Master of Forensic Science
University of Amsterdam
Study of the alterations of microvascularization in
traumatic brain injury and its use in forensic
medicine
Katrin Siri Niedermaier
11865989Supervisor: Marisa Ortega Examiner: Roelof-Jan Oostra January 2019 – July 2019 36 ECTS
Acknowledgments
To begin with, I would like to sincerely thank my supervisor Marisa Ortega for offering me the opportunity to not only work for and write my master thesis on such an interesting topic, but also for her constant support, encouragement and incredible uplifting energy.
Together with my co-supervisor Ignasi Galtés, they built an absolute ideal working environment, where I was able to learn much more than I’d anticipated and gained a great deal of knowledge also offside my actual topic, which made my future academic path much clearer to me.
Thank you both for sharing all your knowledge, curiosity and positivity with me. I would also like to thank Alfonso Rodríguez-Baeza and Santiago Rojas for being so welcoming and for their advice concerning the execution of my work.
Additionally, I am extremely grateful to Anna Garrit for her immense support in the laboratory and her clear organization and planning, as well as to Montserrat Arango Rodríguez, Carmen García Maza, and Rut Maria Rojas Soria, without whom I would have never been able to realize the processing of all my samples.
Another thank you goes to my examiner Prof. Dr. Oostra, who has been a support from afar and made sure that my research was on the right track.
Not to forget, thank you to my family who has always supported me, even being seperated over countries and to my friends who made this time in Barcelona incredible.
Lastly, thank you Michelle Winter Buchwalder for sharing all the new experiences in the autopsies with me and for making every day in the office feel like no work at all.
Table of Contents
Abstract.……… Keywords………. Introduction………. Material & Methods……….
Sample collection……….. Sample preparation……….. Stainings………. Imaging……… Measurements……… Statistical analysis………. Results………
Characteristics of the study subjects………. Description of lesions……… Observing the conservation of brain tissue………
General postmortem tissue alterations……… Description of alterations of brain tissue in TBI………
Neuron density……… Microvessel analysis……… Aquaporin 4 expression analysis………. Discussion……….
Presence of postmortem phenomenons in human brain tissue samples………. Perilesion is part of Lesion……… Alterations of microvascularization……… Edema affects the brain as a whole………. Aquaporin as a marker for ischemia/anoxia……… Future research……… Conclusion………. References………. Supplementary materials……….. List of abbreviations Images Recipes Statistical data 2 2 3 7 7 9 9 10 11 12 13 13 13 16 16 19 19 24 27 29 29 30 30 31 32 33 34 35 38
Abstract
The morbidity of cerebral trauma is determined by the appearance of secondary lesions: ischemia and endocranial hypertension. These lesions appear due to alterations of encephalic microvascularization. The main objective of this work is the study of the alterations that occur in encephalic microvascularization in traumatic brain injury (TBI). From the forensic point of view, the objective has been to assess the role that AQP4 plays as one of the possible markers in TBI, which could be of help in medical forensic studies when lesions are not evident.
Samples of cerebral cortex of 5 deceased individuals without apparent CNS pathology (control group) and of 5 individuals deceased due to severe TBI (TBI group) have been studied. The study has been based on immunolabels for Collagen IV, type 4 aquaporins (AQP4) as well as neuronal Nissl-stains. The samples were visualized by optical microscopy.
The results obtained showed that the study of the areas of injury and perilesion in the TBI group present macroscopic differences which do not correspond to the microscopic study, since there are no significant differences in neuronal or microvascular loss. These findings determine that in TBI microscopic studies should be performed since at macroscopic level lesions can go unnoticed.
The TBI group has presented a decrease in neuron density, as well as a significant decrease in its vascular (capillary) density, both at level of lesion, perilesional focus and in areas at distance, a fact that we correlate with the appearance of secondary lesions such as cerebral edema and ischemia.
We have observed that in the TBI group, capillaries show corrugation phenomena, as well as breakage of intervascular bridges (IBs). AQP4-staining has shown that the TBI group presents an activation of astrocyte feet at distance from the primary lesion, which was also activated early, in the first hours of trauma.
This phenomenon was observed in an individual of the control group where death occurred through mechanical asphyxiation, with the consequent formation of cerebral ischemia and edema. Therefore, we correlate that the activation of AQP4 is not specific to TBI but that these are activated when cerebral edema is present. All the findings observed in the TBI group can be correlated with a disruption of the cerebral parenchyma angioarchitecture, which facilitates tissue ischemia. However, we cannot assure that the activation of AQP4 is a specific TBI-marker and, thus, its value in forensic medicine.
Keywords
Introduction
The human brain is one of the most complex anatomical structures. An injury to the brain can have significant influences on the functions of the whole organism or directly/indirectly lead to its death. Though usually being protected by the skull, many different types of injuries can affect the brain1,2.
Traumatic Brain Injury (TBI) is one of the leading causes of death worldwide3,4. It
contributes to approximately 30% of injury-related deaths5. Following primary trauma
at the contusion site, a secondary auto-destructive insult appears in many patients, leading to serious damage to the central nervous system (CNS)1. These secondary
pathophysiological conditions can be extremely heterogeneous between individuals, showing patterns such as inflammatory responses, disruption of the blood-brain-barrier, excitotoxicity, oxidative stress, astrogliosis, and microglial activation6. Other
important factors that influence the chances of survival of a person are cerebral ischemia and intracranial hypertension3.
In patients with severe TBI (determined by the Glasgow coma scale score < 8), there are primary lesions that occur as a result of the impact itself, and secondary lesions triggered after a variable period of time from the accident, characterized by its progression and determining role in the evolution of these patients, in terms of morbidity or mortality7. Its high mortality rate, prolonged hospitalizations and the
resulting serious consequences, make TBI one of the most important current socio-sanitary problems of our society.
The negative effects of brain trauma are determined not only by the entity of the primary lesion resulting from a direct or indirect biomechanical force on brain matter. Secondary events, such as formation of edema, brain swelling, increase intracranial pressure, changes in cerebral blood flow associated with hypoxia, neuroinflammation, oxidative stress, excitotoxicity, and apoptosis, may occur with some delay and they are the ones that will determine the survival or death of the patient. Regardless of the cause of brain injury, brain edema almost always occurs and is heavily involved in the pathophysiology of brain damage following traumatic injury, being a central part of the vicious cycle of injury in TBI and the major determinant of patient prognosis8.
In the last decade, there have been significant advances both in the clinical management of TBI and in the knowledge of its pathophysiology. The new morphological and pathophysiological discoveries have allowed individualizing and rationalizing therapeutic measures and have contributed to the improvement of the prognosis of these patients. Although there is an important development in neurochemical, histopathological and molecular techniques to study TBI in humans, we still do not have any specific “neuroprotective” therapy, and trauma continues to be one of the main causes of disability and death of young people in industrialized countries9–11.
In the forensic context, it is highly important for forensic pathologists to be able to determine what factors contributed to the death of a person. Sometimes a TBI can be very obvious where the skull is visibly deformed or where a hemorrhage can be observed when the skull is opened during the autopsy. In other cases, especially when the deceased survived for some time after sustaining the injury, a TBI might not be obvious at all. When the forensic pathologist is faced with these cases, he should have tools to be able to determine the type of brain injury that has caused the death of the individual, as well as to determine if survival was possible for a period of time. Our understanding of the pathological mechanisms of TBI has substantially increased. However, many in-depth studies are needed, since the complex cascade of physiological and biochemical mechanisms in TBI are only partially understood. 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-brain barrier function as so-called primary injury), secondary insults, such as increased intracranial pressure, brain edema, ischemia due to irregular cerebral blood flow and metabolic disturbances, oxidative stress or inflammatory reactions, are common following a TBI. Primary injuries on cellular levels include immediate neuronal and glial cellular damage (as necrosis), and axonal ruptures (as axonal injury)12,13.
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 resulting from the trauma14 and to differentiate between
polytraumatized cadavers with and without head injuries15.
The scientific community is currently on the line of trying to elucidate which markers are specific for brain damage, but postmortem biochemistry of trauma cases is still in an early phase of its investigation and, thus, reasonable values have not yet been established to determine such injuries. Moreover, no single biomarker has so far shown to be both sensitive and specific for detection of brain damage in everyday clinical practice16. Therefore, we believe that current research should be first directed
to determine which is the specific cerebral marker to asses traumatic brain injuries, and subsequently test if these can be detected in different body fluids such as blood, saliva, etc.
Some further promising markers of different cell forms (e.g. neurons, glial cells) that have been clinically researched in TBI cases include the glial fibrillary acidic protein, brain-derived neurotrophic factor, tau proteins, ubiquitin carboxy-terminal hydrolase L1, neurofilaments and alpha-II-spectrin breakdown products17.
The cerebral blood capillaries have a basic histological structure, consisting of endothelial cells, pericytes and basement membrane. All these elements are part of the so-called blood-brain barrier (BBB).
BBB consists of a monolayer of endothelial cells that covers the surface of capillary light and restricts the movement of small polar molecules and macromolecules
between blood and cerebral interstitial fluid18. This endothelial barrier is
complemented by capillary pericytes, which share the basement membrane with endothelial cells. Finally, there are the foot processes of astrocytes, which surround approximately 95% of the abluminal surface of the perivascular basement membrane19.
Brain edema is associated with TBI and denotes the swelling of the brain through excess accumulation of fluids. Water transport across microvessels is a main cause of brain edema20. There is growing evidence that members of aquaporins (small,
hydrophobic membrane proteins regarded as a ‘pure water channel family’ that facilitate water transport) play an important role in traumatic brain edema21,22. Of
these, aquaporin-4 (AQP4), the most abundant aquaporin in the brain23, is the
dominant contributor to the regulation of cerebral edema24. It is expressed at
astrocyte foot processes adjacent to vascular endothelial cells25 (Figure 1). Studies in
rats have shown that AQP4 is increased at the injury site after TBI26.
Moreover, cross-talk between astrocyte activation, AQP expression, brain inflammation, and hypoxia exists in brain response to trauma27,28 and this needs to
be further clarified to gain a better understanding of the roles played by several mediators in the regulation of ion/water homeostasis in brain contusive injury. To achieve this depth of knowledge, it is of the utmost importance to consider that, in the brain, a complex network of neurons acts as a functional structure interacting with glial cells (i.e., astrocytes), cerebral blood vessels, and endothelial cells, and as part of a single physiological entity, the neurovascular unit (NVU)29.
Collagen IV stains the basal membrane of microvessels. This staining is important to visualize the morphology of the microvessels.
Research on TBI has been widely conducted; much of it in animal models, other on human brain samples. Most of the studies, however, only concern the lesion site and the areas adjacent to it. This research project aims to study the alterations of brain morphology and function after being subjected to a traumatic injury in areas all around the brain, including areas far from the lesion site, concentrating on microvessels and certain proteins connected to TBI.
Figure 1: Astrocyte foot processes AQP4 is expressed at astrocyte foot processes. Many of these are in direct contact with vessels in the brain (circle), forming part of the BBB.
If histological differences between the TBI and control group are found, it would give knowledge about the effect of trauma to other areas far from the lesion site. This could aid in tackling these aggravating secondary responses to enhance the treatment for people that sustained a TBI. For legal purposes, it would be valuable to distinguish between possible causes of death, where not macroscopically visible, for example, Asphyxia. In some cases, prudence is necessary, to exclude that multiple assaults had happened. The research could also be of great help in deciding if the injury occurred peri- or post-mortem, so if it was related to the death of the person or an event that only happened after the person had already deceased. This would aid in the reconstruction of certain events around the circumstance of death, while also opening the possibility of estimating the age of the injury, giving a minimum amount of survival time.
The main objectives of this study are to determine the existence of morphological alterations in brain microvascularization in deceased individuals as a result of a TBI and if these alterations are specific. Another objective is to detect if differences in the activation pattern of AQP4 of the TBI group occur, and therefore, if this biomarker can be used to differentiate between different injuries and to determine if the changes in the expression of this biomarker could provide reliable information on the age of cortical contusions.
Material & Methods
Sample collectionThe project has been approved by the teaching and research committee of the Institute of Legal Medicine and Forensic Sciences of Catalonia – (IMLCFC) in order to comply with the ethical requirements. For this study, we have used samples of cerebral cortex tissue belonging to 10 human corpses. The samples were obtained during the mandatory medical-legal autopsy of the IMLCFC and according to the usual protocol consisting of the opening and evaluation of the three cavities: thorax, abdomen, and skull.
A criterion for sample selection was namely the execution of the autopsy no later than 24 hours after the person deceased to prevent autolysis of the tissue. TBI-samples had to be taken of persons that died due to a traumatic injury to the brain but showed a minimum survival time of two hours for the secondary injury response to commence. In case of samples of the control group, death had to be unrelated to any damage to the brain. Significant information about the individuals whose brains were collected for sampling can be seen in Table 1.
For sampling, the brain was divided into areas according to Brodmann’s map of cytoarchitectonics (Figure 2). Different areas were selected that constitute primary areas of the respective lobes, namely A312 & A4 - parietal lobe, A10 - frontal lobe, A17 - occipital lobe, A22 - temporal lobe, as well as the Hippocampus and in case of TBI brains macroscopic lesion and peri-lesion sites. In traumatic brains, each area was sampled on both hemispheres; whereas in control brains only one hemisphere was sampled since it was assumed that the right and left hemisphere would show equal signs of autolysis (p-value= 0,6999 -> no significant difference). Each sample was named after the individual it was taken from (OTx), the area it belonged to (Ax) and subsequently the hemisphere (“D”-derecho/right, “I”-izquierda/left). Dices of approximately 1x1cm were cut out of the brain with a scalpel and emerged in Zamboni solution, which was subsequently changed to 70% EtOH for storage in the fridge until carved and stained.
Figure 2: Brodmann’s map of cytoarchitectonics
Brodmann assigned numbers to various brain regions by analyzing each area's cellular structure starting from the central sulcus. Exterior view of left hemisphere [A]. Interior view of right hemisphere
Table 1: Individuals information Age Sex Brain
weight Date/time trauma Date/time death Survival time Date/time autopsy Cause of death Zones primary lesions Autopsy nr.
OC1 52 1440 g - 07/03/18 3 - 4 pm - 08/03/18 9:30 am Lung tumor - 537/18 OC2 29 1305 g - 16/12/18 ~ 5 pm - 17/12/18 10 am respiratory Severe insufficiency - 2566/18 OC3 62 1300 g - 09/01/19 ~ 4 pm - 10/01/19 11 am Adrenal glands insufficiency - 0082/19 OC4 26 1560 g - 20/01/19 3 – 4 pm - 21/01/19 12 am (Hanging) Asphyxia - 0184/19 OC5 83 1230 g - 22/01/19 ~ 11 am - 23/01/19 10 am Cardiac arrhythmia - 0197/19 OT1 59 1500 g 16/04/18 5 pm 17/04/18 11 am 18 h 18/04/18 10 am
Severe TCE frontal lesions, sub-arachnoid hematoma left frontoparietal hemisphere 820/18 OT2 14 1665g 10/06/18 3 pm 11/06/18 9:40 am 18:40 h 12/06/18 10 am Encephalic death fracture cranial base axonal lesion diffuse edema, hemorrhagic contusion in ganglia left
base- COS-mesencephalon
1186/18
OT3 29 1600 g 10/11/18
~ 9 am 10/11/18 ~ 3 pm 30 h 11/11/18 10 am herniation of HSA & amygdala
Multiple fractures parieto-temporal right, HAS diffuse,
left frontal base, left hippocampus 2281/18 OT4 63 1640 g 19/09/18 ~ 9 am 20/09/18 ~ 10 am 25 h 21/09/18 ~ 9 am Encephalic death membrane subdural left/HAS 1933/18 OT5 54 1520 g 26/01/19 11:40 pm 28/01/19 1 pm 25:20 h 29/01/19 11 am Encephalic death subdural hemorrhage right, parenchyma cerebral frontal right and temporal
right
Sample preparation
The collected samples were cut in half to be further processed for stainings. One half of each sample was embedded in paraffin. Therefore, the tissue was slowly deprived of water by emerging it in 70%, 80%, 90% EtOH (2x), absolute EtOH (2x) and Xylem (2x). Subsequently, the tissue was embedded into Paraffin and the blocks were cut into slices of 10µm thickness by the use of a microtome. These slices were further fixed onto glass slides and stained with Hematoxylin & Eosin (H&E) and Nissl.
The other half of each sample was directly cut with a vibratome into slices of 50µm thickness and further stained with antibodies anti-ColIV and anti-AQP4. To obtain better results of the AQP4-staining, it was later conducted on paraffin slides as well.
Stainings
H&E-staining (Table 2), as well as Nissl-staining (Table 3), were performed to characterize the condition of each samples tissue, including visible postmortem phenomenons. Immunostaining with anti-ColIV (Table 4) and anti-AQP4 (Table 5) was performed to visualize the structure of microvessels and the pattern of AQP4 expression.
Table 2: Hematoxylin & Eosin staining protocol
H&E-staining
Step Action Time
1 Xilol 2x 7 min
2 EtOH Absolut 5 min
3 EtOH 96% 5 min
4 EtOH 70% 5 min
5 H2O dest. 5 min
6 Hematoxylin 10 min
7 H2O (tap) 5 min
8 Acid EtOH (HCL + EtOH) 5 sec
9 H2O dest. 30 min
10 Eosin 30 sec
11 H2O dest. 5 sec
12 EtOH 70% 6 min
13 EtOH 96% 6 min
14 EtOH absolute 2x 4 min
15 Xilol 2x 5 min
Table 3: Nissl-staining protocol
Nissl-staining
Step Action Time
1 Rehydrate with dest. H2O -
2 Coloring solution 15 min
3 Wash with demineralized H2O -
Table 4: Collagen immunostaining protocol
Immunostaining Col IV
Step Action Time
1 Clean or hydrate the sample with PBS 2x 5 min
2 Blockage of endogenous Peroxidase H2O2 30 min
3 PBS+ 0,1 Triton 2x 5 min
4 Blockage of unspecific unions (horse serum) 20 min
5 Primary Antibody reaction 1 hr
6 PBS+ 0,1 Triton 3x 5 min
7 Secondary Antibody reaction 1hr
8 PBS- Triton 4x 5 min
9 Strept-Peroxidase in PBS 30 min
10 PBS 5 min
11 DAB 1 min
12 Dest. H2O
13 Hx Mayer 25% in demineralized H2O Put on and off
immediately
14 H2O current x3
15 DAM
16 -> fixation on glass slide
Table 5: Aquaporin immunostaining protocol
Immunostaining AQP IV
Step Action Time
1 Deparaffinate and hydrate the sample in dest. H2O
2 Wash in PBS
3 Blockage of endogenous Peroxidase H2O2 30 min
4 PBS+ 0,5 Triton 5 min
5 -> assemble cover plate
6 PBS 5 min
7 Blockage of unspecific unions (horse serum) 45 min
8 Primary Antibody reaction 1 hr
9 PBS+ 0,5 Triton 3x 5 min
10 Secondary Antibody reaction 1h
11 PBS- 0,5 Triton 3x 5 min
12 Strept-Peroxidase in PBS 45 min
13 PBS 2x 5 min
14 -> remove cover plate
15 DAB 2-25 min
16 Dest. H2O
17 DAM
Imaging
Images of all stainings were taken with a Leica DMD108 microscope (4x, 10x and 20x magnification) and analyzed by the use of the ImageJ and Angiotools software.
Measurements
Neuron density
Images captured from Nissl-stainings were used to calculate the neuron density of TBI-cases compared to controls by the ImageJ analyzing tools. Every image had the same size of 20x magnification. To exclude glial cells being counted as neurons their pixel areas were measured to be able to raise the threshold in ImageJ above them. The stained areas of the glial cells had a mean of 270 pixels, but other glial cells were found which areas included above 370 pixels. Therefore, the threshold for particle analysis was raised to 400-Infinity (pixels). After converting the original image type into an 8-bit and adjusting the threshold, only the cells were counted which pixel areas exceeded the selected threshold (Figure 3). In control samples, only the neurons of each area of one hemisphere were counted, as no difference between the two hemispheres was expected since the cause of death was unrelated to any brain damage. In traumatic samples, the areas of both hemispheres were analyzed.
Microvessel density
The images of the Collagen immunostaining were captured in 10x magnification and converted into black and white images by the use of the ImageJ software. Subsequently, those images were uploaded into the Angiotools software, which recognizes the structure of microvessels, marks them and analyzes the image (Figure 4). The calculated vessel percentage was used to compare the microvessel density in different areas and between control and trauma samples.
Figure 3: Neuron density measurement
Example of measurement of number of neurons in A312D of OC3; original image of Nissl-staining [A], image converted to 8-bit and threshold application [B], particle analysis [C].
Statistical analysis
For each variable, the number of observations (n), mean, standard deviation, standard error of mean, minimum and maximum values, and range were calculated. Comparisons between means of “Control” and “TBI” groups were determined through Mann-Whitney tests and between subgroups through Spearman correlation coefficient test; comparisons between the means of “Left” and “Right” hemisphere groups and subgroups were determined through unpaired student t-tests with Welch´s correction or Kruskal-Wallis with Dunn´s multiple comparison correction test in non-normally distributed variables; correlations between hemispheres was performed through Pearson correlation coefficient test. Comparisons between the means of the “Areas” groups and subgroups were performed through Kruskal-Wallis tests; correlation between “Areas” subgroups was realized through Spearman correlation coefficient test. “Lesion” and “Perilesion” groups mean comparisons was realized through unpaired student t-tests with Welch´s correction, and between subgroups in both variables, One-way ANOVA with Tukey correction for multiple comparisons was performed; Correlation between “Lesion” and “Perilesion” subgroups was performed through a two-tailed bivariate Pearson correlation coefficient test. In all cases, a confidence interval (CI) of 95% was established, and a p˂0.05 was considered to be statistically significant. Statistical analysis and graphics of the data obtained were performed using GraphPad Prism v8.00 (GraphPad Software, La Jolla, California USA) and R v3.53- R package version 2.5-3 (The R Foundation for Statistical Computing, Vienna, Austria).
Figure 4: Microvessel density measurement
Example of microvessel density measurement in A4D of OT1; original image of ColIV- immunostaining [A], image converted to 8-bit and threshold application [B], vessel analysis [C].
Results
Characteristics of the study subjects
The samples obtained of the cerebral cortex for this study derived from 10 subjects: 9 men and 1 woman. 5 subjects belonged to the control group and the other 5 to the TBI group. The ages ranged between from 26 to 83 years in the control group (mean age of 50,5 years) and from 14 to 63 years in the TBI group (mean age of 44 years). The weight of all brains of the TBI group exceeded 1500g, which is a first indication of the presence of cerebral edema.
Description of lesions
To be able to characterize the alterations both postmortem and between controls and traumatics special attention has to be paid to the location of the respective lesions. Control samples were taken of individuals that didn’t show any trauma to the head. The causes of death were lung tumor (OC1), severe respiratory insufficiency (OC2), insufficiency of adrenal glands (OC3), asphyxiation (OC4), cardiac arrhythmia (OC5). The individuals of whom the traumatic samples were taken, however, showed various sites and types of lesions to the brain (Figure 5). For analysis, these had to be categorized as either diffuse or focal injuries. Where multiple lesion sites were present, the most severe one was sampled as the ‘’lesion’. All traumatic brains presented brain swelling (edema), where the brain exceeded a weight of 1500g.
• Focal injuries
Two sample groups were categorized as containing focal injuries, where clear regions could be attributed to having sustained contusions.
OT1 showed two frontal lesions and subarachnoid hematoma on the left frontoparietal hemisphere, with its primary lesions located around A10D and A10I. The individual survived 18 h after sustaining the injury.
The brain of OT2 presented hemorrhage in the ganglia of the left base, as well as in the corpus callosum and mesencephalon. The right hippocampus (HIPOD) was located closest to the lesion, thereby forming part of the macroscopic perilesion. The survival time of OT2 was around 18,5 hours.
• Diffuse injuries
The other three sample groups were categorized as containing diffuse injuries, where no clear region or multiple different regions were affected by the injury.
Multiple cranial fractures of OT3 led to diffuse subarachnoid hemorrhage and lesions in the left frontal base and the left hippocampus. Therefore, A10I and HIPOI can both be attributed to the lesion site. The individual survived these lesions for 30 hours.
OT4 showed hemorrhage in the left subdural membrane and zones of multiple contusions, namely A312D, A312I, A4D, A4I, A22I, HIPOI. The survival time was 25 hours.
OT5 presented subdural hemorrhage on the right hemisphere and general subarachnoid hemorrhage and lesions in the cerebral parenchyma of the right frontal and temporal areas. The only areas not affected by the lesions seemed to be A312I and A4I. The individual survived for around 25,5 hours.
All brains of the traumatic group and their respective affected areas can be seen in Figure 5 on the following page.
OT1
OT2
OT3
OT4
OT5
Figure 5: Images of the lesions of the traumatic brain samples
OT1- frontal lesions and subarachnoid hematoma on the left frontoparietal hemisphere[A]; OT2- Lesion axonal grd. 2, diffuse cerebral edema, hemorrhagic contusion in ganglia of left base-COS-mesencephalon [B]; OT3- Multiple fractures parietotemporal right, HAS diffuse, left frontal base, left hippocampus [C]; OT4- membrane subdural left/HAS [D]; OT5- subdural hemorrhage right, parenchyma
L R R L
L R R L
L R R L L R R L L R R L
Observing the conservation of brain tissue
The samples of the deceased individuals of the control and traumatic group were all collected between 19 hr and 23 hr after the person’s death. Therefore, it is important to characterize which histological findings on the respective tissues were most likely due to the general commencing putrefaction and which ones were seen in the TBI group only. Histological analysis of the cerebral tissue was performed by H&E-staining.
General postmortem tissue alterations
In all sampled individuals phenomenons were found that account for postmortem alterations, which were expected in brain tissues derived from autopsies.
Both groups (controls & traumatics) showed vacuolization around neurons, as well as around vessels in some samples. Many neurons presented elevated eosinophilia.
• Vacuolization
Vacuolization, the formation of vacuoles around neurons, was present in control samples as well as in traumatic samples.
In the control group, it was found in different areas. OC1 showed the least of it, whereas the other controls (OC2-OC5) showed mild to strong vacuolization in certain areas that differed between the samples.
Most areas of the traumatic group showed strong vacuolization around the neurons. No direct connection could be drawn between the lesion site and areas distant to it. It was evident that more areas of traumatic samples with diffuse injuries showed vacuolization than those with focal injuries. However, the vacuolization of samples with focal injuries seemed to be distributed randomly between the areas.
Examples of vacuolization in brain tissues of controls and traumatics are displayed in Figure 6.
• Eosinophilic neurons
Similar to the phenomenon of vacuolization, neuronal cell death was observed in both control and traumatic sample groups. When a neuron is injured it shrinks in size and becomes eosinophilic, expressed through intense red color. In the process of dying their nuclei become pyknotic, which is visible through being darkly stained. Pyknosis is defined as the ‘irreversible condensation of chromatin in the nucleus of a cell undergoing necrosis or apoptosis’30. The mentioned injury can be directly caused
through trauma to the brain or through hypoxia or ischemia. Figure 7 illustrates these phenomenons.
Figure 6: H&E-staining
Visualization of cerebral brain tissue of the control group [A; C] and the traumatic group [B; D] (20x magnification). In both groups, vacuolization can be seen around the neurons without significant differences.
OT4 A10D
OC3 HIPOD OT5 HIPOI
OC2 A10I
A
OT4 A312D
Figure 7: Eosinophilia
Human brain tissue that is derived from autopsies shows neurons in different health states. Injured neurons show eosinophilia (arrow), whereas some neurons already shrunk completely, presenting pyknotic nuclei (triangles).
• Hemorrhage
Blood invasion in cerebral tissue was solely present in samples of the traumatic group. Thus, it cannot be accounted to as a postmortem phenomenon but as a perimortem process that was induced by the traumatic injury. The traumatic brains of the sampled individuals displayed hemorrhage in various areas, each of them including the respective lesion site (Figure 8). Figure 9 illustrates subarachnoid hemorrhage, which was present in OT1, OT3, and OT4.
The following areas contained blood in the cerebral tissue: OT1 – Lesion
OT2 – A4I, A10I, A22I, HIPOD, HIPOI, Lesion OT3 – A22D, Lesion, Perilesion
OT4 – A17I, HIPOD, Lesion, Perilesion OT5 – A22I, HIPOD, Lesion
OT1 Lesion A OT2 Lesion B OT3 Lesion C OT4 Lesion D OT5 Lesion E
Figure 8: Hemorrhage at the lesion site of the traumatic brains
Blood was observed in all traumatic brain samples at the lesion site. Lesion OT1 (4x) [A], lesion OT2 (4x) [B], lesion OT3 (20x) [C], lesion OT4 (20x) [D], lesion OT5 (4x) [E].
Figure 9: Subarachnoid hemorrhage
Blood invasion visible in the subarachnoid space of OT1, OT3 and OT4.
The cortex is coated by the pia mater, which makes up the innermost meninges of the Arachnoid mater
Pia mater Cortex
19 Description of alterations of brain tissue in TBI
Neuron density
To be able to measure the neuron density of the sampled individuals the brain slices were stained with Nissl. This staining allows for analysis of cellular patterns of certain brain areas. Additionally, it allows for a better analysis of the cells, as the background is lower than with H&E-staining.
• Neuron morphology
Microscopically it was already visible that in certain areas the original shape of the neurons was lost, their size was reduced, and that the overall neuron count had decreased when comparing trauma to control (Figure 10).
OC5 A22D A OT4 A22D B OC5 A4D C OT4 A4D D OC3 A10D E OT4 A10D F
Figure 10: Nissl-staining images of neurons
Visualization of neurons of the cerebral cortex in control [A; C; E] and traumatic brains [B; D; F]. Some neurons of the control group show dark staining while maintaining their structure. Compared to controls, the traumatic group shows shrunken neurons that are reduced in size (20x) [B; D]. Overview of areas of the
• Density measurement
Since a traumatic injury to the brain can lead to loss of neuronal mass, the density of neurons in each area was measured through cell counting of multiple images with 20x magnification. Regarding the control group, it was expected that both hemispheres would show similar neuron densities between the respective areas, thus, only one hemisphere was analyzed in this group per individual. To randomize the analysis, the hemisphere that was focused on was alternated between each control sample. Lesion sites in the traumatic group varied strongly, therefore both hemispheres were analyzed.
Overall, it is evident that the neuron density does not show a marked trend between the control and traumatic group. Still, most controls contain slightly more neurons per area than the traumatics. The statistical analysis shows that the neuron count in the different respective areas in both control and trauma group is statistically significant. Although the difference is lesser in the trauma group (Table 6 & 7). The p-value for the control areas is 0,0268 and the one for the areas of the trauma group is 0,0454. The comparison between the neuron density of all samples and their including areas can be seen in Figure 11. A comparison between the areas of the control group and the trauma group showed a p-value of 0,0033 and therefore the difference is statistically significant.
Control neuron means: A312: ~ 217 [220] A4: ~ 206 [200] A17: ~ 398 [400] A22: ~ 283 [290] HIPO: ~ 211 [210]
Figure 11: Neuron density measurement
Selected areas of the samples plotted against the number of neurons.
The control group shows a slightly higher neuron count in most areas compared to the traumatic group. 0 50 100 150 200 250 300 350 400 450 500 OC 1 OC 2 OC 3 OC 4 OC 5 OT 1 OT 2 OT 3 OT 4 OT 5
Table 6: Statistical data of neuron count per area of Control group
CONTROL A312D A4D A10D A17D A22D HIPOD A312I A4I A10I A17I A22I HIPOI
Number of values 3 3 3 3 2 3 2 2 2 2 2 2 Minimum 167 170 220 389 309 194,0 227 202 167 278 228 200 Maximum 251 238 341 497 314 224 247 203 186 404 280 239 Range 84 68 121 108 5 30 20 1 19 126 52 39 Mean 203 208,3 276,7 436,7 311,5 205,3 237 202,5 176,5 341 254 219,5 Std. Deviation 43,27 34,82 60,86 55,1 3,526 16,29 14,14 0,707 13,44 89,1 36,77 27,58 Std. Error of Mean 24,98 20,1 35,14 31,81 2,5 9,404 10 0,5 9,5 63 26 19,5 Kruskal-Wallis test p-value = 0,0286
Table 7: Statistical data of neuron count per area of TBI group
TBI A312D A4D A10D A17D A22D HIPOD A312I A4I A10I A17I A22I HIPOI
Number of values 5 5 5 5 5 5 5 5 5 5 5 5 Minimum 142 135 107 211 139 101 114 134 114 168 168 146 Maximum 252 209 204 422 247 268 292 273 225 375 234 247 Range 110 74 97 211 108 167 178 139 111 207 66 101 Mean 179 180,4 147,8 306,6 187,6 193 212 201,6 158,8 276,8 199,2 177 Std. Deviation 43,45 27,62 38,35 91,24 45,01 62,89 70,31 58,35 45,96 79,69 23,53 42,98 Std. Error of Mean 19,43 12,35 17,15 40,80 20,13 28,12 31,44 26,09 20,56 35,64 10,52 19,22 Kruskal-Wallis test p-value = 0,0454
Figure 12: Neuron density comparison
The boxplot shows a larger distribution in the TBI samples. The neuron count of the control samples per area seems to be more similar to one another.
A comparison of the total neuron density between control and TBI was conducted. The result can be seen in Table 8 and Figure 13. Overall, this comparison showed a significant difference between both groups with a p-value of 0,0009.
Table 8: Statistical data for total comparison of Control & TBI
Control TBI Number of values 29 60 Minimum 167 101 Maximum 497 422 Range 330 321 Mean 257,7 201,7 Std. Deviation 83,46 67,31 Std. Error of Mean 15,50 8,690
Mann Whitney test p-value = 0,0009
Sample collection of the lesion and perilesion site are very subjective since the collection is based on the scientist’s microscopical judgment of what appears to be the area with most damage and where approximately the surrounding region starts. Therefore, neuron count in the lesion and perilesion site were compared to each other to gain knowledge about differences or similarities. Table 9 and Figure 14 show the outcome of the statistical analysis. The p-value is high (0,3095) and therefore the neuron count can be seen as not having a statistically significant difference.
Figure 13: Total neuron density comparison
Both the chart and the boxplot exemplify the greater neuron density in the control group compared to lesser neuron density in the TBI group.
Table 9: Comparison of neuron count in lesion and perilesion Lesion Perilesion Number of values 5 5 Minimum 64 101 Maximum 275 365 Range 211 264 Mean 141,2 213,8 Std. Deviation 81,05 96,63 Std. Error of Mean 36,25 43,22
Mann Whitney test p-value = 0,3095
Figure 14: Neuron count comparison lesion and perilesion
The graph shows a higher neuron count in the perilesion area, while it becomes apparent in the boxplot that the range largely overlaps. Therefore, the significance cannot be attributed to as statistically significant.
Microvessel analysis
• Microvessel morphology
Immunostaining of the samples with anti-Collagen IV (ColIV) allowed for detailed analysis of the microvessel structure. This antibody stains collagen that forms the skeleton of the basal membrane of the capillaries. The basal membrane is an extracellular matrix that separates endothelial cells and pericytes of the extracellular space. All images were captured at 10x magnification. Morphological findings are shown in Figure 15.
In the control group, the vessels seemed more homogenously distributed; in some areas of the traumatic group the distribution seemed more random. Both groups showed many microvessels with a smooth and intact membrane, but also corrugation of the basal membrane could be observed. Corrugation seemed to appear more often in the traumatic group compared to the controls. Intervascular bridges (they were presented as filaments that united different vascular structures, often capillaries, although it was not unusual to find them joining vessels of greater caliber, such as arterioles and venules) were abundant in the control group, most of them intact, some of them broken, whereas the traumatic group showed significantly less intervascular bridges and most of the ones that could be observed were broken. In previous studies the appearance of microaneurysms is described, however, those could not be found in our samples.
A B
C D
OC1 A312D OT1 A10I
OT1 A10D
OT1 A10I OT3 A4D
Figure 15: Microvessel morphology
The same morphological features were observed in the control and trauma group. Overall, the control group presented more healthier looking and homogenously distributed microvessels,
A B
D C
• Microvessel density
Angiotools was used to measure the density of the microvessels in the cortex and white mater of all areas of traumatic and control samples. Images of 10x magnification were used. Both comparisons clearly showed a higher degree of microvascularization of the cortex compared to the white mater. In the same time, the control group showed a higher degree of microvascularization in cortex and white mater when compared to the traumatic group (Figure 16, Table 8 & 9). The cortex of the controls had a mean microvessel percentage area of 28,3%, whereas the traumatics showed less with 24,6%. Similarly, the white mater of the controls had a mean microvessel percentage area of 11,5% and the white mater of the traumatics showed only 9,5%.
Table 8: Statistical data of cortex comparison
Control Cortex Trauma Cortex
Number of values 57 137 Minimum 15,26 14,83 Maximum 41,53 36,50 Range 26,27 21,67 Mean 28,31 24,64 Standard Deviation 6,395 4,266
Standard Error of Mean 0,8470 0,3645
Mann Whitney Test p-value < 0,0001 Figure 16: Microvessel density measurement
Comparison of microvascularization of the cortex of all selected areas between control and trauma samples [A]; and comparison of the microvascularization of the white mater of all selected areas between control and trauma [B]. Both comparisons show a higher microvessel density in the control samples.
Control Trauma
Table 9: Statistical data of white substance comparison Control WS Trauma WS Number of values 35 87 Minimum 6,417 4,280 Maximum 20,33 31,56 Range 13,92 27,28 Mean 11,48 9,465 Standard Deviation 3,071 3,403
Standard Error of Mean 0,5191 0,3648
Mann Whitney Test p-value < 0,0001
Again, sample collection of the lesion and perilesion site are very subjective since the collection is based on the scientist’s microscopical judgment of what appears to be the area with most damage and where approximately the surrounding region starts. Therefore, the percentage vessel area of the lesion and perilesion site of the traumatic samples was compared to each other. The results (Figure 17; Table 10) show that indeed the difference is only minor and even though the areas appear to be different macroscopically, microscopically the perilesion site seems to be part of the actual sustained lesion site.
Table 10: Statistical data of microvessel area comparison between Lesion and Perilesion site Lesion Cortex [%] Perilesion Cortex [%] Lesion WS [%] Perilesion WS [%] Number of values 8 9 5 7 Minimum 16,55 19,53 6,670 6,141 Maximum 29,31 29,38 10,68 12,20 Range 12,75 9,852 4,012 6,062 Mean 21,86 24,0 8,151 8,700 Std. Deviation 4,409 3,419 1,877 2,362 Std. Error of Mean 1,559 1,140 0,8394 0,8927 Figure 17: Microvessel density comparison
Vessel percentage area of the perilesion site shows slightly higher values than the lesion site. However, the differences are minor and cannot be seen as statistically significant. The perilesion site appears to be part of the microscopic lesion site.
27 Aquaporin 4 expression analysis
AQP4 expression was analyzed in 4 samples: OC1, OC4, OT1, and OT5. Mostly, astrocytes were marked since AQP4 is expressed at their foot processes. The control sample OC1 showed a homogenous distribution of astrocytes in the cortex and white mater. In the traumatic sample OT1, however, the astrocytes were distributed heterogeneously. More astrocytes were found in the cortex of both control and trauma. They appeared to be bigger than the astrocytes in OC1 and showed significantly more foot processes. In the traumatic samples, astrocytes were often located around microvessels and seemed to occur in groups. An exception to the described findings was control sample OC4; the asphyxiation (Figure 20). The AQP4 expression through the astrocytes was much more similar to the expression in the traumatic cases. Very big and distinct astrocytes could be found that were mostly located in groups and in the cortex. Images of astrocytes in the cortex and the white mater can be seen in Figure 18 and Figure 19 respectively.
OC1_A312D OC1_A312D
OT1_A17I OT1_A17I
OT5_Lesion OT5_PERI
Figure 18: AQP4 expression in the cortex
Homogenous distribution of astrocytes in cortex of OC1 (10x) [A] (20x) [B]. Heterogenous astrocyte distribution in cortex of OT1 (10X) [C], (20x) [D]. Heterogenous distribution in
A B
D C
OC1_COS OT1_COS OT5_COS Figure 19: AQP4 expression in the white substance
Distribution of fewer astrocytes in corpus callosum compared to cortex of OC1 [A]. Astrocyte in corpus callosum of OT1 [B] and OT5 [C] showing attachment to microvessels. A B C OC4_A312 I OC4_COS OC4_HIPOI OC4_A10I
Figure 20: AQP4 expression in cortex & white substance of OC4
OC4 showed an AQP4 expression pattern much more similar to the traumatic samples. Astrocytes surround microvessels [A] & [D]. The size of them is increased and they possess many foot processes [B] & [C].
A B
Discussion
This conducted study has two important aspects in terms of forensic medicine; the first being that the institutes of legal and forensic medicine are the ones that perform the medical-legal autopsies of dead individuals due to a traumatic brain injury. Thus, they are the ones who can contribute new data with their research both at the epidemiological level and at the physiopathological level. These data are of great interest for clinical care, due to the great morbidity and mortality that TBI presents in the population. Second, this study attempted to evaluate the postmortem changes induced by TBI to the human brain and intended to characterize microvessel morphology and evaluate AQP4 expression. It is important to be able to determine, in the field of forensic medicine, if lesions are produced antemortem or postmortem. For this, the best specific markers of TBI must be determined. Bedsides the pathological study of the alterations of microvascularization produced in TBI, we can determine a characteristic pattern of the type of lesion3,31.
The number of samples in this study is limited since it is not easy to obtain these samples in humans. Even so, we have been able to study in a short period of time 5 TBI samples and 5 controls. However, it can give a general idea about the extension of trauma to the brain including different types of injuries. It was not possible to draw clear paths of how far the lesions extend to since most of the TBIs consisted of diffuse injuries where multiple contusion sides or widespread subarachnoid hemorrhage were present.
Presence of postmortem phenomenons in human brain tissue samples
It is evident that, when working with human brain samples obtained during a necropsy, there are structural changes characteristic of the phase of suffering from brain death in many patients, to which the post-mortem time is added until the extraction and the tissue fixation. These changes are inevitable and therefore inherent in these types of studies. In this context it is necessary to consider that in the postmortem period there are morphostructural alterations of the tissues that must be considered in the interpretation of the results, this having the ability to differentiate these alterations from those produced by the pathological process ‘per se’.
To make a general assessment of these changes, we can resort to classical histological techniques, such as staining the tissue with H&E and study using conventional optical microscopy. In samples of human cerebral cortex obtained in necropsies and studied with this methodology, these changes, amongst others, have been evidenced through high eosinophilia of the neurons, the condensation of its nucleus, the vacuolization of the tissue, the contraction of cell bodies (especially pyramidal neurons)32.
Our study, based on samples of human cerebral cortex of deceased patients, had allowed us to have control samples and TBI samples, trying to discern with different methodologies the TBI alterations with respect to controls.
Vacuolization was observed in the control samples but to an even stronger degree in the TBI group, the traumatic impact seems to have deteriorated the tissue. Nevertheless, the areas that showed most vacuolization did not present a direct correlation concerning the distance to the actual lesion site.
The observed hemorrhage in the traumatic samples could not be found in any control samples, therefore, it cannot be attributed to as a postmortem phenomenon but as being the result of the traumatic impact connected to the TBI individuals’ deaths.
Perilesion is part of Lesion
In the process of sample collection, the most obvious contusion side was sampled as the primary lesion. Surrounding tissue was sampled as the perilesion, although this was done by macroscopic observation and, thus, strongly influenced by observer-based judgment. Statistical analysis of neuron count and microvessel density let us conclude that microscopically the perilesion side is still part of the lesion. Statistically, no significant difference was present. Thus, the lesion seems to be much more extensive than originally thought by macroscopic examination and seems to take over surrounding parts of the brain that visually appear to not have suffered any obvious damage. We must take these findings into account when estimating the cause of death in individuals with TBI since although the lesion is grossly small, microscopic studies will reveal the true extent of the primary lesion.
Alterations of microvascularization
Morphologically, microvessels of the traumatic group appeared to be less healthy and more unstructured than the ones in the control group. Through the traumatic impact, the blood-brain-barrier (BBB) seems to have suffered strong damage. Anti-Collagen IV staining visualized the second layer of the BBB, the basal membrane of the vessels, and it became apparent that the traumatic group showed more corrugation of the microvessels and less healthy smooth vessel surfaces as the controls. Intervascular bridges, as thoroughly described by Ortega et al.31, were only
rarely present in the TBI samples, and if so, mostly broken. This is most likely a consequence of the contusion to the head that the individuals suffered before their death. Previous studies on TBI described the presence of microaneurysms in TBI cases, however, these could not be observed in this study. The studies that observed microaneurysms had an older mean age of their individuals (~60-70 years)31, which
makes it possible that microaneurysms oftentimes appear in individuals of an older age group, not only due to a sustained trauma. In this study, the mean age of the traumatic group was ~44 years and the mean age of the control group was ~50 years. Neither of them presented any microaneurysms.
In the study conducted by Rodríguez-Baeza of the vessels of the human cerebral cortex in deceased patients due to the evolution of secondary lesions after suffering a severe TBI, a series was also described of alterations in microvascularization that
Among these morphological alterations were the so-called ‘corrugated vessels’, which showed an evident decrease in the lumen. These alterations were observed in both cerebral hemispheres diffusely, and not related to the area of direct trauma, the primary lesion. One possible explanation for these structures is the lack of tension in the vascular walls, caused by broken intervascular bridges, which causes the walls to wrinkle31.
Edema affects the brain as a whole
The comparison of the neuron count and microvessel density measurement clearly showed a plummeting trend in the TBI group. Overall, less neurons were found in the respective monitored areas compared to the control group and the density of the microvascularization is likewise lower than that of the controls. Most likely, this can be explained by the presence of edema in all TBI samples. Through the traumatic impact, excess water accumulated in the interstitial space in the brain and, therefore, the space between neurons and microvessels increased. In this study, not only the lesion side and areas surrounding it were monitored, but also characteristic areas distributed all over the brain. The neuron count and microvessel density lowered in each area respectively compared to the control group, thus, we conclude that edema affects the brain as a whole.
Two types of edema are present in our samples. Vasogenic edema is caused by a failure of the BBB, which is a characteristic response for hemorrhagic contusions seen in trauma. This was observed through the Collagen IV staining of the basal membrane that showed strong corrugation in TBI cases. The deteriorating effect of edema to the basal membrane of capillaries in the human brain has been previously described by Castejón33.
Cytotoxic edema is caused by the failure of sodium-calcium-pumps in the cell membrane, which makes water enter the cells. Through Nissl-staining, this could be indirectly observed, as many cells in the TBI group lost their shape, altered their size and presented broken membranes.
The preliminary assessment of the morphological alterations of microvessels observed highlights the role that they can play in the progression of the feared secondary lesions (cerebral ischemia, intracranial hypertension), which would determine the clinical evolution of the patients with the final result of neuronal death. In our study, we found differences in microvascularization, with the TBI group presenting a decrease in density and morphological alterations with respect to the control group. These results are not consistent with the variations by other authors, where no obvious differences in microvascularization are observed by the staining with Collagen IV. We, on the contrary, believe that the morphological alterations described, as well as the difference in quantification, can be of great help when assessing whether a TBI has existed34.
Aquaporin as a marker for ischemia/anoxia
AQP4 has been previously described as a promising marker for TBI. It is expressed on astrocyte foot processes, which form the third layer of the BBB. AQP4 expression was elevated in the TBI group, with a heterogeneous distribution of astrocytes. It could be observed that the astrocytes were located much more directly at microvessels, compared to the control sample. However, special attention has to be paid to the control case OC4. This individual deceased through hanging, which is a type of asphyxiation. AQP4 expression in OC4 behaved much more similar to the traumatic samples than to the other control sample. Astrocytes were larger, possessed more food processes and appeared in groups. All the studied TBI cases showed a survival time of the injury of more than 18 hours, whereas the individual C4 died without delay through the asphyxiation. Thus, AQP4 expression showed a rapid onset. Asphyxiation can cause hypoxia/ischemia in the brain that can lead to edema. AQP4, the most abundant aquaporin in the brain23 is the dominant contributor to the
regulation of cerebral edema24. Therefore, care has to be taken when calling AQP4 a
marker for TBI, as previously done by different authors8, since it is not only specific
for traumatic injuries but also seems to be elevated in other conditions, such as asphyxia. We, therefore, suggest that AQP4 is a marker specific for alterations to the brain that are connected to ischemia/anoxia.
Following TBI, brain edema occurs with vasogenic edema occurring rapidly after the injury, primarily in the center of the lesion, whereas cytotoxic edema has a later onset and is predominant35. There is growing evidence that members of aquaporins (small,
hydrophobic membrane proteins regarded as a ‘pure water channel family’ that facilitate water transport) play an important role in traumatic brain edema22,36.
Other authors have studied the expression of AQP4 at different times after suffering a TBI, concluding that AQP4 is expressed in fatal human TBI at several evolution phases8. The pathophysiology of traumatic brain injury is complex. Some studies
show an overexpression AQP4 in the reactive astrocytes after a brain injury. AQP4 acts as a bidirectional transport of water, which allows it to contribute to both edema and the elimination of water from the brain into blood vessels. It is fundamental both in the movement of water towards swollen astrocytes, as seen in the cellular cytotoxic edema, and in the resolution of vasogenic edema after brain injury.
The present study provides evidence that there are alterations in the level of cerebral microvascularization in TBI. In addition, these alterations occur diffusely and are not specific to areas of primary injury. The activation of AQP4 at a distance to the lesion focus in the TBI is indicative of the role it plays in the production of the dreaded secondary lesions such as edema. And we have shown that it is not specific to TBI since AQP4 has been activated in patients whose cause of death has been cerebral ischemia with subsequent production of edema, as in the case of mechanical asphyxiation.
Future Research
Histological research on TBI is a promising topic since clear differences were observed between the traumatic and control group. To continue this study, it would be interesting to investigate more possible markers, which are specific for TBI. If some can be found that are only expressed after actual contusion to the head, these markers could be experimentally combined to raise the strength of the results. Additionally, their expression pattern in body fluids should be tested. Finding reliable TBI markers in fluids like saliva or blood would be of immense help in establishing a protocol to test for a TBI in a much quicker and easier way than through histological analysis, where the sample processing takes up a lot of time. Physiopathologically, finding therapeutic targets could aid in the prevention of deterioration of a TBI and possibly increase the chances of health improvement of the individuals.
Conclusion
1. Based on the macroscopic difference between lesion and perilesion focus, but the lack thereof microscopically, we conclude that microscopic analysis is of importance to not underestimate the actual size of the injury affected cerebral tissue.
2. Neuron density, as well as vascular density, both decreased in the TBI group compared to the control group, not only in the lesion surrounding areas but also at distance to it, which we correlate with secondary injury phenomenons like edema and ischemia.
3. The observation of broken IBs, corrugation of the microvessel membrane and increased intercapillary distance can be attributed to edema and the disruption of the cerebral parenchyma angioarchitecture, which facilitates ischemia.
4. The existence of significant differences between both groups, in the capillary density and in the density of neurons, both in the grey substance and in the white substance, suggests that cerebral edema occurs diffusely affecting both cerebral cortex and white matter.
5. In addition, this activation occurs in areas distal to the focus of primary lesion, which indicates that these are activated and contribute to the formation of secondary lesions such as edema and hypoxic-ischemic lesions.
6. Early activation of AQP4 on the astrocyte foot processes in the TBI group was visible, but also the mechanical asphyxiation control sample presented this phenomenon. Thus, we conclude that the activation of AQP4 is not a TBI specific biomarker but indicates the presence of edema and ischemia.
7. Alterations of capillary microvascularization and activation of APQ4 in the TBI group demonstrate that encephalic microvessels and astrocytes form a complex functional unit and we should treat this as such.
8. If APQ4 activation is visible in the samples, it is indicative of a lesion that was produced antemortem.
9. More studies are needed to be able to elucidate which would be the specific biomarker of traumatic brain damage and if these can be detected in other body fluids.
References
1. Velázquez A, Ortega M, Rojas S, González-Oliván FJ, Rodríguez-Baeza A. Widespread microglial activation in patients deceased from traumatic brain injury. Brain Inj. 2015;29(9):1126-1133. doi:10.3109/02699052.2015.1018325
2. Oehmichen M, Auer RN, König HG. Forensic Neuropathology and Associated Neurology. Springer-Verlag; 2006. doi:10.1360/zd-2013-43-6-1064
3. Rodríguez-Baeza A, Reina-de la Torre F, Poca A, Martí M, Garnacho A.
Morphological features in human cortical brain microvessels after head injury: A three-dimensional and immunocytochemical study. Anat Rec Part A Discov Mol Cell Evol Biol. 2003;273A(1):583-593. doi:10.1002/ar.a.10069
4. Wang Q, Ishikawa T, Michiue T, Zhu BL, Guan DW, Maeda H. Quantitative immunohistochemical analysis of human brain basic fibroblast growth factor, glial fibrillary acidic protein and single-stranded DNA expressions following traumatic brain injury. Forensic Sci Int. 2012;221(1-3):142-151. doi:10.1016/j.forsciint.2012.04.025 5. Taylor CA, Bell JM, Breiding MJ, Xu L. Traumatic Brain Injury–Related Emergency Department Visits, Hospitalizations, and Deaths — United States, 2007 and 2013. MMWR Surveill Summ. 2017;66(9):1-16. doi:10.15585/mmwr.ss6609a1
6. Andriessen TMJC, Jacobs B, Vos PE. Clinical characteristics and pathophysiological mechanisms of focal and diffuse traumatic brain injury. J Cell Mol Med.
2010;14(10):2381-2392. doi:10.1111/j.1582-4934.2010.01164.x
7. Graham DI, Lawrence AE, Adams JH, Doyle D, McLellan DR. Brain Damage In Non-Missile Head Injury Secondary To High Intracranial Pressure. Neuropathol Appl Neurobiol. 1987;13.3.
8. Neri M, Frati A, Turillazzi E, et al. Immunohistochemical Evaluation of Aquaporin-4 and its Correlation with CD68, IBA-1, HIF-1α, GFAP, and CD15 Expressions in Fatal Traumatic Brain Injury. Int J Mol Sci. 2018;19(11). doi:10.3390/ijms19113544
9. Launer LJ. Demonstrating the case that AD is a vascular disease: epidemiologic evidence. ageing Res Rev. 2002.
10. Sharp DJ, Scott G, Leech R. Network dysfunction after traumatic brain injury. Nat Rev Neurol. 2014;10:156. https://doi.org/10.1038/nrneurol.2014.15.
11. McGinn, M.J.; Povlishock JT. Pathophysiology of Traumatic Brain Injury. Neurosurg Clin N Am. 2016.
12. Krohn M, Dreßler J, Bauer M, Schober K, Franke H, Ondruschka B.
Immunohistochemical Investigation of S100 and NSE in Cases of Traumatic Brain Injury and Its Application for Survival Time Determination. J Neurotrauma.
