The lingual nerve: overview and new insights into anatomical variability based on fine dissection using human cadavers.

This article, for both researchers and clinicians, presents an overview of the lingual nerve and highlights how new insights into human anatomical variability can be gained by integrating fine dissection of cadavers with neuroanatomical approaches, microscopic studies, and morphometric techniques. Textbooks mainly provide descriptions of the typical or common gross anatomical appearance of structures in the human body with little reference to the nature and extent of variation that may be encountered within and between populations. Furthermore, few texts attempt to integrate descriptions of the regional distribution and branching of neural structures with their central connections or their microscopic anatomy. Using the lingual nerve as an example from the head and neck region, we show that there is still an important place for detailed fine dissections of human cadavers when they are also integrated with morphometric techniques applied to data representing observed variation at both macro- and micro-levels. It is essential that health professionals have a sound understanding of the nature and extent of anatomical variation displayed normally by their patients so that they can perform procedures, such as local anaesthesia and surgery, safely and also be able to correctly diagnose pathology when it is present.

The facial nerve: anatomy and associated disorders for oral health professionals.

The facial nerve, the seventh cranial nerve, is of great clinical significance to oral health professionals. Most published literature either addresses the central connections of the nerve or its peripheral distribution but few integrate both of these components and also highlight the main disorders affecting the nerve that have clinical implications in dentistry. The aim of the current study is to provide a comprehensive description of the facial nerve. Multiple aspects of the facial nerve are discussed and integrated, including its neuroanatomy, functional anatomy, gross anatomy, clinical problems that may involve the nerve, and the use of detailed anatomical knowledge in the diagnosis of the site of facial nerve lesion in clinical neurology. Examples are provided of disorders that can affect the facial nerve during its intra-cranial, intra-temporal and extra-cranial pathways, and key aspects of clinical management are discussed. The current study is complemented by original detailed dissections and sketches that highlight key anatomical features and emphasise the extent and nature of anatomical variations displayed by the facial nerve.

Magnesium enhances the beneficial effects of NK1 antagonist administration on blood-brain barrier permeability and motor outcome after traumatic brain injury.

The current study investigated whether adding magnesium to an NK1 tachykinin receptor antagonist after traumatic brain injury would enhance efficacy to further reduce blood-brain barrier permeability and improve functional recovery compared to either treatment alone. Sprague-Dawley rats were injured using the impact acceleration model of diffuse brain injury, and received either no treatment, MgSO4 (30 mg/kg IV), the NK1 antagonist n-acetyl L tryptophan (2.5 mg/kg IP), or both agents combined. Animals were then killed at either 1, 5, or 24 h postinjury for determination of blood-brain barrier permeability using previously administered Evans blue dye or assessed for functional outcome over a 1-week period using the rotarod motor test. As expected, both MgSO4 and n-acetyl L tryptophan significantly reduced blood-brain barrier permeability and improved functional outcome. However, combined n-acetyl L tryptophan and MgSO4 was more effective at reducing blood-brain barrier permeability (P < 0.05) and improving functional outcome (P < 0.001) compared to the individual compounds. Our results demonstrate that combination therapy with magnesium and an NK1 antagonist may be a more effective therapy for TBI than either compound administered alone.

Aberrant distribution of the trochlear nerve: A cadaveric study supported by immunohistochemistry.

The trochlear nerve is generally considered to be a purely motor nerve supplying one extraocular muscle, the superior oblique. In the current study, 28 orbits were dissected and in one orbit (3.6%), the trochlear nerve divided into two main branches. The medial branch followed the classical course, entered the superior oblique muscle and was presumed to be motor in function. However, before entering the muscle, it partially fused with the frontal nerve, and gave a bundle of nerve fibres to the frontal nerve. The lateral branch gave a communication to the frontal nerve, travelled along the lacrimal nerve, received a branch from the lacrimal nerve then penetrated the lacrimal gland. The lateral branch was presumed to be sensory. Paraffin sections from the two branches were stained using immunohistochemistry. The two branches had different nerve fibre populations and showed distinct differences in neurofilament proteins (NFP) immuno-labelling. While both branches showed intense labelling for NFP-H, the lateral branch showed no staining or faint staining for NFP-M and NFP-L respectively, but the medial branch showed moderate labelling for both the NFP-M and NFP-L. Staining for substance P, a marker for nociceptive fibres, showed intense staining in a subset of fibres in the lateral branch, but no staining in the medial branch. Calcitonin gene-related peptide labelling was evident in some axons and some Schwann cells in the medial branch but widespread, weak and fine granular in the lateral branch. These findings indicate that, in some individuals (3.6%), the trochlear nerve may contain motor and sensory fibres, suggesting inter-nuclear communication within the brainstem during embryogenesis or mixing of nerve fibres in their extra-axial pathways.

Bilateral aberrant infratentorial vasculature: a rare cadaveric encounter

During a cadaveric dissection course, abnormalities of the infratentorial vasculature were noted. Following removal of the brain, the posterior cranial fossa showed the left labyrinthine artery splitting into two branches: one entered the internal auditory meatus (IAM) with the eighth cranial nerve, and the other pierced the petrous temporal bone just posterior to the IAM in the region of the subarcuate fossa. On the right side, the anterior inferior cerebellar artery formed a loop that was embedded in the dura just posterior to the IAM, but no vessels were seen entering the IAM. Further dissection into the petrous temporal bone showed the loop directed towards the region of the subarcuate fossa, and the injection of coloured latex confirmed fine arterial distribution of the latex on the surface of the deep part of the eighth cranial nerve. Eight other cadavers dissected in the same course did not show any such anomalies. Knowledge of these vascular variations is important for surgical exposure of the posterior cranial fossa

No disponible

Treatment with the NK1 antagonist emend reduces blood brain barrier dysfunction and edema formation in an experimental model of brain tumors.

The neuropeptide substance P (SP) has been implicated in the disruption of the blood-brain barrier (BBB) and development of cerebral edema in acute brain injury. Cerebral edema accumulates rapidly around brain tumors and has been linked to several tumor-associated deficits. Currently, the standard treatment for peritumoral edema is the corticosteroid dexamethasone, prolonged use of which is associated with a number of deleterious side effects. As SP is reported to increase in many cancer types, this study examined whether SP plays a role in the genesis of brain peritumoral edema. A-375 human melanoma cells were injected into the right striatum of male Balb/c nude mice to induce brain tumor growth, with culture medium injected in animals serving as controls. At 2, 3 or 4 weeks following tumor cell inoculation, non-treated animals were perfusion fixed for immunohistochemical detection of Albumin, SP and NK1 receptor. A further subgroup of animals was treated with a daily injection of the NK1 antagonist Emend (3 mg/kg), dexamethasone (8 mg/kg) or saline vehicle at 3 weeks post-inoculation. Animals were sacrificed a week later to determine BBB permeability using Evan's Blue and brain water content. Non-treated animals demonstrated a significant increase in albumin, SP and NK1 receptor immunoreactivity in the peritumoral area as well as increased perivascular staining in the surrounding brain tissue. Brain water content and BBB permeability was significantly increased in tumor-inoculated animals when compared to controls (p<0.05). Treatment with Emend and dexamethasone reduced BBB permeability and brain water content when compared to vehicle-treated tumor-inoculated mice. The increase in peritumoral staining for both SP and the NK1 receptor, coupled with the reduction in brain water content and BBB permeability seen following treatment with the NK1 antagonist Emend, suggests that SP plays a role in the genesis of peritumoral edema, and thus warrants further investigation as a potential anti-edematous treatment.

Applied anatomy of the lingual nerve: relevance to dental anaesthesia.

OBJECTIVES: (1) to classify the external morphology of the lingual nerve and investigate any relationship between its external and internal morphology, (2) to explore the fascicular structure, nerve tissue density and capillary density of the lingual nerve, and (3) to provide an anatomical explanation as to why adverse clinical outcomes more commonly affect the lingual nerve following local dental anaesthesia. Where possible, comparisons were made between the lingual and inferior alveolar nerves. MATERIALS AND METHODS: The lingual and inferior alveolar nerves were examined in 23 hemi-sectioned heads macroscopically and microscopically 2mm above the lingula. The lingual nerve was also examined in the regions of the third and second molars. Specimens underwent histological processing and staining with Haematoxylin & Eosin, Masson's Trichrome, anti-GLUT-1 and anti-CD 34. RESULTS: The lingual nerve became flatter as it traversed through the pterygomandibular space. There was an increase in the connective tissue and a decrease in nerve tissue density along the lingual nerve (p<0.001). At 2mm above the lingula, the lingual nerve was uni-fascicular in 39% of cases, whilst the inferior alveolar nerve consistently had more fascicles (p<0.001). The lingual nerve fascicles had thicker perineurium but the endoneurial vascular density was not significantly different in the two nerves. CONCLUSIONS: The greater susceptibility of lingual nerve dysfunction during inferior alveolar nerve blocks may be due to its uni-fascicular structure and the thicker perineurium, leading to increased endoneurial pressure and involvement of all axons if oedema or haemorrhage occurs due to trauma.

NK1 receptor antagonists and dexamethasone as anticancer agents in vitro and in a model of brain tumours secondary to breast cancer.

Emend, an NK1 antagonist, and dexamethasone are used to treat complications associated with metastatic brain tumours and their treatment. It has been suggested that these agents exert anticancer effects apart from their current use. The effects of the NK1 antagonists, Emend and N-acetyl-L-tryptophan, and dexamethasone on tumour growth were investigated in vitro and in vivo at clinically relevant doses. For animal experiments, a stereotaxic injection model of Walker 256 rat breast carcinoma cells into the striatum of Wistar rats was used. Emend treatment led to a decrease in tumour cell viability in vitro, although this effect was not replicated by N-acetyl-L-tryptophan. Dexamethasone did not decrease tumour cell viability in vitro but decreased tumour volume in vivo, likely to be through a reduction in tumour oedema, as indicated by the increase in tumour cell density. None of the agents investigated altered tumour cell replication or apoptosis in vivo. Inoculated animals showed increased glial fibrillary acidic protein and ionized calcium-binding adapter molecule 1 immunoreactivity indicative of astrocytes and microglia in the peritumoral area, whereas treatment with Emend and dexamethasone reduced the labelling for both glial cells. These results do not support the hypothesis that NK1 antagonists or dexamethasone exert a cytotoxic action on tumour cells, although these conclusions may be specific to this model and cell line.

Characterisation of Walker 256 breast carcinoma cells from two tumour cell banks as assessed using two models of secondary brain tumours.

BACKGROUND: Metastatic brain tumours are a common end stage of breast cancer progression, with significant associated morbidity and high mortality. Walker 256 is a rat breast carcinoma cell line syngeneic to Wistar rats and commonly used to induce secondary brain tumours. Previously there has been the assumption that the same cancer cell line from different cell banks behave in a similar manner, although recent studies have suggested that cell lines may change their characteristics over time in vitro. METHODS: In this study internal carotid artery injection and direct cerebral inoculation models of secondary brain tumours were used to determine the tumorigenicity of Walker 256 cells obtained from two cell banks, the American Type Culture Collection (ATCC), and the Cell Resource Centre for Medical Research at Tohoku University (CRCTU). RESULTS: Tumour incidence and volume, plus immunoreactivity to albumin, IBA1 and GFAP, were used as indicators of tumorigenicity and tumour interaction with the host brain microenvironment. CRCTU Walker 256 cells showed greater incidence, larger tumour volume, pronounced blood-brain barrier disruption and prominent glial response when compared to ATCC cell line. CONCLUSIONS: These findings indicate that immortalised cancer cell lines obtained from different cell banks may have diverse characteristics and behaviour in vivo.

Walker 256 tumour cells increase substance P immunoreactivity locally and modify the properties of the blood-brain barrier during extravasation and brain invasion.

It is not yet known how tumour cells traverse the blood-brain barrier (BBB) to form brain metastases. Substance P (SP) release is a key component of neurogenic inflammation which has been recently shown to increase the permeability of the BBB following CNS insults, making it a possible candidate as a mediator of tumour cell extravasation into the brain. This study investigated the properties of the BBB in the early stages of tumour cell invasion into the brain, and the possible involvement of SP. Male Wistar rats were injected with Walker 256 breast carcinoma cells via the internal carotid artery and euthanised at 1, 3, 6 and 9 days post tumour inoculation. Culture medium-injected animals served as controls at 1 and 9 days. Evidence of tumour cell extravasation across the BBB was first observed at 3 days post-inoculation, which corresponded with significantly increased albumin (p < 0.05) and SP immunoreactivity (p < 0.01) and significantly reduced endothelial barrier antigen labelling of microvessels when compared to culture medium control animals (p < 0.001). By day 9 after tumour cell inoculation, 100 % of animals developed large intracranial neoplasms that had significantly increased albumin in the peri-tumoral area (p < 0.001). The increased SP immunoreactivity and altered BBB properties at 3 days post-inoculation that coincided with early tumour invasion may be indicative of a mechanism for tumour cell extravasation into the brain. Thus, extravasation of tumour cells into the brain to form cerebral metastases may be a SP-mediated process.

Targeting classical but not neurogenic inflammation reduces peritumoral oedema in secondary brain tumours.

Dexamethasone, the standard treatment for peritumoral brain oedema, inhibits classical inflammation. Neurogenic inflammation, which acts via substance P (SP), has been implicated in vasogenic oedema in animal models of CNS injury. SP is elevated within and outside CNS tumours. This study investigated the efficacy of NK1 receptor antagonists, which block SP, compared with dexamethasone treatment, in a rat model of tumorigenesis. Dexamethasone reverted normal brain water content and reduced Evans blue and albumin extravasation, while NK1 antagonists did not ameliorate oedema formation. We conclude that classical inflammation rather than neurogenic inflammation drives peritumoral oedema in this brain tumour model.

A morphometric analysis of the septal nuclei in schizophrenia and affective disorders: reduced neuronal density in the lateral septal nucleus in bipolar disorder.

The septal nuclei are assumed to play a significant role in the pathophysiology of schizophrenia and affective disorders. The aim of this study was to morphometrically characterize the septal nuclei in patients with schizophrenia, bipolar disorder, and major depressive disorder, when compared with healthy control subjects. We analyzed the septal nuclei by determining the density and size of the neurons in postmortem brains in 17 patients with schizophrenia, 8 patients with bipolar disorder, 7 patients with major depressive disorder, and 14 control subjects matched for age and gender. There was a significant reduction in the neuronal density, but not in the mean cross-sectional area, in the lateral septal nucleus (P = 0.013) in patients with bipolar disorder when compared with control subjects. There were no significant changes in the neuronal density of the septal nuclei of the medial and lateral cell groups in patients with schizophrenia and major depressive disorder when compared with control subjects. There was a significant negative correlation between neuronal density in the lateral septal nucleus and disease duration in patients with major depressive disorder (P = 0.037, r = -0.9). The histopathological abnormality of the decreased neuronal density in the lateral septal nucleus, which is an important limbic region involved in emotions, might be a neuropathological correlate of bipolar disorder.

Anatomical relationships within the human pterygomandibular space: Relevance to local anesthesia.

A thorough understanding of the anatomy of the pterygomandibular space is fundamental to the successful administration of inferior alveolar nerve (IAN) blocks, which are frequently used in dentistry for mandibular anesthesia. However, the nature and extent of anatomical variations and relationships within this space are not well documented, and descriptions vary within the literature. This study analyzed the anatomical patterns and relationships of structures in the pterygomandibular space of 56 human cadaver hemi-heads, with both left and right sides from 10 individuals being available, and described the range of variability. In most cases, the IAN was anterior to the inferior alveolar vasculature with the inferior alveolar vein(s) being closest to the bone. On average, there were two veins per specimen. The position of the inferior alveolar neurovascular bundle expressed as the ratio of its distance from the anterior border of the ramus to the total anteroposterior length of the ramus was 0.60 (standard deviation [SD] = 0.07). The distance of anterior and medial displacement of the lingual nerve to the IAN was 7.3 mm (SD = 2.4 mm) and 3.9 mm (SD = 1.6 mm), respectively. The direct distance between the IAN and lingual nerve was 8.5 mm (SD = 2.4 mm). The sphenomandibular ligament always appeared dense and fibrous, medial to the neurovascular bundle. This anatomical study highlights the extent of variations in the positioning of anatomical structures directly relevant to IAN blocks, and reassesses the rationale for the direct approach, which is currently taught and practiced throughout many countries.

Changes in the blood-CSF barrier in experimental traumatic brain injury.

PURPOSE: Elevation of intracranial pressure (ICP) is a major complication of traumatic brain injury (TBI), and cerebrospinal fluid (CSF) volume is a key factor in ICP regulation. Choroidal epithelial cells (CEC) form the blood-CSF barrier and their integrity is essential for controlling CSF production. In the current study, the morphology of the CEC was studied from 5 h to 28 day after TBI in the rat. METHODS: Male Sprague-Dawley rats were subjected to severe TBI using the impact-acceleration model, and the ultrastructure of the CEC was studied using transmission (TEM) and scanning (SEM) electron microscopy. RESULTS: Radical ultrastructural changes were seen by TEM in CEC in injured animals. At 5 h post-injury cell swelling and incipient cytoplasmic vacuoles were seen. At 24 h most severe changes were noted with extensive widening of intercellular clefts. At 7 day and 14 day post-injury, increased cytoplasmic electron density was evident. At 21 day, most microvilli had bulbous ends, and at 28 day cytoplasmic vacuoles were numerous with widened intercellular clefts. SEM revealed a continuum of changes in all injured animals and most conspicuous was the heterogeneity of surface features, with most cells showing bulbous and cup-shaped microvilli, burr-like processes and pits. Epiplexus cells were hypertrophic and more numerous. CONCLUSION: At 4 weeks after trauma, choroidal epithelial cells continued to show morphological alterations suggesting that brain homeostasis was still not restored.

Hemoglobin crystals: a pro-inflammatory potential confounder of rat experimental intracerebral hemorrhage.

In vivo rat hemoglobin crystallization has been reported in lung, liver and kidney, but never following central nervous system injury. In the present study, we examined hemoglobin crystallization following experimental intracerebral hemorrhage (ICH) and its effects on inflammation. Ninety-one rat brains, subjected to either autologous or collagenase ICH, and vehicle controls, were retrospectively examined. In both models, hemoglobin crystals were present in most brains at 24 and 48 h. They were especially prominent at 24 h in autologous ICH brains (2.5% of the hematoma vs 0.6% in collagenase animals; p=0.0001) and, at 5 h, were only present in autologous ICH brains. Crystals were diminishing at 48 h and were absent at 7 days. Crystals appeared in clusters around blood vessels. In both models, at 24 h, crystals appeared strongly chemotactic for neutrophils. This effect was most pronounced in autologous ICH brains (2628+/-182 neutrophils/mm(2) hematoma crystals vs 327+/-54 neutrophils/mm(2) hematoma; p<0.0001). In these animals up to 30% of the total neutrophilic infiltrate was located around crystals. A greater overall neutrophilic infiltrate was seen in autologous ICHs with higher percentages of crystalline hemoglobin (p=0.04 for trend). Although hemoglobin crystallization occurs in both models of ICH, it is particularly prominent following autologous ICH. Accordingly, hemoglobin crystallization may exaggerate the importance of inflammation in this model.

Anatomical variations in the branches of the human aortic arch: a recent study of a South Australian population.

Variations of the branches of the aortic arch are likely to occur as a result of the altered development of certain branchial arch arteries during the embryonic period of gestation. In the present investigation the pattern of branches of the aortic arch was studied in 81 cadavers from a recent South Australian population of European descent, who have migrated to (n = 38) or were born and lived in (n = 43) South Australia during the twentieth century. Two principal variations were noted in the present study. Firstly, in 6 cadavers, the left vertebral artery originated directly from the arch of the aorta, between the left common carotid and the left subclavian arteries. The 6 subjects were among the subgroup born in South Australia, giving an incidence of 13.95%, which is much higher than in previous reports. The overall incidence of 7.41%, when related to the whole group, is also higher than incidences reported in other populations. The presence of this variation suggests that in some individuals part of the aortic arch is formed from the left 7(th) inter-segmental artery. Secondly, none of the cadavers examined had the thyroidea ima artery, contrasting with previously reported incidences that varied between 4% and 10%. Since all 6 cadavers with the left vertebral artery variant were born in South Australia, it is suggested that environmental factors may have contributed to this variation. Significant environmental changes in South Australia around the turn of the twentieth century are discussed. This study represents the first systematic investigation of the branches of the aortic arch in a South Australian population and provides data relevant to the practice of medicine.

Severity-dependent expression of pro-inflammatory cytokines in traumatic spinal cord injury in the rat.

The post-traumatic inflammatory response in acute spinal cord contusion injury was studied in the rat. Mild and severe spinal cord injury (SCI) was produced by dropping a 10 g weight from 3 and 12 cm at the T12 vertebral level. Increased immunoreactivity of TNF-alpha in mild and severe SCI was detected in neurons at 1 h post-injury, and in neurons and microglia at 6 h post-injury, with a less significant increase in mild SCI. Expression was short-lived and declined sharply by 1 d post-injury. RT-PCR showed an early significant up-regulation of IL-1 beta, IL-6 and TNF-alpha mRNAs, maximal at 6 h post-injury with return to control levels by 24 h post-injury, the changes being less statistically significantly in mild SCI. Western blot showed early transient increases of IL-1 beta, IL-6 and TNF-alpha proteins in severe SCI but not mild SCI. Immunocytochemical, western blotting and RT-PCR analyses suggest that endogenous cells (neurons and microglia) in the spinal cord, not blood-borne leucocytes, contribute to IL-1 beta, IL-6 and TNF-alpha production in the post-traumatic inflammatory response and that their up-regulation is greater in severe than mild SCI.

Peripheral nervous system and central nervous system pathology in rapidly progressive lower motor neuron syndrome with immunoglobulin M anti-GM1 ganglioside antibody.

Pathological studies, including novel teased peripheral nerve fiber studies, were performed in a patient who presented with a rapidly progressive, lower motor neuron syndrome and high titer of immunoglobulin M anti-GM1 ganglioside antibody. In the central nervous system, there was a severe loss of motor neurons and central chromatolysis with ubiquitin immunopositive cytoplasmic inclusions in residual motor neurons. In the peripheral nervous system, axonal degeneration of myelinated fibers in the anterior nerve roots was evident. Pathologic evidence of sensory nerve involvement was also found despite the absence of clinical or electrophysiological sensory abnormalities. Sectional studies of single myelinated nerve fibers from an antemortem sural nerve biopsy showed remyelination and globular paranodal swellings due to focal complex myelin folding and degeneration in 13% of fibers. Postmortem studies of the sural nerves 4 weeks later showed paranodal demyelination (90% of fibers), but no paranodal swellings and similar findings were present in samples of the ulnar, radial, median, tibial, and common peroneal nerves. Paranodal abnormalities of enlargement of the adaxonal space, myelin degeneration, and axonal compaction were found on cross-sectional studies of individual teased fibers, which on conventional light microscopic assessment appeared normal. These changes suggest a disturbance of paranodal axonal-myelin adhesion due to binding of the anti-GM1 ganglioside antibody to the common epitope known to be present on the myelin sheath and nodal axolemma in the paranodal region of both motor and sensory nerves.

Early expression and cellular localization of proinflammatory cytokines interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha in human traumatic spinal cord injury.

STUDY DESIGN: Post-traumatic inflammatory response was studied in 11 human cases of acute spinal cord contusion injury. OBJECTIVES: To examine the inflammatory cellular response and the immunocytochemical expression and localization of interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha in human spinal cord after contusion injury. SUMMARY OF BACKGROUND DATA: : The post-traumatic inflammatory response plays an important role in secondary injury mechanisms after spinal cord injury, and interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha are key inflammatory mediators. METHODS: : The study group comprised 11 patients with spinal cord contusion injury and 2 normal individuals. Histologic and immunocytochemical assessments were undertaken to evaluate the inflammatory cellular response and the immunoexpression of interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha in the injured human spinal cord. The cellular sources of interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha were elucidated by immunofluorescence double-labeled confocal imaging. RESULTS: : Increased immunoreactivity of interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha was detected in neurons 0.5 hour after injury, and in neurons and microglia 5 hours after injury, but the expression of these proinflammatory cytokines was short-lived and declined sharply to baseline by 2 days after injury. In the inflammatory cellular response, as early as 0.5 hour after spinal cord injury, activated microglia were detected, and axonal swellings and axons were surrounded by microglial processes. Numerous neutrophils appeared in the injured cord 1 day after injury, and then their number declined dramatically, whereas macrophages progressively increased after day 1. CONCLUSIONS: Endogenous cells (neurons and microglia) in the human spinal cord, not the blood-borne leukocytes, contribute to the early production of interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha in the post-traumatic inflammatory response, and microglia are involved the early response to traumatic axonal injury.