Accepted Manuscript

Post-mortem biochemistry of NSE and S100B: A supplemental tool for detecting a

lethal TBI?

Monique Sieber, Jan Dreßler, Heike Franke, Dirk Pohlers, Benjamin Ondruschka

PII: S1752-928X(18)30032-5
DOI: 10.1016/j.jflm.2018.02.016
Reference: YJFLM 1630

To appear in: Journal of Forensic and Legal Medicine

Received Date: 20 November 2017

Revised Date: 7 February 2018
Accepted Date: 11 February 2018

Please cite this article as: Sieber M, Dreßler J, Franke H, Pohlers D, Ondruschka B, Post-mortem
biochemistry of NSE and S100B: A supplemental tool for detecting a lethal TBI?, Journal of Forensic
and Legal Medicine (2018), doi: 10.1016/j.jflm.2018.02.016.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
Research Paper

Post-mortem biochemistry of NSE and S100B: a

supplemental tool for detecting a lethal TBI?
Monique Siebera, Jan Dreßlera, Heike Frankeb, Dirk Pohlersc, Benjamin Ondruschkaa,*

PT
a
Institute of Legal Medicine, University of Leipzig, Leipzig, Germany
b
Rudolf Boehm Institute of Pharmacology and Toxicology, University of Leipzig,

RI
Leipzig, Germany
c
Center of Diagnostics GmbH, Klinikum Chemnitz, Chemnitz, Germany

SC
* Correspondence to: Dr. med. B. Ondruschka
University of Leipzig

U
Institute of Legal Medicine
AN
Johannisallee 28
04103 Leipzig, Germany
E-mail-address: benjamin.ondruschka@medizinuni-leipzig.de
M

Telephone: +49 (0341) 9715152

D
TE
C EP
AC

1
 ACCEPTED MANUSCRIPT
Research Paper

Post-mortem biochemistry of NSE and S100B: a

supplemental tool for detecting a lethal TBI?
ABSTRACT

PT
Purpose: Traumatic brain injury (TBI) is a very common entity that leads to numerous
fatalities all over the world. Therefore, forensic pathologists are in desperate need of
supplemental methodological tools for the diagnosis of TBI in everyday practice besides the

RI
standard autopsy. The present study determined post-mortem neuron specific enolase (NSE)
and S100 calcium-binding protein B (S100B) levels as biological markers of an underlying
TBI in autopsy cases.

SC
Methods: Paired serum and CSF samples of 92 fatalities were collected throughout routine
autopsies. Afterwards, the marker levels were assessed using commercially available

U
immunoassays (ECLIA, Roche Diagnostics). For statistical analysis, we compared the TBI
cases to three control groups (sudden natural death by acute myocardial infarction, traumatic
AN
death without impact on the head, cerebral hypoxia). Moreover, the TBI cases were
subdivided according to their survival time of the trauma. Brain specimens have been
collected and stained immunohistochemically against the aforementioned proteins to illustrate
their typical cellular staining patterns with an underlying TBI compared to non-TBI fatalities.
M

Principal results: CSF NSE and S100B levels were elevated after TBI compared to all
control groups (p < 0.001). Although this finding can already be investigated among the TBI
D

cases dying immediately subsequent to the trauma, the marker levels in CSF increase with
longer survival times until a peak level within the first three days after trauma. There is a
TE

strong correlation between both marker levels in CSF (r=0.67). The presence or absence of
cerebral tissue contusion following the initial trauma does not seem to affect the CSF levels of
both proteins (p > 0.05). Post-mortem serum levels of both proteins were not elevated in TBI
cases compared to controls (p > 0.05). Former elaborated cut-off values in CSF were
EP

confirmed and were only exceeded when a TBI survival time of at least 30 min was reached.

Major conclusions: The present results report that post-mortem NSE and S100B CSF levels
C

are significantly elevated subsequent to a fatal TBI.

AC

Key words: autopsy, cerebrospinal fluid, NSE, post-mortem biochemistry, S100B, traumatic brain
injury.

1
 ACCEPTED MANUSCRIPT
Introduction

Traumatic brain injuries (TBIs) constitute a major issue of public health. Current
epidemiological data estimates more than two million hospital discharges and 82,000 TBI-
related deaths in Europe in 20121. There were 2.8 million reported cases with 50,000 fatalities
in the United States in 2013.2 Regarding such high incidences per year, it is not surprising that
the diagnosis of TBI is also a common finding in legal medicine.3 The traumatic lesions with

PT
bleedings and cortical contusions can, thus, be detected during the autopsy as well as by
subsequent microscopic investigations in daily routine.4,5

RI
It is known that beside the initial impairment to the head and brain (leading to immediate
stretching, compression and ruptures of vessels and brain tissue, and damage to the blood-

SC
brain barrier function as so-called primary injury), secondary insults, such as increased
intracranial pressure, brain oedema, ischemia due to irregular cerebral blood flow and

U
metabolic disturbances, oxidative stress or inflammatory reactions, are common following a
TBI as so-called secondary injury.6,7 Primary injuries on cellular levels include immediate
AN
neuronal and glial cellular damage (as necrosis), and axonal ruptures (as axonal injury). The
secondary phase of injury can last for hours, days or even months, and is known to be
M

associated with subacute or delayed fatalities resulting from apoptosis, neuronal degeneration
or cell loss (for neurons), a so-called ‘secondary axotomy’ for axons and astrogliosis as well
D

as glial scar formation.7-9

TE

Additional biochemical investigations may sometimes be beneficial and necessary to deduce

trauma survival times, determine the extent of the secondary hypoxic damage to the brain
EP

resulting from the trauma10 and to differentiate between polytraumatised cadavers with and
without head injuries.11
C

Nevertheless, there have been only a few trials to develop an approach to identify TBI cases
as such via biochemical analysis of post-mortem body fluids.7,11-14
AC

It needs to be considered that analyses like these are only reasonably possible when
examining samples of fatalities with short post-mortem intervals (PMIs) to prevent
falsification of measurable concentrations due to decomposition and putrefaction.15 However,
post-mortem biochemistry of trauma cases is still in an early phase of its investigation and,
therefore, reasonable values for a maximum PMI have not yet been established. Moreover, no
single biomarker has yet been shown to be both sensitive and specific for detection of brain

2
 ACCEPTED MANUSCRIPT
damage in everyday clinical practice.16 It would constitute a great simplification and
amendment to the present methods of detecting a TBI in post-mortem setting used routinely if
it were possible to establish a method displaying an existing trauma in body fluids.

Two biomarkers, NSE and S100B, have been broadly researched and investigated among
living patients and could be proven to be rather sensitive for the impairments in question.17-19

Some further promising markers of different cell forms (e.g. neurons, glial cells) that have

PT
been clinically researched in TBI cases include the glial fibrillary acidic protein, brain-derived
neurotrophic factor, tau proteins, ubiquitin carboxy-terminal hydrolase L1, neurofilaments

RI
and alpha-II-spectrin breakdown products (see Lorente 2017 for a current review).20 However,
as stated before, there is no thoroughly accepted single biomarker of TBI or indeed of brain

SC
injury of any type in a clinical setting. The very few existent results in a post-mortem setting
have been summarized recently.21

U
Neuron specific enolase (NSE) can be found predominantly in cell bodies and axons of
AN
neurons, and is, thus, one of the leading proteins of the human nervous system.22 It
participates in axonal transport and is upregulated to maintain homeostasis after axonal
injuries.23 It is then released from the damaged axons into the extracellular space and, from
M

there, into the CSF, for example via a damaged ventricular wall or blood-brain barrier. Hence,
it can reach the blood system via intracranial veins24 using the glymphatic system25 or
D

potentially by phagocytic macrophages.26 Therefore, it is a valuable marker for detecting

TE

axonal injury27 and fatal TBIs.5,11,28 It also occurs in platelets and erythrocytes, and, moreover,
in an exiguous amount in various human organs.29 However, NSE levels among living
EP

patients have been proven to possibly increase in serum samples after traumas without brain
damage and after heavy blood loss situations as well.30
C

S100 calcium-binding protein B (S100B) can be found predominantly in glial cells

(astrocytes, oligodendrocytes).31 The occurrence in the nervous system represents up to 90 %
AC

of all S100B sources in the human body.32 The S100B operates mainly by contributing to
signal transduction from glial to other nervous cells33 and it takes part in the regulation of cell
remodelling.32 The protein may pass to the bloodstream after its paracrine excretion into the
extracellular space following damage or death of astrocytes via arachnoid villi or through a
disrupted blood-brain barrier or via more complex intracellular transport forms.32,34,35 It has
been proven to be useful for estimating prognosis and outcome after TBIs in intensive care

3
 ACCEPTED MANUSCRIPT
patients19,36 and after different causes of secondary central hypoxia without initial impact on
the head.37,38

The aim of the study presented was to test the relevance and practical applicability of NSE
and S100B as useful postmortem biochemical markers of brain damage after TBI in forensic
practice.

PT
RI
U SC
AN
M
D
TE
C EP
AC

4
 ACCEPTED MANUSCRIPT
Experimental procedure

Sampling

We collected serum and CSF samples of 92 routine autopsy cases that were examined at the
Institute of Legal Medicine of the University of Leipzig.

The material collected includes samples of 23 females (25 %) and 69 males (75 %), dying at

PT
an age of between 18 and 95 years (median: 60 years). The PMI was defined as the time
interval between the moment of death and the freeze-storage of the samples. The median
amounted to 58 h (range: 5 to 148 h). Any macroscopic signs of putrefaction led to immediate

RI
rejection of the specific case; any macroscopic signs or the knowledge of existent neoplasms

SC
also led to rejection. Cases with known neurodegenerative diseases were excluded as well.

Blood samples were obtained from the femoral veins. In order to acquire the CSF samples, the

U
cerebellar tentorium was removed carefully to prevent iatrogenic impurity of the sample with
blood. Subsequently, the CSF could be collected from the suboccipital subarachnoid space.
AN
Immediately afterwards, both blood and CSF samples were centrifuged at 5000 rpm at 4 °C
for 5 min. The supernatant was subsequently stored at -80 °C, strictly avoiding any thawing
M

process until further analysis.

Regarding the causes of death, the cases were divided into four groups: traumatic brain injury
D

(TBI, n = 45), isolated torso trauma without traumatic impact to the head (ITT, n = 18),
TE

diffuse cerebral hypoxia (DCH, n = 15) and acute myocardial infarction as a specific example
of any sudden death originating from an inner cause (AMI, n = 14). The non-TBI cases did
EP

not show any injury to the head or the brain.

We decided to define a blunt force injury induced intracranial bleeding and/or cortical
C

contusions as a macroscopic equivalent for a TBI and, therefore, as objective inclusion

AC

criterion for our case group. The TBI cases were graded into three subgroups according to
their individual survival time of the trauma. Following related investigations of this topic, the
subcategories were defined as acute death (survival time < 2h; n=23), subacute death (survival
time 2h – 3d, n=11) and delayed death (survival time 3 – 19 d, n=11). Table 1 summarizes
the patients’ demographic characteristics, including percentages of cases who underwent final
cardiopulmonary resuscitation and the circumstances of death of the cohort studied.

5
 ACCEPTED MANUSCRIPT
Furthermore, all TBI cases were divided by their distinct injury patterns into those with
intracranial bleedings (e.g. epidural, subdural, subarachnoid; total n = 17) compared to those
with additional cerebral tissue damage (cortical contusion, intracerebral bleeding, tissue
laceration; total n = 29).

Survival times after the traumatic impacts for both TBI and ITT cases were estimated
according to the reports of paramedics and emergency physicians, the records of the police

PT
investigations and based on pathological findings during the autopsy. Only single cases of
ITT exceeded the acute death scenario.

RI
As DCH cases showed only very short agony phases in general and all included AMI cases
died suddenly or were found dead after a short period of time unseen, the survival time was

SC
defined as ‘none’ for all cases among those groups. All DCH and AMI cases were, therefore,
defined as acute deaths.

U
The analysis of blood alcohol concentration was performed in 62 of all 92 cases, whereupon
AN
12 cases showed measurable levels of blood alcohol (range: 0.15 – 2.73 g/l), among which
only three cases showed a relevant blood alcohol concentration above 1.5 g/l. In addition,
toxicological examination showed findings of narcotic agents in 7 out of 61 samples tested
M

(Zolpidem, n = 1; morphine derivates, n = 2; methamphetamine, n = 4), none of them in lethal

concentrations.
D

Measurement
TE

The Elecsys ElectroChemiLuminescence ImmunoAssay (ECLIA; Roche Diagnostics,

Mannheim, Germany) was used to analyse the serum and CSF levels of NSE and S100B on a
EP

Modular E170 analyser (Roche Diagnostics) as described before in detail.11 Please see the
online fact sheets for technical details
C

(https://www.roche.de/res/content/7854/nse_factsheet.pdf and
https://www.roche.de/res/content/7854/s100_factsheet.pdf). The limits of detection are
AC

reported as 0.05 ng/ml for NSE and 0.005 ng/ml for S100B. The serum samples were diluted
1:10 for NSE and up to 1:100 for S100B before measurement. The CSF samples had to be
thinned down 1:100 for NSE and up to 1:500 for S100B before the analysis.

The haemolytic index of the samples was quantified using a cobas c701 analyser (Roche
Diagnostics) by measuring absorbance at 570/600 nm of saline diluted samples to estimate the
grade of red blood cells destroyed within the single sample.

6
 ACCEPTED MANUSCRIPT
All investigators were entirely blinded to all patients’ data while carrying out the assays.

Histology and immunohistochemistry

Brain cortex tissue samples were collected during autopsies and fixed in 4 % neutral buffered
formalin. The samples were stored in small cubes and rested for at least one week for
complete fixation. Afterwards, they were embedded in paraffin. Sections were standardly
stained by haematoxylin-eosin for morphological evaluation of different cell types and in

PT
individual cases, immunohistochemically with anti-NSE (monoclonal mouse; Agilent Dako)
and anti-S100 (polyclonal rabbit; Agilent Dako), as described before in detail.5

RI
Statistics

Statistical analysis was performed using Microsoft Excel (2010; Bellevue, Washington) and

SC
the open source software R (version 3.4.0, 2017).
All correlations were computed using bivariate correlation (r). Categorial variables were

U
analysed using the Chi-square test. The Kolmogorov-Smirnov test was used to assess whether
AN
the biomarker levels were normally distributed. As they were not, the groups were tested for
statistically significant differences using the non-parametric Mann-Whitney U test. With that,
we compared all TBI cases to all controls, as well as all TBI cases to every single control
M

group. This test was also used to compare the different TBI survival time categories to one
another and to check for differences between TBI and control cases with similar survival
D

periods. The resulting p-values were adjusted for multiple testing using the Benjamini-
TE

Hochberg procedure.

Adjusted p-values of p < 0.05 were considered as statistically significant and p < 0.001 as
EP

highly statistically significant.

C
AC

7
 ACCEPTED MANUSCRIPT
Results

Acceptance sampling

The haemolytic index, as a marker for free haemoglobin in the sample, was used to determine
the quality of all 184 samples. The serum samples showed marginal correlation of protein
levels with the haemolytic indices (r = 0.29 for NSE; r = 0.08 for S100B) and no correlation
with the brain weight (r = 0.05 for NSE; r = 0.03 for S100B). The results of those

PT
measurements in CSF displayed that most of those samples were crystal clear and blood-free.
There was a moderate correlation of the haemolytic indices in the CSF samples with

RI
concentrations of the proteins (r = 0.36 for NSE; r = 0.24 for S100B) and concerning the PMI
(r = 0.18 for NSE; r = 0.43 for S100B). Interestingly, there was also a moderate correlation

SC
between the brain weight of the deceased (weighed at autopsy using a calibrated scale) and
the corresponding CSF levels to be measured (r = 0.34 for NSE; r = 0.38 for S100B).

U
There was practically no relevant interrelation between the PMI and the haemolytic index in
AN
all cases (r = 0.19 for CSF; r = 0.03 for serum). Both proteins could be proven to correlate
moderately to the PMI in serum samples (r = 0.32 for NSE; r = 0.46 for S100B). Neither
serum nor CSF levels showed any age- (r = 0.04 to 0.13) or sex-dependency (r = 0.01 to
M

0.28).
D

None of the cases with drug influence or relevant alcohol blood concentration was within the
TE

extreme values of the groups.

Cause of death differentiation

EP

Please see Table 2 for a tabular overview of the measured median marker levels of the case
and control groups. In addition, please find histologically and immunohistochemically
C

achieved pictures in Figure 1 displaying the cerebral occurrence of both proteins studied in
AC

their typical cellular staining patterns, hereby illustrating the differences between TBI cases
and controls.

There were generally much higher levels of NSE and S100B to be found in CSF than in the
corresponding serum samples, with moderate correlations between both compartments (r =
0.28 for NSE, r = 0.27 for S100B). Statistically, there is a strong correlation of the CSF levels
of both proteins (r = 0.67).

8
 ACCEPTED MANUSCRIPT
When comparing CSF levels of both proteins in all TBI cases to all controls, they were much
higher after TBI (p < 0.001 for both NSE and S100B). Similar results were obtained
comparing all TBI cases with non-head trauma fatalities, the ITT cases (p < 0.05 for NSE, p <
0.001 for S100B) and with the DCH group (p < 0.05 for both NSE and S100B). A clear
discrimination between TBI and AMI cases was shown for CSF concentrations of S100B (p <
0.05), but not for NSE (p = 0.17). Comparable significances are obtained when comparing the
control groups to the acute TBI cases individually to check similar survival periods.

PT
No significant differences could be detected in serum samples at all (p > 0.05).

RI
There were 28 cases (62.2 % of all TBI cases) among the TBI casualties showing cerebral
tissue contusion along with the intracranial bleeding that could be found in all TBI cases as

SC
primary inclusion criterion. There were more cases with cerebral tissue damage than without
in every survival time category, although not reaching statistical significance (p > 0.05).
However, statistical analysis showed no significant differences between the marker levels

U
comparing both groups. Please see Table 3 for details.
AN
All TBI cases were then sorted into three subgroups considering their survival times. The only
significant difference could be detected when comparing acute and subacute death cases
M

measuring CSF S100B. All other comparisons remained without significant differences in
marker levels. The highest CSF NSE levels were measured within TBI survival times of 3 to 4
D

d. The CSF S100B levels showed elevated levels above all reference values in TBI survival
TE

times between 4 h and 4 d. The CSF levels of both proteins only exceeded the empirically
elaborated cut-off values11 when they evinced a minimum trauma survival time of 30 min.
EP

Please see Figure 2 for details of post-mortem measurement in serum and CSF of NSE. In
analogy, detailed results of S100B analysis in both body fluids are depicted in Figure 3.
C
AC

9
 ACCEPTED MANUSCRIPT
Discussion

The objective of the present study was to determine biochemical changes in body fluids after
a TBI in the post-mortem examination. The main observations were that post-mortem levels
of NSE and S100B in CSF are significantly related to underlying causes of death and are
excessively elevated after fatal TBI whereas until now, post-mortem serum levels of neither
protein could be shown to discriminate between TBI cases and other fatalities. Furthermore,

PT
even TBI cases with only short survival times < 2 h evinced higher CSF values than the
controls checked.

RI
Although the cases studied evince a maximum PMI of 6 d with transfer of the deceased into
adequate cool chambers within hours after death, it is unfeasible to avoid unspecific post-

SC
mortem changes in the form of autolytic processes of the corpse and, thus, of the body fluids.
However, we decided to avoid sampling from corpses showing any signs of putrefaction and

U
decay rather than applying an arbitrarily determined maximum PMI for supplying an actual
transferability of the obtained results to daily medicolegal practice. Therefore, the materials
AN
used in our study represent the quality of body fluid samples that can be found in the reality of
forensic practice, as they exhibited PMIs that can be regarded as moderate for our home
M

country. The analysis of haemolytic indices can only be a rough guideline for the extent of
post-mortem alterations, and yet it represents an internal quality control per sample.
D

Nevertheless, the possible impact of the methodologically necessary dilution should also be
TE

considered. Regarding further endeavours to identify lethal TBI cases biochemically,

additional investigation of the cytolysis and proteolysis of the central nervous system would
be preferable for a better understanding of possible post-mortem fluctuations in biomarker
EP

profiles.

It was proven that S100B does not increase in blood samples collected for 2 d at room
C

temperature.11,13 The current results confirm these observations, yet they implicate an
AC

enlarged probability of measuring false high concentrations after this time interval. Therefore,
the authors would advise further studies on the topic to be carried out within 2 d after the
individual’s death whenever serum analyses are scheduled.

Both NSE and S100B are known to be stable during long-term freeze-storage39,40 and do not
appear to show any dependency on sex or age in either serum or CSF. While S100B is also
known to be rather independent of haemolytic changes,41 there is an increase of NSE in serum
and CSF levels following breakdown of red blood cells, which contain the enzyme
10
 ACCEPTED MANUSCRIPT
themselves. This could lead to difficulties in interpreting the findings.42 These unpredictable
changes, even after a short PMI, are one of the main reasons to reject measurement of serum
NSE in samples of fatalities as a marker for TBI in daily routine. The fluctuations were too
high for linear regression estimation exploitable as a ‘corrective factor’.

The alterations of CSF NSE levels are supposedly due to the fact that haemolysis only occurs
in CSF when there is blood admixture, thus, causing a quite similar dependency to that as

PT
observed in serum. Nevertheless, it remains suitable for detecting TBI cases, as the amount of
CSF clearly outweighs the extent of the admixture in the sample and because of the sample
preparation used with centrifugation that separated the bloody content from the clear

RI
supernatant fluid.

SC
Cause of death differentiation

The CSF NSE and S100B levels were the highest in fatalities after suffering TBI, but were

U
also found elevated in ITT, CVI and AMI cases when compared to clinical references and
AN
studies.43,44 This could hint at an unspecific passive release of the proteins in the agony phase
or even in the early post-mortem period,15 or the fact that the process of death is always
accompanied by a certain extent of cerebral hypoxia and, thus, cell death of neuronal and glial
M

tissue, regardless of its definite cause.45 There is still not enough knowledge of agonal and
early autolytic processes to provide a satisfactory answer to this question, since no
D

comprehensive forensic analysis deals with subsequently stored body fluid samples from both
TE

ante-mortem and post-mortem collection from an individual.

However, the extent of elevation could be shown to be significantly higher in TBI cases that
EP

likely had a more distinct impact on brain tissue compared to all other groups investigated.
Biochemically assessed marker levels could be found elevated in both TBI cases with and
C

without cortical contusions, which is in line with the results obtained from studies in clinical
settings.46-48 The CSF levels of both markers peaked within the first days after TBI; this is
AC

also known for living patients.18,44 The DCH cases especially evinced high amounts of both
proteins NSE and S100B in CSF samples as well, which should be interpreted as a
consequence of the hypoxic damage of cells due to the causes of death presented herein,
although dysfunction of the central nervous system is inevitable in the process of dying
caused by final circulatory and respiratory failure. Therefore, the release of NSE and S100B
in the CSF does not seem to be an only trauma-induced phenomenon, but is also associated

11
 ACCEPTED MANUSCRIPT
with severe hypoxic episodes, as Li et al. stated previously.13 This is in line with the current
literature in clinical studies.44,49

We compared the control groups that exceeded 2 h of survival time only in individual cases
(among the ITT group) to the TBI cases that died in the same time post-injury to assess to
which extent the survival time itself influences the marker levels and to rule out the possibility
of statistical distortion when comparing acute deaths of the control group to TBI cases dying

PT
in a subacute or delayed time period subsequent to the trauma. However, we still found
comparable differences between acute TBI cases and the control groups, indicating that the
difference is not only caused by the prolonged hypoxic phase that goes along with a more

RI
extensive survival time, but also to the possibly very intense and fast cellular damage
succeeding the cerebral trauma.

SC
Contrary to the findings of Daoud and colleagues,50 the current data does not suggest
limitations regarding the validity of S100B in fatalities that suffered a multisystem trauma

U
along with a TBI, at least not in CSF. The ITT group showed significantly lower levels of the
AN
proteins in CSF than the TBI group even with partly investigated fatal polytraumas.

Former post-mortem data

M

To the best of the authors’ knowledge, there have only been a few studies on the post-mortem
D

measurement of NSE or S100B observing the cause of death.11-14 The present data supports
the former findings concerning elevated post-mortem CSF levels of both proteins. The
TE

moderate correlation between the CSF levels of both markers and the brain weight at the time
of the autopsy had not been reported before, but can be explained well by the cytotoxic
EP

oedematous reaction of the brain after a TBI, although the brain weight is only indicative of
the extent of brain damage in the final result of swelling/herniation in the agonal phase as
C

manifestation of cerebral and cellular dysfunction. However, the induced oedema in TBI
cases leads to a generalisation of the initially located damage caused by the traumatic impact.
AC

The concordant link between the CSF levels of NSE and S100B, which can assumingly be
explained by the count of neuronal respectively glial cells, that can possibly be harmed in the
course of any impact to the brain and, therefore, elevate the protein levels is in line with this
aspect.

12
 ACCEPTED MANUSCRIPT
Limitations

A rather wide range of both NSE and S100B protein levels was detected among the single
groups in both paired samples. This might be due to the fact of the existing heterogeneity of
the corpses observed as well as of the samples’ different PMIs, or even merely because of
different injury patterns among the TBI and the ITT group that caused the individual’s death.
In addition, neither protein is known to be brain-specific. It would be preferable to examine an

PT
even larger cohort in order to rule out or define specific confounders and, consequently,
elaborate the true utility of the markers.51

RI
On top of that, one must consider the primarily excluded patients with neurodegenerative
diseases or malignant tumours. Our data does not provide enough information about the exact

SC
influence of those diseases on the biomarker levels, thus, we cannot estimate the possible
impact of these clinical conditions or others that we did not even consider. On the other hand,
it cannot safely be ruled out that some of the patients included actually had or developed such

U
diseases to a pre- or subclinical extent. Family testimonies and medical documentation for all
AN
cases did not report neurodegenerative diseases. Moreover, when comparing clinical results of
biomarker elevation due to chronical brain diseases and the extent of elevation perceived
M

when analysing samples of fatal TBI cases, the former seems to be of minor significance.

It appears reasonable to obtain CSF and blood samples via external punctures on the arrival of
D

the corpse at the morgue in the future, as it is increasingly carried out before post-mortem
TE

radiological investigations take place. If legally possible, it also appears appropriate to collect
the samples during the external post-mortem examination on-site. This offers the opportunity
EP

to reduce the influence of post-mortem changes within the body fluids to an insignificant
minimum.
C

Consequences for legal medicine

AC

The authors would not recommend NSE or S100B measurement in serum knowing the
statistical results that do not seem to provide valuable information about the cause of death at
this point. Other body fluids that can be collected during routine forensic autopsies, such as
vitreous humour, pericardial fluid or other blood sources (subclavian, heart chamber), can be
interesting subjects for further research on the topic of TBI-related biomarkers in the post-
mortem setting. This time, we limited our research activities to the most relevant biofluids

13
 ACCEPTED MANUSCRIPT
serum and CSF, because both can also be archived easily in living patients and have,
therefore, been researched broadly among those.

Further studies on post-mortem biochemical measurement may include associated case-by-

case immunohistochemical brain investigations to illustrate the morphological changes after
the trauma to the central nervous system that constitutes the cause of the changes in biomarker
levels, as Olczak and colleagues have already done for the microtubule-associated protein
Tau.7 Immunohistochemical investigation of NSE and S100B, even in combination, has

PT
already been performed.5 Both markers are used in our forensic institute as routine
microscopic staining parameters in questions of vitality and wound age estimation after TBI.

RI
As CSF is a body fluid easily extracted in every autopsy, it represents an ideal material to

SC
evaluate the cause of death of a deceased, especially in cases where a certain cause of death
cannot be found from the macroscopic and microscopic examination or whenever two-stage
injuries are detected as the main mechanism of a traumatic death. The data presented suggests

U
that both NSE and S100B occur in significantly higher levels after a TBI than in all other
AN
cases of injury to the body, cerebral hypoxia or sudden natural death, thus providing a high
selectivity for those cases, most commonly because of the CSF’s proximity to the injured
M

brain itself.52 Therefore, post-mortem biochemistry of CSF samples can be considered a

worthwhile distinguishing feature and an easy, quick and inexpensive diagnostic tool even in
D

cases where an autopsy is not (yet) ordered by the local prosecutors or the relatives, and
could, thus, be established as a valuable supplemental tool in everyday practice of a forensic
TE

pathologist, although it clearly cannot serve as an equivalent substitute for a full autopsy to
identify a definite cause of death. In the course of this, CSF sampling should be performed as
EP

soon as possible after death, for example via suboccipital puncture, whenever comprehensive
corresponding laws are available in the countries in question to minimize potential post-
C

mortem alterations.
AC

14
 ACCEPTED MANUSCRIPT
References

1. Majdan M, Plancikova D, Brazinova A, Rusnak M, Nieboer D, Feigin V, Maas A.

Epidemiology of traumatic brain injuries in Europe: a cross-sectional analysis. Lancet Public
Health. 2016;1:e76-e83.

2. Taylor CA, Bell JM, Breiding MJ, Xu L. Traumatic brain injury-related emergency
department visits, hospitalizations, and deaths – United States, 2007 and 2013. MMWR

PT
Surveill Summ. 2017;66:1-16.

3. Faul M, Coronado V. Epidemiology of traumatic brain injury. Handb Clin Neurol.

RI
2015;127:3-13.

4. Hausmann R, Betz P. The time course of the vascular response to human brain injury: an

SC
immunohistochemical study. Int J Legal Med. 2000;113:288-292.

5. Krohn M, Dreßler J, Bauer M, Schober K, Franke H, Ondruschka B. Immunohistochemical

U
investigation of S100 and NSE in cases of traumatic brain injury and its application for
AN
survival time determination. J Neurotrauma. 2015;32:430-440.

6. Thelin EP, Nelson DW, Bellander BM. Secondary peaks of S100B in serum relate to
M

subsequent radiological pathology in traumatic brain injury. Neurocrit Care. 2014;20:217-

229.
D

7. Olczak M, Niderla-Bielinska J, Kwiatkowska M, Samojlowicz D, Tarka S, Wierzba-

TE

Bobrowic, T. Tau protein (MAPT) as a possible biochemical marker of traumatic brain injury
in postmortem examination. Forensic Sci Int. 2017;280:1-7.

8. Franke H, Verkhratsky A, Burnstock G, Illes P. Pathophysiology of astroglial purinergic

EP

signaling. Purinergic Signal. 2012;8:629-657.

C

9. Simon DW, McGeachy MJ, Bayir H, Clark RS, Loane DJ, Kochanek PM. The far-reaching
scope of neuroinflammation after traumatic brain injury. Nat Rev Neurol. 2017;13:171-191.
AC

10. Oehmichen M, Meissner C, von Wurmb-Schwark N, Schwark T. Methodical approach to

brain hypoxia/ischemia as a fundamental problem in forensic neuropathology. Legal Med.
2003;5:190-201.

11. Ondruschka B, Pohlers D, Sommer G, Schober K, Teupser D, Franke H, Dreßler J. S100B and
NSE as useful postmortem biochemical markers of traumatic brain injury in autopsy cases. J
Neurotrauma. 2013;30:1862-1871.

15
 ACCEPTED MANUSCRIPT
12. Osuna E, Perez-Carceles MD, Luna A, Pounder DJ. Efficacy of cerebro-spinal fluid
biochemistry in the diagnosis of brain insult. Forensic Sci Int. 1992;52:193-198.

13. Li DR, Zhu BL, Ishikawa T, Zhao D, Michiue T, Maeda H. Postmortem serum protein S100B
levels with regard to the cause of death involving brain damage in medicolegal autopsy cases.
Legal Med. 2006;8:71-77.

14. Li DR, Michiue T, Zhu BL, Ishikawa T, Quan L, Zhao D, Yoshida C, Chen JH, Wang Q,
Komatsu A, Azuma Y, Maeda H. Evaluation of postmortem S100B levels in the cerebrospinal

PT
fluid with regard to the cause of death in medicolegal autopsy. Legal Med. 2009;11:S273-
S275.

RI
15. Maeda H, Zhu BL, Ishikawa T, Quan L, Michiue T. Significance of postmortem biochemistry
in determining the cause of death. Legal Med. 2009;11:S46-S49.

SC
16. Zurek J, Fedora M. The usefulness of S100B, NSE, GFAP, NF-H, secretagogin and Hsp70 as
a predictive biomarker of outcome in children with traumatic brain injury. Acta Neurochir.
2012;154:93-103.
U
AN
17. Hajdoková L, Sobek O, Prchalová D, Bilková Z, Koudelková M, Lukásková J, Matuchová I.
Biomarkers of brain damage: S100B and NSE concentrations in cerebrospinal fluid – a
M

normative study. Biomed Res Int. 2015;2015:379071.

18. Thelin EP, Jeppsson E, Frostell A, Svensson M, Mondello S, Bellander BM, Nelson DW.
D

Utility of neuron-specific enolase in traumatic brain injury; relations to S100B levels,

TE

outcome, and extracranial injury severity. Crit Care. 2016;20:285.

19. Rodríuez-Rodríguez A, Egea-Guerrero JJ, Gordillo-Escobar E, Enamorado-Enamorado J,

EP

Hernández-Garcia C, Ruiz de Azúa-López Z, Vilches-Arenas Á, Guerrero JM, Murillo-

Cabezas F. S100B and neuron-specific enolase as mortality predictors in patients with severe
traumatic brain injury. Neurol Res. 2016;38:130-137.
C

20. Lorente L. Biomarkers associated with the outcome of traumatic brain injury patients. Brain
AC

Sci. 2017;7:E142.

21. Ondruschka B, Schuch S, Sieber M, Woydt L, Dreßler J. Biochemistry in traumatic brain

injury: even useful in postmortem setting. Rom J Legal Med. 2017;25:301-302.

22. Schmechel D, Marangos PJ, Brightmann M. Neurone-specific enolase is a molecular marker

for peripheral and central neuroendocrine cells. Nature. 1978;276:834-836.

16
 ACCEPTED MANUSCRIPT
23. Kawata K, Liu CY, Merkel SF, Ramirez SH, Thierney RT, Langford D. Blood biomarkers for
brain injury: What are we measuring? Neurosci Biobehav Rev. 2016;68:460-473.

24. Bandyopadhyay S, Hennes H, Gorelick MH, Wells RG, Walsh-Kelly CM. Serum neuron-
specific enolase as a predictor of short-term outcome in children with closed traumatic brain
injury. Acad Emerg Med. 2005;12:732-738.

25. Plog BA, Dashnaw ML, Hitomi E, Peng W, Liao Y, Lou N, Deane R, Nedergaard M.
Biomarkers of traumatic injury are transported from brain to blood via the glymphatic system.

PT
J Neurosci. 2015;35:518-526.

RI
26. Penkola M, Moos T, Carrasco J, Hadberg H, Molinero A, Bluethmann H, Hidalgo J. Strongly
compromised inflammatory response to brain injury in interleukin-6-deficient mice. Glia.
1999;25:343-357.

SC
27. Ogata M, Tsuganezawa O. Neuron-specific enolase as an effective immunohistochemical
marker for injured axons after fatal brain injury. Int J Legal Med. 1999;113:19-25.

U
AN
28. Goede A, Dreßler J, Sommer G, Schober K, Franke H, Ondruschka B. Wound age estimation
after fatal traumatic brain injury. Rechtsmedizin. 2015;25:261-267.

29. Haimoto H, Takahashi Y, Koshikawa T, Nagura H, Kato K. Immunohisto-chemical

M

localisation of γ-enolase in normal human tissues other than nervous and neuroendocrine
tissues. Lab Invest. 1985;52:257–263.
D

30. Pelinka LE, Hertz H, Mauritz W, Harada N, Jafarmadar M, Albrecht M, Redl H, Bahrami S.
TE

Nonspecific increase of systemic neuron-specific enolase after trauma: clinical and

experimental findings. Shock. 2005;24:119-123.
EP

31. Donato R. S100: A multigenetic family of calcium-modulated proteins of the EF-hand type
with intracellular and extracellular functional roles. Int J Biochem Cell Biol. 2001;33:637-668.
C

32. Sen J, Belli A. S100B in neuropathologic states: the CRP of the brain? J Neurosci Res.
AC

2007;15:1373-1380.

33. Steiner J, Bernstein HG, Bielau H, Berndt A, Brisch R, Mawrin C, Keilhoff G, Bogerts B.
Evidence for a wide extra-astrocytic distribution of S100B in human brain. BMC Neurosci.
2007;8:2.

34. Barger SW, van Eldik LJ. S100 beta stimulates calcium fluxes in glial and neuronal cells. J
Biol Chem. 1992;267:9689-9694.

17
 ACCEPTED MANUSCRIPT
35. Lamers KJB, Vos P, Verbeek MM, Rosmalen F, van Geel WJA, van Engelen BG. Protein S-
100B, neuron-specific enolase (NSE), myelin basic protein (MBP) and glial fibrillary acidic
protein (GFAP) in cerebrospinal fluid (CSF) and blood of neurological patients. Brain Res
Bull. 2003;61:261-264.

36. Vos PE, Lamers KJ, Hendriks JC, van Haaren M, Beems T, Zimmermann C, van Geel W, de
Reus H, Biert J, Verbeek MM. Glial and neuronal protein in serum predict outcome after
severe traumatic brain injury. Neurology. 2004;62:1303-1310.

PT
37. Wojtczak-Soska K. Lelonek M. S-100B protein: an early prognostic marker after cardiac
arrest. Cardiol J. 2010;17:532-536.

RI
38. Schiavi P, Iaccarino C, Servadei F. The value of the calcium binding protein S100 in the
management of patients with traumatic brain injury. Act. Biomed. 2012;83:5-20.

SC
39. Raabe A, Kopetsch O, Gross U, Zimmermann M, Gebhart P. Measurements of serum S-100B
protein: effects of storage time and temperature on pre-analytical stability. Clin Chem Lab
Med. 2003;41:700-703.
U
AN
40. Ramont L, Thoannes H, Volondat A, Chastang F, Millet MC, Maquart FX. Effects of
hemolysis and storage condition on neuron-specific enolase (NSE) in cerebrospinal fluid and
M

serum: implications in clinical practice. Clin Chem Lab Med. 2005;43:1215-1217.

41. Beaudeux J, Léger P, Dequen L, Gandjbakhch I, Criat P, Foglietti MJ. Influence of hemolysis
D

on the measurement of S-100beta protein and neuron-specific enolase plasma concentrations

TE

during coronary artery bypass grafting. Clin Chem. 2000;46:989-990.

42. Gao F, Harris DN, Sapsed-Byrne S. Neuron-specific Enolase and Sangtec 100 assays during
EP

cardiac surgery: part III. Does haemolysis affect their accuracy? Perfusion. 1997;12:171-177.

43. Al Nimer F, Thelin E, Nyström H, Dring AM, Svenningsson A, Piehl F, Nelson DW,
C

Bellander BM. Comparative assessment of the prognostic value of biomarkers in traumatic

brain injury reveals an independent role for serum levels of neurofilament light. PLoS One
AC

2015;10:e0132177.

44. Kellermann I, Kleindienst A, Hore N, Buchfelder M, Brandner S. Early CSF and serum
S100B concentrations for outcome prediction in traumatic brain injury and subarachnoid
hemorrhage. Clin Neurol Neurosurg. 2016;145:79-83.

45. Dreßler J, Hanisch U, Kuhlisch E, Geiger KD. Neuronal and glial apoptosis in human
traumatic brain injury. Int J Legal Med. 2007;121:365-375.

18
 ACCEPTED MANUSCRIPT
46. Herrmann M, Jost S, Kutz S, Ebert AD, Kratz T, Wunderlich M, Synowitz H. Temporal
profile of release of neurobiochemical markers of brain damage after traumatic brain injury is
associated with intracranial pathology as demonstrated in cranial computerized tomography. J
Neurotrauma. 2000;17:113-122.

47. Wolf H, Frantal S, Pajenda GS, Salameh O, Widhalm H, Hajdu S, Sarahrudi K. Predictive
value of neuromarkers supported by a set of clinical criteria in patients with mild traumatic
brain injury: S100B protein and neuron-specific enolase on trial. J Neurosurg. 2013;118:1298-

PT
1303.

48. Olivecrona Z, Bobinksi L, Koskinen LOD. Association of ICP, CPP, CT findings and S-100B

RI
and NSE in severe traumatic head injury. Prognostic value of the biomarkers. Brain Inj.
2015;29:446-454.

SC
49. Yan EB, Satgunaseelan L, Paul E, Bye N, Nguyen P, Agyapomaa D, Kossmann T, Rosenfeld
JV, Morganti-Kossmann MC. Post-traumatic hypoxia is associated with prolonged cerebral

U
cytokine production, higher serum biomarker levels, and poor outcome in patients with severe
traumatic brain injury. J Neurotrauma. 2014;31:618-629.
AN
50. Daoud H, Alharfi I, Alhelali I, Charyk Stewart T, Qasem H, Fraser DD. Brain injury
biomarkers as outcome predictors in pediatric severe traumatic brain injury. Neurocrit Care.
M

2014;20:427-435.
D

51. Papa L, Wang KKW. Raising the bar for traumatic brain injury biomarker research: methods
make a difference. J Neurotrauma. 2017;34:2187-2189.
TE

52. Hanrieder J, Wetterhall M, Enblad P, Hillered L, Bergquist J. Temporally resolved differential

proteomic analysis of human ventricular CSF for monitoring traumatic brain injury biomarker
EP

candidates. J Neurosci Methods. 2009;177:469-478.

C
AC

19
 ACCEPTED MANUSCRIPT
Figure legends:

Figure 1: Exemplary microscopic pictures of acute death cases after traumatic brain injuries (on
the left) in comparison to controls (on the right).
A – Fresh cortical contusion with tissue laceration and numerous haemorrhages (double arrow) (H&E,
scale bar: 100 µm). B – Frontal lobe with intact subarachnoid space (single arrow) and uninjured
cortical layers (arrowhead) (H&E, scale bar: 100 µm).
C – Immunopositive neurons with axons (double cross) in a pericontusional zone of a frontal lobe

PT
(anti-NSE, scale bar: 25 µm). D – Negative immunoreaction in neurons (single cross) and glial cells in
an uninjured frontal lobe (anti-NSE, scale bar: 25 µm).

RI
E – S100-immunopositive reaction of glial cells (double asterisk) in a pericontusional zone of a frontal
lobe. Please note the positivity of the glia limitans (double arrow) in close proximity to the

SC
surrounding meninges and cerebrospinal fluid (anti-S100, scale bar: 50 µm). F – Among control
cases, the glia limitans (single arrow) and most of the glial cells stay immunonegative (anti-S100,
scale bar: 50 µm).

U
G – White matter with S100-immunopositive glial cells (including satellitosis; minus) and increased
AN
protein accumulation next to a cerebral vessel wall (double arrow) and, to a lesser extent, even
intravascular (arrowhead) after traumatic impact (anti-S100, scale bar: 50 µm). H – White matter
evincing S100 reactions for glial cells but without staining of vessels (single arrow) (anti-S100, scale
M

bar: 50 µm).
D

Figure 2: Post-mortem NSE values. Boxplots of the results in serum (upper diagram) and CSF
samples (lower diagram) separated by a dashed line into different causes of death (shown left in white
TE

boxes compared to the dark grey box for traumatic brain injury cases) and the trauma survival times
(shown right in light grey boxes).
EP

Legend: NSE, neuron specific enolase; CSF, cerebrospinal fluid; AMI, acute myocardial infarction;
ITT, isolated torso trauma; DCH, diffuse cerebral hypoxia. TBI, traumatic brain injury; aTBI, acute
death after traumatic brain injury (survival < 2h); sTBI, subacute death after traumatic brain injury
C

(survival 2h - 3d); dTBI, delayed death after traumatic brain injury (survival > 3d).
AC

**, p<0.001 after adjustment; *, p<0.05 after adjustment.

Figure 3: Post-mortem S100B values. Boxplots of the results in serum (upper diagram) and CSF
samples (lower diagram) separated by a dashed line into different causes of death (shown left in white
boxes compared to the dark grey box for traumatic brain injury cases) and the trauma survival times
(shown right in light grey boxes).
Legend: S100B, S100 calcium-binding protein B; CSF, cerebrospinal fluid; AMI, acute myocardial
infarction; ITT, isolated torso trauma; DCH, diffuse cerebral hypoxia. TBI, traumatic brain injury;
20
 ACCEPTED MANUSCRIPT
aTBI, acute death after traumatic brain injury (survival < 2h); sTBI, subacute death after traumatic
brain injury (survival 2h - 3d); dTBI, delayed death after traumatic brain injury (survival > 3d).
**, p<0.001 after adjustment; *, p<0.05 after adjustment.

PT
RI
U SC
AN
M
D
TE
C EP
AC

21
 ACCEPTED MANUSCRIPT

Median age Male gender Final resuscitation Median PMI Survival time range Circumstances of death (percentage)
in y (IQR) percentage attempts in h (IQR)
yes / no / positive
percentage

PT
Acute myocardial infarction (AMI)

RI
n=14 75 (25) 71.4% 10 / 4 / 71.4% 57 (19) ‘0’ Internal cause (100%)

Diffuse cerebral hypoxia (DCH)

SC
n=15 53 (30) 66.7% 6 / 9 / 40.0% 53 (33) ‘0’ Suicide (60.0%), brawl (13.3%), internal cause (13.3%),
traffic accident (6.7%), iatrogenic (6.7%)

U
AN
Isolated torso trauma (ITT)

n=18 51 (49) 55.6% 12 / 6 / 66.7% 21 (55) 5min – 13h Brawl (n=44.4%), Traffic accident (33.3%), iatrogenic

M
(11.1%), downfalls (5.6%), suicide (5.6%)

Traumatic brain injury with acute death (aTBI), survival time <2h

D
n=23 56 (34) 82.6% 13 / 10 / 56.5% 49 (39) few sec – 1h 47min Traffic accident (87.0%), downfalls (13.0%)

TE
Traumatic brain injury with subacute death (sTBI), survival time 2h - 3d
EP
n=11 62 (23) 81.8% 4 / 7 / 36.4% 84 (86) 2.33h – 43.5h Downfalls (63.6%), traffic accident (36.4%)

Traumatic brain injury with delayed death (dTBI), survival time >3d
C
AC

n=11 57 (46) 100.0% 5 / 6 / 45.5% 87 (83) 72h – 19d Traffic accident (27.3%), downfalls (9.1%), brawl (9.1%)

Table 1: Summarized characteristics of the investigated fatalities.

Legend: IQR, interquartile range; PMI, postmortem interval; min, minutes; h, hours; d, days; y, years.
22
 ACCEPTED MANUSCRIPT

Acute myocardial Diffuse cerebral Isolated torso TBI with acute TBI with subacute TBI with delayed

PT
infarction hypoxia trauma death death death
Median CSF NSE 1346.5 (659.63) 682 (514) 284 (304.3) 2479 (1785.5) 5776 (2930.3) 2313 (2995.8)

RI
in ng/ml (QD)

SC
Median CSF S100B 4308 (1418.4) 3247 (1655) 1890.5 (1347.8) 5946 (1947.4) 12372 (3246) 8320 (4149.5)
in ng/ml (QD)

U
Median serum NSE 480 (196.1) 307.1 (196.1) 166.8 (77.6) 364.7 (264.5) 547 (172.3) 658.9 (192.1)
in ng/ml (QD)

AN
Median serum S100B 99 (60.4) 281.5 (183.2) 51.7 (53.5) 132 (166.9) 347 (165) 211.1 (125.6)
in ng/ml (QD)

M
D
Table 2: Numerical presentation of median levels and quartile deviation (QD) for all cases and control groups investigated. Traumatic brain injury cases were

TE
listed according to their trauma survival time.
Legend: CSF, cerebrospinal fluid; NSE, neuron specific enolase; QD, Quartile Distribution; S100B, S100 calcium binding protein beta; TBI, traumatic brain
EP
injury.
C
AC

23
 ACCEPTED MANUSCRIPT

TBI without cerebral tissue damage TBI with cerebral tissue damage (Adjusted) p-value
n acute / subacute / delayed 10 / 5 / 2 13 / 6 / 9 0.303
Median brain weight in g 1480 (135) 1410 (130) 0.459

PT
(QD)
Median CSF levels of NSE 1429 (1446) 3452.5 (2919.4) 0.886

RI
in ng/ml (QD)

SC
Median CSF levels of S100B 6634 (3201) 7451 (3385.4) 0.789
in ng/ml (QD)

U
Median serum levels of NSE 615.8 (307.9) 480.3 (221,2) 0.992

AN
in ng/ml (QD)
Median serum levels of S100B 347 (168.1) 163.7 (131.6) 0.237

M
in ng/ml (QD)

D
Table 3: Comparison between traumatic brain injury (TBI) cases with and without cerebral tissue damage.

TE
The statistical tests used were the Chi-square test for distribution of survival time categories and the Mann-Whitney U test for marker levels. The values have
been adjusted using the Benjamini-Hochberg procedure.
EP
Legend: TBI, traumatic brain injury; CSF, cerebrospinal fluid; NSE, neuron specific enolase; QD, Quartile Distribution; S100B, S100 calcium binding protein
beta.
C
AC

24
 ACCEPTED MANUSCRIPT
Acknowledgement:
The authors would like to thank Mr. Philip Saunders (Language Support Services, Berlin,
Germany) for correcting the English language of the manuscript. In addition the authors
sincerely thank both anonymous reviewers for their critical comments on an earlier version of
this paper.
This paper is dedicated to the little daughter of BO, which was born on the day of the first

PT
submission.

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
Highlights

• Post-mortem CSF NSE and S100B levels were elevated in TBI compared to controls
• The CSF levels already differ between acute dying TBI cases vs. controls
• The CSF levels of the biomarkers increase with longer trauma survival times
• Investigating CSF samples with short PMIs is inevitable for appropriate results
• Post-mortem serum levels of NSE and S100B were not elevated in TBI cases

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Conflict of interest: This work was partly supported by the German Ministry of Defense.

PT
RI
U SC
AN
M
D
TE
C EP
AC