BRAIN INJURY, VOL. 18, NO.

7 (JULY 2004), 671683

S100 in mild traumatic brain injury

C A T H A R I N A N Y G R E N D E B O U S S A R D y,
P A M F R E D M A N z, A N D E R S L U N D I N y,
K E R S T I N A N D E R S S O N z,
GUNNAR EDMAN} and JO RGEN BORG
y Department of Rehabilitation Medicine, Karolinska Institutet, Danderyd Hospital,
Stockholm, Sweden
z Institute of Clinical Neuroscience, Section of Experimental Neuroscience,
The Sahlgrenska Academy at Goteborg University, SU/Molndal, Sweden
} R&D unit, Department of Psychiatry, Karolinska Institutet, Danderyd Hospital,
Stockholm, Sweden
Department of Neuroscience, Rehabilitation Medicine, Uppsala University,
Uppsala, Sweden
(Received 17 June 2003; accepted 31 October 2003)

Primary objectives: To examine the diagnostic value of S100 in mild traumatic brain injury (MTBI).
Research design: Prospective cohort study.
Methods and procedures: S100B, S100A1B and S100BB concentrations were examined in sera from
patients with MTBI with an arrival Glasgow Coma Scale score of 15 or 14, patients with orthopaedic
injuries and non-injured subjects.
Main outcome and results: Mean values and proportions of subjects above cut-off limits for S100B and
S100A1B were significantly higher in each trauma group than in non-injured controls, but only for
S100A1B when patients with MTBI were compared with controls with orthopaedic injuries. Using
a 97.5 percentile cut-off limit, the sensitivity of S100A1B for MTBI vs orthopaedic injury was 61%
(95% confidence interval (CI) 4973%), specificity 77% (95% CI 6293%). The area under the ROC
curve did not approach 0.9 for any cut off limit.
Conclusions: Diagnostic validity of S100 in acute MTBI was not demonstrated. S100A1B has merits
for long-term prognostic studies.

Introduction
Most traumatic brain injuries are mild. The reported incidence of patients attending
hospital with mild traumatic brain injury (MTBI) ranges from
100300 or more
per 100 000 population [14]. MTBI includes a history of altered or lost conscious-
ness, post-traumatic amnesia and an emergency room (ER) Glasgow Coma Scale
(GCS) score of 1315 [5]. Several studies during the last decade [610] have
demonstrated that intra-cranial lesions detectable with computerized tomography
(CT) examination are present in
5% of the patients with MTBI presenting with a
GCS score of 15 and in more than 30% in those patients presenting with a GCS

Correspondence to: Catharina Nygren de Boussard, MD, Department of Rehabilitation

Medicine, Karolinska Institutet, Danderyd Hospital, SE-182 88, Stockholm, Sweden. e-mail: catharina.
nygren@reh.ds.sll.se

Brain Injury ISSN 02699052 print/ISSN 1362301X online # 2004 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080/02699050310001646215
672 C. N. de Boussard

score of 13. Recent studies indicate that clinical variables, such as GCS scores, any
sign of basal scull fracture, vomiting  2 episodes, can be used to predict CT scan
abnormalities [6, 7] and short-term clinical outcome [6] in MTBI.
The presence of other than radiologically visible brain tissue damage and its
impact on long-term outcome in MTBI is still poorly understood. Other brain
injury markers have been considered both for diagnostic and prognostic purposes.
One of these is the S100 protein (Soluble in 100% ammonium sulphate) [11, 12].
The S100-proteins comprise 19 members that are differentially expressed in a
large number of cell-types [13]. These proteins have amino acid sequence homo-
logy and similar structural properties. S100 proteins are metabolized in the kidney
and excreted via the urinary tract [14]. They are mainly present intra-cellularly
and are involved in the calcium homeostasis. However, the exact functions of
S100 proteins are not known [13, 15].
Brain S100 protein is a mixture of two closely related protein species, S100A1
and S100B (formerly known as S100 and S100, respectively). S100A1 and S100B
form the homodimers S100A1A1 and S100BB as well as the heterodimer S100A1B
[13]. Immunohistochemical and biochemical studies have shown that S100A1 and
S100B are present in neurones as well as in glial cells, including astrocytes, microglial
cells, ependymal cells and oligodendrocytes [13, 1618]. S100B monomer is pre-
dominantly represented in the brain [19]. The expression of S100A1 and S100B
is not restricted to the nervous system but has also been described in bone marrow,
adipose tissue and the kidney [13, 19, 20].
The S100B immuno-reactivity most commonly described in the literature refers
to the summed concentrations of the S100B monomers in S100A1B and S100BB.
Abnormal S100B concentrations in serum have been reported in patients with head
trauma [2124], subarachnoidal haemorrhages [25], cerebral infarcts [26] and after
cardiac surgery [27]. In patients with traumatic brain injury (TBI), higher serum
levels of S100B have been observed in patients with severe TBI as compared
to those with moderate and mild TBI [28]. S100B has been suggested to predict
both acute and long-term outcome after TBI [21, 22, 2934]. Only a few previous
studies on S100 in sera have measured S100A1B and S100BB separately [35, 36].
S100BB is suggested to be most specific for brain associated diseases, since the
S100B monomer is considered the dominating isoform in CNS [19]. A recent
study by Anderson et al. [35] reported increased levels of S100B as well as
S100A1B and S100BB in sera from patients with body trauma and without signs
of head injury. Thus, the specificity of S100 as a marker for brain trauma has been
questioned [3538].
The aim of the present study was to elucidate the diagnostic validity of S100
for MTBI by examining (1) the levels and time course of S100A1B, S100BB and
S100B in the acute and post-acute phase in patients with MTBI, using patients with
mild orthopaedic injuries and non-injured subjects as controls, and (2) the relation
between the different S100 forms and clinical as well as radiographic findings
in patients with MTBI.

Materials and methods

Consecutive head injury patients from three emergency departments (ED) were
considered for entry in this prospective cohort study. The study was performed
between January 2000 and December 2001. Patients exposed to high-energy
S100 in mild TBI 673

trauma were admitted to a regional level 1 trauma centre and, thus, not eligible for
the study. One hundred and twenty-two patients volunteered to participate in a
follow-up study, including assessment sessions and Magnetic Resonance Imaging
scan (MRI) of the brain. The source population of the study comprises
800 000
inhabitants in northern and central parts of Stockholm County at an age of 1565
(mean 38 years), women being
51%. The local research ethics committee
approved the study.

Patients with MTBI

A majority of the patients were observed at hospital for 1 day only. Longer periods
of hospital stay were due to other concomitant, most commonly orthopaedic
injuries. A minority of the patients were discharged from hospital without further
observation after the examination at the ED.
Inclusion criteria were all of the following: blunt trauma to the head within the
past 24 hours resulting in a witnessed loss of consciousness (LOC) or a reported
amnesia, a GCS score of 1415 at first examination in the ED and age 1565 years.
Exclusion criteria were any of the following: amnesia more than 24 hours, LOC
more than 30 minutes, no clear history of trauma as the primary event (e.g. epileptic
seizures), other major injury or neurological disorder. Patients with previous or
current psychiatric illness or alcohol dependence were not excluded.
Injury mechanisms were fall accident in 59%, traffic accident (pedestrians,
bicyclists and motorcyclists) in 19%, assault in 7% and not clearly defined (e.g.
sport accidents) in 15% of the included patients.
To compare participants and non-participants, the authors retrospectively
reviewed records for all patients fulfilling inclusion criteria during a 3 month
period. During this time-period, 15% of eligible patients accepted participation in
the study. The non-participants had a mean age of 37.1 years and 65% were male.
CT scan had been performed in 60% of the non-participants, none showing
pathological findings. Hospitalization rate for non-participants was 53% and for
participants 80%.
Twenty patients (16%) withdraw from the study during the study period.
Dropouts were compared to those followed up with respect to age, gender,
hospital, education, occupation, PTA, LOC, initial symptom severity, previous
psychiatric history and alcohol intoxication at admission. Except for dropouts
having less years of education (mean difference 2.3 years), there were no significant
differences.

Controls
Inclusion criteria for patients with mild orthopaedic injuries were a distortion
or fracture of the upper or lower limb within the past 24 hours, no signs or history
of head trauma, no other major injuries, an age of 1565 years, living in northern
or central parts of Stockholm County. Patients with major neurological disorders
were excluded.
Non-injured controls were recruited by local advertisement. Inclusion criteria
were an age of 1565 years, living in the northern and central parts of Stockholm
County, a self-reported good health and no history of recent head trauma.
674 C. N. de Boussard

Procedures
The emergency ward staff documented GCS score, amnesia, loss of consciousness,
alcohol level according to a research protocol. Blood samples for S100 analyses were
collected in the ED, the following day and 14 days post-injury. All samples were
centrifuged and serum was separated and stored at 20 C until analyses.
A CT scan was performed within 24 hours post-injury. MRI was performed
within 7 days post injury.
Symptoms were assessed by use of a Swedish version of Rivermead post-
concussion questionnaire (RPQ) [39] and the patients filled in the inventory 1, 7
and 14 days post-injury.
In the patients with orthopaedic injuries, blood-samples were collected at
the ED and no other follow-up was performed. In the non-injured controls,
blood samples were collected at three different occasions, a first visit and 14 days
and 3 months after the first visit.

S100 analyses
S100B was measured using a commercially available immunoluminometric assay
(LIA-mat Sangtec S100 Sangtec Medical, Bromma, Sweden) described elsewhere
[40, 41]. The functional detection limit of this assay is 0.02 mg/L.
S100A1B and S100BB were analysed by an enzyme-labelled immunosorbent
assay (ELISA) method (CanAg Diagnostics AB, Gothenburg, Sweden). Antibodies
reacting with S100BB and S100A1B but not with S100A1 or S100A1A1 dimer
were used in combination with antibodies specific for S100B for development
of the respective immunoassays. The antibodies were established by immunization
of mice with S100BB and S100A1B and the specificity determined in ELISAs
against S100A1A1, S100A1B, S100B and S100BB [42]. The immunoassays for
S100BB and S100A1B were designed as two-step sandwich immunoassays using
biotinylated catching monoclonal antibody (Mab) and HRP labelled detecting
MAb. The different antibodies were conjugated with biotin and horseradish
peroxidase according to standard procedures.
The assay protocols were as follows: 50 mL of sample or standard was incubated
together with 100 mL biotinylated catching MAb ( 1.5 mg/ml in Assay buffer)
in streptavidin coated microtiter plates for 2 hours with constant shaking. After
washing three times with 0.05 M Tris, 0.05% Tween 20, pH 7.5 the wells were
incubated with 100 mL of HRP labelled detecting MAb ( 1.5 mg/ml in Assay
Buffer) for another 1 hour during constant shaking. After washing six times as
above, the bound HRP labelled MAb was detected by incubation with 100 mL
TMB substrate for 30 minutes (Neogen Corp. US). The OD of the wells was
determined in a microtiter plate spectrophotometer at 405 nm after addition of
100 mL Stop solution (0.12 M HCl). The OD was proportional to the amount of
S100 antigen present in the samples and quantified from the standard curve con-
structed from the OD of the S100 standards included in the assay. In the S100BB
assay, biotinylated S10 MAb specific for S100BB dimer was used for catching and
detection was made using HRP labelled S23 MAb specific for the S100B sub-unit.
In the S100A1B assay, biotinylated S23 MAb specific for the S100B sub-unit was
used for catching the S100 antigen and the HRP labelled S35 MAb specific for the
S100A1B dimer was used.
S100 in mild TBI 675

Statistics
All variables were summarized with standard descriptive statistics (e.g. mean standard
deviation and frequency). As the distribution of the S100 variables were severely
and positively skewed, differences between groups were analysed with non-
parametric methods, Kruskal-Wallis for the three-group comparisons and Mann-
Whitney for the two group and post-hoc comparisons. Differences between groups
in categorized variables were analysed with the 2 -method, unless an expected
cell frequency was less than six when Fisher exact test was applied. Relationships
were calculated as Kendalls value order coefficients. The significance level was 5%
and all statistical analyses were two-tailed. For all S100 measurements, cut off limits
were defined as above the 97.5 percentile in the non-injured controls. Receiver
operator characteristic (ROC) curves were calculated with cut-off limits of 97.5
percentile as well as 75, 90 and 99 percentile.

Results
Characteristics of the study samples are presented in table 1. No significant
differences were found between patients with MTBI and the two control groups
concerning age or gender.

Table 1. Sociodemographic and clinical characteristics of patients with MTBI, patients with mild orthopaedic
injuries and non-injured-controls

Patients with Non-injured

Patients with mild orthopaedic controls
Characteristics MTBI (n 122) injuries (n 35) (n 35)

Age Mean 37.3 39.0 39.0

range 1565 1563 1662
Gender n (%)
men 71 (58) 15 (43) 17 (49)
women 51 (42) 20 (57) 18 (51)
Affected by alcohol n (%) 30 (25) 0 (0) 0 (0)
GCS at arrival in hospital n (%)
15 109 (89) N/A N/A
14 13 (11) N/A N/A
Injury-related pathological 8 (7) N/A N/A
findings in CT and/or MRI
Type of orthopaedic injury n(%)
Fracture 6 (4) 17 (49) N/A
Contusion/distortion 51 (42) 18 (51)
Loss of consciousness (minutes) n (%)
0 31 (25) 35 (100) N/A
0.10.9 25 (20) 0 (0) N/A
15 47 (39) 0 (0) N/A
630 19 (15) 0 (0) N/A
Anterograde amnesia (minutes) n (%)
0 2 (0) 35 (100) N/A
0.10.9 18 (15) 0 (0) N/A
15 29 (24) 0 (0) N/A
645 45 (37) 0 (0) N/A
> 45 27 (22) 0 (0) N/A

N/A, not applicable.

676 C. N. de Boussard

Levels and time course of S100A1B, S100BB and S100B

Data of the analyses with the three assays for S100 measurements in sera collected
from patients with MTBI, patients with mild orthopaedic injuries and non-injured
controls are presented in table 2.
The mean values of S100B were significantly higher in patients with MTBI
(z 3.94, p < 0.001) and in patients with mild orthopaedic injuries (z 3.99,
p < 0.001) as compared to non-injured control subjects. The mean values of
S100B in patients with MTBI and in patients with mild orthopaedic injuries
were not significantly different (z 0.00, p 1.00). These findings were confirmed
when the proportions of subjects with concentrations above and below the cut off
limit of S100B in each of the three groups were compared.
Using the 97.5 percentile in the non-injured controls as a cut-off limit, the
sensitivity of S100B was 38% (95% confidence interval (CI) 2848%), the specificity
97% (95% CI 91100%) and negative predictive value 36% (95% CI 2646%)
(patients with MTBI and patients with mild orthopaedic injuries vs non-injured
controls). The area under the ROC curve did not approach 0.9 for this or other
cut-off limits.
The mean values of S100A1B were significantly higher in patients with
MTBI (z 6.53, p < 0.001) and in patients with mild orthopaedic injuries
(z 3.55, p < 0.001) when compared with non-injured control subjects. The
mean value of S100A1B was also significantly higher in patients with MTBI as
compared to patients with mild orthopaedic injuries (z 3.74, p < 0.001). These
findings were confirmed when the proportions of subjects with concentrations
above and below the cut-off limit of S100A1B in each of the three groups were
compared.
Using the 97.5 percentile in the non-injured controls as a cut-off limit, the
sensitivity of S100A1B was 61% (95% CI 4973%), the specificity 77% (95% CI
6293%) and negative predictive value 48% (95% CI 3462%) (MTBI vs mild
orthopaedic injuries). The area under the ROC curve did not approach 0.9 for
this or other cut-off limits.
The S100BB concentrations were close to the detection limit of the assay in
the majority of cases. The mean values, as well as the proportions of subjects above
and below the cut-off level, did not differ significantly between the three groups,
i.e. patients with MTBI, patients with mild orthopaedic injuries and non-injured
controls.
S100B and S100A1B concentrations both decreased over time (figure 1). The
decrease was most pronounced for S100B. Accordingly, the number of patients
with values above cut-off limits for S100B as well as S100A1B also decreased from
the examination in the ED to the examination at day 14. It is noticeable that
S100A1B was the only isoform with a mean concentration above the cut-off
limit and only in the acute sample. S100A1B also showed the highest proportion
of patients with MTBI above cut-off at all three measures. S100BB showed the
same tendency, but all values were close to detection limit and only few patients had
S100BB concentrations above the cut-off limit.
S100B and S100BB were significantly correlated (acute  0.47, p < 0.001; day
1  0.30, p < 0.001; and day 14  0.36, p < 0.001). The correlations between
S100B and S100A1B were (acute  0.19, p < 0.05; day 1  0.05, non-significant
(ns); and day 14  0.09 ns) and between S100A1B and S100BB (acute  0.23,
 S100 in mild TBI
Table 2. S100 in the trauma groups (acute sera) and in the non-injured controls

Patients with mild

Patients with MTBI orthopaedic injuries Non-injured-controls
(n 66) (OC) (n 31) (NC) (n 35)

% above % above % above Mann-Whitney Test

Test M SD cut off * M SD cut off * M SD cut off * (mean values)

S100B 0.145 0.1259 41 0.142 0.1089 32 0.065 0.0334 3 MTBI > NC, OC > NC
S100A1B 0.104 0.0614 64 0.056 0.0453 23 0.023 0.0188 3 MTBI > OC, MTBI > NC, OC > NC
S100BB 0.018 0.0236 3 0.026 0.0316 10 0.023 0.0428 3 NS

* 97.5 percentile of non-injured controls: S100B 0.15, S100A1B 0.085, S100BB 0.08.

677
 678
S 100B S 100A 1B

0,160 70% 0,16 0 70%

0,140 0,14 0

Percentage of patietns with concentrations above cut off

Percentage of patients with concentrations above cut off

60% 60%

0,120 0,12 0
50% 50%
0,100

Concentration ug/L
0,10 0
Concentration ug/L

40% 40%
0,080 0,08 0
30% 30%
0,060 0,06 0
20% 20%
0,040 0,04 0

0,020 10% 10%

0,02 0

0,000 0%
0,00 0 0%
Acute Day 1 Day 1 4
Acute Day 1 Day 14
Tim e
Tim e

Bars: mean values of S100

Line: percentage of patients with S100 concentrations above cut off

C. N. de Boussard
Figure 1. Time course of S100B, S100A1B in patients with MTBI.
S100 in mild TBI 679

Table 3. Radiological abnormalities and S100 concentrations in patients with MTBI

No CT or MRT pathology CT and/or MRT pathology

% above % above p-value

Test M SD cut off * M SD cut off * (mean-values)

S100B
Acute 0.125 0.0907 38 0.252 0.2214 60 < 0.05
Day 1 0.089 0.0900 10 0.173 0.2125 23 < 0.01
Day 14 0.058 0.0679 5 0.056 0.0482 6 > 0.1
S100A1B
Acute 0.102 0.0632 62 0.118 0.0505 80 > 0.1
Day 1 0.070 0.0602 40 0.068 0.0464 38 > 0.1
Day 14 0.063 0.0575 30 0.063 0.0578 31 > 0.1
S100BB
Acute 0.014 0.0157 0 0.043 0.0416 2 < 0.05
Day 1 0.008 0.0108 0 0.036 0.0761 2 < 0.1
Day 14 0.006 0.0080 0 0.08 0.0808 0 > 0.1

* 97.5 percentile of non-injured controls: S100B 0.15, S100A1B 0.085, S100BB 0.08.

p < 0.01; day 1   0.08 ns; and day 14  0.15 ns). Thus, each of the S100
variables has a high proportion of unique variance.

The association between S100 and injury severity

The mean values of S100B and of S100BB at the acute examination, as well as
on day 1, were higher in patients with MTBI with CT and/or MRT abnormalities
as compared to patients without radiological abnormalities, as shown in table 3
(S100B acute z 2.32, p < 0.05, S100B day 1 z 2.97, p < 0.01, S100BB acute
z 1.65, p < 0.05, S100BB day 1 z 2.53, p < 0.1). There was no corresponding
difference for S100A1B (S100A1B acute z 1.50, p > 0.1, S100A1B day 1
z 0.40, p > 0.1). These findings were confirmed when the proportions of subjects
with concentrations above and below the cut-off limit of each S100 form were
compared.
There was no relation between S100B, S100A1B or S100BB concentrations and
symptoms according to the RPQ. There was a significant correlation between time
from accident to the first blood sample and S100BB concentrations ( 0.24,
p < 0.05) but not for S100B or S100A1B. No significant association was found
between the concentrations of S100 variables and alcohol level. There was no
significant difference in mean values of S100B and S100A1B in the patients with
orthopaedic injuries that had fractures when compared to those with soft tissue
contusions or distortions or between patients with MTBI with or without minor
orthopaedic injuries.

Discussion
The most conspicuous finding was that S100A1B seems to be more specific for
brain injury than S100B in patients within the milder part of the MTBI spectrum.
It should be pointed out that the majority of patients with MTBI included
in the study had an arrival GCS score of 15, only few had an intra-cranial lesion
680 C. N. de Boussard

on CT scan examination and none required surgical intervention. S100A1B was

significantly higher in patients with MTBI as compared to the orthopaedic and
the non-injured control groups, while S100B did not differentiate the two trauma
groups.
The majority of patients with MTBI has a benign clinical course with symptoms
resolving within 3 months [4345] but in a proportion, which varies between
studies [43, 45], symptoms persist longer. An association between the protein
S100B and persisting symptoms [46] as well as cognitive deficits [29, 30] and
disability [34] has been reported in TBI of varying severity. The present data do
not favour S100A1B or S100B as an acute diagnostic tool to predict symptoms the
first 2 weeks after the trauma in this milder part of the MTBI spectrum. The data
indicate that S100A1B might be more useful than S100B as a brain injury marker in
studies of long-term prognoses of symptoms as well as cognitive deficits after MTBI.
S100B, i.e. the sum of S100A1B and S100BB isoforms, did not discriminate
patients with MTBI from patients with mild orthopaedic injuries. However,
in patients with MTBI the S100B level was higher in the sub-group of patients
with radiological abnormality than in patients without such abnormalities. This
finding is in accordance with previous reports on S100B, even if the data do not
confirm a high negative predictive value of S100B for CT scan abnormalities [22].
One explanation of this finding might be that patients with MTBI with radiological
abnormalities have been exposed to more intensive trauma and more energy transfer
also to other body tissues causing more non-brain specific S100B elevation, or that
S100B and S100A1B elevations reflect different pathophysiological processes in the
brain tissue.
The cellular mechanisms related to the occurrence of the S100 proteins in the
context of trauma have not been described. It is not known if cell disintegration
is required for a detectable serum elevation of any of the S100-isoforms or if
this might occur without cell damage. The correlation analysis of the S100 forms,
showing a unique variance of each variable, indicates that the pathophysiological
mechanisms are not the same for the three S100 forms. The time patterns also
differed. No association was found between the time interval from the injury to
the first blood-sample and the S100A1B or S100B levels. However, the S100BB
concentrations were low and close to the detection limit, which might indicate
a short biologic half time. Very early, post-injury blood sampling would probably
enhance the information on this isoform. S100A1B exhibited a slower decline than
S100B over the study period. Interestingly, the S100B and S100A1B levels declined
over the 14 day period, although the biological halftime of the isoform S100A1A1
has been estimated to
2 hours [14].
Theoretically, the sum of the S100A1B and S100BB values measured should
equal the value of the S100B assay, which was, however, not the case in this
study. This discrepancy might be related to the properties of the antibodies or to
other methodological factors, such as the calcium concentration of the standards
used. The antibodies used for detection of the S100 molecules recognize the three-
dimensional structure of the molecules. Since this structure is destroyed in the
denaturation process that is required for a Western blot analysis, such analysis
could not be used to determine the antibody specificity.
No association was found between the S100 levels and alcohol, which is in
agreement with previous studies in humans [47, 48].
S100 in mild TBI 681

S100B has been suggested to identify patients with TBI at risk for long-term
cognitive dysfunction [29]. The study suggests that S100A1B might be a more
brain injury specific marker in patients within the mild TBI spectrum and, thus,
a candidate for long-term prognostic studies of cognitive complaints in these
patients.

Acknowledgements
This research was founded by grants from the AFA insurance company.

References
1. KRAUS, J. F. and NOURJAH, P.: The epidemiology of mild, uncomplicated brain injury. Journal
of Trauma, 28: 16371643, 1988.
2. BRITTON, M., BORG, J., COLLIANDER, M. et al.: Concussion; observation at hospital or computer
tomography and discharged home? The Swedish Council on Technology Assessment in Health Care
(SBU) Report 2000; Report no. 153.
3. SERVADEI, F., CIUCCI, G., PIAZZA, G. et al.: A prospective clinical and epidemiological study of
head injuries in northern Italy: the comune of Ravenna. Italian Journal of Neurological Science, 9:
449457, 1988.
4. WRIGHTSON, P. and GRONWALL, D.: Mild head injury in New Zealand: incidence of injury and
persisting symptoms. New Zealand Medical Journal, 111: 99101, 1998.
5. MILD TRAUMATIC BRAIN INJURY COMMITTEE OF THE HEAD INJURY INTERDISCIPLINARY SPECIAL
INTEREST GROUP OF THE AMERICAN CONGRESS OF REHABILITATION MEDICINE: Definition of
mild traumatic brain injury. Journal of Head Trauma Rehabilitation, 8: 8687, 1993.
6. STIELL, I. G., WELLS, G. A., VANDEMHEEN, K. et al.: The Canadian CT Head Rule for patients with
minor head injury. Lancet, 357: 13911396, 2001.
7. HAYDEL, M. J., PRESTON, C. A., MILLS, T. J. et al.: Indications for computed tomography in
patients with minor head injury. New England Journal of Medicine, 343: 100105, 2000.
8. MILLER, E. C., HOLMES, J. F. and DERLET, R. W.: Utilizing clinical factors to reduce head CT scan
ordering for minor head trauma patients. Journal of Emergency Medicine, 15: 453457, 1997.
9. MILLER, E. C., DERLET, R. W. and KINSER, D.: Minor head trauma: is computed tomography
always necessary? Annals of Emergency Medicine, 27: 290294, 1996.
10. LIVINGSTON, D. H., LAVERY, R. F., PASSANNANTE, M. R. et al.: Emergency department discharge
of patients with a negative cranial computed tomography scan after minimal head injury. Annals of
Surgery, 232: 126132, 2000.
11. SINDIC, C. J., CHALON, M. P., CAMBIASO, C. L. et al.: Assessment of damage to the central nervous
system by determination of S-100 protein in the cerebrospinal fluid. Journal of Neurology,
Neurosurgery and Psychiatry, 45: 11301135, 1982.
12. PERSSON, L., HARDEMARK, H. G., GUSTAFSSON, J. et al.: S-100 protein and neuron-specific enolase
in cerebrospinal fluid and serum: markers of cell damage in human central nervous system. Stroke,
18: 911918, 1987.
13. DONATO, R.: Functional roles of S100 proteins, calcium-binding proteins of the EF-hand type.
Biochimica et Biophysica Acta, 1450: 191231, 1999.
14. USUI, A., KATO, K., ABE, T. et al.: S-100ao protein in blood and urine during open-heart surgery.
Clinical Chemistry, 35: 19421944, 1989.
15. DONATO, R.: S100: a multigenic family of calcium-modulated proteins of the EF-hand type with
intracellular and extracellular functional roles. International Journal of Biochemical Cell Biology, 33:
637668, 2001.
16. ISOBE, T., TAKAHASHI, K. and OKUYAMA, T.: S100a0 (alpha alpha) protein is present in neurons
of the central and peripheral nervous system. Journal of Neurochemistry, 43: 14941496, 1984.
17. HEIZMANN, C. W.: Ca2-binding S100 proteins in the central nervous system. Neurochemistry
Research, 24: 10971100, 1999.
18. SCHAFER, B. W. and HEIZMANN, C. W.: The S100 family of EF-hand calcium-binding proteins:
functions and pathology. Trends in Biochemical Science, 21: 134140, 1996.
682 C. N. de Boussard

19. FANO, G., BIOCCA, S., FULLE, S. et al.: The S-100: a protein family in search of a function. Progress
in Neurobiology, 46: 7182, 1995.
20. ZIMMER, D. B., CORNWALL, E. H., LANDAR, A. et al.: The S100 protein family: history, function,
and expression. Brain Research Bulletin, 37: 417429, 1995.
21. BIBERTHALER, P., MUSSACK, T., WIEDEMANN, E. et al.: Rapid identification of high-risk patients
after minor head trauma (MHT) by assessment of S-100B: ascertainment of a cut-off level.
European Journal of Medical Research, 7: 164170, 2002.
22. INGEBRIGTSEN, T., ROMNER, B., MARUP-JENSEN, S. et al.: The clinical value of serum S-100 protein
measurements in minor head injury: a Scandinavian multicentre study. Brain Injury, 14:
10471055, 2000.
23. DE KRUIJK, J. R., LEFFERS, P., MENHEERE, P. P. et al.: S-100B and neuron-specific enolase in
serum of mild traumatic brain injury patients. A comparison with health controls. Acta Neurologica
Scandinavica, 103: 175179, 2001.
24. RAABE, A., GROLMS, C., SORGE, O. et al.: Serum S-100B protein in severe head injury.
Neurosurgery, 45: 477483, 1999.
25. WIESMANN, M., MISSLER, U., HAGENSTROM, H. et al.: S-100 protein plasma levels after aneurysmal
subarachnoid haemorrhage. Acta Neurochirgia, 139: 11551160, 1997.
26. MISSLER, U., WIESMANN, M., FRIEDRICH, C. et al.: S-100 protein and neuron-specific enolase
concentrations in blood as indicators of infarction volume and prognosis in acute ischemic stroke.
Stroke, 28: 19561960, 1997.
27. SHAABAN-ALI, M., HARMER, M. and VAUGHAN, R.: Serum S100 protein as a marker of cerebral
damage during cardiac surgery. British Journal of Anaesthesia, 85: 287298, 2000.
28. ROTHOERL, R. D., WOERTGEN, C., HOLZSCHUH, M. et al.: S-100 serum levels after minor and
major head injury. Journal of Trauma, 45: 765767, 1998.
29. HERRMANN, M., CURIO, N., JOST, S. et al.: Release of biochemical markers of damage to neuronal
and glial brain tissue is associated with short and long term neuropsychological outcome after
traumatic brain injury. Journal of Neurology, Neurosurgery and Psychiatry, 70: 95100, 2001.
30. WATERLOO, K., INGEBRIGTSEN, T. and ROMNER, B.: Neuropsychological function in patients with
increased serum levels of protein S-100 after minor head injury. Acta Neurochirgia, 139: 2631,
1997.
31. PLEINES, U. E., MORGANTI-KOSSMANN, M. C., RANCAN, M. et al.: S-100 beta reflects the
extent of injury and outcome, whereas neuronal specific enolase is a better indicator of neuro-
inflammation in patients with severe traumatic brain injury. Journal of Neurotrauma, 18: 491498,
2001.
32. RAABE, A., GROLMS, C. and SEIFERT, V.: Serum markers of brain damage and outcome prediction
in patients after severe head injury. British Journal of Neurosurgery, 13: 5659, 1999.
33. WOERTGEN, C., ROTHOERL, R. D. and BRAWANSKI, A.: Early S-100B serum level correlates
to quality of life in patients after severe head injury. Brain Injury, 16: 807816, 2002.
34. TOWNEND, W. J., GUY, M. J., PANI, M. A. et al.: Head injury outcome prediction in the
emergency department: a role for protein S-100B? Journal of Neurology, Neurosurgery and
Psychiatry, 73: 542546, 2002.
35. ANDERSON, R. E., HANSSON, L. O., NILSSON, O. et al.: High serum S100B levels for trauma
patients without head injuries. Neurosurgery, 48: 12551260, 2001.
36. ANDERSON, R. E., HANSSON, L. O., NILSSON, O. et al.: Increase in serum S100A1-B and S100BB
during cardiac surgery arises from extra cerebral sources. Annals of Thoracic Surgery, 71: 15121517,
2001.
37. ANDERSON, R. E., HANSSON, L. O., LISKA, J. et al.: The effect of cardiotomy suction on the brain
injury marker S100beta after cardiopulmonary bypass. Annals of Thoracic Surgery, 69: 847850,
2000.
38. JOHNSSON, P.: S100-B in blood: a marker of brain damage or simply a covariate? Scandinavian
Cardiovascular Journal, 34: 548549, 2000.
39. KING, N. S., CRAWFORD, S., WENDEN, F. J. et al.: The Rivermead Post Concussion Symptoms
Questionnaire: a measure of symptoms commonly experienced after head injury and its reliability.
Journal of Neurology, 242: 587592, 1995.
40. NYBERG, L., KROON, R., ULLN, A. et al.: Sangtec 100 LIA a new sensitive monoclonal assay
for measuring protein S100 in patients with malignant melanoma. The XXIV Meeting of the
International Society for Oncodevelopmental Biology and Medicine, Colorado, California Nov. 1722,
1996.
S100 in mild TBI 683

41. STIGBRAND, T., NYBERG, L., ULLN, A. et al.: A new specific method for measuring S100B
in serum (Abstract). International Journal of Biology Markers, 15: 3340, 2000.
42. NILSSON, O., NILSSON, K., DAHLN, U. et al.: Characterisation of antigenic determinants in S100
(Abstract). Tumor Biology, 19: 108, 1998.
43. TEASDALE, T. W. and ENGBERG, A.: Duration of cognitive dysfunction after concussion,
and cognitive dysfunction as a risk factor: a population study of young men. British Medical
Journal, 315: 569572, 1997.
44. MACCIOCCHI, S. N., BARTH, J. T., ALVES, W. et al.: Neuropsychological functioning and recovery
after mild head injury in collegiate athletes. Neurosurgery, 39: 510514, 1996.
45. PONSFORD, J., WILLMOTT, C., ROTHWELL, A. et al.: Factors influencing outcome following mild
traumatic brain injury in adults. Journal of the International Neuropsychology Society, 6: 568579,
2000.
46. SAVOLA, O. and HILLBOM, M.: Early predictors of post-concussion symptoms in patients with mild
head injury. European Journal of Neurology, 10: 175181, 2003.
47. BIBERTHALER, P., MUSSACK T., WIEDEMANN, E. et al.: Influence of alcohol exposure on S-100b
serum levels. Acta Neurochirgia Supplement, 76: 177179, 2000.
48. BIBERTHALER, P., MUSSACK, T., WIEDEMANN, E. et al.: Elevated serum levels of S-100B reflect the
extent of brain injury in alcohol intoxicated patients after mild head trauma. Shock, 16: 97101,
2001.