis of a defective form of the enzyme responsible for
the NAA degradation (N-acetyl-asparto-acylase [ASPA]).
More generally, any major central nervous system disease
involving either direct neuronal and/or axonal damage, sec-
ondary hypoxic-ischemic, or toxic insult will result in abnor-
PM&R Vol. 3, Iss. 10S2, 2011
malities of NAA homeostasis. In the field of TBI, however, a
very innovative hypothesis seemed more fascinating among
proportional to the severity of trauma .
By measuring whole-brain levels of NAA via high-perfor-
mance liquid chromatography  in 3 different levels of ex-
perimental, closed, and diffuse TBI (mild, moderate, and se-
ing spontaneous recovery with lower levels of trauma and irre-
versible decrease in the others . The findings also were
consistent with long-term behavioral observation in animals
injured with the same model of mTBI, showing only slight
differences with sham-injured animals, with the main differ-
ences being present 1 day after injury and showing consistent
improvement over time . All these bench data strongly
supported the indication for a potential role of NAA in quanti-
fying neuronal damage and predicting neuropsychological out-
use of1H-MRS allow to measure NAA noninvasively in vivo
The finding of recovery in the “concussed” animals im-
was attributable to transient biochemical changes and not
simply to cell death. Similar to the previously described
biochemical changes, the striking finding was again the ra-
pidity of the onset of significant NAA reduction, identified as
early as 2 hours after injury, with the lowest values recorded
at 15 hours after impact (?46% compared with control
values). Spontaneous recovery was observed to occur within
48-96 hours, but that took place only in mildly injured rats.
Beyond showing the profound TBI-induced modification in
NAA homeostasis, this finding clearly demonstrated that
different levels of “physical” injury correlated with different
levels and kinetics of “biochemical” damage, which are re-
versible in mTBI and irreversible in severe TBI (sTBI) .
Substantial evidence exists that NAA synthesis takes place
exclusively in neuronal mitochondria, that it is strictly tied to
neuronal energy metabolism, and that the distribution pat-
tern of NAA closely parallels the distribution of “respiratory
activity.” For an overview of the data supporting a bioener-
getic role for NAA in neurons, see Moffet et al .
A close linear relationship has been demonstrated between
the efficacy of ATP synthesis and the ability to synthesize NAA
[60,61]. NAA synthesis is indeed an energy-requiring process
dependent on the availability and the energy of hydrolysis of
acetyl coenzyme A (CoA) used as the acetyl group donor in the
transferase. It is fundamental to understand that when acetyl
CoA is used for NAA synthesis, there is an indirect high-energy
cost to the cell. In fact, because acetyl CoA will not enter the
citric acid cycle (Krebs’ cycle), a decrease will occur in the
production of reducing equivalents (3 reduced nicotinamide
adenine dinucleotide and 1 reduced flavin adenine dinucle-
otide) as the fuel for the electron transport chain. It has been
clearly evidenced that in the general post-traumatic metabolic
injury, following a pattern very similar to those observed for
both ATP and NAA . Therefore in metabolic conditions of
low ATP availability, when all of the pathways and cycles de-
voted to energy supply are operating at their maximal activity
with the aim of replenishing ATP levels, acetyl CoA will not be
accessible for NAA synthesis. Only when the ATP deficiency is
the NAA “production” pathway. It also should be recalled that
synthase, which is the enzyme of the Krebs’ cycle, using acetyl
concentration can be seen as an indirect marker of post-trau-
matic metabolic energy impairment.
recovering after an mTBI, the concussive biochemical de-
rangement (involving more complex pathways than simple
NAA homeostasis ) cannot be considered to be resolved.
Thus NAA embodies a biochemical surrogate marker to
monitor the overall cerebral metabolic status, and it appears
that under conditions of reduced NAA, although the cells are
functional, they are still experiencing energetic imbalance.
POSTCONCUSSIVE BRAIN VULNERABILITY
The basic pathophysiological paths explored thus far have clar-
ified some aspects of this particular clinical entity, suggesting
that even if concussion is considered a form of mTBI, it ought
the exception of the almost always punctual reversibility of all
adjective “mild” when referring to a traumatic event that can
have such consequences to the fundamental metabolic and
energy states of neuronal cells. However, while all of these
biochemical modifications are scientifically interesting, they
might appear, at a first glance, of negligible clinical utility be-
cause they are all spontaneously and fully reversible.
Despite this reversibility, a reasonable body of evidence
clearly demonstrates that the “concussed” brain cells undergo a
peculiar state of “vulnerability,” during which time if they sus-
tain a second, typically nonlethal insult in a close temporal
proximity, they would suffer irreversible damage and die
[65,66]. In the preclinical setting, this period of time has been
well defined in duration thanks to high reproducibility of the
closed-head rat model of mTBI that has been used to demon-
strate biochemically the concept of vulnerability [62,64], origi-
induced pathophysiologic conditions, mainly manifested by
of modest entity, creating a disproportion between the trauma
severity and the subsequent cerebral damage.
Signoretti et al PATHOPHYSIOLOGY OF CONCUSSION
Several studies in animals in which investigators focused
on mTBI-induced dysfunction have been published, and
current data support the concept of transient biochemical
and physiologic alterations that may be exacerbated by re-
peated mild injuries within specific time windows of vulner-
ability [62,64,67]. In a rat weight-drop experiment per-
simulate a “second impact” condition, it was clearly demon-
strated that levels of NAA, ATP, and the ATP/ADP ratio
decreased significantly when measured 2 days after repeated
concussion (Figure 1). Maximal metabolic abnormalities
were seen when the occurrence of 2 mild injuries were
separated by a 3-day interval; in fact, the metabolic abnor-
malities in these animals were similar to those occurring after
sTBI. In a follow-up study, similar perturbations were found
to persist as late as 7 days after double impact, indicating
prolonged metabolic effects from repeat mTBI in the same
a histopathology study in which they described the impor-
in mice, which led to pronounced cellular damage compared
concluded that although the brain was not morphologically
damaged after a single concussive insult, its vulnerability to a
second concussive impact was dangerously increased.
According to Hovda et al  and Doberstein et al ,
metabolic alterations can persist for days after concussion,
creating no morphological damage but representing the
pathological basis of the brain’s vulnerability. All these data
provide the experimental demonstration of the exquisitely
a unique contribution to the complex biochemical damage
underlying the clinical scenario of a repeated concussive
trauma, sometimes leading to catastrophic brain injury.
To explain the differences between the underlying meta-
bolic dysfunction occurring after a concussion and those
and consider the degree of the NAA and ATP reduction,
which is approximately 20% and 50%, respectively .
More importantly, the ADP concentration is only slightly
increased after mTBI but is found to be substantially in-
creased by 35% after sTBI . Despite the significant ATP
reduction by one fifth, if the insult is “mild,” the mitochon-
dria are not yet irreversibly damaged and still possess a
sufficient phosphorylating capacity (ie, a modest decrease of
mitochondrial phosphorylating activity) to allow spontane-
ous complete ATP restoration, which was fulfilled after ap-
proximately 5 days in the aforementioned experiment .
On the contrary, the 35% increase in ADP found after more
severe levels of injury indicates a profoundly different situa-
tion with an altered capacity of mitochondria to support the
cell energy requirements in terms of ATP synthesis (ie, pro-
found decrease in the ATP/ADP ratio).
If after a first mild injury a second concussion finds the
perfectly reversible energetic failure, it will cause further
mitochondrial malfunctioning, leading to the same irrevers-
that occur too close in temporal proximity can simulate the
effects of a single severe injury. The key biochemical issue of
the vulnerable brain lies in the incomplete resolution of the
initially reversible energetic crisis triggered by the first insult.
The foremost clinical implication of these experimental data
is that within days after injury, the metabolic effects of 2 con-
cussions can be dangerously additive. This information might
metabolites currently are not available. The second clinical im-
plication of this notion is again remarkable because it is very
difficult to establish how long the aforementioned period of
trauma would be uneventful.
THE SECOND IMPACT SYNDROME AND THE
HYPOTHESIS OF “THE PERFECT STORM”
A handful of previously published cases have reported on
patients (mostly involved in sports-related activities) who,
while still having symptoms from a previous head injury,
experienced a second injury that unexpectedly and unpre-
dictably led to sustained intracranial hypertension and cata-
strophic outcomes . This entity, also known as the sec-
Figure 1. Concentrations of N-acetylaspartate (NAA) (left
y-axis) and adenosine triphosphate (ATP) (right y-axis) as de-
termined by high-performance liquid chromatography in the
whole brains of rats subjected to repeat diffuse mild traumatic
brain injuries (TBIs) (spaced by 3 or 5 days) or single diffuse TBI
(mild TBI or severe TBI). Control subjects were sham-operated
animals, ie, animals receiving anesthesia and surgical proce-
dures out of injury. Each histogram is the mean of 6 animals. No
significant differences were demonstrated when rats sustain-
ing a single mild TBI and rats sustaining a repeat mild TBI
spaced by 5 days were compared. Similarly, no differences
were observed when rats sustaining a single severe TBI and rats
sustaining a repeat mild TBI spaced by 3 days were compared.
*P ? .05 versus control subjects.
PM&R Vol. 3, Iss. 10S2, 2011
cerebral edema after mTBI/concussion [72-77].
Skepticism about this entity notwithstanding , the
major concern about SIS is that it is an exceedingly rare
clinical condition when compared with the overall incidence
of concussion, even though an elevated risk for subsequent
mTBI exists among persons who are still recovering from a
fact that the resolution of clinical symptoms might not coin-
cide with the “closure” of the temporal window of brain
metabolic imbalance “opened” by the first trauma [82,83].
Thus the question of whether the brain had fully recovered
from the first concussive injury while experiencing the sec-
ond one remains unanswered.
Once again, laboratory data have provided clarification of
some of these complex matters. In a recently published
weight-drop experiment, rats were subjected to 2 diffuse
mTBIs, with the second mTBI delivered after 1, 2, 3, 4, and 5
days, and then all animals were killed 48 hours after the last
impact. Notably, mitochondrial-related changes progres-
sively worsened with the time between concussions up to 3
days apart, when the metabolic abnormalities were similar to
those occurring after a single sTBI . In this model and
with this experimental timeline, the third day after trauma
was the point when the cell’s energy-dependent recovery
processes were at their maximal intensity. However, if the
of the window of vulnerability in the rat, this can not be
affirmed in the case of human beings in which the very many
uncontrolled variables render each impact different from
another. This concept was clearly developed by Giza and
Hovda , who showed that each physiologic parameter
modified by a concussion has its own time frame, and each
head injury can be very different from the next. Therefore,
they concluded that it is difficult to definitively state the true
duration of vulnerability to a second injury . Results of
our studies in concussed athletes strongly corroborated this
50 concussed athletes recovered NAA concentration before
30 days post-impact, it was also evident that the time of NAA
normalization was not identical in each subject, thus render-
ing impossible to define the time of brain vulnerability with
other hand, it was also clearly demonstrated that none of our
concussed patients had clearance of post-concussive clinical
symptoms faster than NAA normalization, ie, disturbance of
[82,83], when post-concussive symptoms are persistent for
weeks after concussion. In an as yet unpublished article, we
describe a group of 6 doubly concussed athletes in which the
post-concussive syndrome persisted up to 2 months post-
injury (Vagnozzi et al., 2011, submitted). Even in this re-
stricted group of patients, recovery of NAA occurred much
later (75 to 120 days), once again suggesting that rescue of
brain metabolism does not correlate with self-reported clear-
ance of post-concussive symptoms. Therefore a second im-
pact occurring at this stage had the most profound effects
because of the minimal “metabolic buffering capacity” to
counteract the known early changes reinitiated by the new
mTBI. With their biochemical homeostasis not yet re-estab-
lished, the ionic imbalance will prevail and massive cerebral
swelling will take place [84,85].
The reason why SIS is, fortunately, an extremely rare
condition is probably because it represents a sort of “perfect
storm,” an extremely random and hardly predictable situa-
tion generated by the odd combination of the severity of the
initial concussion, the time interval between the 2 traumas,
and the metabolic state of the brain at the time of the second
singly and doubly concussed athletes who were examined by
1H-MRS for their NAA cerebral content at different time
points after concussive events demonstrated that the recov-
this study, athletes who experienced a second concussion
between the 10th and the 13th day after the first insult did
not have SIS, nor did they demonstrate signs of sTBI; how-
ever, they all had a significant delay in both symptom reso-
lution and NAA normalization . In other words, the
effects of the second concussion were not fatal, but they were
somehow not proportionate to the entity of the concus-
sive insult. Most likely, the second concussion occurred
when the brain cells were completing recovery of impaired
metabolic functions, and thus it only produced a limited
cumulative effect with moderate worsening of the clinical
pictures. Thus it is conceivable to infer that it is the time
interval between the 2 concussions that drives the clinical
and metabolic evolution.
It is our belief that SIS should not be solely considered as
an “all-or-none” phenomenon and should not be limited to
those instances that result in death from malignant swelling.
The concept of SIS should be extended to include all the
other occurrences in which a disproportion between the
severity of the second injury and the concussive clinical
features (ie, intensity and/or time of resolution) or cerebral
metabolic changes (ie, extent of NAA decrease and/or delay
in its normalization) is clearly observed. The degree of this
type of SIS will depend on which phase of the metabolic
recovery the brain is in at the time of the second concussion.
UNDERSTANDING THE DEGREE OF
MILDNESS OF AN mTBI: CHANGES IN
With use of the same model of experimental repeat mTBI,
studies from our laboratories  demonstrated that there
was an effect of the time interval between concussions on
ASPA gene expression. A progressive increase in the messen-
Signoretti et al PATHOPHYSIOLOGY OF CONCUSSION
ger ribonucleic acid transcript of the ASPA gene was ob-
sustained the 2 injuries 3 days apart . Animals reinjured
past 5 days had values of messenger ribonucleic acid for
ASPA comparable with those recorded in control animals.
Based on these data, it appears that TBI-induced NAA varia-
tions may not be attributable simply to a decreased rate of
The aforementioned results allowed researchers to hy-
phases with 2 different mechanisms. Initially, independent
neurons to the extracellular space. Simultaneously, mito-
chondrial impairment causes a cell energy deficit with con-
sequent diminution in NAA synthesis. In the case of revers-
ible brain damage, such as single mTBI or repeat mTBI, in
which the second impact occurs outside the brain’s vulnera-
bility “window,” recovery of mitochondrial functions will
allow restoration of ATP homeostasis and the subsequent
normalization in the rate of NAA efflux and biosynthesis (ie,
NAA levels close to those of control subjects with no increase
in ASPA expression). In single sTBI or in repeat mTBI in
which the second impact occurs within the brain vulnerabil-
mitochondrial malfunctioning induces a constant NAA out-
flow toward the oligodendrocytes, which, as an adaptive
mechanism, increases the expression of ASPA. This phenom-
enon, combined with the decreased rate of NAA biosynthesis
responsible for the dramatic NAA depletion.
These results were immediately followed by a collabora-
tive study on transcriptomics in which the authors studied
the simultaneous expression of approximately 30,000 rat
genes whose products are involved in a variety of cellular
processes . With the use of complementary deoxyribo-
nucleic acid microarray technology, it was reported that after
stretch injury to hippocampal slice cultures (as a suitable cell
model to induce graded TBI), the expression of 999 genes
was altered in mTBI compared with control patients. The
altered genes in mTBI-stretched cells clustered in the so-
called “biological process” group, which was shown to be
involved in the structural damage of cellular architecture.
Most of these genes are indeed involved in signal trans-
ducer activity, regulation of transcription, and cell commu-
nication. This finding indicated that even after a mild stretch
injury, as compared with a closed, diffuse mTBI, intense
activity involving transcription and signaling exchange is
initiated. In addition, it has been found that certain genes
involved in the apoptotic process, such as voltage-dependent
anion-selective channel protein 1 (ie, VDAC1), SH3-domain
GRB2-like endophilin B1 (SH3GLB1), pleckstrin homology-
like domain, family A, member 1 (PHDLA1), Rho-associated
coiled-coil containing protein kinase 1 (ROCK1), and eu-
karyotic translation initiation factor 4 gamma, 2 (EIF4G2-
predicted), were down-regulated. Furthermore, an up-regu-
lation was seen in genes involved in the antiapoptotic
process, such as chemokine (C-C motif) ligand 2 (CCL2),
vascular endothelial growth factor A (VEGFA), baculoviral
IAP repeat-containing 3 (BIRC3), TSC22 domain family,
ing protein 3 (BNIP3), and nuclear receptor subfamily 4,
group A, member 1 (NR4A1).
Most of these expression changes were only found after
mild stretch injury, indicating that these hippocampal cell
cultures have activated protective and repair mechanisms.
The most interesting finding was that more genes were dif-
ferentially expressed after mild brain injury than after severe
injury, further supporting the notion that even after mTBI,
characterized by the absence of radiological and clinical
abnormalities, a complex cellular response is initiated and
distinct neuronal dysfunction occurs. This finding corrobo-
rates previous findings that these effects are “primary” cellu-
lar effects not determined by local blood flow or oxygen
delivery or by any systemic factors .
The overall doubt that might be generated from the com-
bination of these studies on gene expression with the bio-
chemical works previously cited is that, apparently, not
much rationale is left to justify the adjective “mild” when
dealing with a concussive injury. It is undeniable that all the
aforementioned changes are fully reversible, but it must be
kept in mind that this reversibility is true only if a second
“equally mild” TBI does not occur within the temporal win-
dow of metabolic brain vulnerability.
In the 18th century, Alexis Littre performed a famous post-
occur without obvious anatomic damage to the brain. He
performed an autopsy on one particular patient who had
wall. Littre detected no cerebral injury, a finding consistent
with the 16th-century Ambroise Pare’s notion, according to
which the symptoms of concussion “. . . reflected a func-
tional disturbance rather than structural damage such as
contusion, hemorrhage or laceration of the brain” .
More than half a millennium since Pare’s first intuition,
basic science data collected thus far have clarified only some
of the many aspects of this particular clinical entity, suggest-
ing that short-term as well as long-term consequences may
very well be overcome simply by understanding the meta-
bolic conditions of the injured brain cells.
ing NAA after an initial concussion and monitoring it until
normalization might represent a significant step forward in
quantifying the objective nature of postconcussive metabolic
disturbances. Because of its high concentration within neurons
PM&RVol. 3, Iss. 10S2, 2011
MR signal into a specific volume of tissue, thus providing a
real-time “image” of the brain neurochemistry. At the present
time,1H-MRS offers a unique opportunity to endeavor to “bio-
actual metabolic dysfunction, apart from signs and symptoms,
because often clearance of clinical disturbances does not coin-
cide with full cerebral metabolic recovery.
The results of a multicenter clinical trial  involving 40
concussed athletes and 30 healthy volunteers recently have
been published and reveal that despite different combina-
tions of magnetic field strengths (1.5 or 3.0 T) and modes of
spectrum acquisition (single- or multi-voxel) among the MR
scanners currently in use in most neuroradiology centers,
NAA determination represents a quick (15-minute), easy to
perform, noninvasive tool to accurately measure changes in
cerebral biochemical damage that occur after a concussion.
Patients exhibited the most significant alteration of metabo-
initially in a slow fashion and, after day 15, more rapidly. At
30 days after the injury, all subjects exhibited complete
recovery, that is, having metabolite ratios similar to values
detected in control subjects (Figure 2).
Interestingly, patients self-declared a clearance of their
symptoms between 3 and 15 days after concussion. To have
a snapshot of the degree of energetic impairment and to
monitor the eventual recovery curve might represent a useful
strategy to avoid a second mTBI soon afterward that could
lead to a more severe injury.
Finally, the combination of metabolic regional data ob-
tained with longitudinal1H-MRS studies, serial neuropsy-
responsible for the cumulative impairments of cerebral func-
tion and cognition, including early onset of memory distur-
bances, early depression, and even dementia.
Sudden and profound biochemical changes occur after a con-
cussive trauma. These changes are activated by the mechanical
insult itself and lead to ionic disturbance, EAA “neurotoxicity,”
initial mitochondrial dysfunction, ROS-mediated damage, en-
ergy metabolism depression, alteration of gene expression, and
ultimately variation of NAA concentration, the “surrogate”
tation of concussion—a capricious combination of headache,
iting, blurred vision, attention difficulty, concentration prob-
lems, memory problems, orientation problems, self-appraisal
problems, expression and speech or language problems, irrita-
bility, depression, anxiety, sleep disturbance, problems with
emotional control, loss of initiative, blunted affect, somatic pre-
occupation, hyperactivity, disinhibition, or problems related to
employment, marriage, relationships, and home and or school
More problematically, within days after a simple blow to
the head, this intricate biochemical derangement can result
in a dangerous state for the brain, generating a situation of
metabolic vulnerability to the point that if another equally
“mild” injury were to occur, the 2 concussions would show
the biochemical equivalence of a severe brain trauma. The
immediate clinical implication derived from this evidence is
that trials are warranted to investigate the application of
1H-MRS for measurement of NAA and to monitor the full
recovery of brain metabolic functions.
1. Bruns J Jr., Hauser WA. The epidemiology of traumatic brain injury: A
review. Epilepsia 2003;44(Suppl 10):2-10.
2. Tagliaferri F, Compagnone C, Korsic M, Servadei F, Kraus J. A system-
atic review of brain injury epidemiology in Europe. Acta Neurochir
3. van der Naalt J. Prediction of outcome in mild to moderate head injury:
A review. J Clin Exp Neuropsychol 2001;23:837-851.
Figure 2. Metabolite ratios of N-acetylaspartate/creatine-
containing compounds (NAA/Cr) and choline-containing
healthy control patients and concussed athletes. Histograms
are the means of 30 healthy control patients and 40 con-
cussed athletes. Standard deviations are represented by ver-
tical bars. Data were collected in 3 different neuroradiological
centers with the use of either the “single-voxel” mode through
a 3-T apparatus, the “single-voxel” mode through a 1.5-T
apparatus or the “multivoxel” mode through a 3-T apparatus.
No differences were observed when data collected in the 3
neuroradiology centers were compared. At 3 days after injury
the NAA/Cr ratio decreased by 17.6% and gradually recov-
ered to complete normalization at 30 days. The Cho/Cr ratio
did not show any significant variation. *P ? .01 with respect to
control patients. **P ? .01 with respect to values determined at
the previous time points.
Signoretti et al PATHOPHYSIOLOGY OF CONCUSSION
4. Vos PE, Battistin L, Birbamer G, et al. European Federation of Neuro-
of an EFNS task force. Eur J Neurol 2002;9:207-219.
5. Yates D, Aktar R, Hill J. Guideline Development Group. Assessment,
investigation, and early management of head injury: Summary of NICE
guidance. BMJ 2007;6;335:719-720.
6. Swann IJ, MacMillan R, Strong I. Head injuries at an inner city accident
and emergency department. Injury 1981;12:274-278.
7. Shackford SR, Wald SL, Ross SE, et al. The clinical utility of computed
tomographic scanning and neurologic examination in the management
of patients with minor head injuries. J Trauma 1992;33:385-394.
8. Taheri PA, Karamanoukian H, Gibbons K, Waldman N, Doerr RJ,
Hoover EL. Can patients with minor head injuries be safely discharged
home? Arch Surg 1993;128:289-292.
skull radiography for intracranial injury in children with blunt head
injury. Lancet 1997;349:821-824.
10. Livingston DH, Lavery RF, Passannante MR, et al. Emergency depart-
ment discharge of patients with a negative cranial computed tomogra-
phy scan after minor head injury. Ann Surg 2000;232:126-132.
11. Fabbri A, Servadei F, Marchesini G, Negro A, Vandelli A. The changing
face of mild head injury: Temporal trends and patterns in adolescents
and adults from 1997 to 2008. Injury 2010;41:968-972.
12. Kay T, Newman B, Cavallo M, Ezrachi O, Resnick M. Toward a
neuropsychological model of functional disability after mild traumatic
brain injury. Neuropsychology 1992;6:371-384.
13. Gouvier WD, Cubic B, Jones G, Brantley P, Cutlip Q. Postconcussion
symptoms and daily stress in normal and head-injured college popu-
lations. Arch Clin Neuropsychol 1992;7:193-211.
14. Alexander MP. Mild traumatic brain injury: Pathophysiology, natural
history, and clinical management. Neurology 1995;45:1253-1260.
15. Ingebrigtsen T, Romner B, Kock-Jensen C. Scandinavian guidelines for
initial management of minimal, mild, and moderate head injuries.
J Trauma 2000;48:760-766.
16. Bigler ED. Neurobiology and neuropathology underlie the neuropsy-
chological deficits associated with traumatic brain injury. Arch Clin
17. Esselman PC, Uomoto JM. Classification of the spectrum of mild
traumatic brain injury. Brain Inj 1995;9:417-424.
18. De Kruijk JR, Twijnstra A, Leffers P. Diagnostic criteria and differential
diagnosis of mild traumatic brain injury. Brain Inj 2001;15:99-106.
19. McCrory P, Meeuwisse W, Johnston K, et al. Consensus statement on
Concussion in Sport 3rd International Conference on Concussion in
Sport held in Zurich, November 2008. Clin J Sport Med 2009;19:185-
20. McCrory P, Johnston K, Meeuwisse W, et al. Summary and agreement
Prague 2004. Br J Sports Med 2005;39:196-204.
21. Aubry M, Cantu R, Dvorak J, et al. Summary and agreement statement
of the First International Conference on Concussion in Sport, Vienna
2001. Recommendations for the improvement of safety and health of
athletes who may suffer concussive injuries. Br J Sports Med 2002;36:
22. Ommaya AK, Goldsmith W, Thibault L. Biomechanics and neuropa-
thology of adult and paediatric head injury. Br J Neurosurg 2002;16:
23. Margulies SS, Thibault LE. A proposed tolerance criterion for diffuse
axonal injury in man. J Biomech 1992;25:917-923.
24. Walilko TJ, Viano DC, Bir CA. Biomechanics of the head for Olympic
boxer punches to the face. Br J Sports Med 2005;39:710-719.
25. Barkhoudarian G, Hovda DA, Giza CC. The molecular pathophysiol-
ogy of concussive brain injury. Clin Sports Med 201;30:33-48.
26. Farkas O, Lifshitz J, Povlishock JT. Mechanoporation induced by
diffuse traumatic brain injury: An irreversible or reversible response to
injury? J Neurosci 2006;26:3130-3140.
27. Faden AI, Demediuk P, Panter SS, et al. The role of excitatory amino
acids and NMDA receptors in traumatic brain injury. Science 1989;
28. Katayama Y, Becker DP, Tamura T, et al. Massive increases in extracel-
lular potassium and the indiscriminate release of glutamate following
concussive brain injury. J Neurosurg 1990;73:889-900.
29. Xiong Y, Gu Q, Peterson PL, Muizelaar JP, Lee CP. Mitochondrial
dysfunction and calcium perturbation induced by traumatic brain
injury. J Neurotrauma 2007;14:23-34.
30. Nicholls DG, Budd SL. Mitochondria and neuronal glutamate excito-
toxicity. Biochim Biophys Acta 1998;1366:97-112.
31. Khodorov B, Pinelis V, Storozhevykh T, Vergun O, Vinskaya N. Dom-
inant role of mitochondria in protection against a delayed neuronal
Ca2? overload induced by endogenous excitatory amino acids follow-
ing a glutamate pulse. FEBS Lett 1996;393:135-138.
32. Zoratti M, Szabò I. The mitochondrial permeability transition. Biochim
Biophys Acta 1995;1241:139-176.
tion is a primary event in glutamate neurotoxicity. J Neurosci 1996;16:
34. Nanavaty UB, Pawliczak R, Doniger J, et al. Oxidant-induced cell death
in respiratory epithelial cells is due to DNA damage and loss of ATP.
Exp Lung Res 2002;28:591-607.
35. Nojiri H, Shimizu T, Funakoshi M, et al. Oxidative stress causes heart
failure with impaired mitochondrial respiration. J Biol Chem 2006;
36. Pacher P, Liaudet L, Mabley J, Komjáti K, Szabó C. Pharmacologic
resent a novel therapeutic approach in chronic heart failure. J Am Coll
37. Vagnozzi R, Marmarou A, Tavazzi B, et al. Changes of cerebral energy
metabolism and lipid peroxidation in rats leading to mitochondrial
dysfunction after diffuse brain injury. J Neurotrauma 1999;16:903-
38. Palozza P, Moualla S, Krinsky NI. Effects of ?-carotene and ?-tocoph-
erol on radical-initiated peroxidation of microsomes. Free Radic Biol
39. Thies RL, Autor AP. Reactive oxygen injury to cultured pulmonary
artery endothelial cells: Mediation by poly(ADP-ribose) polymerase
activation causing NAD depletion and altered energy balance. Arch
Biochem Biophys 1991;286:353-363.
40. Morgan WA. Pyridine nucleotide hydrolysis and interconversion in rat
hepatocytes during oxidative stress. Biochem Pharmacol 1995;49:
41. Janero DR, Hreniuk D, Sharif HM, Prout KC. Hydroperoxide induced
muscle cells. Am J Physiol 1993;264:C1401-C1410.
42. Lautier D, Hoflack JC, Kirkland JB, Poirier D, Poirier GG. The role of
poly(ADP-ribose) metabolism in response to active oxygen cytotoxic-
ity. Biochim Biophys Acta 1994;1221:215-220.
43. Tavazzi B, Di Pierro D, Amorini AM, et al. Direct NAD(P)H hydrolysis
into ADP-ribose(P) and nicotinamide induced by reactive oxygen spe-
cies: A new mechanism of oxygen radical toxicity. Free Radic Res
44. Tavazzi B, Di Pierro D, Bartolini M, et al. Lipid peroxidation, tissue
rat heart as a function of increasing ischemia. Free Radic Res 1998;28:
45. Solaroglu L, Okutan O, Kaptanoglu E, Beskonakli E, Kilinc K. In-
creased xanthine oxidase activity after traumatic brain injury in rats.
J Clin Neurosci 2005;12:273-275.
46. Kawamata T, Katayama Y, Hovda DA, et al. Administration of excit-
glucose utilization seen following concussive brain injury. J Cereb
Blood Flow Metab 1992;12:12-24.
PM&R Vol. 3, Iss. 10S2, 2011
47. Yoshino A, Hovda DA, Kawamata T, et al. Dynamic changes in local Download full-text
cerebral glucose utilization following cerebral conclusion in rats: Evi-
dence of a hyper and subsequent hypometabolic state. Brain Res
48. Andersen BJ, Marmarou A. Post-traumatic selective stimulation of
glycolysis. Brain Res 1992;585:184-189.
experimental head injury. Neurosurg Rev 1989;12(Suppl 1):400-411.
50. Yoshino A, Hovda DA, Katayama Y, et al. Hippocampal CA3 lesion
prevents postconcussive metabolic dysfunction in CA1. J Cereb Blood
Flow Metab 1992;12:996-1006.
51. Tavazzi B, Signoretti S, Lazzarino G, et al. Cerebral oxidative stress and
injury in rats. Neurosurgery 2005;56:582-589.
52. Wu A, Ying Z, Gomez-Pinilla F. Vitamin E protects against oxidative
damage and learning disability after mild traumatic brain injury in rats.
Neurorehabil Neural Repair 2010;24:290-298.
53. Moffett JR, Ross B, Arun P, Madhavarao CN, Namboodiri AM. N-
Acetylaspartate in the CNS: From neurodiagnostics to neurobiology.
Prog Neurobiol 2007;81:89-131.
54. Truckenmiller ME, Namboodiri MA, Brownstein MJ, Neale JH. N-
Acetylation of L-aspartate in the nervous system: Differential distribu-
tion of a specific enzyme. J Neurochem 1985; 45:1658-1662.
Evidence for cellular damage in normal-appearing white matter corre-
lates with injury severity in patients following traumatic brain injury: A
magnetic resonance spectroscopy study. Brain 2000;123:1403-1409.
56. Tavazzi B, Vagnozzi R, Di Pierro D, et al. Ion-pairing high-performance
liquid chromatographic method for the detection of N-acetylaspartate
and N-acetylglutamate in cerebral tissue extracts. Anal Biochem 2000;
57. Signoretti S, Marmarou A, Tavazzi B, Lazzarino G, Beaumont A, Vag-
mitochondrial dysfunction following diffuse traumatic brain injury.
J Neurotrauma 2001;18:977-991.
58. Beaumont A, Marmarou A, Czigner A, et al. The impact-acceleration
model of head injury: Injury severity predicts motor and cognitive
performance after trauma. Neurol Res 1999;21:742-754.
59. Friedman SD, Brooks WM, Jung RE, Hart BL, Yeo RA. Proton MR
spectroscopic findings correspond to neuropsychological function in
traumatic brain injury. AJNR Am J Neuroradiol 1998;19:1879-1885.
60. Bates TE, Strangward M, Keelan J, Davey GP, Munro PM, Clark JB.
Inhibition of N-acetylaspartate production: Implications for1H-MRS
studies in vivo. Neuroreport 1996;7:1397-1400.
61. Baslow MH. NAAG peptidase as a therapeutic target: Potential for
regulating the link between glucose metabolism and cognition. Drug
News Perspect 2006;19:145-150.
62. Vagnozzi R, Tavazzi B, Signoretti S, et al. Temporal window of meta-
bolic brain vulnerability to concussions: Mitochondrial-related impair-
ment—part I. Neurosurgery 2007;61:379-389.
63. Harford S, Weitzman PD. Evidence of isosteric and allosteric nucleo-
tide inhibition of citrate synthease from multiple-inhibition studies.
Biochem J 1975;151:455-458.
64. Tavazzi B, Vagnozzi R, Signoretti S, et al. Temporal window of meta-
bolic brain vulnerability to concussions: Oxidative and nitrosative
stresses—part II. Neurosurgery 2007;61:390-396.
65. Hovda DA, Badie H, Karimi S, Thomas S, Yoshino A, Kawamata T.
Concussive brain injury produces a state of vulnerability for intracra-
nial pressure perturbation in the absence of morphological damage. In:
VIII. New York, NY: Springer-Verlag; 1983, 469-472.
66. Giza CC, Hovda DA. The neurometabolic cascade of concussion. J Athl
67. Longhi L, Saatman KE, Fujimoto S, et al. Temporal window of vulner-
ability to repetitive experimental concussive brain injury. Neurosur-
68. Laurer HL, Bareyre FM, Lee VM, et al. Mild head injury increasing the
brain’s vulnerability to a second concussive impact. J Neurosurg 2001;
70. Vagnozzi R, Signoretti S, Tavazzi B, et al. Hypothesis of the postcon-
rence. Neurosurgery 2005;57:164-171.
sports head trauma. JAMA 1984;252:538-539.
72. Cantu RC. Second-impact syndrome. Clin Sports Med 1998;17:37-44.
73. Bowen AP. Second impact syndrome: A rare, catastrophic, preventable com-
74. Cantu RC. Malignant brain edema and second impact syndrome. In:
Cantu RC, ed. Neurologic Athletic Head and Spine Injuries. Philadel-
phia, PA: WB Saunders; 2000, 132-137.
75. Cobb S, Battin B. Second-impact syndrome. J Sch Nurs 2004;20:262-
76. Mori T, Katayama Y, Kawamata T. Acute hemispheric swelling associ-
injury in sports. Acta Neurochir Suppl 2006;96:40-43.
77. Cantu RC, Gean AD. Second-impact syndrome and a small subdural
hematoma: An uncommon catastrophic result of repetitive head injury
with a characteristic imaging appearance. J Neurotrauma 2010;27:
78. McCrory P. Does second impact syndrome exist? Clin J Sport Med
79. United States Centers for Disease Control and Prevention. Sports-
related recurrent brain injuries. MMWR Morb Mortal Wkly Rep 1997;
80. Guskiewicz KM, Weaver NL, Padua DA, Garrett W Jr. Epidemiology of
concussion in collegiate and high school football players. Am J Sports
81. Zemper ED. Two-year prospective study of relative risk of a second
cerebral concussion. Am J Phys Med Rehabil 2003;82:653-659.
82. Vagnozzi R, Signoretti S, Tavazzi B, et al. Temporal window of meta-
bolic brain vulnerability to concussion: A pilot 1H-magnetic resonance
spectroscopic study in concussed athletes—part III. Neurosurgery
83. Vagnozzi R, Signoretti S, Cristofori L, et al. Assessment of metabolic
brain damage and recovery following mild traumatic brain injury: A
multicentre, proton magnetic resonance spectroscopic study in con-
cussed patients. Brain 2010;133:3232-3242.
84. Signoretti S, Marmarou A, Aygok GA, Fatouros PP, Portella G, Bullock
using high-resolution proton magnetic resonance spectroscopy. J Neu-
85. Marmarou A, Signoretti S, Fatouros PP, Portella G, Aygok GA, Bullock
MR. Predominance of cellular edema in traumatic brain swelling in
patients with severe head injuries. J Neurosurg 2006;104:720-730.
86. Fiskum G. Mitochondrial participation in ischemic and traumatic
neural cell death. J Neurotrauma 2000;17:843-855.
87. Di Pietro V, Amin D, Pernagallo S, et al. Transcriptomics of traumatic
brain injury: Gene expression and molecular pathways of different
grades of insult in a rat organotypic hippocampal culture model.
J Neurotrauma 2010;27:349-359.
88. Levasseur JE, Alessandri B, Reinert M, Bullock R, Kontos HA. Fluid
percussion injury transiently increases then decreases brain oxygen
consumption in the rat. J Neurotrauma 2000;17:101-112.
89. Shaw NA. The neurophysiology of concussion. Prog Neurobiol 2002;
Signoretti et al PATHOPHYSIOLOGY OF CONCUSSION