The Neuropathological and Clinical Diagnostic Criteria of Chronic Traumatic Encephalopathy: A Critical Examination in Relation to Other Neurodegenerative Diseases

Abstract

This work critically reviews chronic traumatic encephalopathy (CTE), with a specific focus on the single criterion necessary and sufficient for diagnosis. Herein, CTE is compared to other well-established neurodegenerative entities including Alzheimer’s disease and dementia with Lewy bodies. Each neurodegenerative disorder is reviewed in five pertinent areas: 1) historical perspective, 2) guideline formation process, 3) clinical diagnostic criteria, 4) pathological diagnostic criteria, and 5) validation of previously described diagnostic criteria (e.g., sensitivity and specificity). These comparisons indicate that CTE is a disease in the earliest stages of formation and has yet to undergo rigorous development and refinement similar to other neurodegenerative diseases. Suggested future revisions to the diagnostic criterion of CTE include establishing a lower threshold for accumulation of pathology, as well as accounting for the presence of concomitant neuropathology and comorbid neurodegenerative disorders. Currently, while initial efforts have been attempted, agreed upon antemortem clinical criteria do not exist. As has been the scientific standard with similar neurodegenerative disorders, antemortem diagnostic guidelines should first be refined through subcommittees of neuroscientists from diverse institutional backgrounds with a subclassification of levels of diagnostic certainty (possible, probably, and definite). Validation studies should then assess the predictive value and accuracy of proposed antemortem diagnostic criteria in relation to potential pathological criteria.

Keywords

Chronic traumatic encephalopathy concussion football neurodegenerative diseases sports traumatic brain injury

INTRODUCTION

The potential for long-term neurologic impairment and chronic traumatic encephalopathy (CTE) from contact sport participation has taken hold in the public narrative. The perceived risk of CTE in youth, high school, and collegiate athletes exposed to concussive and sub-concussive impacts has resulted in policy changes at the local, state, and national level [1, 2]. Based on a small number of case reports and convenience samples [3, 4], it appears to have become widely accepted that repetitive neurotrauma of any kind suffered in sport or military activities has deleterious effects on long-term neurologic function.

Players, parents, coaches, educators, and healthcare providers represent key stakeholders impacted by this potential relationship, which has far-reaching public health implications. Experts have postulated that we may be prematurely assuming a causal relationship between sporting head impacts and neurodegeneration [5 –7]. Popularized accounts of CTE in professional athletes [8, 9] may be responsible for the current public and scientific disconnect.

A more measured and scientific approach calls for a step back in order to better understand all that CTE represents. Relevant questions include: how long has CTE existed? How accurately can a diagnosis be made ante- and postmortem? How does CTE compare to other neurodegenerative disorders? Though first mentioned in 1928 by Martland, the 21st century version of CTE bears little resemblance to its original description. Recent reports have acknowledged a “classic” [3] versus “modern” form of CTE [10]. Currently, the only pathologic criterion required to diagnose CTE is accumulation of a single focus of abnormally phosphorylated tau protein (p-tau) in an irregular pattern at the depths of one or more cortical sulci. Moreover, few to no qualifiers in the current diagnostic criteria [10] are given for the fact that abnormal tau proteins are also found in other neurodegenerative processes or if CTE is seen in tandem with other neurodegenerative diseases [6 , 11].

Upon further examination, one finds that modern CTE is a young, complex, and often confounded disorder. A recent review by Iverson and colleagues [12] called for the separation of CTE neuropathology from the various clinical correlates often attributed to the disorder. The authors meticulously highlight that an extensive variety of cognitive, behavioral, and emotional sequelae are often attributed to CTE without observational relation to neuropathology, and narrowing of the clinical symptoms associated with its presentation is needed [13, 14]. Moreover, a recent study reported that the wide range of mood, behavioral, and cognitive symptoms were reported at similar rates among CTE-negative individuals (97.14%) compared with CTE-positive individuals (96.67%) [15].

To fully understand the science of CTE amidst the new wave of public health concern, a critical examination is warranted. The objective of the current review was to critically compare CTE to similar neurodegenerative diseases in the following areas: 1) historical diagnosis, 2) guideline formation, 3) current clinical and 4) pathologic diagnostic criteria, and 5) validation studies. By better understanding CTE among its peer neurodegenerative disorders, the public and scientific community can begin to contextualize the implications of CTE.

METHODS

To compare diagnostic criteria across multiple neurodegenerative disorders, a selective review of each disease entity was performed. Each neurodegenerative disorder selected for this review has well-established, consensus-driven diagnostic criteria that have undergone at least one revision. Secondly, while all of these diseases are neurodegenerative in nature, these diagnoses possess some diversity in epidemiology, neuropathology, laboratory/neuroimaging techniques involved in arriving at the diagnosis, and clinical presentation/symptomology, which allows for determination of common elements often utilized in development of diagnostic criteria. In addition to CTE, the two neurodegenerative disorders included for comparison were: Alzheimer’s disease (AD) and dementia with Lewy bodies (DLB).

For each degenerative disorder, five distinct areas were reviewed:

Historical diagnosis – first characterization of the disease until formation of initial diagnostic guidelines, including early case reports and initial case series proposing the clinical entity.

Guideline formation process – description of how the first official diagnostic guidelines were proposed, including number of iterations, number of experts, types of specialties included, different institutions involved, number of subcommittees, and how committees were assembled.

Clinical diagnostic criteria – description of current clinical criteria needed to establish a diagnosis.

Pathologic diagnostic criteria – description of current pathologic findings needed to establish a diagnosis.

Validation studies – discussion of validation studies that have examined the efficacy of the current diagnostic guidelines (e.g., correlational studies, sensitivity and specificity, etc.).

In completing each of the five sections, selective reviews were performed. In the absence of a systematic review for each neuropathologic entity, the largest studies and those cited frequently as guideline formations were included. For validation studies, commonly cited efforts with large sample sizes were chosen. References of all current and former guidelines were searched for relevant articles that may have been missed.

RESULTS

Alzheimer’s disease

Historical diagnosis

In 1906, Alois Alzheimer first described AD pathology, including neurofibrillary tangles (NFTs) and neuritic plaques, in the autopsy case report of Auguste Deter [16]. Interestingly, his index patient was in her early 50s and presented with primary psychiatric symptoms. The 1980s marked the initial establishment of a standardized approach to diagnosing AD. In the fall of 1983, 23 experts met as part of the first National Institute of Neurological and Communicative Disorders and Stroke (NINCDS) and the Alzheimer’s Disease and Related Disorders Association (ADRDA) panel [17]. The group’s purpose was to establish clinical diagnostic criteria for AD. Their efforts yielded specific clinical criteria (published in 1984) that estimated probabilities of “possible”, “probable”, and “definite” AD. The number of cases studied to arrive at such proposed criteria was not mentioned. However, at this very early stage, the group responsibly noted that such criteria could not be fully operationalized due to insufficient knowledge about the disease. With the first AD clinical diagnostic criteria proposed, the experts regarded the criteria as tentative and subject to modification pending the emergence of empirical studies. In 1985, a group of 37 experts, convened by the same NINCDS and ADRDA aimed to further identify the most pressing areas for study, and to outline clinical and technical issues within research on AD diagnosis. The multi-disciplinary group was organized into 6 panels representing neurochemistry, neuropathology, neuroradiology, neurology, neuropsychology, and psychiatry [18].

That same year (1985), the original pathologic guidelines for diagnosing AD were published. Referred to as the Khachaturian criteria, these neuropathological guidelines were based on age-dependent numerical cutoffs of senile plaque counts and the density of plaques per field. In 1986, 24 institutes (and under the auspices of the National Institute of Aging [NIA]) prospectively followed 1,094 patients, collecting information such as clinical symptoms, neuropsychological symptoms, and neuropathologic findings. This approach set the standard for establishing a diagnosis of AD for many years to come. These criteria were subsequently challenged in the following decade due to a lack of clinical correlation with the Khachaturian criteria— postmortem studies show that some individuals with pathologic findings of AD often had no clinical manifestations of disease [19, 20]. Additional critiques of the Khachaturian criteria were that while based on the density of neuritic plaques, they did not consider NFTs, nor did they consider correlations with the varying stages of clinical dementia in AD patients.

Guideline formation process

Following publication of the Khachaturian criteria, findings from further clinical-pathological studies facilitated the criteria published by the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) in 1991. A group of neuropathologists from nine university medical centers were tasked with creating a neuropathological protocol consisting of an illustrated guidebook and data entry form to be linked with clinical information on demented and non-demented patients. From 15 CERAD centers, neuropathology data from 142 consecutive brain autopsies of patients clinically diagnosed as having probable AD, and from 8 subjects deemed as having no evidence of cognitive impairments (150 total), were analyzed and validated in the formation of a standardized protocol for the neuropathologic assessment of AD [21]. CERAD criteria were based on the semi-quantitation of neuritic plaque density in key topographic regions of the brain. The semi-quantitated plaque score, in consideration of patient age, would then permit estimation of the likelihood that a patient’s antemortem dementia was in fact due to AD, versus other dementing illnesses. The CERAD effort revolved solely around pathologic diagnosis, no clinical correlation, and also had a poor balance of cases (142) to controls (8).

Recognizing the need for a comprehensive, unified clinical-pathologic criteria, Braak and Braak [22] published a combined staging system based on the distribution of NFTs in key topographic regions of the brain that demonstrated differential and hierarchical vulnerability to AD. This system permitted classification of AD into six stages, correlating to three groups of progressive symptomatic manifestation (discussed below). This Braak and Braak study examined 83 brains, which included patients with a clinical diagnosis of dementia, but whose neuropathologic exam did not show features of developed AD (“demented old-aged patients”), as well as patients with a clinical diagnosis of dementia who did demonstrate sufficient isocortical neurofibrillary changes to substantiate the clinical diagnosis of AD [22].

Subsequently in 1997, 17 experts from 16 different U.S. and European institutions met as part of the NIA and Reagan Institute Working Group. Following a two-day workshop, recommendations for the postmortem assessment of AD were published. Incorporating both CERAD and Braak and Braak criteria (described above), protocols were proposed that considered cases encountered in daily routine practice settings, as well as in AD research settings [23].

Clinical diagnostic criteria

The clinical diagnostic criteria for AD have traditionally revolved around the progressive dementia, namely memory loss and global cognitive decline. However, it has been increasingly recognized over the past three decades that symptomatology and neuropathologic findings exist on a heterogeneous spectrum and, while pure AD may be present, clinical and pathologic overlap with other neurodegenerative entities may occur [24]. In recent years, mild cognitive impairment (MCI) has been described as a form of prodromal AD, and it has been acknowledged that a better understanding of conversion from MCI to AD is needed. Conversely, AD neuropathologic changes in the brain may occur in the absence of clinical cognitive impairment and may possibly reflect AD early in clinical evolution [25].

Revision of the 1984 NINCDS-ADRDA clinical AD criteria occurred in 2009 when the NIA and Alzheimer’s Association sponsored an international group from academia and industry. The guideline formation involved 4 separate committees tasked with formulating diagnostic criteria for both AD dementia and symptomatic predementia of AD, revised pathologic criteria for AD, and a research agenda to improve knowledge of asymptomatic preclinical AD. The process consisted of committee members meeting in person and via conference calls. Feedback gathered from public symposia held during the 2010 International Conference on Alzheimer’s Disease were incorporated. A fifth subcommittee was tasked with reviewing biomarker recommendations and harmonizing discussion across workgroups in the fall 2010. The final, peer-reviewed documents were published in early 2011.

The revised NINCDS-ADRDA criteria expanded beyond memory loss to include a diagnosis of all-cause dementia, which required impaired abilities in at least two cognitive domains (memory, executive functioning, visuospatial abilities, language, and/or social comportment) established by clinical examination and confirmed by neuropsychological testing or other objective cognitive assessment. Deficits must impact social or occupational functioning, represent a decline from a prior level of neurobehavioral functioning, and symptoms cannot be due to delirium or psychiatric conditions. Probable AD criteria additionally include an insidious onset, worsening of cognitive symptoms, and impairment in a core cognitive domain (i.e., memory, language, visuospatial abilities, and executive functioning) and one other cognitive domain. Features of cerebrovascular disease, DLB, and frontotemporal dementia must be absent. Finally, clinical symptoms cannot be attributable to another medical condition or medication. The possible AD distinction is given when all other criteria for probable AD are met, but the course is atypical or there is an etiologically mixed presentation [26].

Biomarkers were omitted from clinical diagnostic guidelines but included for research purposes. Probable AD dementia with evidence of the AD pathophysiological process is applied when biomarkers or autopsy confirms AD pathology in an individual meeting clinical criteria for probable AD. Individuals with AD-positive biomarkers or autopsy confirmed AD pathology, but who meet criteria for a non-AD dementia are designated as possible AD dementia with evidence of the AD pathophysiological process [17].

Prior to the first consensus meeting in 1983 described above, the Diagnostic and Statistical Manual of the American Psychological Association (DSM) independently attempted to describing clinical criteria for AD in its third edition, published in 1980 [27]. The diagnosis of dementia was solely based on clinical presentation and required impairment in memory and 1 other cognitive domain. Currently in its fifth edition published in May of 2013 (DSM-5) and its third iteration of AD clinical diagnostic criteria, these revisions classify AD as a subtype of Major Neurocognitive Disorder [28]. The new criteria mirror those outlined by NINCDS-ADRDA for all-cause dementia; however, impairment in only one cognitive domain is required (complex attention, executive function, learning and memory, language, perceptual-motor, or social cognition). The DSM-5 defers description of subtype-specific cognitive, behavioral, and functional symptoms to the respective consensus group criteria.

Pathologic diagnostic criteria

The most recently established criteria and guidelines for diagnosing and reporting AD were published in 2012, approximately 27 years after publication of the original Khachaturian criteria. A total of 562 cases selected from autopsies collected in the National Alzheimer’s Coordinating Center Uniform Data Set from 2005 to 2010 were stratified and analyzed on the basis of CERAD scores, Braak stage, and Clinical Dementia Rating Scale (CDR) by experts from the U.S. and Europe. These cases came from approximately 30 NIA-funded AD Centers across the U.S. Reflecting the collaboration of the NIA and the Alzheimer’s Association, the standardized criteria proposed recognize pre-clinical stages of AD, and incorporate three key pathologic parameters into the AD assessment. Referred to as the “ABC” score, they reflect the amyloid-β (Aβ) plaque score based on Thal phases (A), the NFT score based on Braak and Braak stages (B), and the CERAD neuritic plaque score (C), with each parameter assessed on a scale of 0 to 3 and reported regardless of clinical history [29].

ABC scores are used to classify “None,” “Low,” “Intermediate,” and “High” pathologic evidence of AD. None is designated in the absence of Aβ plaques (no Thal phase). When Aβ deposition is confined to the neocortical or allocortical regions (Thal phases 1-2) and neuritic plaques are infrequent (CERAD score of none to sparse), the Low classification is used. Low also captures Braak stages 0 to II. Intermediate designation is given to Braak stages III to VI, except for those with Thal phases 4-5 and CERAD scores of moderate to frequent, which are instead classified as High [29].

Validation studies

The above discussion illustrates how diagnostic criteria for AD evolved progressively over several decades, not only with accumulation of clinical pathologic correlative data, but also in applying and building upon the progressive foundation of previous working groups. This, in effect, forms the basis of continuous validation, which started in 1983 by a group of experts convened by the NINCDS and ADRDA. Throughout the last 35 years, with the collaboration of additional work groups and institutes, accrual and maturation of robust, generalizable patient registries led way to enhancement of diagnostic criteria, thus reflecting further validation and refinement of existing criteria and standards.

A synthesis review of several meta-analyses revealed that the original NINCDS-ADRDA clinical criteria confirmed against neuropathological AD diagnosis yielded variable sensitivity (76–93%) and specificity (55–91%) [30]. Using the updated NINCDS-ADRDA criteria described above, preliminary studies have revealed good diagnostic accuracy of 90% or greater [31]. Regarding efficacy of neuropathological criteria, evaluation by 10 AD Center neuropathologists revealed excellent agreement in ABC score classification (weighted κ= 0.88) [32].

Dementia with Lewy bodies

Historical diagnosis

DLB was first characterized by Fritz Jakob Heinrich Lewy in 1912 while studying Parkinson’s disease (PD) neuropathology in Munich, Germany [33]. At this time, Lewy recorded cellular inclusions outside of the substantia nigra (dorsal motor nucleus of the vagus nerve, the nucleus basalis of Meynert, and some thalamic nuclei of PD patients) that were characteristic of PD [34]. The inclusions were first termed Lewy bodies (corps de Lewy) in 1919 by Konstantin Nikolaevich Tretiakoff, who postulated that nerve cell loss due to these inclusions may be associated with motor symptoms of tremor and rigidity [35]. The further connection between Lewy body (LB) pathology and current clinical features of DLB was presented by Okazaki and colleagues in 1961, who reported postmortem findings of diffuse LB-like pathology in two individuals with dementia [36]. One of these cases included a 69-year-old male, who exhibited gradual deterioration in mental status, along with visual hallucinations, and flexion rigidity. Postmortem examination revealed numerous LBs in the brain stem and diencephalon; however, what was particularly unique about this case was the cerebral cortical presence of intracytoplasmic, eosinophilic, and argentophilic inclusions (i.e., Lewy-like-bodies).

A thoroughly presented case study by Kosaka et al. [37] could be considered as the first report of DLB, listed as an “unclassifiable dementia” in an autopsy of a 65-year-old female with gradual progression of memory difficulties, rigidity, and psychomotor “restlessness.” Kosaka and colleagues [38] then published a case series of three individuals demonstrating similar subcortical (basal ganglia) and cortical distribution of LBs, most especially in the anterior frontal, temporal, insular, and cingulate cortex. Two years later, Kosaka published a series of 20 individuals with similar neuropathological distributions where he introduced the term Lewy body disease [39] and proposed three main classifications based on distribution location: 1) brain stem type (now corresponding best with PD), 2) transitional type, and 3) diffuse type. Prior to the current day diagnostic term of DLB, the disease was known as “diffuse Lewy body disease” [40] and cerebral type of Lewy body disease [41].

The next era or major step in the formation of DLB as a distinct diagnostic entity occurred approximately a decade later with the formation of the first consensus guidelines for clinical and pathological diagnosis of DLB as part of the First International Workshop of the Consortium on Dementia with Lewy Bodies [42]. Given the heterogeneity of findings from case reports described above, the consortium set out to establish concrete diagnostic criteria for the clinical symptoms (antemortem) and postmortem identification of DLB. Additionally, the consortium established strict pathological criteria for a diagnosis of DLB, which included guidelines regarding: 1) the morphology, 2) sampling distribution, and 3) and frequency of LBs and a diagnostic rating protocol (also described below). Simultaneously with the formation of the first consortium, another significant event in the evolution of Lewy body disease was also taking place, as α-synuclein was identified as the hallmark of the disease by Spillantini and associates [43], who demonstrated similar staining of α-synuclein in LBs across idiopathic PD and DLB patients.

Guideline formation process

The initial collective attempt at establishing consensus diagnostic criteria for DLB first occurred in 1995 as part of the First International Workshop of the Consortium on DLB mentioned above. This consensus meeting attempted to improve upon and synthesize two prior efforts to operationalize criteria for DLB as a distinct clinical and neuropathological diagnostic entity [44, 45]. This earlier attempt involved retrospective reviews of 21 neuropathologically confirmed LB cases and 37 neuropathologically confirmed AD cases as a comparison group; however, this effort towards operationalization was limited by a lack of institutional and subject matter diversity. Attempting to expand upon this initial attempt, the first consensus meeting consisted of 26 subject matter experts across diverse disciplines (neurology, neuropathology, psychiatry, neuropsychology, neuroscience) from 21 different institutions. While the number of cases evaluated was not listed, the initial consensus criteria aimed to critically evaluate prior attempts to establish criteria, as well as peer-reviewed studies based on these criteria. Since the original consensus meeting in 1995, three additional meetings have taken place as part of the continual refinement and improvement of the DLB diagnostic criteria [46 –48]. Review of all consensus criteria and changes with each subsequent iteration is beyond the scope of the current paper. The following content within the section will review the process in which the most recent guidelines were formulated.

The most recent consensus meeting was conducted in 2015 and organized by the Mayo School of Continuous Professional Development (MSCPD). Meeting support and co-sponsorship included organizations such as the Lewy Body Dementia Association, the Lewy Body Society, Alzheimer’s Association, the National Institute on Aging, and the National Institute on Neurologic Disease and Stroke. Sixty-three invited multi-disciplinary subject matter experts from over 40 institutions/organizations and 10 countries attended the consortium and participated in one or more working groups in some capacity during the international DLB conference.

The process of guideline formation was completely transparent and involved four stages of development; this included 1) pre-conference scoping, 2) pre-conference working groups, 3) conference activity, and 4) post-conference activity. Pre-conference scoping involved two individuals from two different institutions and countries inviting a group of international experts to review the current literature around DLB, as well as determining the agenda for the DLB consortium meeting. Based on these pertinent topics, four pre-conference working groups consisting of international experts were established in order to review the current consensus guidelines. The four work groups, which met multiple times by email or teleconference, included areas of 1) clinical diagnosis, 2) clinical management and trial design, 3) pathology, genetics, biofluids and basic science and 4) global harmonization. In examining these four areas, work groups were instructed to adhere to a consistent review framework of 1) identifying elements in need of amendment, and those not needing to be changed; 2) potential new topics for inclusion, and 3) identify anticipated future developments.

Guideline formation continued through conference activities, in which the four working groups conducted live review sessions. An educational program also took place in conjunction with the consensus meeting and guideline formation, and any individual attending that portion of the event was able to register and participate in the discussions of the four different working groups. Drafts of recommendations for each of the four live review sessions were constructed and led by chairs of working groups. Following the conference (post-conference activities), working group chairs circulated final reports among members for review and solicitation of feedback. Final drafts of working groups were submitted to the conference chair, who drafted an integrated consensus report. The draft of the integrated report was circulated to members for review and a final revised version was submitted for publication [48].

Clinical diagnostic criteria

As highlighted above, the consensus guidelines for the diagnosis and management of DLB have undergone three iterations and the fourth edition, most recently published in 2017, maintains several aspects of the previous versions with several modifications/additions [48]. According to the consensus, a diagnosis of DLB requires the presence of a “progressive cognitive decline of sufficient magnitude to interfere with normal social or occupational functions, or with usual daily activities” [48]. Levels of diagnostic certainty are delineated as probable and possible DLB, similar to AD. Specific combinations of core clinical features, supportive clinical features, indicative biomarkers, and supportive biomarkers can be used in determining a diagnosis at different levels of diagnostic certainty.

A diagnosis of probable DLB requires the presence of two or more core clinical features with or without the additional presence of indicative biomarkers. A diagnosis of probable DLB can also be made for instances in which there is the presence of a one core clinical feature and one or more indicative biomarker. According to the guidelines, probable DLB should not be diagnosed solely on the presence of biomarkers. A diagnosis of possible DLB can be made if only one core clinical feature is present with no indicative biomarker evidence. Possible DLB can also be diagnosed if one or more indicative biomarkers are present, but no core clinical features. Below, we further describe the core clinical features, indicative biomarkers, and supportive features.

Core clinical features

Four core clinical features of DLB were listed as: 1) fluctuating cognition with pronounced variations in attention and alertness, 2) recurrent visual hallucinations that are typically well formed, 3) REM sleep behavior disorder (RBD; may precede cognitive decline), and 4) one or more spontaneous cardinal features of parkinsonism (bradykinesia, rest tremor, or rigidity). Aspects of each of the core clinical features, as they relate to DLB, are reviewed in depth by the consensus guidelines. Additionally, likelihood of each symptom is provided based on prevalence studies (e.g., visual hallucinations observed in 80% of DLB patients and parkinsonian features in 85%). Suggestions for specific instruments/inventories and laboratory studies that can aid in the identification of core clinical features are also provided.

Indicative biomarkers

Three indicative biomarkers in the diagnosis of DLB which are provided include: 1) SPECT or PET demonstrating reduced dopamine active transporter (DAT) uptake in basal ganglia; 2) Low uptake on 123iodine-MIBG myocardial scintigraphy; 3) Polysomnographic confirmation of REM sleep without atonia. These biomarkers are classified as indicative, rather than supportive, due to their clinical utility and diagnostic sensitivity and specificity. For example, PET reduced DAT uptake in the basal ganglia on PET imagining has demonstrated effective sensitivity (78%) and specificity (90%) in differentiating DLB from AD [49]. Similarly, reduced uptake on ¹²³iodine-MIBG myocardial scintigraphy has also demonstrated clinically useful sensitivity (77%) and specificity in differentiating DLB from AD (94%) [50].

Supportive DLB features

Supportive clinical features and biomarkers are also provided by the guidelines in order to aid clinicians in diagnosing DLB. These indicators were not included as part of the required criteria, as the experts responsibly acknowledged the lack of diagnostic specificity [51 –53]. Supportive clinical features that are commonly observed in DLB include 1) general sleep disturbance (hyper- and hyposomnia), 2) transient episodes of unresponsiveness, and 3) severe sensitivity to antipsychotic medication. Supportive biomarkers are also provided and include 1) relative preservation of medial temporal lobe structures on CT/MRI scan, 2) generalized low uptake on SPECT/PET perfusion/metabolism scan with reduced occipital activity accompanied by the cingulate island sign on FDG-PET imaging, and 3) prominent posterior slow-wave activity on EEG with periodic fluctuations in the pre-alpha/theta range. Additionally, clinical features that indicate decreased likelihood of DLB presence are discussed in the guidelines in order to assist with differential diagnosis. For example, when parkinsonism occurs at or after the onset of dementia, or is the sole clinical feature, DLB is regarded as less likely.

Pathologic diagnostic criteria

The First International Workshop of the Consortium on DLB in 1995 [42] provided a descriptive morphology of LBs, highlighting conventional characteristics and locations of the disease pathology. For example, the authors distinguish “classic” LB inclusions with a hyaline core and pale halo typically seen in the brain stem from cortical LBs, which typically involves less well-defined spherical inclusions seen in cortical neurons. Further, the guidelines provide direction for effective detection of LBs such as effective histological stains (i.e., brainstem LB is H-E and for cortical LB H-E and/or ubiquitin with tau immunostaining to differentiate cortical LBs from small tangles). Classification of co-occurring pathology (i.e., CERAD guidelines for AD and vascular pathology, especially as they relate to the presence of LBs) is also discussed. Since the initial guidelines, immunohistochemistry for α-synuclein has been a major advancement in this area and is now the preferred method of identification of LBs [43].

Comprehensive guidelines for brain sampling were put forth by the original guidelines, which included sampling of specific cortical sections at appropriate coronal levels (see McKeith et al. [42] for full description of sampling guidelines). The CERAD guidelines for brainstem sampling were adopted. The protocol for frequency scoring of LBs involves a numerical score ranging from 0 to 2 for each region sampled. Scores are assigned 0 for no observed LB count, 1 for up to 5 LBs, and 2 for greater than 5 LBs. Scores for each area are then summed based on distribution or density classification for five main cortical regions. These summed scores provide the basis for a severity grading system across three subtypes (neocortical, limbic, brainstem-predominant), which correspond with Kosaka’s original subtypes [41].

Since the formation of the initial guidelines, subtypes of LB pathology have expanded to five classifications (diffuse neocortical, limbic [transitional], brainstem-predominant, amygdala-predominant, and olfactory bulb only) [48]. Additionally, the most recent guidelines provide a system of assessment, which aids in the attribution of dementia-related symptoms to LB pathology findings. Specifically, the system grades the likelihood that postmortem pathology findings are associated with DLB clinical observations as low, intermediate, and high. The five aforementioned LB subtypes are classified based on the location of recorded pathology and the presence of AD pathology based on the NIA-AA guidelines described above. These two factors determine the low, intermediate, and high classification of pathology as predictor of clinical presentation (probable DLB). For example, diffuse neocortical LB pathology in the presence of none/low AD pathology (Braak stage 0– II) would be classified as “high,” essentially predicting a high likelihood of antemortem diagnosis of probable DLB, assumed to be associated with the observed pathology.

Validation studies

Given that the most current guidelines were published very recently, validation studies have yet to be performed. However, multiple validation studies for all prior DLB consensus criteria have been performed in order to continually refine and sharpen each iteration of diagnostic criteria. A recent systematic review and meta-analysis by Rizzo and colleagues [54] investigated the course of DLB criteria, examining changes in accuracy criteria with each subsequent consensus conference. Pooled sensitivity, specificity, and accuracy were calculated from 22 selected studies that consisted of 1,585 patients. Based on the systematic review [54], the first consensus criteria for possible DLB [42] yielded pooled sensitivity, specificity, and accuracy of 72.3%, 64.3%, and 66%, respectively. Pooled sensitivity, specificity, and accuracy for probable DLB were 48.6%, 88%, and 79.2%, respectively. The revised iteration of guidelines’ criteria for possible DLB yielded pooled sensitivity, specificity, and accuracy of 91.3%, 66.7%, and 81.6%, respectively. Studies examining probable DLB criteria produced pooled sensitivity, specificity, and accuracy of 88.3%, 80.8%, and 90.7%, respectively. Based on the pooled diagnostic metrics, changes and refinements to the first consensus guidelines improved overall diagnostic accuracy of the second iteration.

Chronic traumatic encephalopathy

Historical diagnosis

Many believe the first mention of long-term brain damage induced by contact sport was in 1928 by Harrison Martland who published “Punch Drunk” in the Journal of American Medical Association, detailing the neurocognitive and neurobehavioral condition of 23 professional boxers [55]. The term CTE was coined in 1940 by Bowman and Blau in the case report of a 28-year-old boxer [56]. Roberts et al. [57] in 1969 and Corsellis et al. [3] in 1973 provided the foundation for what is now referred to as “classic” CTE. From a list of 16,781 retired boxers in the UK, Roberts selected an age-stratified random sample of 250; he was able to locate and clinically examine 224 of these men, and found that 17% exhibited some form of neurologic dysfunction (11% with a mild form, and 6% with a severe form). Four years later, Corsellis et al. published the first gross, neuropathologic criteria for diagnosing CTE based on autopsies of 15 boxers, which included: cerebral atrophy, enlarged lateral/third ventricles, thinning of the corpus collosum, a cavum septum pellucidum with fenestrations, cerebellar scarring, and agyrophilic neurofibrillary degeneration. Roberts et al. [58] later examined the 15 specimens used by Corsellis and nearly all had amyloid beta deposition suggestive of AD, confounding aspects of the initial formulated diagnostic criteria.

After a nearly 30-year period, it was not until 2005 that a second wave of CTE studies came about, but for the first time in non-boxing athletes, subsequently dubbed “modern” CTE [6]. Omalu and colleagues [59] published case reports of former NFL players in 2005 [59], 2006 [60], and 2010 [61], and one former wrestler in 2010 [62] presenting with CTE; however, the pathological and clinical presentations were substantially different from the earlier accounts published by Roberts in 1969 and Corsellis and colleagues in 1973. As CTE in non-boxers became more acknowledged, the “modern” CTE entity was established. Though the “classic” and “modern” CTE entities are not formally recognized in any guidelines, further discussion is necessary to contextualize the history of modern CTE.

Dr. Ann McKee and colleagues of the Boston University Center for the Study of Traumatic Encephalopathy (BU-CSTE) group published their first case series of modern CTE in 2009 [63]. The BU-CSTE group proposed the first contemporary CTE staging system based on four primary criteria: 1) perivascular foci of p-tau immunoreactive NFTs and astrocytic tangles (ATs); 2) irregular cortical distribution of p-tau immunoreactive NFTs and ATs at the depth of cerebral sulci; 3) clusters of subpial and periventricular NFTs in the cerebral cortex, diencephalon, basal ganglia, and brainstem; and 4) NFTs in the cerebral cortex preferentially in the superficial layers [63]. Based on these criteria and autopsies of 51 individuals (convenience sample of athletes and civilians) with a history of mTBI, of which 80% possessed the criteria described above, McKee and colleagues described a progressive, 4-tiered neuropathologic staging system of CTE, and each stage had a corresponding clinical signs and symptoms description.

As the BU-CSTE diagnostic criteria was proposed, a dissimilar set of diagnostic criteria was proposed by Omalu et al. [64] based on autopsies of 17 contact sport athletes. Omalu et al. proposed four non-progressive, distinct neuropathologic phenotypes of CTE. Unlike McKee’s classification, “stages” of disease progression were not described. Additionally, Omalu, did not include clinical presentation as part of his criteria, and did not propose that clinical signs and symptoms were associated with a particular phenotype.

Guideline formation process

The first consensus guidelines for CTE were first published in 2016 based on a National Institute of Health (NIH) consensus meeting aimed towards determining the nature of CTE and long-term TBI outcomes [10]. Central to this aim was to determine whether CTE was a distinctive tauopathy that could be reliably distinguished from other tauopathies. Four neuropathologists not involved in the examination and grading of subjects (two from the BU-CSTE group, one from Mt. Sinai, and one from the Mayo Clinic Jacksonville) selected 25 cases of various tauopathies, all of which originated from the BU-CSTE brain bank. Seven neuropathologists from seven different institutions (Washington University School of Medicine, Mayo Clinic Jacksonville, Brigham and Women’s Hospital, University of Washington School of Medicine, Uniformed Services University of Health Sciences, Boston University, and Columbia University) rated each specimen as 1-unsure, 2-possible, 3-probably, or 4-definite after microscopic inspection. All cases (according to the staging criteria of BU-CSTE) were late-stage disease, with 8 being presumptive stage IV CTE and 2 being presumptive stage III CTE. No stage I or stage II cases were evaluated. When assessing agreement for any tauopathy in the 25 cases, which included 6 other tauopathy diagnoses (AD, argyophilic grain disease, corticobasal degeneration, Guamanian Parkinson’s dementia complex, primary age-related tauopathy, and progressive supranuclear palsy), the agreement level was 67%, meaning that 33% of tauopathy experts disagreed on a tauopathy diagnosis. Even with examination of only late-stage cases— the most severe forms— agreement for CTE was 78%. A closer inspection of the online supplement of published consensus guidelines shows that all CTE cases were diagnosed with other disease “co-morbidities” by some respondents. Further, CTE was the sole diagnosis in only 27 out of 70 interpretations.

After initial evaluations, evaluators were then provided with the gross findings and clinical summaries for each case, and asked to revisit their initial diagnosis and provide a second level of diagnostic certainty. On the aforementioned 1–4 scale, the degree of diagnostic certainty rose from 3.1 to 3.7 when a clinical history and pathologic summary was provided. In contrast, greater agreement was observed among non-CTE cases. For example, there was 97.1% agreement among reviewers for AD cases.

This seminal publication represents the initial and only official clinical guideline formation process, involving samples taken from a single institution and evaluated by 7 neuropathologists [10]. Further, there has yet to be any follow-up multi-institutional meetings for refinement of the criterion based on new research since this initial consensus conference held in 2015 and published in 2016. Of additional concern is the lack of academic transparency or listing of any discussions that occurred during the process of guideline formation outside of the final publication. It is common practice for consensus papers to highlight discussions that occurred during the consensus meeting in narrative form, especially those in which there may have been disagreement among participants. Both consensus guidelines described above (AD and DLB) have adhered to that practice. For example, as part of the NIA-AA consensus meeting for AD, the paper highlighted that “a major point of discussion among committee members was the relative value of evaluation Aβ/amyloid plaque phase and neuritic plaque score in the assessment of AD neuropathological change;” and go on to discuss this in further detail [29]. The CTE consensus paper is solely a discussion of pathology, not consensus dialogue, and contains little content regarding the implications, or lack thereof, the diagnosis actually possesses. Without knowledge of these discussions, the reader is left uncertain regarding consensus members discussion around a potential lower pathological threshold, disagreements on staging, and the purpose of omitting stage I and II cases from the rating procedure.

Clinical diagnostic criteria

No consensus clinical diagnostic criteria currently exist for CTE. However, several authors have attempted to coalesce clinical and/or research diagnostic criteria based on earlier studies. In the first contemporary staging system published in 2013 [65], McKee and colleagues attached clinical symptoms to each one of their I– IV CTE Stages: Stage I: asymptomatic, mild short-term memory difficulties, depression, and mild aggression; Stage II: mood lability, explosivity, loss of attention, and depression; Stage III: visuospatial difficulties, memory loss, executive dysfunction, cognitive impairment, and apathy; and Stage IV: profound attention loss, language difficulties, paranoia, dysarthria, and parkinsonism. These clinical variants were based on anecdotal inferences from case studies rather than expert consensus or higher order statistical analyses.

In their 2011 study, Omalu and colleagues [64] outlined 9 syndromic profiles of CTE positive cases; however, no attempt was made to correlate symptoms with specific neuropathology due to the postmortem, retrospective nature of their study. The 9 syndromic domains were: deterioration in (1) social and (2) cognitive functioning, (3) mood and (4) behavioral disorders, (5) deterioration of interpersonal relationships, (6) criminal/violent tendencies, (7) alcohol/drug abuse, (8) religiosity, and (9) generalized head and body aches. No attempts to validate or replicate these syndromic profiles have been made since their proposal in 2011.

Based upon the work of Omalu, McKee, and review of available literature, Jordan [13] attempted to delineate between symptom types of CTE as either behavioral/psychiatric, motor, or cognitive in nature. Additionally, the same review provided a diagnostic classification system of improbable, possible, probable, and definitive based on numbers and clusters of symptoms. For example, ‘probable CTE’ consisted of any neurological process characterized by two or more of symptom types described above, which is also distinguishable from any known disease process (e.g., not consistent with PD). Considerable overlap between other neurogenerative disorders and this classification system exists. In fact, possible CTE is described as “any neurological process that is consistent with the clinical description of CTE, but can be potentially explained by other known neurological disorders.” This classification has yet to be empirically examined or validated.

Another attempt was made to describe the clinical symptoms of CTE by Stern et al. [66], with 36 CTE positive male subjects without comorbid neurodegenerative or motor neuron disease. After retrospective reports from next-of-kin informants, two distinct clinical presentations were proposed: 1) younger age with behavioral/mood disturbances, and 2) older age with cognitive impairment. In addition to the methodological limitations of retrospective reports, these clinical presentations entailed interpretations and impressions from both next-of-kin and clinical evaluators. Additionally, only simple descriptive reports were provided, rather than utilizing rigorous statistical methods such as factor analyses or other data reduction methods that have been used elsewhere [67 –69].

Further classification of clinical presentation for research, termed traumatic encephalopathy syndrome (TES) was attempted by Montenegro [14], based on a range of clinical symptoms associated with the disorder. This classification system was based on reviews of prior literature and not an independent sample of subjects. A diagnosis of TES required five core general criteria (e.g., history of multiple head impacts), one core clinically-determined cognitive, behavioral, or mood feature, and at least two supportive features (e.g., impulsivity, anxiety, headache). Montenegro also attempted to establish a sub-classification system, delineating TES into four variants: TES behavioral/mood, TES cognitive, TES mixed, and TES dementia. Another review was performed by Reams et al. [70], who modified the previous TES criteria. These modified criteria contained seven required features and three variant-based (emotional dysregulation, behavioral change, and mood disturbance) supportive features. Again, empirical investigation and validation of these criteria have yet to be performed.

In the largest and most recent report to date, Mez and colleagues [4] attempted to correlate clinical symptoms to pathologic disease stages I– IV by dichotomizing groups into mild (Stage I-II) and severe (Stage III-IV) CTE. Retrospective next-of-kin interviews were also employed to collect antemortem cognitive, behavioral, and mood symptoms. These were essentially descriptive reports with minimal pathologic correlation and no mention of sensitivity and specificity compared to controls. For example, the authors note in the discussion that “Participants with mild CTE pathology often had these symptoms [impulsivity, depressive symptoms, apathy, anxiety, explosivity, episodic memory symptoms, and attention and executive function symptoms] despite having relatively circumscribed cortical pathology and absence of p-tau pathology in the hippocampus, entorhinal cortex, or amygdala.” Additionally, patients in both the mild and severe CTE groups exhibited the majority of reported symptoms at virtually the same rate.

Pathologic diagnostic criteria

Based on the established consensus guidelines, a diagnosis of CTE requires meeting only a single criterion: at least one perivascular p-tau lesion consisting of p-tau aggregates in neurons, astrocytes, and cell processes around small blood vessels found at the depths of cortical sulci. There are other supportive findings, such as p-tau pretangles and NFTs in superficial cortical layers (layers II/III) of the cerebral cortex; pretangles, NFTs, or extracellular tangles in CA2 and CA4 of the hippocampus; subpial p-tau astrocytes at the glial limitans; and dot-like p-tau neurites [4]. Furthermore, CTE stages of worsening severity are also reported. However, the only necessary and sufficient pathologic criterion required is a solitary perivascular p-tau lesion located in the depth of a sulcus.

Validation studies

There is currently an absence of validation studies for any of the staging systems or clinical phenotypes described above. Ideally, validation of any of these antemortem classification systems would involve prospective enrollment of former contact sport athletes of all playing levels (youth, high school, collegiate, professional) and controls (non-contact sports and/or non-athletes) in order to measure symptoms and changes over time and then utilize these recordings to attempt to predict postmortem neuropathological findings. Prospective studies of this nature could also be effective validation of the staging system or phenotypes presented in prior work. This would involve investigating the sensitivity and specificity of antemortem clinical criteria, or even characteristics in differentiating those exhibiting pathological findings, especially as compared to controls. With this, further refinement of the first iteration of pathological and proposed clinical criteria can occur. For example, DLB diagnostic criteria have demonstrated improvement in sensitivity and specificity with each iteration and refinement of consensus guidelines (see above). However, pooled specificity of the newest guidelines is still approximately 66% and continued improvement is likely to occur with further consensus efforts and refinements.

Prospective multi-institution studies of this nature would also allow for a more precise estimate of CTE prevalence. The generalizability of the current proposed clinical diagnostic criteria is extremely limited, as they are derived primarily from professional male athletes. The degree to which these apply to women, or those who played contact sports at lower than elite levels is unknown. Furthermore, while the reports continue to diagnosis and attribute proposed clinical symptoms of CTE across all populations (i.e., high school and collegiate athletes) [71, 72], the prevalence of and degree to which the pathological diagnostic and staging criteria apply to those below the age of 60 is particularly unclear, as none of the 10 cases that comprised the “CTE as submitted diagnosis” group used to develop the pathological criteria was younger than 60 [10]. Additionally, at this time, the single current diagnostic criterion is based upon 25 cases, only 3 of which did not contain AD pathology (Aβ). This raises concerns regarding the true prevalence of CTE, as that degree of sparseness of non AD-related cases is derived from an institution, which currently possesses the largest repository of donated CTE cases. Multi-center prospective enrollment would offer the opportunity to determine the true prevalence of the disease.

DISCUSSION

Although CTE has been described in various forms for approximately 90 years, its neuropathologic delineation remains in evolution, and the efforts at arriving at consensus agreement remain in infancy. To date, a single, preliminarily agreed upon consensus diagnostic criterion based on few selected late-stage cases of CTE is available, and this criterion awaits cross-validation and empirical extension. Without a universally accepted, independently established diagnostic staging system, clinical-pathologic correlations of CTE is likely to be fraught with error. Unlike AD or DLB, a diagnosis of CTE currently requires only a single microscopic focus of perivascular tau at the depth of a sulcus. At present, the presence of a solitary (or even multiple foci) of p-tau has been invoked to explain a wide range of neuropsychiatric symptoms ranging from headache to suicidality [4 , 66].

Development of reliable pathological criteria levels

Under the currently proposed staging method, where there is no lower threshold for the diagnosis of CTE, brains with one, two, or three foci of p-tau accumulation are counted as CTE cases. This potentially leads to an overestimate of CTE case numbers, a circumstance where specificity is sacrificed for sensitivity. This problem was highlighted by a recent prospective case series of 111 brains published by Noy and colleagues [73], who demonstrated that only 4% of cases were positive for CTE. When the authors included ‘tiny’ amounts of pathology consistent with stage I CTE, the number of CTE-positive cases jumped to 30.6%, resulting in a sample prevalence of over one-third. The inclusion of these minimal cases in CTE populations clouds the ability to accurately detect signs and symptoms which may in fact be specific to CTE, thus leading to over-inclusion of symptoms that are common and nonspecific in psychiatric and neurodegenerative disorders. Current efforts are underway to identify other possible neuropathological features that may be more prevalent in CTE, such as astrocytic degeneration [74], which could provide more desirable neuropathological specificity of CTE. However, the current pathological criterion (≥single p-tau lesion at the depths of a sulcus) may be over inclusive and any attempts to develop diagnostic clinicopathological criteria in relation to neuropathology will likely be fruitless.

In other neurodegenerative diseases where tau is a prominent feature, cognitive impairment and behavioral changes are associated with the distribution of tau throughout the brain, which forms the basis of the Braak and Braak staging in AD. However, as stated above, staging approaches in AD take into account the significant diagnostic overlap with individuals who display no cognitive symptoms despite high loads Aβ plaques and tau-positive NFTs [25]. Unlike CTE, the autopsy diagnosis of AD is based on an age-adjusted numerical score with an age-related minimum number of lesions in key regions of the brain from the limbic system to the neocortex. A 4-grade scale counting classic AD lesions from none to frequent is then integrated into an age-related score where the diagnosis of AD is said to be possible, probable, or definite. This approach not only relies on autopsy findings, but the age of the patient and the clinical suspicion of AD.

Development of reliable pathological criteria levels helps to minimize some of the subjectivity in disease diagnosis (i.e., interpretation of neuropathological findings). Postmortem diagnoses involve interpretation of neuropathological findings and not a quantified diagnostic test that definitively computes and classifies disease presence. These neuropathological interpretations can be subject to individual preferences/biases, as well as other factors [75]. Within the field of pathology, this is commonly referred to as “diagnostic review bias.” As part of this bias, the pathologist is aware of a final result and may search more thoroughly for evidence, or be more likely to interpret borderline histological findings [76]. The subjectivity in the pathological interpretation of the presence of a disease was observed in the NINDS/NIBIB consensus meeting, as at least one expert on the panel diagnosed CTE in 8 of the non-CTE cases. In contrast, one expert failed to identify CTE as the sole diagnosis in any of the cases, which was still true for 4 of the 10 cases, even after unblinding. Given this subjectivity in pathological interpretation, it is important to highlight that most of the evidence linking, contact sport participation, a neuropathological diagnosis of CTE, and adverse long-term outcomes have all been derived from a single institution/disease related study center. By developing reliable pathological criteria that includes pertinent factors (e.g., age, co-occurring pathology) derived from standard clinical protocols, reduction of subjectivity in neuropathological interpretation can be achieved. As highlighted above, inter-rater agreement for CTE diagnosis (78% agreement among the original consensus panel), is inferior to AD, based only on a sequential accumulation of NFTs in the cortex without taking into clinical variables (i.e., age), which has yielded a kappa of 0.90 [77].

Consideration of concurrent neurodegenerative diseases

The classification of neurodegenerative diseases such as AD also take into account the presence of other coexisting pathologies. Even after decades of research, the diagnosis of AD is often still problematic due to the diverse morphologic heterogeneity of AD, including the numerous subtypes and presence of coexisting neurodegenerative findings such as LBs, cerebrovascular disease, and hippocampal sclerosis. In these situations, AD may be the primary diagnosis, but the comorbid neurodegenerative pathologies and diseases are recognized as existing and possibly contributing to dementia. As such, the standardized approach to reporting of autopsy findings are reflective of this and not only describe the scores, but also clinical correlations and the description of other coexisting pathologies. The experts involved in these sister diseases acknowledge and embrace this uncertainty. To date, such is not the case in CTE. Instead we see the converse, where scientific experts (and the lay public) embrace preliminary and immature data, all with the presumption of scientific certainty.

While the CTE consensus panel recommended that the presence of pathology consistent with another primary, comorbid neurodegenerative disease excludes CTE as a single diagnosis, cases in the literature are still reported or emphasized as solely CTE [4 , 78]. There is little to no consideration given to the other neurodegenerative disease processes that may be present and contributing to symptom burden. For example, in a most recent 2017 publication, 98 of the 177 brains with CTE diagnoses had other primary neurodegenerative diseases [4]. However, this is not reflected in the case reporting, and all clinical findings of subjects are attributed to CTE and no other comorbid neurodegenerative disease processes.

Refinement of criteria through diverse sample and panels

Unlike the lengthy processes and large case numbers used in AD and DLB described above, none of the CTE cases were acquired prospectively and all clinical information relied on self-reporting or recollections of next of kin, which reflects study design limitations; including the potential for heuristic errors, such as confirmation bias and congruence bias. Valid and reliable pathological and clinical diagnostic criteria should be based on the rigorous processes broadly implemented by the two other neurodegenerative disorders discussed in this paper. While the two degenerative disorders varied in the degree to which they adhered to the established consensus process, the development of each criteria set adhered to a few common principles. These include 1) multiple iterations or refinements of originally developed criteria, 2) samples taken from large registries of specimens, in order to obtain a clearer estimation of disease characteristics (e.g., true prevalence, risk factors, etc.), 3) consensus meetings consisting of diverse subject matter experts from different institutions and disciplines, and 4) validation studies as a means to evaluate and refine the previously established criteria. Perhaps the most important of these principles is number 4, as it involves refinement and attempts to improve proposed criteria. Given that consensus statements inherently suggest that the science on the issue is not yet settled, results are often provisional and will likely require modification, which has been observed among the large majority of consensus statements. The CTE diagnostic neuropathologic criterion (single focus of p-tau), in its current form, will continue to lack diagnostic specificity until a more comprehensive consensus process is performed that at least broadly adheres to these common principles of neurodegenerative disease criteria development, particularly further refinement.

Table 1

Disease history and consensus derived criteria for AD, DLB, and CTE

	Alzheimer’s Disease	Dementia with Lewy Bodies	Chronic Traumatic Encephalopathy
Evidence
a) Earliest year	1906	1912/1961	2005
b) Iterations	8	4	1
c) Years	35 y	22 y	3 y
d) Cases	562	Literature Reviewed	25^*
e) Experts	18	63	7
f) Institutions	18	>40	7
g) Subcommittees	Yes	Yes	No
Pathologic Diagnostic Criteria	•ABC score of Aβ, NFTs, CERAD neuritic plaques, each with scale of 0 to 3	•Score from 0 to 2 in each region based on number of LBs	At least 1 perivascular p-tau lesion consisting of p-tau aggregates in neurons, astrocytes, and cell processes around small blood vessels found at the depths of cortical sulci
	•ABC score classified as None, Low, Intermediate, High pathologic evidence of AD based on 0 to 3 in all 3 aspects.	•Scores summed based on distribution and density in five main regions
		•Low, intermediate, and high classification based on above and concomitant AD pathology
Clinical Diagnostic Criteria	•Meets requirements for All-cause dementia (function decline, poor activities of daily living, psychiatric ruled out, cognitive/behavioral impairment)	•Progressive cognitive decline that interferes with function; attention, executive, visuoperceptual deficits; persistent memory impairment later	NA
	•Probable AD: insidious, worsened cognition, two impaired ‘core’ cognitive domains•Possible AD: most above criteria met, atypical course, mixed presentation	•Probable DLB: 2 + clinical features with or without biomarkers (reduced dopamine transporter uptake in the basal ganglia on PET or SPECT); or, 1 + clinical feature and 1 + biomarker
		•Possible DLB: 1 + biomarker without clinical features
Validation
a) Sensitivity	76–93%	88% – 91%	NA
b) Specificity	55–91%	67% – 81%	NA
c) Accuracy	90%	82% – 91%	NA

^*25 cases of tauopathies; cases of stage III/IV CTE = 10; cases of stage I/II CTE = 0.

The study of CTE is in its infancy. Despite wide use of the terminology in publications, there are still no universally accepted or validated diagnostic and/or staging criteria. The neuropathologic diagnostic criterion currently in use is based on small numbers of cases, many with other prominent neurodegenerative pathology and/or diseases, and are the products of just a select group of neuropathologists. The CTE literature lacks a nomenclature that is reflective of the other more prevalent neurodegenerative diseases, and there is no use of terminology such as ‘unlikely,’ ‘possible,’ or ‘probable’, which is typical of that used in other neurodegenerative disease classifications. Additionally, antemortem clinical criteria with distinct profiles for each stage of CTE should involve rigorous processes that involves refining specific clinical symptoms associated with each classification, rather than efforts involving simple correlational or frequency-based retrospective neurobehavioral observations. However, as highlighted above, attempting to establish clinical criteria without further refinement of the single neuropathological criterion will likely prove futile. The use of pathologic staging to infer disease progression and explain clinical symptoms suffers from a lack of a lower threshold for diagnosis and as result, lacks diagnostic specificity. The lack of a lower threshold inflates the number of CTE cases and inhibits the clarification of purported clinical symptoms associated with this disease. Until consensus- and collaboration-derived diagnostic and staging criteria are established by the international medical community, we will not have a clear understanding of CTE. Going forward, we suggest further involvement with international consensus panels with the inclusion of cases reflecting a range of disease severity and cases from more than one study center in order to aid in the development of the optimal neuropathological and clinical criteria.

DISCLOSURE STATEMENT

Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/18-1058r2).

References

National Football League, NFL retired playersMediation Agreement, http://static.nfl.com/static/content/public/photo/2013/08/29/0ap2000000235504.pdf, Accessed February 8th, 2017.

Morris C, Illinois Proposes Law to Ban Kids Under 12 From Playing Tackle Football, http://fortune.com/2018/01/26/illinois-ban-kids-tackle-football/, Accessed February 8th, 2018.

Corsellis

, Bruton

, Freeman-Browne

(1973) The aftermath of boxing. Psychol Med 3, 270–303.

Mez

, Daneshvar

, Kiernan

, Abdolmohammadi

, Alvarez

, Huber

, Alosco

, Solomon

, Nowinski

, McHale

, Cormier

, Kubilus

, Martin

, Murphy

, Baugh

, Montenigro

, Chaisson

, Tripodis

, Kowall

, Weuve

, McClean

, Cantu

, Goldstein

, Katz

, Stern

, Stein

, McKee

(2017) Clinicopathological evaluation of chronic traumatic encephalopathy in players of American football. JAMA 318, 360–370.

Castellani

, Perry

, Iverson

(2015) Chronic effects of mild neurotrauma: Putting the cart before the horse? J Neuropathol Exp Neurol 74, 493–499.

Gardner

, Iverson

, McCrory

(2014) Chronic traumatic encephalopathy in sport: A systematic review. Br J Sports Med 48, 84–90.

Iverson

, Gardner

, McCrory

, Zafonte

, Castellani

(2015) A critical review of chronic traumatic encephalopathy. Neurosci Biobehav Rev 56, 276–293.

Barr

(2013) An evidence based approach to sports concussion: Confronting the availability cascade. Neuropsychol Rev 23, 271–272.

Solomon

, Sills

(2014) Chronic traumatic encephalopathy and the availability cascade. Phys Sportsmed 42, 26–31.

10.

McKee

, Cairns

, Dickson

, Folkerth

, Keene

, Litvan

, Perl

, Stein

, Vonsattel

, Stewart

, Tripodis

, Crary

, Bieniek

, Dams-O’Connor

, Alvarez

, Gordon

, group

(2016) The first NINDS/NIBIB consensus meeting to define neuropathological criteria for the diagnosis of chronic traumatic encephalopathy. Acta Neuropathol 131, 75–86.

11.

Manley

, Gardner

, Schneider

, Guskiewicz

, Bailes

, Cantu

, Castellani

, Turner

, Jordan

, Randolph

, Dvořák

, Hayden

, Tator

, McCrory

, Iverson

(2017) A systematic review of potential long-term effects of sport-related concussion. Br J Sports Med 51, 969–977.

12.

Iverson

, Keene

, Perry

, Castellani

(2018) The need to separate chronic traumatic encephalopathy neuropathology from clinical features. J Alzheimers Dis 61, 17–28.

13.

Jordan

(2013) The clinical spectrum of sport-related traumatic brain injury. Nat Rev Neurol 9, 222–230.

14.

Montenigro

, Baugh

, Daneshvar

, Mez

, Budson

, Au

, Katz

, Cantu

, Stern

(2014) Clinical subtypes of chronic traumatic encephalopathy: Literature review and proposed research diagnostic criteria for traumatic encephalopathy syndrome. Alzheimers Res Ther 6, 68.

15.

Alosco

, Mez

, Tripodis

, Kiernan

, Abdolmohammadi

, Murphy

, Kowall

, Stein

, Huber

, Goldstein

, Cantu

, Katz

, Chaisson

, Martin

, Solomon

, McClean

, Daneshvar

, Nowinski

, Stern

, McKee

(2018) Age of first exposure to tackle football and chronic traumatic encephalopathy. Ann Neurol 83, 886–901.

16.

Alzheimer

, Stelzmann

, Schnitzlein

, Murtagh

(1995) An English translation of Alzheimer’s 1907 paper, “Uber eine eigenartige Erkankung der Hirnrinde”. Clin Anat 8, 429–431.

17.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939–944.

18.

Khachaturian

(1985) Diagnosis of Alzheimer’s disease. Arch Neurol 42, 1097–1105.

19.

Davis

, Schmitt

, Wekstein

, Markesbery

(1999) Alzheimer neuropathologic alterations in aged cognitively normal subjects. J Neuropathol Exp Neurol 58, 376–388.

20.

Knopman

, Parisi

, Salviati

, Floriach-Robert

, Boeve

, Ivnik

, Smith

, Dickson

, Johnson

, Petersen

, McDonald

, Braak

, Petersen

(2003) Neuropathology of cognitively normal elderly. J Neuropathol Exp Neurol 62, 1087–1095.

21.

Mirra

, Heyman

, McKeel

, Sumi

, Crain

, Brownlee

, Vogel

, Hughes

, van Belle

, Berg

(1991) The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology 41, 479–486.

22.

Braak

, Braak

(1991) Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82, 239–259.

23.

(1997) Consensus recommendations for the postmortem diagnosis of Alzheimer’s disease. The National Institute on Aging, and Reagan Institute Working Group on Diagnostic Criteria for the Neuropathological Assessment of Alzheimer’s Disease. Neurobiol Aging 18, S1–2.

24.

Lopez

, Litvan

, Catt

, Stowe

, Klunk

, Kaufer

, Becker

, DeKosky

(1999) Accuracy of four clinical diagnostic criteria for the diagnosis of neurodegenerative dementias. Neurology 53, 1292–1299.

25.

Jansen

, Ossenkoppele

, Knol

, Tijms

, Scheltens

, Verhey

, Visser

, Aalten

, Aarsland

, Alcolea

, Alexander

, Almdahl

, Arnold

, Baldeiras

, Barthel

, van Berckel

, Bibeau

, Blennow

, Brooks

, van Buchem

, Camus

, Cavedo

, Chen

, Chetelat

, Cohen

, Drzezga

, Engelborghs

, Fagan

, Fladby

, Fleisher

, van der Flier

, Ford

, Förster

, Fortea

, Foskett

, Frederiksen

, Freund-Levi

, Frisoni

, Froelich

, Gabryelewicz

, Gill

, Gkatzima

, Gómez-Tortosa

, Gordon

, Grimmer

, Hampel

, Hausner

, Hellwig

, Herukka

, Hildebrandt

, Ishihara

, Ivanoiu

, Jagust

, Johannsen

, Kandimalla

, Kapaki

, Klimkowicz-Mrowiec

, Klunk

, Köhler

, Koglin

, Kornhuber

, Kramberger

, Van Laere

, Landau

, Lee

, de Leon

, Lisetti

, Lleó

, Madsen

, Maier

, Marcusson

, Mattsson

, de Mendonça

, Meulenbroek

, Meyer

, Mintun

, Mok

, Molinuevo

, Møllergård

, Morris

, Mroczko

, Van der Mussele

, Na

, Newberg

, Nordberg

, Nordlund

, Novak

, Paraskevas

, Parnetti

, Perera

, Peters

, Popp

, Prabhakar

, Rabinovici

, Ramakers

, Rami

, Resende de Oliveira

, Rinne

, Rodrigue

, Rodríguez-Rodríguez

, Roe

, Rot

, Rowe

, Rüther

, Sabri

, Sanchez-Juan

, Santana

, Sarazin

, Schröder

, Schütte

, Seo

, Soetewey

, Soininen

, Spiru

, Struyfs

, Teunissen

, Tsolaki

, Vandenberghe

, Verbeek

, Villemagne

, Vos

, van Waalwijk van Doorn

, Waldemar

, Wallin

, Wiltfang

, Wolk

, Zboch

, Zetterberg

(2015) Prevalence of cerebral amyloid pathology in persons without dementia: A meta-analysis. JAMA 313, 1924–1938.

26.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

Jr. , Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263–269.

27.

American Psychiatric Association (1980) Diagnostic and Statistical Manual of Mental Disorders, New York, NY.

28.

American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental Disorders, Washington, D.C.

29.

Hyman

, Phelps

, Beach

, Bigio

, Cairns

, Carrillo

, Dickson

, Duyckaerts

, Frosch

, Masliah

, Mirra

, Nelson

, Schneider

, Thal

, Thies

, Trojanowski

, Vinters

, Montine

(2012) National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement 8, 1–13.

30.

Gaugler

, Kane

, Johnston

, Sarsour

(2013) Sensitivity and specificity of diagnostic accuracy in Alzheimer’s disease: A synthesis of existing evidence. Am J Alzheimers Dis Other Demen 28, 337–347.

31.

Dubois

, Feldman

, Jacova

, Hampel

, Molinuevo

, Blennow

, DeKosky

, Gauthier

, Selkoe

, Bateman

, Cappa

, Crutch

, Engelborghs

, Frisoni

, Fox

, Galasko

, Habert

, Jicha

, Nordberg

, Pasquier

, Rabinovici

, Robert

, Rowe

, Salloway

, Sarazin

, Epelbaum

, de Souza

, Vellas

, Visser

, Schneider

, Stern

, Scheltens

, Cummings

(2014) Advancing research diagnostic criteria for Alzheimer’s disease: The IWG-2 criteria. Lancet Neurol 13, 614–629.

32.

Montine

, Monsell

, Beach

, Bigio

, Bu

, Cairns

, Frosch

, Henriksen

, Kofler

, Kukull

, Lee

, Nelson

, Schantz

, Schneider

, Sonnen

, Trojanowski

, Vinters

, Zhou

, Hyman

(2016) Multisite assessment of NIA-AA guidelines for the neuropathologic evaluation of Alzheimer’s disease. Alzheimers Dement 12, 164–169.

33.

Goedert

, Spillantini

, Del Tredici

, Braak

(2013) 100 years of Lewy pathology. Nat Rev Neurol 9, 13–24.

34.

Lewy

(1912) Paralysis agitans. I. Pathologische Anatomie. In, Handbuch der Neurologie, Lewandowsky

, Abelsdorff

, eds. Springer, Verlag, Berlin, pp. 920–933.

35.

Tretiakoff

(1919) Contribution à l’étude de l’anatomie pathologique du locus niger de Soemmering avec quelques déductions relatives à la pathogénie des troubles du tonus musculaire et de la maladie de Parkinson. University of France, Paris.

36.

Okazaki

, Lipkin

, Aronson

(1961) Diffuse intracytoplasmic ganglionic inclusions (Lewy type) associated with progressive dementia and quadriparesis in flexion. J Neuropathol Exp Neurol 20, 237–244.

37.

Kosaka

, Oyanagi

, Matsushita

, Hori

(1976) Presenile dementia with Alzheimer-, Pick- and Lewy-body changes. Acta Neuropathol 36, 221–233.

38.

Kosaka

(1978) Lewy bodies in cerebral cortex, report of three cases. Acta Neuropathol 42, 127–134.

39.

Kosaka

, Matsushita

, Oyanagi

, Mehraein

(1980) Clinicopathological study of Lewy body disease. Psychiatr Neurol Jpn 82, 292–311.

40.

Kosaka

, Yoshimura

, Ikeda

, Budka

(1984) Diffuse type of Lewy body disease: Progressive dementia with abundant cortical Lewy bodies and senile changes of varying degree–a new disease? Clin Neuropathol 3, 185–192.

41.

Kosaka

, Iseki

, Odawara

, Yamamoto

(1996) Cerebral type of Lewy body disease. Neuropathology 1, 32–35.

42.

McKeith

, Galasko

, Kosaka

, Perry

, Dickson

, Hansen

, Salmon

, Lowe

, Mirra

, Byrne

, Lennox

, Quinn

, Edwardson

, Ince

, Bergeron

, Burns

, Miller

, Lovestone

, Collerton

, Jansen

, Ballard

, de Vos

, Wilcock

, Jellinger

, Perry

(1996) Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): Report of the consortium on DLB international workshop. Neurology 47, 1113–1124.

43.

Spillantini

, Schmidt

, Lee

, Trojanowski

, Jakes

, Goedert

(1997) Alpha-synuclein in Lewy bodies. Nature 388, 839–840.

44.

Byrne

, Lennox

, Godwin-Austen

, Jefferson

, Lowe

, Mayer

, Landon

, Doherty

(1991) Diagnostic criteria for dementia associated with cortical Lewy bodies. Dementia 2, 283–284.

45.

McKeith

, Perry

, Fairbairn

, Jabeen

, Perry

(1992) Operational criteria for senile dementia of Lewy body type (SDLT). Psychol Med 22, 911–922.

46.

McKeith

, Perry

(1999) Report of the second dementia with Lewy body international workshop: Diagnosis and treatment. Consortium on Dementia with Lewy Bodies. Neurology 53, 902–905.

47.

McKeith

(2006) Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): Report of the Consortium on DLB International Workshop. J Alzheimers Dis 9, 417–423.

48.

McKeith

, Boeve

, Dickson

, Halliday

, Taylor

, Weintraub

, Aarsland

, Galvin

, Attems

, Ballard

, Bayston

, Beach

, Blanc

, Bohnen

, Bonanni

, Bras

, Brundin

, Burn

, Chen-Plotkin

, Duda

, El-Agnaf

, Feldman

, Ferman

, Ffytche

, Fujishiro

, Galasko

, Goldman

, Gomperts

, Graff-Radford

, Honig

, Iranzo

, Kantarci

, Kaufer

, Kukull

, Lee

VMY

, Leverenz

, Lewis

, Lippa

, Lunde

, Masellis

, Masliah

, McLean

, Mollenhauer

, Montine

, Moreno

, Mori

, Murray

, O’Brien

, Orimo

, Postuma

, Ramaswamy

, Ross

, Salmon

, Singleton

, Taylor

, Thomas

, Tiraboschi

, Toledo

, Trojanowski

, Tsuang

, Walker

, Yamada

, Kosaka

(2017) Diagnosis and management of dementia with Lewy bodies: Fourth consensus report of the DLB Consortium. Neurology 89, 88–100.

49.

McKeith

, O’Brien

, Walker

, Tatsch

, Booij

, Darcourt

, Padovani

, Giubbini

, Bonuccelli

, Volterrani

, Holmes

, Kemp

, Tabet

, Meyer

, Reininger

, DLB Study Group (2007) Sensitivity and specificity of dopamine transporter imaging with 123I-FP-CIT SPECT in dementia with Lewy bodies: A phase III, multicentre study. Lancet Neurol 6, 305–313.

50.

Yoshita

, Arai

, Asada

, Fujishiro

, Hanyu

, Iizuka

, Iseki

, Kashihara

, Kosaka

, Maruno

, Mizukami

, Mizuno

, Mori

, Nakajima

, Nakamura

, Nakano

, Nakashima

, Nishio

, Orimo

, Samuraki

, Takahashi

, Taki

, Tokuda

, Urakami

, Utsumi

, Wada

, Washimi

, Yamasaki

, Yamashina

, Yamada

(2015) Diagnostic accuracy of 123I-meta-iodobenzylguanidine myocardial scintigraphy in dementia with Lewy bodies: A multicenter study. PLoS One 10, e0120540.

51.

Lobotesis

, Fenwick

, Phipps

, Ryman

, Swann

, Ballard

, McKeith

, O’Brien

(2001) Occipital hypoperfusion on SPECT in dementia with Lewy bodies but not AD. Neurology 56, 643–649.

52.

Harper

, Fumagalli

, Barkhof

, Scheltens

, O’Brien

, Bouwman

, Burton

, Rohrer

, Fox

, Ridgway

, Schott

(2016) MRI visual rating scales in the diagnosis of dementia: Evaluation in 184 post-mortem confirmed cases. Brain 139, 1211–1225.

53.

O’Brien

, Firbank

, Davison

, Barnett

, Bamford

, Donaldson

, Olsen

, Herholz

, Williams

, Lloyd

(2014) 18F-FDG PET and perfusion SPECT in the diagnosis of Alzheimer and Lewy body dementias. J Nucl Med 55, 1959–1965.

54.

Rizzo

, Arcuti

, Copetti

, Alessandria

, Savica

, Fontana

, Liguori

, Logroscino

(2018) Accuracy of clinical diagnosis of dementia with Lewy bodies: A systematic review and meta-analysis. J Neurol Neurosurg Psychiatry 89, 358–366.

55.

Martland

(1927) Punch drunk. J Am Med Assoc 91, 1103–1107.

56.

Bowman

, Blau

(1490) Psychotic states following head and brain injury in adults and children. In Injuries of the Skull, Brain and Spinal Cord. Neuropsychiatric, Surgical, and Medico-Legal Aspects, Brock

, ed. Williams & Wilkins, Baltimore, MD, pp. 309–359 .

57.

Roberts

(1969) Brain damage in boxers: Study of the prevalence of traumatic encephalopathy among ex-professional boxers. Pitman Medical & Scientific Publishing Co., Turnbridge Wells, Kent, England.

58.

Roberts

, Allsop

, Bruton

(1990) The occult aftermath of boxing. J Neurol Neurosurg Psychiatry 53, 373–378.

59.

Omalu

, DeKosky

, Minster

, Kamboh

, Hamilton

, Wecht

(2005) Chronic traumatic encephalopathy in a National Football League player. Neurosurgery 57, 128–134; discussion 128-134.

60.

Omalu

, DeKosky

, Hamilton

, Minster

, Kamboh

, Shakir

, Wecht

(2006) Chronic traumatic encephalopathy in a national football league player: Part II. Neurosurgery 59, 1086–1092; discussion 1092–1083.

61.

Omalu

, Hamilton

, Kamboh

, DeKosky

, Bailes

(2010) Chronic traumatic encephalopathy (CTE) in a National Football League Player: Case report and emerging medicolegal practice questions. J Forensic Nurs 6, 40–46.

62.

Omalu

, Fitzsimmons

, Hammers

, Bailes

(2010) Chronic traumatic encephalopathy in a professional American wrestler. J Forensic Nurs 6, 130–136.

63.

McKee

, Cantu

, Nowinski

, Hedley-Whyte

, Gavett

, Budson

, Santini

, Lee

, Kubilus

, Stern

(2009) Chronic traumatic encephalopathy in athletes: Progressive tauopathy after repetitive head injury. J Neuropathol Exp Neurol 68, 709–735.

64.

Omalu

, Bailes

, Hamilton

, Kamboh

, Hammers

, Case

, Fitzsimmons

(2011) Emerging histomorphologic phenotypes of chronic traumatic encephalopathy in American athletes. Neurosurgery 69, 173–183; discussion 183.

65.

McKee

, Stern

, Nowinski

, Stein

, Alvarez

, Daneshvar

, Lee

, Wojtowicz

, Hall

, Baugh

, Riley

, Kubilus

, Cormier

, Jacobs

, Martin

, Abraham

, Ikezu

, Reichard

, Wolozin

, Budson

, Goldstein

, Kowall

, Cantu

(2013) The spectrum of disease in chronic traumatic encephalopathy. Brain 136, 43–64.

66.

Stern

, Daneshvar

, Baugh

, Seichepine

, Montenigro

, Riley

, Fritts

, Stamm

, Robbins

, McHale

, Simkin

, Stein

, Alvarez

, Goldstein

, Budson

, Kowall

, Nowinski

, Cantu

, McKee

(2013) Clinical presentation of chronic traumatic encephalopathy. Neurology 81, 1122–1129.

67.

, Chaudhuri

, Bielza

, de Pedro-Cuesta

, Larranaga

, Martinez-Martin

(2017) Parkinson’s disease subtypes identified from cluster analysis of motor and non-motor symptoms. Front Aging Neurosci 9, 301.

68.

Cheng

, Cheng

, Liou

(2013) Group factor analysis for Alzheimer’s disease. Comput Math Methods Med 2013, 428385.

69.

Collerton

, Burn

, McKeith

, O’Brien

(2003) Systematic review and meta-analysis show that dementia with Lewy bodies is a visual-perceptual and attentional-executive dementia. Dement Geriatr Cogn Disord 16, 229–237.

70.

Reams

, Eckner

, Almeida

, Aagesen

, Giordani

, Paulson

, Lorincz

, Kutcher

(2016) A clinical approach to the diagnosis of traumatic encephalopathy syndrome: A review. JAMA Neurol 73, 743–749.

71.

Belson

(2017) “Family sues Pop Warner over suicide of player who had brain disease.”. New York Times. 7 February 2017.

72.

Boston University Research: CTE Center, 18 year old high school football player, http://www.bu.edu/cte/our-research/case-studies/18-year-old/.

73.

Noy

, Krawitz

, Del Bigio

(2016) Chronic traumatic encephalopathy-like abnormalities in a routine neuropathology service. J Neuropathol Exp Neurol 75, 1145–1154.

74.

Hsu

, Gangolli

, Su

, Holleran

, Stein

, Alvarez

, McKee

, Schmidt

, Brody

(2018) Astrocytic degeneration in chronic traumatic encephalopathy. Acta Neuropathol 136, 955–972.

75.

Fandel

, Pfnur

, Schafer

, Bacchetti

, Mast

, Corinth

, Ansorge

, Melchior

, Thuroff

, Kirkpatrick

, Lehr

(2008) Do we truly see what we think we see? The role of cognitive bias in pathological interpretation. J Pathol 216, 193–200.

76.

Schmidt

, Factor

(2013) Understanding sources of bias in diagnostic accuracy studies. Arch Pathol Lab Med 137, 558–565.

77.

Nagy

, Vatter-Bittner

, Braak

, Yilmazer

, Schultz

, Hanke

(1997) Staging of Alzheimer-type pathology: An interrater-intrarater study. Dement Geriatr Cogn Disord 8, 248–251.

78.

Alosco

, Kasimis

, Stamm

, Chua

, Baugh

, Daneshvar

, Robbins

, Mariani

, Hayden

, Conneely

, Au

, Torres

, McClean

, McKee

, Cantu

, Mez

, Nowinski

, Martin

, Chaisson

, Tripodis

, Stern

(2017) Age of first exposure to American football and long-term neuropsychiatric and cognitive outcomes. Transl Psychiatry 7, e1236.