Histological changes in thalamus in short term survivors following traumatic brain injury: An autopsy study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.125256
Source of Support: None, Conflict of Interest: None
Background: Severe traumatic brain injury (TBI) is a major cause of morbidity and mortality. Reduction of thalamic volumes were seen in upto 80% of patients who survived for more than 3 months after TBI. However, the same may not be true in patients who died earlier following TBI. Aims: To to study the thalamus for evidence of any injury in short term survivors of TBI (<5 days) using immunohistochemistry to look for evidence of acute thalamic injury. Materials and Methods: A cross sectional prospective study was done in which autopsy specimens of short term survivors of TBI (<5 days) were studied for histopathological changes. Results: A total of 16 patients with a mean age of 37.8 years were included in the study. CT scan revealed acute subdural haematoma in 10, contusions in 4 patients, extradural haematoma and depressed fracture in 1 each, and diffuse axonal injury in 1 patient. Seven patients required surgery in the form of a decompressive hemicraniectomy. The histopathological analysis of the bilateral thalami showed evidence of congestion of the cerebral capillaries in 8 patients. Axonal retraction balls were seen in 8 patients, myelin breakdown products were seen in 14 patients and axonal swelling was seen in 14 patients. Conclusions: Thalamic injury is universal in the setting of severe TBI in patients who have decreased survival and may be a significant factor for the poor outcome in these patients.
Keywords: Axonal retraction balls, neurofilament protein, thalamus, traumatic brain injury
Severe traumatic brain injury (TBI) is the major cause of morbidity and mortality in patients younger than 40 years of age. The incidence varies from 50 to 546/lac population with mortality rates of 18-50% for severe TBI. ,, The last 25 years have seen a substantial increase in our understanding of the pathological mechanisms of TBI, though large gaps still remain. Though there is a large body of literature available on the pathological sequelae of trauma to the human brain in the form of long-term survivors of TBI, ,, the literature on those who survive <1 week is sparse , Reduction of thalamic volumes were seen in up to 80% of patients who survived for more than 3 months after TBI  However, the same may not be true in patients who died earlier than this following TBI, because with time trans neuronal degeneration secondary to cortical injury may be the cause of volume loss. Detection of acute thalamic injury is difficult as computed tomography (CT) is unable to detect injuries to the thalamus unless there are contusions or infarcts and it is the only imaging study feasible in unstable TBI patients. Magnetic resonance imaging (MRI) ,, is able to detect these changes, but it is not feasible to do a MRI in these sick patients.
Definite evidence of the entire spectrum of brain injury can only be obtained by histopathological examination of the brain, which can only be carried out post mortem. Conventional staining (hematoxylin and eosin [H and E], silver stains etc.) are able to detect evidence of injury to the neurons only after 12-24 h and hence may miss injury in patients who survive for <12 h.  Various proteins that are transported in the axoplasm get accumulated at the site of axonal injury and can be detected by various staining techniques. Amyloid precursor protein (APP) staining is able to detect damaged axons within 35 min of injury.  Other proteins including neurofilament (NF) have also been reported to accumulate and can be used to determine axonal injury. , NF in micro dialysis has also been studied and found to be of prognostic significance. 
Experimental studies in animals have shown evidence of injury to the thalamocortical connections and brainstem in TBI, ,, which may be the cause of loss of consciousness and loss of cortical control due to injury to the thalamic relay.
We believe that patients with severe TBI who have diffuse injury (absence of structural brain lesions on CT scan) and have short survival, have injury to the thalamus which is not picked up on routine imaging with CT scan. Acute thalamic injury may be responsible (at least in part) for the poor neurological status of patients with severe TBI. We propose to study the thalamus for evidence of any injury in short term survivors of TBI (<5 days) using immunohistochemistry to look for evidence of acute thalamic injury.
We did a cross-sectional prospective study at our trauma center to study the evidence of thalamic injury patients with severe TBI who survived for <5 days following injury. All patients with severe TBI (Glasgow coma score [GCS] ≤8) were screened and the eligible were included in the study. Inclusion criteria were (1) Patients with severe TBI (post resuscitation GCS ≤8), (2) CT scan was done at time of admission, (3) died within 5 days of injury and (4) underwent an autopsy study. Exclusion criteria were (1) Patients with post ressuscitation GCS score >8, (2) patients dead at arrival to our hospital, (3) died more than 5 days after injury, (4) if admission time CT scan was unavailable and (5) patients with no clinical evidence of head trauma.
Management of all patients was consistent with the standard guidelines. Surgery was carried out if indicated. Repeat imaging studies were carried out whenever deemed necessary. After the death of the patient the body was stored in a refrigerated body cabinet and an autopsy was carried out.
At autopsy the brain was removed as one block and cut in the mid sagittal plane superiorly at the corpus callosum and anteriorly at the anterior commissure. This block of brain, which contained the thalamus was preserved in 10% formalin in saline until the histopathological analysis.
Thin sections of the thalamus were made and stained with H and E, NF (NF protein), myelin basic protein (MBP), glial fibrillary acidic protein (GFAP) and myelin. H and E is the basic stain used for all histopathology. Blocks from regions of interest on H and E were processed for immunohistochemistry with GFAP to look for evidence of glial cell injury, MBP and myelin stains to look for evidence of myelin degeneration (granulation) and demyelination. NF stains the NF protein, which is structural proteins of axons and is released after trauma to the axons.
The clinical records of the patients were analyzed for demographics, mechanisms of injury, clinical status at presentation, surgery carried out if any and time to death. The CT scans were analyzed for presence of any structural lesions and evidence of diffuse axonal injury (DAI).
A total of 16 patients were included in the study. Of these, 12 males and 4 females. The average age was 37.8 years (9-60 years). The survival following injury ranged from 1 to 5 days with a mean of 2 days [Table 1].
Fall from height was the most common mechanism of injury (in 7 patients), motor vehicle accident was seen in 5 patients [Table 2]. Lucid interval was present in 1 patient. The average GCS was 4.12 (range : 3-7). Five patients had additional systemic injuries.
CT scan revealed acute subdural hematoma in 10, contusions in 4 patients, extradural hematoma and depressed fracture in 1 each and DAI in 1 patient. The average time between injury and CT scan was 7 h (range 90 min to 48 h). Basal cistern effacement, suggestive of raised ICP, was seen in 10 patients. Subfalcine herniation was seen in 9 patients. Midline shift ranging from 8 to 21 mm was seen in 8 patients [Table 3]. Repeat CT scans were carried out in 7 patients and findings are described in [Table 4].
Seven patients required surgery in the form of a decompressive hemicraniectomy. Intraoperative brain bulge was seen in 6 patients.
Post-mortem examination confirmed the findings of the last imaging study. The histopathological analysis of the bilateral thalami showed evidence of congestion of the cerebral capillaries in 8 patients [Table 5]. Axonal retraction balls were seen in 8 patients [Figure 1], myelin breakdown products were seen in 14 patients [Figure 2] and [Figure 3] and axonal swelling was seen in 14 patients [Figure 1]. When the axon was in continuity with the swelling in the middle it was classified as axonal swelling and when there was discontinuity in the axon it was classified as retraction bulb [Figure 2]. All patients had either axonal swelling or evidence of myelin breakdown on immunohistochemistry. On myelin and NF stains Grade 1 was taken as evidence of myelin granular degeneration and grade 2 as appearance of axonal retraction balls, the higher of the two was taken as the grade for a particular patient [Table 5].
Consciousness is mediated by neuronal networks that include the thalamic and extra thalamic ascending arousal systems, which are responsible for wakefulness and damage to these can result in varying grades of coma.  The damage to these networks may not always be seen on CT scan, whereas MRI is able to detect injury to these thalamocortical connections but is usually not feasible in the acute setting.  The cause of DAI was thought to be mechanical till recently,  when it has been demonstrated that trauma induced focal axolemmal permeability leads to local influx of Ca 2+ which activates cysteine proteases and cause proteolytic digestion of brain spectrin, a major constituent of the subaxolemmal cytoskeleton, leading to DAI. , The neuronal loss that occurs can be measured in vivo by analyzing the NF heavy chain NfH476-986 and Nfh476-1026 from micro dialysis samples and its levels have been shown to have prognostic significance. 
Warner et al.  prospectively studied brain volumes after TBI using MRI (three dimensional T1 weighted magnetization-prepared rapid acquisition with gradient echo images) and found a regional variations in the volume loss [Table 6]. The amygdala, hippocampus, thalamus were among the regions that had a reduction in volume at a repeat MRI between 6 and 14 months (initial MRI done at 0.5-9 days) whereas caudate nucleus and inferior temporal cortex were relatively resistant to atrophy. The reduced fractional anisotropy of the thalamic projection fibers has been demonstrated and this may be of relevance in altered cognition. 
Gusmγo et al.  in an autopsy study of 120 patients (83 deaths within 24 h) found hypoxic damage in 19.2% (4.8% in those surviving <24 h, 51.4% in those surviving >1 day) mainly in the hippocampus and subiculum (65.2%), thalamus (34.8%), neocortex (26.1%) and basal ganglia (21.7%).
Conti et al.  studied the temporal pattern of apoptosis in rats using Terminal deoxynucleotidyl transferase dUTP nick end labeling histochemistry assay and found that the apoptotic response to trauma is regionally distinct. Apoptosis in the thalamus peaked by 2 weeks, the white matter at 1 week and the hippocampus at 48 h.
The histopathological analysis revealed evidence of axonal injury in all of our patients either in the form of axonal swelling or evidence of myelin breakdown. Axonal retraction balls were seen in 8 patients with the earliest appearance seen in a patient who had survived 6 h post trauma. Evidence of congestion of capillaries was seen in 8 patients. There is anterograde axonal transport of structural proteins like β-APP, NF after synthesis in the cell body. Any breakdown in the integrity of the axon leads to a time bound accumulation of these proteins at the site of the break and these can then be studied by using antibodies to these proteins. β-APP (transported by the fast transport by a kinesin-based motor) has been extensively studied after TBI and is used as a marker for evidence of trauma to the axons. ,,,
Grady et al.  in their study have reported the accumulation of immune positivity to 68 KD NF in axonal injury, which was seen at about 6 h post trauma. Dunn-Meynell et al.  demonstrated NF immunoreactivity 3 h post trauma in animal studies. NF is also demonstrable in the serum with first peak appearing at 16 h after injury and can be used to monitor neuronal damage ,, NF is transported by slow transport in the axons. Sherriff et al.  NF reactivity was demonstrated after microwave antigen retrieval. [Table 7] shows the various studies done in human TBI.
Axonal retraction balls the markers of axonal injury appear 12-24 h after trauma  when studied with conventional stains. However with the use of immunohistochemistry the presence of axonal injury can be detected much earlier. The minimum time between trauma and death in the present study was 6 h which is well within the reported time at which axonal injury can be demonstrated with immunohistochemistry by staining for NF (being part of the neuronal cytoskeleton) or β-APP (transported in the axoplasm).
NF has been infrequently used in the setting of trauma, though studies have shown its ability to pick up lesions within 3 h of trauma. The novelty of demonstrating evidence of injury with NF lies in the fact that it can be quantitatively measured in the serum and cerebrospinal fluid and correlation of its levels with thalamic injury on autopsy studies may be of some role as a biomarker with prognostic significance in TBI.
The present study has demonstrated evidence of thalamic injury using NF and myelin stain in 87.5% patients with severe TBI who died within 5 days of trauma and none of these patients had any abnormalities in the thalamus on the CT scan. However on combining the two stains, all (100%) patients had evidence of thalamic injury. This suggests that thalamic injury is universal in the setting of severe TBI in patients who have decreased survival. Thalamic injury may be a significant factor for the poor outcome in these patients.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]