Traumatic Brain Injury and Risk of Long-Term Brain Changes,Accumulation of Pathological Markers,and Developing Dementia: A Review

Introduction

Traumatic brain injuries (TBI) are common, resulting in an estimated 10 million hospitalizations worldwide, but the actual prevalence of TBI is difficult to estimate because many milder injuries do not come to medical attention [1]. The Centers for Disease Control and Prevention reported that rates of TBI have increased more than 50% in the United States over the past decade, with falls, being struck by an object, and vehicle crashes being the most common causes of injury [2]. However, it is likely that the dramatic increase in rates of TBI reflects a greater awareness for milder injuries among the general population and a much more conservative approach to diagnosing mild TBI than has existed in the past. Diffuse axonal injury, involving damage to cerebral white matter tracts, occurs in nearly all severities of TBI, acutely disrupting biochemical and cytoskeletal functions to produce physical, emotional, and/or cognitive symptoms. Diffuse axonal injury can be induced during a TBI through several mechanisms, including axonal stretching or shearing, swelling, or secondary physiological changes downstream from neuronal damage (e.g., Wallerian degeneration) [3]. Although many individuals fully recover from brain trauma, especially milder injuries, there is a growing public health concern about the potential long-term consequences from TBI. Widespread media coverage following the suicide of several high-profile professional football athletes with a history of head impacts and reported cognitive and emotional changes in middle-age has contributed to fears of developing neurodegenerative disorders many years after TBI among not only athletes, but also military personnel and members of the general community. However, there is currently a debate in the scientific literature relating to whether TBI is a static injury or possibly a dynamic injury process, in which TBI leads to chronic neuroinflammation or progressive neurodegeneration linked to later-in-life functional changes. This article provides a critical review of the evidence to date surrounding TBI as a risk factor for long-term brain changes and cognitive decline, in addition to an increased risk of neurodegenerative conditions.

Classification of TBI

TBI is often graded on the severity of symptoms resulting from a blow to the head or body that affects the brain. Alteration in mental status, such as disorientation and cognitive difficulties, loss of consciousness (LOC), a loss of memory for events occurring before or after the injury (i.e., posttraumatic amnesia), and the Glasgow Coma Scale score, when available, have historically been used to characterize whether an injury is mild, moderate, or severe. While there is no universally agreed upon system for classifying TBI severity, in general, cutoffs for the duration or severity of each of these elements are often used in clinical practice (See Table 1) [4, 5]. However, not all individuals fit neatly into these categories, and as a result, such classification systems are waning in popularity. For example, individuals who meet criteria for mild TBI generally do not show intracranial abnormalities (e.g., cerebral hemorrhage or contusion) on standard neuroimaging techniques. However, there is a subset of mild TBI individuals that do show one or more brain abnormalities, and this is often associated with poorer recovery outcomes relative to individuals with mild TBI without such complications [6]. Thus, a further classification widely used is uncomplicated mild TBI when less severe injuries do not result in intracranial abnormalities and complicated mild TBI when these are present.

A methodological challenge that has limited research on TBI as a risk factor for long-term brain changes and cognitive decline relates to TBI classification. Despite some studies having TBI criteria similar to those used in clinical practice, many have relied on broad definitions of TBI that are based on presence/absence of LOC, intracranial abnormalities, or reported number of cumulative head impacts. Duration of LOC, when available, has often times been divided into arbitrary intervals (i.e., <5 and ≥5 minutes or <1 and ≥1 hour) that do not correspond well to common rubrics used to reflect mild, moderate, or severe injuries. This is because research on the long-term effects of TBI frequently rely on retrospective reports derived from medical records or data from longitudinal aging studies, where TBI-related questions were limited since that was not the purpose of the investigation. Heterogeneity in the classification of TBI in research studies has likely contributed to at least some of the mixed findings in the literature, as different associations may be possible depending on how TBI was defined (e.g., LOC >1 hour versus any head impact). Whereas several meta-analyses have examined the link between TBI and dementia [7–9], to date these also have relied on broad classifications of TBI and have the potential to mask possible associations that may be related to TBI severity, multiple injuries, and age at injury, among other factors. Because TBI is a heterogeneous condition, and neurodegenerative processes may develop over decades, it is important to consider how methodological factors and differences could influence study findings.

ARE THERE LONG-TERM BRAIN CHANGES FROM TBI?

DOES TBI CAUSE AN ACCUMULATION OF NEUROBIOLOGICAL MARKERS RELATED TO NEURODEGENERATIVE DISEASES?

TBI has been associated with a buildup of neurobiological markers that are also found in brains of patients with AD, Lewy body disease, and other neurodegenerative disorders (see Table 4). Amyloid-β (Aβ) plaques, one of the pathognomonic markers of AD, were reported in 60% of patients (n = 39) surviving a moderate-to-severe TBI for at least a year prior to autopsy [38]. Greater Aβ accumulation on PET neuroimaging has also been seen in the posterior cingulate region in individuals surviving a moderate-to-severe TBI within the past 1–17 years compared to healthy controls, with greater accumulation of amyloid found to be associated with more extensive white matter injury [39]. Hyperphosphorylated tau (pTau) is the other primary pathognomonic biomarker for AD and is found in other tau-related neurodegenerative diseases called tauopathies (e.g., frontotemporal dementia). pTau has also been observed at autopsy in a small sample of moderate-to-severe TBI patients surviving for at least 1-year post-injury (n = 39), with tau deposits being more widespread in comparison to a healthy control group [38]. Transactive response DNA protein (TDP–43) is a pathognomonic marker for many cases with frontotemporal lobar degeneration and amyotrophic lateral sclerosis and has been observed in multiple brain regions of small samples of retired professional athletes from contact sports [40]. Greater cytoplasmic TDP–43 has been seen at autopsy in patients sustaining a moderate-to-severe TBI in comparison to a healthy control group [41]. However, only 4% of moderate-to-severe TBI patients (N = 62) had phosphorylated TDP–43, which was equivalent to the proportion of healthy controls, indicating that non-phosphorylated TDP–43 may increase after TBI and be a general marker of neuronal injury. Furthermore, alpha-synuclein, a protein associated with Parkinson’s disease and Lewy body dementia, has been found at autopsy in 20% of patients with a prior history of TBI who had LOC of 1 hour or less and 16% of patients who had LOC for more than 1 hour, though interestingly, only the less severe injuries (LOC 1 hour or less) were associated with the presence of alpha-synuclein pathology [42].

While these studies suggest a relationship between TBI and presence of pathological processes in neurodegenerative conditions, particularly in moderate-to-severe injuries, the scientific community has not established whether the accumulation of these proteins is progressive after TBI. Moreover, the cascade of events responsible for a buildup of pathological markers in AD and related dementias in some individuals with a TBI history is unknown. There has been some indication that inflammation may play a role in the abnormal accumulation of these proteins. For example, microglial cells can be activated by TBI. In a preliminary PET imaging study of 10 patients with a moderate-to-severe TBI between 1–17 years post-injury, increased microglial activation was seen in subcortical brain regions (e.g., thalamus, basal ganglia, internal capsule), but not around the area of primary injury [43]. No relationship was observed with time since injury or extent of neuronal injury, suggesting that TBI may activate a chronic inflammatory response that can persist years after injury (at least in moderate-to-severe TBIs) that does not correlate with a “progressive-like process.” Alternatively, it has been posited that TBI may alter the brain’s ability to integrate neuronal networks when performing cognitive demands, reducing “cognitive/neuronal reserve” [44], which could interact with neurodegenerative changes to increase the risk for cognitive decline rather than contributing to pathological burden per se.

Despite several studies having documented an association between moderate-to-severe TBI and accumulation of pathological markers in AD and related conditions, there is a lack of research examining whether similar changes occur with mild TBI or subclinical head impacts. Clinicians and researchers would be misguided to assume that mild TBI simply carries the same amount of risk as moderate-to-severe TBI. This is in part because a buildup of the proteins that are precursors to AD and related conditions (e.g., Aβ and pTau) have been shown to peak within weeks after mild injuries and rapidly clear thereafter. For example, among 30 Olympic boxers, a large percentage (80%) showed elevated cerebrospinal fluid tau concentrations 1–6 days following a bout, but all boxers’ tau levels returned to normal when re-assessed 14 days later [45]. It is also not known whether mild TBI can produce a persistent inflammatory response in individuals to confer a vulnerability to worse aging outcomes. However, there is a growing body of literature suggesting that the neurofilament light (NF-L) protein may be a biomarker of neuronal injury after TBI. NF-L has been found to be increased following a bout in boxers [46, 47], concussion in professional hockey players [46], and a single season in collegiate football athletes [48, 49]. However, similar to tau, NF-L levels appear to normalize at some point following injury. Retired professional boxers (N = 471) having their last bout 2 years prior did not differ significantly from healthy controls on NF-L concentrations nor show significant changes in NF-L over a 2-year period [47]. It is unclear at present how the reparative process may be overcome in some individuals with a mild TBI to potentially become associated with a greater accumulation of Aβ, pTau, alpha-synuclein, and/or markers of inflammation/neuronal injury years later. Therefore, potential mechanisms to explain how mild TBI may be a risk factor for AD or other neurodegenerative disorders remain unknown and is an area that will be important for future research to address. The reader is referred to recent publications by Wang et al. (2018) and Zetterberg, Smith, and Blennow (2013) for an in-depth review of neurobiological markers post-TBI [50, 51].

DOES TBI INCREASE THE RISK FOR DEMENTIA?

TBI has been found to be a risk factor for later development of dementia in several retrospective cohort studies using medical records (see Table 5). In a large Swedish sample, individuals with a single mild TBI, more severe TBI, and multiple brain injuries (n = 164,334) were found to have a 1.6, 2.0, and 2.8 fold higher risk, respectively, for developing dementia [52]. Risk for dementia following a TBI was still observed even among those who were matched to a sibling without any TBI history (n = 46,970). Risk was greatest 1-year post-injury (odds ratio [OR] = 3.5) but remained significant for up to 30 years later (OR = 1.6). A similar study of over 2.7 million individuals in Denmark found largely the same results, with both mild (hazard ratio [HR] = 1.1) and severe (HR = 1.3) TBI associated with an increased risk for dementia [53]. Risk was again highest within 1 year of injury (HR = 1.2) and greater with multiple injuries (i.e., more than 4 prior TBI, HR = 2.8). The prevalence of dementia in this study was 5.3% for individuals with a history of TBI versus 4.7% for those without. Another investigation examining the incidence of dementia in United States veterans with (n = 178,779) and without history of TBI (n = 178,779) in their medical records, found that individuals with a history of TBI had a higher incidence of dementia (6.1%) over an average of 4 years post-injury versus 2.6% for those without a TBI history [54]. All levels of injury severity, including mild TBI without LOC (HR = 2.3), mild TBI with LOC (HR = 2.5), and moderate-to-severe TBI (HR = 3.7), were associated with an increased likelihood of receiving a dementia diagnosis in that investigation. However, in a separate study of California health data for persons aged 55 years and older (N = 164,661), a history of moderate-to-severe TBI was associated with a higher likelihood of developing dementia over 5–7 years in all participants (HR = 1.7), whereas a single, mild TBI (HR = 1.3) was only associated with increased risk among those older than 64 years at the time of injury [55].

Whereas many of these studies suggest a relationship between TBI of any severity and risk for developing dementia, it is important to consider the methodological challenges of epidemiological studies when interpreting any findings. Dementia within a few years of injury would be expected to relate to cognitive impairment directly from an injury or due to an underlying neurodegenerative disorder that was already present, but perhaps not obvious at the time of injury. Data on TBI and dementia are often based on clinical diagnostic codes, but the actual classification of TBI and cognitive status and the criteria used to make these determinations are not standardized, can vary between clinicians, and may not necessarily be accurate, particularly in large hospital-based data sources that were not designed for research. Dementia is also a non-specific clinical classification, and can be associated with medical conditions that could potentially be reversed or stabilized with treatment, unlike progressive neurodegenerative diseases. Because retrospective cohort studies sometimes rely on medical records, it is difficult to know the proportion of individuals with a dementia diagnosis that actually reflect neurodegenerative disease versus those related to direct cognitive impairments from TBI and/or a medical condition unrelated to the injury (e.g., cerebrovascular accident, toxin exposure, metabolic imbalance). More importantly, when one considers the overall risk for cognitive changes, less than 7% of individuals with a history of TBI have been found to develop dementia for any reason [54, 55], including moderate-to-severe injuries that are typically associated with residual cognitive deficits. Regardless of study limitations and methodological differences, this demonstrates that most persons who sustain one or several injuries, despite level of severity, do not experience significant cognitive changes or dementia with aging.

IS TBI RELATED TO CHRONIC TRAUMATIC ENCEPHALOPATHY?

Multiple mild TBIs and repetitive head impacts have been posited to cause chronic traumatic encephalopathy (CTE), a neuropathological syndrome reported to involve a progressive decline in emotional and cognitive functioning [78]. Nonspecific factors such as impulsivity, irritability, depression, suicidality, substance misuse, gait instability, motor slowness, and attention and memory changes have been used to characterize cases with CTE pathology based on postmortem interviews with informants [79]. Early exposure to head trauma has also been described to be associated with onset of symptoms in cases with CTE pathology, with retrospective reports of emotional and cognitive changes said to start approximately 13 years earlier for individuals who began playing football before the age of 12 in one sample [80]. This is 10-11 years earlier than the association generally found between TBI resulting in LOC and onset of AD and FTD as reviewed above. It is important to note, however, that retrospective interviews are extremely vulnerable to recall bias, allowing for symptoms to be described that may or may not actually have been present or may reflect longer-standing proclivities. In fact, many studies have demonstrated that self- and informant-reported cognitive complaints have a weak association with performance on objective neuropsychological tests, and more strongly relate to mood symptoms [81–84]. This raises questions about whether symptoms reported in cases with CTE may be misattributed to CTE pathology. In fact, depression and substance abuse can develop for a variety of reasons such as genetic predisposition, adverse life events, and other health-related factors, and may be a source of cognitive difficulties or complaints. An investigation of Wisconsin high school graduates from 1957 examined whether playing football during that time was a risk factor for having the symptoms previously described in cases with CTE pathology 30–45 years later [85]. Depression, anxiety, anger, heavy alcohol consumption, and neuropsychological functioning were evaluated at several time points (ages 52, 65, and again at age 72), comparing individuals who played high school football (n = 834) with individuals who played a non-contact sport (n = 800) or did not play any sport (n = 1,951). Football players had similar patterns of cognitive performance, alcohol use, and mood symptoms to each of the comparison groups in that large study, and actually demonstrated lower depressive symptoms overall than the other groups. These data contrast reports that football is linked to worse cognitive and emotional outcomes later in life, as previously described in cases with CTE. The reader is referred to recent publications by Ling, Neal, and Revesz (2017), Iverson et al., (2017), and Smith et al., (2019) for in-depth reviews of the history of CTE, neuropathological characteristics, and controversies [86–88].

Because a progressive neurobehavioral decline in individuals diagnosed with CTE pathology has been reported in several studies, CTE is believed by some to be a neurodegenerative disease. Along these lines, a few studies have staged CTE pathology based on the amount and location of the characteristic tau patterns present [89], and have reported a gradual change in functioning for some cases with CTE pathology at all levels of pathological burden [79]. As defined by McKee and colleagues, Stage I requires 1-2 CTE lesions at the depths of cortical sulci. Stage II requires ≥3 lesions across multiple cortical regions. Stage III requires multiple lesions and widespread neurofibrillary degeneration, and Stage IV requires diffuse lesions and neurofibrillary degeneration across the cerebral cortex, thalamus, and brainstem. Although the descriptions of progressive clinical change are again limited by use of retrospective interviews only, as outlined above, the criteria for what has been considered CTE lesions have varied over time and may have resulted in some cases being misdiagnosed. In 2015, a consensus conference was convened by the NINDS/NIBIB to define the neuropathological criteria for CTE. Brain tissue from 10 cases with presumed CTE and 15 cases from various tau neurodegenerative disorders, all considered at least moderate in disease severity, were selected (though specific selection methods were not specified) and then sent for blind evaluation by a group of seven neuropathologists [90]. Proposed pathological criteria for CTE and well-established criteria for various other tauopathies including AD, progressive supranuclear palsy, corticobasal degeneration, argyrophilic grain disease, primary age-related tauopathy, and Parkinson dementia complex Guam were provided to the evaluators. For the brain slices presumed to be CTE, 91% of the neuropathologists’ reviews classified CTE as the diagnosis. In the end, a consensus panel further refined the neuropathological criteria for CTE to establish criteria that did not clearly overlap processes from the other tau-related neurodegenerative conditions. CTE criteria were defined as pTau aggregation in an irregular pattern in neurons, astrocytes, or cell processes around small blood vessels at the depths of sulci within the brain [90]. However, the consensus group did not incorporate any staging system for CTE, which may have been at least in part due to restricting the selection of patients to the moderate-to-severe end of pathological burden (e.g., Stage III and IV). As a result, it is unknown if CTE lesions thought to represent Stage I/II can be consistently recognized by neuropathologists and distinguished from other conditions since there has only been one reliability study for CTE to date. Despite a previous staging system being available, it would seem possible that there could be alterations or modifications to staging CTE given that prior criteria for recognizing the syndrome consisted of many features that overlapped other neuropathological conditions as well as normal aging. This may have important implications, and it is unclear how many cases in the literature previously diagnosed with CTE would in fact continue to receive a neuropathological diagnosis if re-examined using the new criteria. Additionally, no threshold of pathological burden was incorporated into the CTE criteria to represent when CTE lesions may be considered “significant,” akin to other well-established neurological conditions. For example, when pathologists are diagnosing AD, the criteria for diagnosing AD relies on staging the pathological changes to reflect a No, Low, Intermediate, or High likelihood for AD to be associated with clinical symptoms [87], as we can see some accumulation of these pathological markers of AD in individuals without cognitive symptoms or a dementia diagnosis. In an autopsy study of 2,232 individuals aged 1 to 100 years, only 10 cases were negative for tau pathological changes and these were not older than 24 years of age at death, which raises a question of whether some pathological changes are simply a part of “normal” aging [92]. As such, it would seem likely that the presence of a single or a few CTE/tau lesions might not be associated with any clinical symptoms, and other processes, either neurological, medical, or psychiatric, may instead be the source of symptoms.

To complicate matters further, many cases identified with CTE also show a significant number of neuropathological lesions from other conditions, including AD and related dementias, often leading to a co-morbid diagnosis of AD or other types of neurodegenerative diseases [87]. Although these conditions would be linked to clinical symptoms for patients without CTE, often times, cases with alternative neuropathological diagnoses and CTE pathology are primarily classified as a “CTE case” rather than having the other well-established condition as the primary diagnosis. Due to an absence of systematic clinicopathological studies, it is unknown if CTE is associated with clinical symptoms that are distinct from alternative neuropathological diagnoses or contributes to the clinical profile in a meaningful way to allow for CTE to explain cognitive/behavioral changes when co-occurring pathological processes are present. Furthermore, CTE pathology has been reported in numerous other conditions, including epilepsy and opioid use, in addition to cases with no known history of TBI [93]. Between 1954 and 2013, a total of 153 patients with CTE were reported, and 20% had a documented history of a substance use disorder [94]. It is unknown how substance use and other conditions may influence CTE pathology, but a greater aggregation of pTau lesions has been observed in those with a history of opioid use when compared to healthy age-matched control subjects [95]. At the very least, these findings suggest that CTE pathology is not seen exclusively within the context of a history of TBI and may relate to other factors.

While the neuropathological features of CTE have been recently published, the clinical implications, if any, of CTE pathology remain unclear. Thus, future research is clearly needed to systematically correlate clinical symptoms during life (using standardized neurological and neuropsychological assessments) with CTE pathology at autopsy in order to better understand CTE and isolate risk factors associated with its development.

ARE THERE LONG-TERM COGNITIVE CHANGES FROM TBI?

Potential long-term consequences on cognitive functioning related to a history of TBI have gained increased attention in recent years, even aside from the issue of CTE. In a study that examined neuropsychological performances of 61 patients with a history of TBI of any severity on average 30 years after initial evaluation, cognitive decline was observed in 56% of subjects [96]. It is important to note, however, that moderate-to-severe TBI accounted for more than 70% of the injuries, indicating that severity of TBI is important to consider when evaluating later risk of cognitive decline. While follow-up neuropsychological scores were significantly lower in TBI subjects compared to a control group, not surprisingly, generalized cognitive impairment often occurs with moderate-to-severe TBI and thus, differences may have been directly related to these injuries. Nonetheless, neuropsychological profiles among veterans aged 50 years and older with (n = 88) and without (n = 81) a TBI history, where the majority of injuries were mild in nature (68%), showed significant differences [97]. Individuals with a history of multiple mild TBIs (n = 41) or moderate-to-severe TBI (n = 27) were found to show lower scores on tasks of processing speed and executive functioning, with mean performances falling nearly 1 standard deviation below the normative sample for both injury groups, whereas those with a single, mild TBI (n = 19) did not show this trend. Another study of former retired professional football athletes (n = 28), many of whom self-reported a history of multiple concussions, were found to have lower verbal memory performances compared to healthy controls, although mean performances were still within the normal range [30, 98]. Although these findings suggest that a history of multiple mild TBIs or moderate-to-severe TBI may be associated with a vulnerability to diminished cognitive functioning in aging for some individuals, it is important to consider that these studies contained limited sample sizes and require replication in future research investigations before the findings can be appropriately generalized more broadly to other athletes, military personnel, or the public.

Along these lines, two recent multi-center studies provide support for using caution when making assertions about the influence of TBI on cognitive decline later in life. An investigation of 432 participants with normal cognition and 274 with probable AD in the NACC compared performances on a neuropsychological battery over an 8-year time span between those with and without a history of TBI with LOC [99]. A history of TBI was not associated with a greater rate of cognitive decline on any neuropsychological test for those with normal cognition, nor was TBI history associated with more rapid declines for those with AD. Another investigation using the NACC dataset examined the risk for progression to AD and course of decline among 2,719 participants with prodromal AD (i.e., MCI) with and without a TBI history with LOC [100]. Despite a history of TBI with LOC being linked to an earlier onset of symptoms, TBI history was not associated with faster conversion to AD or a greater rate of decline on a measure of global cognition. Also, the annual rate of progression from MCI to AD was similar, at 8% for individuals with a TBI history versus 10% for those without.

Conclusion

Moderate-to-severe TBI has garnered several lines of evidence for TBI being a risk factor for the later development of dementia and neurodegenerative disorders. Degeneration of neuronal structures and white matter connections seen on MRI scans and Aβ and pTau found at autopsy have been observed years after injury in those with moderate-to-severe injuries that is greater than would be expected for aging-related effects. Taken together with studies reporting an increased incidence of dementia, there is a significant amount of support to suggest that moderate-to-severe TBI can increase the risk for AD, FTD, and other types of dementia later-in-life in some individuals. Mechanism(s) underlying this risk remains unclear at present, though triggering of a chronic inflammatory process and/or an accumulation of pathological proteins in these conditions may be important. Generally, studies investigating whether TBI is associated with a progressive accumulation of pathological markers seen in AD and related dementias, structural/functional brain-related changes, and cognitive decline many years after injury have been few. For example, the majority of studies have only investigated structural brain changes at two time points, which limits interpretations given that it is difficult to determine whether any changes may have stopped shortly after the initial scan, plateaued after a few years, or progressed for several years thereafter. Although some studies have included additional serial scans to assess for brain changes, the duration of these have been limited to <5 years following a moderate-to-severe injury, and it is unclear at present whether neurodegenerative changes can continue beyond this timeframe. It is thus unknown whether moderate-to-severe TBI may increase the risk for developing neurodegenerative conditions by contributing to the accumulation of pathological proteins of an already present neurodegenerative disorder such as AD, lowering one’s cognitive/neuronal reserve, and/or may initiate a progressive process ultimately leading to the development of a neurodegenerative dementia in some individuals.

Mild TBI has received some support for increasing the risk for disrupted brain connections/function and neurodegenerative conditions. More complex injuries that include brain hemorrhage, post-traumatic stress disorder, and/or blast exposure as well as presence of cognitive impairment later-in-life (MCI and dementia) may explain the majority of the findings in structural/functional imaging studies. Thus, later-in-life brain changes in individuals having a history of mild TBI without these complications have not been established at present. Furthermore, methodological challenges in studies examining risk for dementia following TBI limit the conclusions that can be drawn, and therefore provide modest or equivocal evidence for mild TBI as an established risk factor at present. Questions about whether a dementia process was already present at the time of injury or related to some condition completely separate from the injury remain to be resolved. Unlike moderate-to-severe TBI, there is also an absence of studies demonstrating whether mild TBI is associated with a greater buildup of pathological markers in AD and related dementias to suggest a pathological link between mild TBI and risk for dementia. Future research investigating presence of AD-related pathological biomarkers and longitudinal structural/functional imaging after mild TBI is needed to better understand the long-term implications in aging.

When one considers that less than 7% of individuals with TBI of any severity (i.e., mild and moderate-to-severe) later develop dementia across the literature, most people have a very high chance for not developing dementia as they age following TBI. Thus, the growing public health concern about developing neurodegenerative disorders after a TBI seems disproportionate to the risk estimates currently reported in the scientific literature. Widespread media coverage about CTE pathology in former professional athletes has no doubt contributed to the public’s fear about later-in-life effects of TBI. However, there is much that is still unknown about CTE despite it being an established neuropathological syndrome. Questions about the clinical significance of CTE pathology, whether a pathological threshold is needed, and possible confusion of symptoms with alternative neurological, medical, psychiatric, and other neuropathological conditions remain to be answered in future studies. It will be particularly important to clarify whether CTE pathological lesions are related to clinical symptoms before identifying potential risk factors, and a review as well as validation of the initial consensus CTE diagnostic criteria is needed. TBI as a risk factor for long-term brain changes and cognitive decline is a rapidly growing area of research and future studies are needed to answer many of the questions and concerns expressed by military personnel, athletes, parents of young athletes, and the general public.