Chronic traumatic encephalopathy (CTE) is a progressive degenerative disease and progressive tauopathy of the brain caused by sequelae of head trauma and is common in athletes or patients with recurrent brain trauma such as boxers, professional soccer players, professional wrestlers, epilepsy patients, head bangers and domestic violence victims [1,2,3,4,5,6]. The name CTE was created in 1949 by Critchley. Critchley has defined this entity as a punch-drunk syndrome [7,8]. McKee et al. had made a pathological criteria and progressive pathological stages for CTE, that have been used for decades for the evaluation of neuropathology of this disease (Table 1) [3,7]. It is a progressive global atrophy of the brain including the cerebral hemispheres, medial temporal lobe, thalamus, mammillary bodies and brainstem, with ventriculomegaly and fenestration of cavum septum pellucidum. For decades CTE's neuropathology has been well established [1,3,5,7,9]. Microscopically, extensive p-tau immunoreactive neurofibrillary tangles (NFTs) and dot-like and thread-like neurites are patchy and irregularly distributed throughout the brain. The difference in NFT in other types of tauopathy and CTE is the distribution pattern of tau protein; 1) the significant involvement of the depth of the cerebral sulci, 2) the remarkable involvement of the periventricular, perivascular and subpial area, and 3) tau-immunoreactive astrocytes throughout the brain are exclusively found in CTE [6]. There has been a possibility of child and adolescent onset traumatic brain injury or chronic recurrent concussion. However, so far there was no autopsy-proven childhood-onset CTE cases [2,10]. We report a CTE autopsy case in a young man who died at the age of 36. He suffered severe dementia since he was 10 years old and had a history of febrile convulsions when he was 4 years old.

The deceased subject was 36-year-old male. In his age of 4 years old, he had a febrile convulsion. Since then, he lived without any special illness, but there was self-harmful behaviors such as frequent hitting his head on the wall and his mental condition gradually worsened. In the fourth grade in elementary school (10 years old), abnormal symptoms of mental illness, amnesia, cognitive impairment, poor orientation, and personality changes appeared. He was admitted to a psychiatric hospital with profound mental retardation. He continued to hurt. He spent a long time in the bed and died. His brain was donated to the brain bank of Seoul National University Hospital and an autopsy was carried out. The weight of brain was 1170 gm and the cerebrum was 19.1×17.5×17.4 cm. Grossly, the outer surface of the brain is unremarkable but corpus callosum and lateral geniculate body was diminished in size (Fig. 1). Bilateral hippocampi were atrophic. Neither infarction nor severe bleeding was observed. There was a slit-like discoloration that seemed to have small bleeding. The major cerebrovascular structures showed nonspecific changes. Microscopically, gray matter was mildly atrophic and white matter showed multifocal rarefaction and perivascular widening. There were only a few hemosiderin deposits around the blood vessels. GFAP immunostaining revealed reactive gliosis in the gray matter, especially in the subpial area and the molecular layer. P-tau (AT8, ThermoFisher, Waltham, USA, 1: 100 dilution) positive flame-shaped neurofibrillary tangles (NFTs) and neuropil threads were observed in the cerebral neocortex, hippocampus, basal ganglia and, thalamus, hypothalamus, entorhinal cortex, amygdala, nucleus basalis of Meynert and locus coeruleus, olfactory bulbs, mammillary bodies, especially around the blood vessels and the depth of the cerebral sulci (Fig. 2). The midbrain and the pons (substantia nigra and dorsal and median raphe nuclei) showed p-tau positive globular NFTs and neuropil threads. Tufted astrocytes did not exist. The olfactory bulb also showed p-tau positive neurons, neuropil threads and astrocytes, but cerebellum had no p-tau-positive cells or neurites. There was no significant pathology in the dentate gyrus. Lewy body or α-synuclein positive glia were not observed. A rare and tiny β-amyloid positive diffuse plaques were present in the motor and the sensory cortices. 4 repeat (4R) tau (Millipore, Ontario, Canada, X100) immunostaining revealed positivity in the perikarya of the neurons and neuropil threads, but their number was smaller than the number of the p-tau positive neurons and neuropil threads (Fig. 2). There was no TDP43 positive abnormal neurons. These findings were consistent with the pathology of the CTE.

CTE is a long-term neurological and neuropathologic sequelae associated with repetitive brain injury to athletes, epileptics, head bangers and the victims of abuse [5,6,9]. The exact incidence of CTE caused by repetitive head injury is unknown, but McKee et al. speculated that it will be much higher than we expect [6]. So far there was no autopsy-proven childhood-onset CTE cases in English literature. However, children and adolescents may have traumatic brain injury or chronic recurrent concussion [2,10].

The long-term effect of repetitive concussion of boxers and other athletes showed a unique pattern of pathology first described in 1973 by Corsellis et al. [11] It is characterized by irregularly distributed NFTs and NTs with phosphorylated and 4R tau protein accumulation especially in the depth of cerebral sulci and perivascular area [1,3,5,9]. Ghajari et al. reported the computerized modelling images of CTE [12]. The neuropathological diagnostic criteria and progressive staging pathology from I-IV were made in 2013 by McKee et al. [7] Grossly, CTE pathology involves atrophy of the cerebral cortex, especially the frontal and temporal lobes, diencephalon and mammillary bodies, and cavum septum pellucidum or septal fenestrations [7]. Microscopically, CTE is characterized by p-tau and 4R-tau positive neurofibrillary, neurites, and astrocytic tangles, which are patchy distributed around small blood vessels, and are preferentially distributed in the depth around the cerebral sulci. In well-established disease, tau pathology is most prominent in the frontal and temporal lobes, hippocampus, amygdala, and entorhinal cortex [7]. Abnormal accumulation of malformed TAR DNA-binding protein 43 (TDP-43) is also observed in CTE and is associated with CTE symptoms [3].

According to McKee et al., brain is not atrophic and the medial temporal lobe is preserved from p-tau pathology until stage 2. In stage 3, the amygdala, hippocampus, and entorhinal cortex show substantial p-tau pathology in addition to throughout the neocortex [3,7]. Our case also showed all the unique neuropathologies mentioned above. With McKee's stage, our case is compatible with stage 3, because NFTs present widespread throughout the cortex, especially at the depth of the cerebral sulci, hippocampus, entorhinal cortex, amygdala, nucleus basalis of Meynert and locus coeruleus, olfactory bulbs, hypothalamus, mammillary bodies, substantia nigra and dorsal and median raphe nuclei as well as mild cerebral atrophy and mild ventriculomegaly (Fig. 2).

Martland reported “punch drunk” in 1928, but the reported patient's brain pathology was different from that of CTE. Because the autopsy showed only punctate hemorrhage due to concussion injury [13]. In that paper, there was no mention about tau-positive neurofibrillary tangles. Critchley used the term CTE in 1949 as a punch drunk syndrome of traumatic brain injury, but it was before the tau protein was discovered [8].

Recently Edwards III et al. reported that transient brain injury (TBI) can contribute to the formation of misfolded oligopeptides, mainly tau and β-amyloid, in studies on experimental animal models [14].

CTE is a distinct clinical and pathological entity, which is tau related neurodegeneration. Symptoms of CTEs include insidious onset of deterioration of attention, loss of concentration, memory loss, disorientation, sometimes dizziness and headaches. Gradual deterioration can lead to additional symptoms such as insufficient insight, poor judgment, and dementia. In severe cases, the movement of the muscles is slowed by propulsion gait, and masked facies, stuttering, tremors, vertigo, and hearing loss appear [6].

Because the clinical features may be similar to Alzheimer's disease, differential diagnosis is required [2,3]. In our case, we did not see any β-amyloid positive neuritic plaques characteristic of Alzheimer's disease (AD). We found rare and tiny plaques in the motor and sensory cortex. Another differential diagnosis is neurofibrillary tangle-predominant dementia (NFTPD or Tangle-only dementia), which also lacks β-amyloid positive neuritic plaques. NFTPD is clinically and pathologically different from ours. NFTPD is a late-onset progressive dementia with a shorter duration than AD and occurs in the elderly with an average age of 79.7 years [15]. Pathologically, the NFT distribution pattern of p-tau protein in NFTPD is similar to the distribution, spread and severity of AD. However, it does not prefer localization to the perivascular, periventricular and subpial regions or the depth of the cerebral sulci [15]. P-tau positive astrocytic tangles are not present in AD. And the disease that causes similar symptoms in children is lysosomal storage disease like Gangliosidosis type II and neuronal ceroid lipofuscinosis. However, in our case, there was no lipid vacuolization, severe neuronal loss and diffuse astrogliosis with macrophage accumulation which is characteristic of this disease. So our case does not fit into these diseases.

Studies of clinicopathological correlation, such as the Understanding Neurologic Injury and Traumatic Encephalopathy (UNITE) Study, can help identify sensitive clinical features of CTE pathology [16]. Future improvements for CTE diagnosis require prospective studies including neuropsychological tests using imaging and fluid biomarkers.

The problem of our case is that there is no 6-year history between febrile seizures and dementia. However, symptoms of CTE, such as mental retardation and behavioral problems, was already appeared when he was admitted to a mental hospital for symptoms in his age of 10 and showed patterns of self-harm. He lived in bed for a long time and died at the age of 36. Autopsy of our case revealed a typical pathology of CTE.

Here we report a case of CTE that was proven by autopsy. We report this case because it is interesting that CTE can be diagnosed only by autopsy without a clear history of repeated brain damage. This example shows that CTE can occur at young age and that even children can experience CTE dementia.