Chronic traumatic encephalopathy (CTE) is a neuropathological condition characterized by the deposition of phosphorylated tau in neurons around small blood vessels at the depths of cortical sulci (
1). CTE has been reported to be common in persons with previous participation in contact sports, primarily American football (
2), and concerns about the implications of CTE lesions have been increasing among former professional athletes and the general population. As a result, efforts to identify clinical features that might be associated with CTE have been growing. CTE has been thought to involve a neurodegenerative course based on retrospective descriptions of progressive declines in cognitive, behavioral, or mood functioning by the next of kin of persons with pathognomonic lesions of CTE (
3). Within the past decade, five sets of diagnostic criteria have been developed for classifying symptoms that could be attributed to CTE lesions in living persons. Most sets of criteria use the term
traumatic encephalopathy syndrome (TES) to distinguish between the neuropathological condition and proposed clinical presentation, as well as to minimize confusion, as the term
chronic in CTE denotes a stable, static process, and the clinical features are thought to be progressive. TES criteria have been broad and include a wide range of symptoms across neurobehavioral domains, including problems with attention, memory, executive function, visuospatial ability, suicidality, aggression, substance misuse, paranoia, gambling, slurred speech, gait instability, and headache.
The first set of TES criteria was published in 2013 (
4) and contain classifications of “possible” or “probable” TES that are based on whether cognitive impairment, behavioral or mood symptoms, and motor dysfunction could be distinguished from other neurological disorders as determined with neuroimaging (e.g., negative amyloid positron emission tomography scan to help rule out Alzheimer’s disease). The second set of TES criteria was also published in 2013 (
5) and similarly involves characterizing symptoms as “possible” or “probable” TES, with symptoms selected if they were reported in at least 7% of CTE cases. The criteria require a history of at least one head trauma; having three or more symptoms or signs (a total of five for probable) such as headache, cognitive impairment, behavioral or mood symptoms, paranoia, motor dysfunction, or alcohol use disorder; and various neurological abnormalities (i.e., nystagmus, hemiparesis, reduced facial expression) that are persistent for at least 2 years and not accounted for by well-known medical-psychiatric syndromes. The third set of TES diagnostic criteria was published in 2014 and is based on symptoms or signs reported in at least 70% of CTE cases (
6). These criteria involve five features: history of repetitive head trauma; cognitive, behavioral, or mood problems; at least two supportive characteristics (e.g., delayed onset of problems, co-occurrence of headache, suicidality); persistence of symptoms for at least 1 year; and the absence of a well-known neurological disorder that could account for the symptoms. The fourth set of TES diagnostic criteria emerged in 2016 (
7) and characterizes symptoms as “possible” or “probable” TES, similar to earlier sets of criteria, although it requires a history of repetitive head trauma; delayed onset of impairment in cognition and behavior, mood, or motor functioning; persistence of symptoms for at least 2 years; a progressive course; and absence of another neurological disorder that could account for the symptoms. The most recent set of TES criteria (
8) involves features selected by a committee convened as part of a workshop for the National Institute of Neurological Disorders and Stroke (NINDS). The NINDS consensus criteria exclude many symptoms with high base rates in the general population as core features and require “substantial” repetitive head trauma; impairment in memory, executive function, or neurobehavioral regulation (e.g., explosiveness); a progressive course; and that another neurological or medical-psychiatric disorder does not fully account for the symptoms.
Despite different sets of TES diagnostic criteria being developed over the past decade, none were developed on the basis of rigorous empirical data. Specifically, each set of TES criteria was based on literature reviews of cases in which patients were described to have CTE or on postmortem interviews with patients’ next of kin, rather than on clinical data from longitudinal evaluations during life. Also, although each set of TES criteria specifies that well-known medical or mental health–related causes should not account for the symptoms, satisfying this condition can be difficult, given that each published set of criteria has symptoms that overlap multiple possible etiologies (
9,
10). Furthermore, no studies have validated any of the proposed clinical features. Only one study to date has assessed the classification accuracy of TES criteria (the 2014 set) for diagnosing CTE, and although sensitivity was high (97%), specificity was low (21%) (
11).
Although some authors have noted that TES criteria should be used for research purposes only (
8,
9,
12), this line has been blurred in that some criteria were developed to provide clinicians a framework for diagnosing TES in living persons (
4,
7). In fact, a recent survey of more than 3,000 former National Football League (NFL) players revealed that 108 reported receiving a diagnosis of TES or CTE by a health care provider (
13). Interestingly, among those reporting a TES or CTE diagnosis, not only were cognitive complaints frequent, but medical, physical, and mental disorder comorbidities were also common, including sleep apnea, vascular risk factors, a history of depression, and chronic pain. Another survey study of over 600 former NFL players, using a comprehensive set of measures, performed a profile analysis to characterize functioning later in life among those who were reported to be at high risk for TES and CTE (
14). Two groups were identified that reported significant functional problems: one group reported very low functioning across measures of cognition, behavioral regulation, physical function, pain, sleep, and psychological symptoms, and the other group reported difficulties primarily associated with cognitive, physical, and mood (i.e., anxiety) functioning. Not surprisingly, greater medical and mental disorders were seen in these groups, which is similar to findings among persons who reported receiving a TES or CTE diagnosis. As such, doubts and concerns about TES criteria have been raised that are largely due to the fact that the criteria were derived from postmortem interviews with the next of kin (rather than objective clinical, prospectively acquired data) about a group of individuals (mostly former professional football athletes) who experienced a complex and broad range of problems that could contribute to experiencing nonspecific cognitive, behavioral, and mood difficulties.
Serious adverse effects can occur from disregarding or overlooking the impact of medical, physical, and psychiatric difficulties of persons described to be at high risk for TES and CTE. Erroneously giving living persons a diagnosis of TES could result in an increase in psychological distress, significant worsening of any mood or behavioral problems, and an unwillingness to engage in interventions for symptoms that might be treatable. Because several patients have already been diagnosed with TES in clinical settings and the TES criteria lack validation, the purpose of this article is to provide clinicians and scientists with a case report to inform how applying the sets of TES criteria carries a risk for misdiagnosis. This case report presents neuropsychological data from a former NFL player who received a clinical diagnosis of TES outside of the University of Texas Southwestern Medical Center before presenting for re-evaluation 7 years later. By reviewing this case vignette, clinicians and scientists should have a practical understanding of how TES criteria can misguide a clinical diagnostic evaluation and an understanding of the importance of considering appropriate and common differential diagnoses.
CASE REPORT
The patient was a middle-aged African American man who played American football in middle school, high school, and college and at the professional level for a total of more than 10 years. He was in the NFL for >5 years, playing in over 100 games. Throughout his lifetime of playing football, he reported experiencing a “countless” number of concussions but reported five injuries that involved momentary loss of consciousness followed by altered mental state, nausea, and headache that resolved within a couple of weeks of each episode. When he retired from the NFL, he had concerns about the long-term effects of concussions and CTE. Approximately 5 years later, he began experiencing difficulties with cognitive, behavioral, and mood functioning, with a progressive course of decline described. Initial cognitive difficulties involved misplacing common objects (e.g., wallet, keys) and forgetting where he parked his car. With time, he was having word-finding problems, was losing his place in conversations, was repeating questions, was forgetting to close exterior doors to the home, and was having problems tracking his appointments. He was working for a major corporation, but he failed certifications for his position multiple times because of reported cognitive difficulties, and he eventually resigned. Moreover, his wife started managing his calendar and all of his appointments because of his cognitive difficulties. Significant life stress was also described around that time, involving discord with his ex-wife, some family issues, and financial strain. Although there was no history of psychiatric disorders or symptoms reported before then (including substance use disorders), significant mood and behavioral changes occurred, including depressed mood, suicidal thoughts, anger, physical and verbal aggression, and anxiety. In fact, he was coaching youth athletics, but this ended because of the children and their parents being concerned about his anger and aggression. With regard to medical history, he had high blood pressure, osteoarthritis, chronic back pain, headaches, and sleep disturbance (insomnia with frequent awakenings). He noted a long history of high levels of daily pain (rated 7 out of 10 in severity). Family medical history was negative for any dementia syndromes. Furthermore, his previous history was negative for problems learning or performing in school, and he made average grades in college (not leading to a degree).
INITIAL NEUROPSYCHOLOGICAL EVALUATION OUTSIDE OF OUR INSTITUTION
At his initial neuropsychological evaluation, the patient was alert, but fluctuations in his attention were observed, involving instances of staring off during the evaluation. With regard to the validity of the test results, no measures of engagement were reported. Despite attentional fluctuations, he was thought to have put forth his best effort throughout testing, with no validity concerns noted.
Neuropsychological results from this evaluation are presented in
Table 1. General cognitive abilities fell in the average range, but moderately low scores were seen on measures of simple auditory attention, divided attention, working memory, and visuomotor speed. Low scores were also observed on tasks of executive function, such as phonemic fluency, design fluency, concept formation, card sorting, as well as on the Trail Making Test. In terms of memory, few details were provided to fully characterize functioning in this domain, and unfortunately, no test scores were provided. Nonetheless, indices of delayed memory were reportedly low, but these were noted to be primarily for free recall of material, with his performance significantly improving with recognition cues and choices. Results on the Minnesota Multiphasic Personality Inventory–2 (MMPI-2) indicated severe levels of depression and anxiety, characterized by feelings of hopelessness and worthlessness and suicidal thoughts. The clinician concluded that the findings reflected dysfunction of frontal lobe systems and, in combination with depression and problems regulating his mood, were noted to be consistent with what has been described about CTE. Thus, a clinical diagnosis of TES was made with no other potential diagnoses listed.
APPLICATION OF TES CRITERIA
On the basis of data collected at the time of his initial evaluation, the patient met all published sets of TES criteria (
Table 2). In accordance with the newest NINDS criteria, the patient had extensive repetitive head trauma exposure and low test scores that were potentially indicative of executive dysfunction, as well as features consistent with what has been termed
neurobehavioral dysregulation, a pattern of emotional instability and physical or verbal aggression. Although the criteria state that progression should be observed with neuropsychological reevaluation, it allows a self- or informant report to be relied on if formal testing is unavailable. In this case, there were strong self- and informant descriptions of a progressive decline in both cognitive and behavioral functioning for more than a year. However, the exclusion criterion was difficult to apply, in large part because disentangling the underlying source of TES symptoms is challenging, especially considering that emotional instability may occur with treatable conditions.
NEUROLOGICAL WORK-UP AND MANAGEMENT
The patient later presented to a cognitive neurologist at our institution for clinical management. An MRI was completed and revealed no brain abnormalities, but it did show some degenerative changes in the spine. Although a sleep study was ordered, this was never completed, despite the patient sleeping an average of only 4 hours per night and experiencing daytime drowsiness (with indications of potential sleep apnea). Antidepressant medication and a sleep aid were prescribed, and although the patient did not want to take medication at that time, he eventually agreed several years later. Over time, his cognitive difficulties were reported to be worsening, and he was referred for another neuropsychological evaluation 7 years after his initial assessment.
NEUROPSYCHOLOGICAL RE-EVALUATION AT OUR INSTITUTION
At his second neuropsychological evaluation, the patient described instances of forgetting to pick up his kids at events and often misplacing his wallet and keys. More frequent word-finding problems were also reported. As a result, his wife was providing a great deal of support, managing all of his appointments. He began taking antidepressant medication 2 months before the evaluation, and he noted a significant reduction in depressive feelings and in anger and aggression. However, high anxiety was still present, which included frequently worrying about being the next former athlete to “die from CTE.” His medical history was unchanged from his initial neuropsychological evaluation.
Despite appearing motivated to do his best during testing, he was tired because of his sleep problems and demonstrated reduced alertness during some tasks. This was seen during a formal measure of engagement and on an auditory naming measure, and low scores were interpreted cautiously. The neuropsychological results from this evaluation are presented in
Table 1 for a comparison with his initial scores. Mildly low scores were seen on measures of working memory and immediate verbal recall. He was slow to learn on a verbal list-learning test and had a pronounced tendency to immediately recall the last few words he was told (recency recall) as well as repeat responses during learning trials (perseverations). However, there were minimal intrusion errors, and he recalled nearly all of the words he learned after a delay (75% retention). Similarly, his immediate recall of details was mildly low on a story memory test, but he recalled most of what he learned after a delay (88% retention). Overall, significant improvements in many abilities were seen in comparison with his previous evaluation, with improved scores in simple attention, processing speed, executive functions, and memory, whereas scores in working memory, expressive vocabulary, and visuospatial skills were unchanged. No scores reflected a decline from his previous evaluation results. The patient endorsed moderate levels of anxiety but only mild depressive symptoms, which was also an improvement from his previous evaluation and likely corresponded to starting medication. Although the patient previously met each set of TES diagnostic criteria, the results showed no evidence of a neurodegenerative disorder, given that there were no declines in his cognitive or behavioral functioning over the course of 7 years. Rather, the evidence suggested significant improvement in both of these areas, potentially related to taking antidepressant medication. Effects from inadequate sleep were thought to explain the four low test scores, especially given the fluctuating engagement, although chronic pain and high anxiety were also seen as potential contributors.
DISCUSSION
The cognitive, behavioral, and mood symptoms described postmortem in persons with CTE lesions have been posited by some to reflect a clinical syndrome related to CTE pathology (i.e., TES). However, among individuals having reported such symptoms and considered high risk for CTE, multiple medical, physical, and psychiatric problems exist. Because there have been no systematic studies showing that CTE lesions are associated with clinical symptoms, TES diagnostic criteria have been controversial (
15,
16).
Several studies have applied previous TES diagnostic criteria (developed before the recent NINDS criteria) to various data sets and have found that a sizable proportion of individuals with physical or psychiatric problems in the general population met criteria for TES (
17,
18). Even among individuals considered at high risk for CTE, the severity of depression symptoms was found to be the only predictor for meeting previous TES criteria in one investigation (
19), with no relationship with neuropsychological performances or head injury exposure. Although these studies illustrate how earlier TES criteria involve high risk for misdiagnosing symptoms from physical and psychiatric problems, the new NINDS TES criteria attempted to resolve such limitations by removing as core features many symptoms with high base rates in the general population. With the new NINDS guidelines, no studies have yet examined the sensitivity and specificity of the criteria to inform whether the risk of misdiagnosis is diminished. To our knowledge, this case presentation is the first neuropsychological report of an individual diagnosed with TES in a clinical setting who fulfills all sets of TES criteria, including the newest NINDS guidelines. However, the patient in this case later showed substantial cognitive and behavioral improvements after the initiation of psychiatric treatment. The case report suggests that the NINDS guidelines, similar to previous criteria, contain features that overlap with difficulties related to physical and psychiatric problems, which will likely continue to limit specificity.
A progressive course of cognitive and behavioral problems is considered a core feature in the NINDS criteria, similar to previous TES criteria. With the NINDS criteria, the definition of cognitive impairment was narrowed to deficits in memory, executive function, or both, as documented by neuropsychological measures. Similarly, the NINDS guidelines replaced some of the nonspecific symptoms (e.g., depression, irritability) with “neurobehavioral dysregulation,” although this remains a broad symptom that can be interpreted in various ways. The guidelines further specify that progression should be observed with neuropsychological re-evaluation, but they allow subjective methods (self- or informant report) to be relied on if formal testing is unavailable. This case report serves as a reminder that subjective reporting of progressive cognitive and behavioral decline as well as low test scores can and do occur for nonneurodegenerative reasons. Disentangling whether the source of complaints and low test scores may relate to a treatable condition or a pathological process can be challenging. Indeed, patients with mental disorders often are referred for neuropsychological evaluation because of concerns for a neurodegenerative condition, and seeing the initial profile of low scores in this case report would raise at least some concerns for cognitive decline. However, most established neurodegenerative disorders require extra clinical criteria to rule a condition in or out (e.g., parkinsonism in Lewy body dementia). With TES, the only other NINDS criterion to be used is the level of exposure to repetitive head trauma, which has questionable clinical significance based on previous work (
12,
20) as well as this case report. Even if a less common presentation is suspected (e.g., a nonamnestic variant of Alzheimer’s disease), the best course diagnostically would be a cautious approach and a recommendation of re-evaluation once psychological symptoms (e.g., depression) have been treated.
The low neuropsychological test scores likely played a significant role in this patient receiving a TES diagnosis at the initial evaluation outside of our institution. When considering a rationale for the poor performances, one explanation might relate to suboptimal engagement. Although exaggeration or malingering cannot be excluded, the patient had severe depression and inadequate sleep, both of which can be associated with detrimental effects on engagement in neuropsychological tests (
21,
22). Unfortunately, no measures of effort were administered to evaluate the patient’s level of engagement. Alternatively, the poor performances at his initial evaluation may have been related to reduced scores from the combined impact of severe depression and inadequate sleep. Depression has been linked with lower scores on measures of attention, processing speed, executive dysfunction, and verbal memory in some individuals (
23), and inadequate sleep has been associated with reduced attention and processing speed (
24,
25). Because the patient started antidepressant medication nearly 2 months before we re-evaluated him, it seems reasonable to assume that an apparent decrease in mood or behavioral symptoms may be related to the significant improvements on scores with our examination because of one or both reasons. Nevertheless, this case illustrates that TES criteria can misdiagnose cognitive and behavioral difficulties in individuals with treatable physical and psychiatric problems. For this reason, and the lack of acceptance and validity of TES, TES criteria should not be applied in non-research settings.
Multiple TES frameworks have been proposed to characterize the clinical symptoms and anticipated progressive course of problems that may correspond with the pathological diagnosis of CTE. These clinical profiles have been predominantly derived from retrospective questioning of the family members of individuals who received a diagnosis of CTE at autopsy, not from prospectively collected, clinically acquired data. At present, none of the TES criteria have been clinicopathologically validated. In this case, we have demonstrated how a patient thought to have a neurodegenerative syndrome related to CTE was misdiagnosed by applying TES criteria, even with the most recent NINDS guidelines. Therefore, a clinical diagnosis of TES should not be made until clinicopathologically validated criteria have been established.