Neuropathology of Epilepsy

Abstract

Epilepsy is the fourth most common neurological disorder in the world, affecting people of all ages. By definition it is a disorder marked by sudden recurrent episodes of sensory disturbances, consciousness, or convulsions, and associated with abnormal electrical activity in the brain. The effects from seizures on the brain are complex and often need to be separated from any other underlying neurological diseases that may have caused increased susceptibility to seizures. There is evidence to support detrimental effects of seizures on brain histology; this is not inevitable (Haut, 2004; Blumcke, 2009). The decrease in oxygen to vital portions of the brain caused by prolonged seizures results in neuronal death by way of apoptotic or necrotic pathways while activating gliosis and microglial pathways. These histological changes are best identified through the subcellular, synaptic and molecular level outside of normal histology. The most effective way to see the effects is through post mortem examinations. Following epilepsy related deaths such as Status epilepticus, traumatic brain injury, neurodegeneration, sudden death in epilepsy, and examining the different aspects of the brain during Hippocampal sclerosis.  

POST MORTEM EXAMINATIONS IN EPILEPSY

A post mortem examination, or autopsies, is an examination of the body once a patient is deceased. The goal is for pathologists to remove all organs and examine for abnormalities to determine a cause of death. Pathologists specialize in the cause and effects of various diseases. In the case of Epilepsy during a post mortem examination, the neuropathologist aims to answer these three questions: Can I identify the exact cause for epilepsy? Did secondary issues occur because of seizures? Is there any direct correlation between Epilepsy and the cause of death? From my experience in the Anatomic Pathology department at the University of Central Arkansas, I know that after post mortem examinations, the brain is harvested in 20% neutralized buffered formalin for at least seven days before dissection. During dissection residents are encouraged to take sections to seek abnormal changes in the regions of the hippocampus, neocortex, thalamus, amygdala, and cerebellum.

Epilepsy-related deaths

Status epilepticus: In status epilepticus, normal inhibitory mechanisms are unsuccessful and epileptic activity becomes dominant. These mechanisms include endocytosis and decrease of GABA receptors, increase of AMPA and NMDA receptors. This disorder is defined as a seizure that last longer than 5 minutes or having more than one seizure within a 5 minutes period, without returning to a normal level of consciousness between episodes (Status Epileptius, n.d.).  Depletion of inhibitory peptides dynorphin, galanin, somatostatin and neuropeptide Y occurs with pro-convulsants like substance P increase, acting to self-sustain seizures over an extended period of time (Chen, 2007). In the hippocampus proper, there are four regions or subfields titled CA1, CA2, CA3 and CA4. When following the post mortem examination neuropathological findings, few studies have found the following areas had been diminished in these areas: hippocampal CA1, CA3 and the hilus – dentate granule cells may be spared (Pohlmann-Eden, 2004), corticomedial and basolateral nuclei of the amygdala, neocortex mid-cortical layers, the entorhinal cortex, Purkinje cell layer of the cerebellum (Corsellis, 1983), mammillary bodies (Tsuchida, 2007), the dorsal medial nuclei of the thalamus (Fujikawa, 2000), and the basal ganglia (Tan, 1984). The neuronal damage is mostly unilateral but in some cases of extended hemi-convulsions cerebral hemiatrophy can occur with laminar necrosis of the second to fourth cortical layers. But in all, neuronal loss is not an inevitable consequence of status epilepticus.

Traumatic brain injury: Traumatic brain injury or TBI happens when a head injury causes damage to the brain. Every year millions of people in the United States suffer from brain injuries that are severe enough for hospital admission. The more extensive injuries can cause permanent brain damage or death. Lesions caused by the injury can be a cause and also a consequence of epilepsy. Patients with generalized seizures have a higher risk of minor and severe cerebral injuries including cerebral hemorrhage and contusions. All of this is in correlation with the type of seizure, frequency, and control. A post mortem examination can identify cystic cortical contusions, mostly located in the frontal-temporal regions. In the frontal-temporal region, about 30% of cystic cortical contusions may increase vulnerability to age-accelerated neurofibrillary tangle pathology (Thom, 2011).

Picture 1

 Neurodegeneration: Neurodegeneration is the progressive loss of structure or function of neurons, including death of neurons. Many diseases including amyotrophic lateral sclerosis, Parkinson’s disease, Alzheimer’s disease, and Huntington’s disease are a result of this disease. The underlying epileptogenic mechanism is unknown for these diseases and their susceptibility for epilepsy but the toxic effects of amyloid-β on synaptic transmission have been considered (Friedman, 2012).

Sudden death in Epilepsy

Neuropathology: In the unfortunate event that a patient passes away unexpectedly from epilepsy, a post mortem examination is mandatory. It’s referred to as Sudden Unexpected Death in Epilepsy (SUDEP) and it is to exclude an anatomical (e.g. cardiac) or other cause of death. The gross examination of the brain might seem to be more “full” than a normal brain or swollen giving a higher than normal brain weight. Analysis from SUDEP series report macroscopic abnormalities include cerebral traumatic lesions (contusions, gliosis, previous craniotomy sites), hippocampal or cortical atrophy, haemangiomas, low-grade tumors, and cortical malformations (Shields, 2002). To date there does not seem to be any accurate data in terms of the risk or association of any of these specific pathological lesions for SUDEP.  A further investigation to every lesion is always required because some lesions, including the acquired old injuries and cortical neuronal damage may give an inaccurate measure of the clinical severity of epilepsy. Histology is mandatory in SUDEP cases to confirm the macroscopic lesions and gather further information for any unsuspected pathology e.g. meningo-encephalitis.

An autopsy on every patient with epilepsy is not necessary, because of availability and the amount of patients that pass away with this disease would not be feasible. Ideally, a neuropathologist will be consulted for the brain examination portion of the autopsy. The Royal College of Pathologists’ guidelines on autopsy practice in epilepsy recommends that a case should be made to the Coroner and relative for retention of the whole brain for fixation. Optimal fixation time is 2-3 weeks, but a lot of the time this timeline is not available. Another option would be to fix coronal slices of the brain (taken 1.5 cm thick in front of the midbrain and behind the midbrain for about 2-3 days with photography and histology sampling. If this timeline again is not permissible ten small tissue samples must be selected and trimmed for histopathological analysis and the brain immediately returned to the body during the time of autopsy. Studies show that in SUDEP cases if the brain is cut and examined in a fresh state rather than a fixed state, important pathology can be overlooked (Black, 2002)

SUDEP: recognition, likely mechanisms and future directions: While there are no neuropatholigcal diagnostic features of SUDEP, distinguishing this category from other cases allows us to establish a pattern that can be grouped together. SUEDP can be further classified as ‘definite’, ‘possible’, or ‘probable’ depending on whether the autopsy data is complete (Nashef, 2012). Studies of epidemiology and other current research support that SUDEP is an ictal event and that cardiac, pulmonary or autonomic dysfunction concurrent with a seizure are the main mechanisms contributing in each case. SUDEP has many complicated layers to the diagnosis with many mechanisms contributing to each case, which are the cause for the numerous obstacles prior to the 2002 National Sentinel Audit in the UK. The standing guidelines for best practice in epilepsy deaths were issued by the Royal College of Pathologists. To continue the interest in the progress of this category, the causes and preventions must be highlighted and global action should be required between multidisciplinary professionals including neuropathologists.

HIPPOCAMPAL SCLEROSIS

Hippocampal Sclerosis, also referred to mesial temporal sclerosis, MTS, or Ammon’s horn sclerosis, or AHS, characterized pathologically by loss of pyramidal neurons, granule cell dispersion and gliosis in the hippocampus (International League Against Epilepsy, n.d.). The pattern can be seen on neuroimaging or macroscopic examination as a reduction in the volume of structure in Picture 2. The hippocampal sclerosis shown is seen post mortem from a sliced level of the thalamus. Another disorder with strong association to Hippocampal sclerosis is the clinical syndrome of mesial temporal lobe epilepsy (MTLE). The similar features have been recognized for more than one hundred years (Thom, 2009). As with other disorders, its cause and the pathways of which it causes epilepsy is still unknown and the focus of further research with both human tissue and experimental processes.

Picture 2

Patterns of sclerosis: To accurately diagnosis Hippocampal sclerosis, there must be definite identification of pyramidal neuronal loss and gliosis involving CA1, CA4 and CA3 subfields of the hippocampal formation as shown in Table 1. One can microscopically see the pattern f neuronal loss histologically and possibly upon macroscopic examination, depending on the integrity of preserved specimen and severity of the disorder. The CA2 sector is more resistant to neuronal loss because the pyramidal cells in this region tend to be better preserved. Upon first look, the granule cell layer seems preserved but in 40-50% of cases will reveal a weakening package of neurons. ‘End folium sclerosis’ is used to describe the pattern of neuronal damage to the hilar and CA4 pyramidal cells in some patients. There is no research explaining the phenomena of regional selectivity of pyramidal cell loss, excitatory pathways and networks, transformed inhibitory input, and the variability of endogenous neuroprotective mechanisms that are usually involved.

 Some organizations have been used previously for the sub classification of neuronal loss. The Wyler system and other quantitative evaluations have done their best to categorize them. The ILAE system was introduced in 2013 and has been shown to provide useful diagnostic information, see Table 1. Clincopathological correlations show an association between ILAE subtype and pre-operative memory (Coras, 2014) Different Hippocampal sclerosis subtypes may be seen with high field MRI.

Table 1

“Table 1: Semi-quantitative microscopic examination based on formalin-fixed, paraffin-embedded surgical tissue (4–7 μm section thickness), hematoxylin and eosin staining, cresyl-violet combined with Luxol-fast blue staining GFAP immunohistochemistry and NeuN immunohistochemistry. The scoring system corresponds to neuronal cell loss (NeuN staining) and is defined for Subiculum and CA1 to CA4 as following: 0 = No obvious neuronal loss or moderate astrogliosis only; 1 = Neuronal loss and gliosis (GFAP) is moderate or patchy is patchy more or less than moderate; 2 = severe neuronal loss and fibrillary astrogliosis. Scores for the Dentate Gyrus (DG): granule cell layer is normal (score = 0), dispersed (score 1; can be focal) or shows severe granule cell loss (score 2; can be focal)” (Blumcke et al, 2013).

The ILAE subtypes are shown in Picture 3; pyramidal neurones of hippocampal subfields and subiculum are shown in red, granule cells in blue and astrocytosis in green (Thom 2014)

Picture 3

Picture 4

“Picture 4: Patterns of hippocampal sclerosis (all NeuN-stained sections). (A) No MTS. No loss of neurons detected in all subfields (CA1-CA4). (B) MTS type 2 (CA1-sclerosis). Predominant neuronal loss in CA1 subfield. (C) MTS type 1a (classic hippocampal sclerosis). Significant neuronal loss from CA1 and CA4 with better preservation of neurons in CA2. (D) MTS type 1b (severe hippocampal sclerosis). Total HS with neuronal loss in all subfields and in the dentate gyrus. Scale bars 1mm. *For regions with significant cell loss. MTS: mesial temporal sclerosis; HS: hippocampal sclerosis” (Jardim, 2012).

Associated findings with hippocampal sclerosis

An increased number of granule cells in the molecular layer is associated with hippocampal sclerosis, a discovery by CR Houser in 1990. This dispersion seems unusual for this disorder yet it makes up for 40-50% of Hippocampal sclerosis cases in surgical series (Lurton, 1998; Lurton, 1997; Wieser, 2004; Thom, 2000). The granules appear enlarged and more fusiform in shape, increased cytoplasm, and neuropil separating neurons. As the granules migrate to the border cell layer, making the granule clusters less defined, a classification for the thickness of range of dispersion is proposed. Cell layers thicker than 10 cells or 120 μm was proposed. Some cases the thickness ranged from 200 μm or greater compared to the mean control widths, which were closer to 100 μm. There was no association found between granule cell dispersion and early onset of seizures. Picture 5 shows examples of different categories and grades of dispersion. Because there was no association proven, the functional significance of the mechanism for dispersion is also unclear. The dispersion has also shown up in experimental models of Temporal Lobe Epilepsy (TLE). For example, the cells appear approximately four days following seizures and increasing in size and variability over eight weeks and persist for at least six months (Suzuki, 2005). Another proposal has been made to suggest that persistent radial glial processes guide the migration f granule cells through the dentate gyrus.

Picture 5

“Picture 5: Grading of dentate gyrus reorganisation. (a–c) Granule cell dispersion (Cresyl Violet stain). Grade 0 (a) normal; Grade 1 (b) mild dispersion; Grade 2 (c) severe dispersion. (SGZ=sub granular zone). (d–g) Calretinin patterns in the dentate gyrus. (d) Normal (Grade 0) with dense bundles of fibers in the immediate inner molecular layer just above the granule cell layer and similar fibers in the sub granular zone (SGZ). (e) Loss of CR fibers in SGZ (Grade 1), (f) sprouting fibers in the inner molecular layer (Grade 2). (g) Extensive sprouting in outer molecular layer (Grade 3). (h–k) Neuropeptide Y. (h) Normal pattern with clear definition between the inner and outer molecular layer and absent fibers sprouting (Grade 1), (i) distinction between inner and outer molecular layer fibers visible but increased radial fiber networks (Grade 1), (j) loss of boundary between inner and outer molecular layer and increased radial fibers (Grade 2), (k) as for Grade 2 but increased fiber plexus in SGZ region in addition (Grade 3). (l–n) Calbindin in the dentate gyrus. The patterns of labeling in the granule cells was graded as recently described (Martinian et al., 2012): (l) normal pattern with labeling of majority of granule cell neurones and apical dendrites (Grade 0), (m) loss of CB expression (Grade 1), (n) basal granule cells negative and dispersed granule cells positive (Grade 2). Bar=45μm.” (Thom, 2012)

Mossy fiber sprouting: In animal models of MTLE (the kainite model) and hippocampal sclerosis in humans a process called mossy fiber sprouting (MFS) occurs. The mossy fiber projections collaborate and extend to the molecular layer. Approximately 90% of these projections make excitatory synapses with apical dendrites and spines of granule cells inside the molecular layer. Through these connections, a small circuit is created and is potentially pro-epileptogenic. More recently post mortem studies support MFS is probably an epiphenomenon rather than actually being the cause of seizures. The best way to visualize this process is using a Timm silver method because mossy fibers contain large levels of zinc. Another way to visualize MFS is through immunohistochemistry for dynorphin A. This is an opioid neuropeptide that should be present in granule cells and terminal fields of mossy fibers. Picture 6 visualizes the demonstration of Mossy Fiber sprouting in the hippocampus. The upper left picture is a normal slide with Timms stain dominating in the gyrus. The upper right picture is Hippocampal Sclerosis with Mossy fiber sprouting with Timms stain. The lower two pictures are Dymorpin stain of the entire hippocampus.

Picture 6

Widespread change in association with hippocampal sclerosis in TLE: Pathology suggests that clues to seizure onset can be found in structures adjacent to the hippocampus.

Amygdala: In patients with temporal lobe epilepsy studies have shown gliosis and neuronal loss at the lateral nucleus of the amygdala as well as the parvicellular division of the basal ganglia. Amygdala sclerosis might even be used to describe this phenomenon but no definite amount and extent of neuronal loss defines this term. Therefore reports of amygdala sclerosis have been between 35-76%. Any amount of neuronal loss larger than 76% is usually defined as hippocampal sclerosis, even with the amygdala sclerosis reported in isolation.

Neocortex: Neuronal loss in the neocortex is most apparent in the mid-cortical layers associated with gliosis. Other prominent features may include: subcortical white matter gliosis, atrophy, Chaslin’s superficial cortical gliosis, and an increased amount of deposits of corpora amylacea. Following a seizure mild, focal leptomeningeal chronic inflammatory infiltrates may occur. An extensive amount of gliosis in the temporal lobe can be accompanied by dysplasia, this is now termed FCD type IIIa. Patients with hippocampal sclerosis that had post mortem studies add to the data that support more widespread neocortical pathology.

Thalamus:  In many studies involving MRI scans, post mortem examinations and experimental studies of epilepsy, it is shown that there is a volume reduction within the thalamus. This is thought to be in association with hippocampal and amygdala atrophy along with TLE. Atrophy of the thalamus is more often seen unilaterally and in association with atrophy of the fornix and mammillary bodies. The after effect of seizures and transneuronal degradation are proposed pathways for this occurrence.

Cerebellum: Cerebellar atrophy has a good amount of research that supports acquisition during the course of epilepsy rather than a predisposing factor for seizures. MRI volumetric studies reveal an increased amount of atrophy in patients with an established timeline of epilepsy versus patients who are newly diagnosed with the disease. This has been observed in both focal and generalized seizure studies in patients with Temporal Lobe Epilepsy noting a 4-6% reduction in size. During the sectioning of the brain after post mortem examination neuropathologists may find symmetric atrophy of the anterior or posterior lobes as shown in Picture 7. In less severe cases the damage can be seen in a limited amount in a folium.

Picture 7

References

1. (n.d.). Retrieved from https://www.epilepsydiagnosis.org/aetiology/hippocampal-sclerosis-overview.html
2. Black M, Graham DI. Sudden unexplained death in adults caused by intracranial pathology. J Clin Pathol 2002;55:44-50.
3. Blumcke I, Kistner I, Clusmann H, et al. Towards a clinico-pathological classification of granule cell dispersion in human mesial temporal lobe epilepsies. Acta Neuropathol 2009;117:535-544.
4. Chen JW, Naylor DE, Wasterlain CG. Advances in the pathophysiology of status epilepticus. Acta Neurologica  Scandinavica 2007;115:7-15.
5. Coras R, Pauli E, Li, J, et al. Differential influence of hippocampal subfields to memory formation: insights from patients with temporal lobe epilepsy. Brain 2014;137(Pt 7):1945-1957.
6. Corsellis JA, Bruton CJ. Neuropathology of status epilepticus in humans. Adv Neurol 1983;34:129-139.
7. Friedman D, Honig LS, Scarmeas N. Seizures and epilepsy in Alzheimer’s disease. CNS Neurosci Ther 2012;18:285- 294.
8. Fujikawa DG, Itabashi HH, Wu A, Shinmei SS. Status epilepticus-induced neuronal loss in humans without systemic  complications or epilepsy. Epilepsia 2000;41:981-991.
9. Haut SR, Veliskova J, Moshe SL. Susceptibility of immature and adult brains to seizure effects. Lancet Neurol 2004;3:608-617.
10. Jardim, Anaclara & Neves, Rafael & Caboclo, Luis & Lancellotti, Carmen & Martinez Marinho, Murilo & Centeno, Ricardo & Cavalheiro, Esper & Scorza, Carla & Márcia Targas Yacubian, Elza. (2012). Temporal lobe epilepsy with mesial temporal sclerosis: Hippocampal neuronal loss as a predictor of surgical outcome. Arquivos de neuro-psiquiatria. 70. 319-24. 10.1590/S0004-282X2012000500003.
11. Lurton D, El Bahh B, Sundstrom L, Rougier A. Granule cell dispersion is correlated with early epileptic events in human temporal lobe epilepsy. J Neurol Sci 1998;154:133-136.
12. Nashef L, So EL, Ryvlin P, Tomson T. Unifying the definitions of sudden unexpected death in epilepsy. Epilepsia 2012;53:227-233.
13. Pohlmann-Eden B, Gass A, Peters CN, Wennberg R, Blumcke I. Evolution of MRI changes and development of bilateral hippocampal sclerosis during long lasting generalised status epilepticus. J Neurol Neurosurg Psychiatry 2004;75:898-900.
14. Shields LB, Hunsaker DM, Hunsaker JC, 3rd, Parker JC, Jr. Sudden unexpected death in epilepsy: neuropathologic findings. Am J Forensic Med Pathol 2002;23:307-314.
15. Status Epilepticus. (n.d.). Retrieved from https://www.hopkinsmedicine.org/healthlibrary/conditions/nervous_system_disorders/status_epilepticus_134,42
16. Tan N, Urich H. Postictal cerebral hemiatrophy: with a contribution to the problem of crossed cerebellar atrophy. Acta Neuropathol (Berl) 1984;62:332-339.
17. Thom M, Liu JY, Thompson P, et al. Neurofibrillary tangle pathology and Braak staging in chronic epilepsy in relation to traumatic brain injury and hippocampal sclerosis: a post-mortem study. Brain 2011;134:2969-2981.
18. Thom M, Martinian L, Catarino C, et al. Bilateral reorganization of the dentate gyrus in hippocampal sclerosis: a postmortem study. Neurology 2009;73:1033-1040.
19. Thom M, Sisodiya SM, Beckett A, et al. Cytoarchitectural abnormalities in hippocampal sclerosis. J Neuropathol Exp Neurol 2002;61:510-519.
20. Thom M. Hippocampal sclerosis: progress since Sommer. Brain Pathol 2009;19:565-572.
21. Thom, M., Review: Hippocampal sclerosis in epilepsy: a neuropathology review. Neuropathol Appl Neurobiol 2014;40(5):520-543.
22. Thom, Maria & Liagkouras, Ioannis & Martinian, Lillian & Liu, Joan & Catarino, Claudia & M Sisodiya, Sanjay. (2012). Variability of sclerosis along the longitudinal hippocampal axis in epilepsy: A post mortem study. Epilepsy research. 102. 10.1016/j.eplepsyres.2012.04.015.
23. Traumatic Brain Injury | TBI. (2018, September 07). Retrieved from https://medlineplus.gov/traumaticbraininjury.html
24. Tsuchida TN, Barkovich AJ, Bollen AW, Hart AP, Ferriero DM. Childhood status epilepticus and excitotoxic neuronal injury. Pediatr Neurol 2007;36:253-257.