Appendix B

Biomarker Development for Diagnosis, Prognosis, and Monitoring of Traumatic Brain Injury

A number of biomarkers are being evaluated for or show promise in providing improved sensitivity for diagnosis, prognosis, and monitoring of traumatic brain injury (TBI). This appendix summarizes developments in this area.

NEUROIMAGING BIOMARKERS

Neuroimaging is an important component in the evaluation of patients with a suspected TBI. Imaging is a critical tool in identifying TBI pathologies that increase a patient’s risk of mortality or further neurologic deterioration, and that indicate the need for urgent medical intervention.

Current Options

In clinical settings, computed tomography (CT) remains the standard for imaging of acute TBI because of its ability to identity gross pathologies, such as intracranial hemorrhage, swelling, or presence of foreign objects; its noninvasive nature; and its wide availability across nearly all care settings (e.g., in urban and rural hospitals, trauma centers, and community hospitals). There are established guidelines for diagnosing TBI with CT (Freire-Aragón et al., 2017). In patients with more severe injuries, head CT is preferable in characterizing the type and extent of specific TBI pathologies (e.g., diffuse axonal injury), as well as quantifying the combined contribution of multiple pathologies (e.g., subarachnoid hemorrhage and cortical contusion). However, brain CT scans are limited in detecting and quantifying more subtle neuronal injury.

Magnetic resonance imaging (MRI), the second common neuroimaging method for TBI, can better characterize the nature, locale, and extent of TBI pathology and track the course of pathologies over time relative to CT. Clinicians may use routine MRI when a patient exhibits persistent or worsening symptoms because MRI can provide enhanced anatomic detail

compared with CT, giving clinicians greater diagnostic clarity. Currently, MRI is an adjunctive tool in the evaluation of patients whose clinical outcome appears more severe than what would be expected based on brain CT (Currie et al., 2016). Susceptibility artifact-sensitive sequences (e.g., susceptibility-weighted imaging) are also available and provide information on the presence of microhemorrhages within the brain that are less visible on more conventional T2 sequences (Blennow et al., 2016). Given the coexistence of white matter microhemorrhage and diffuse axonal injury, this approach may furnish a better understanding of patients who are symptomatic but have a normal CT. Additional investigation is warranted before changing current clinical practice guidelines, given the lower frequency of these findings in mild TBI and their inconsistent relation to functional outcome (Currie et al., 2016; Pavlovic et al., 2019). However, the subgroup of TBI patients who benefit the most from MRI and optimal time between injury and brain MRI evaluation have not yet been well defined.

In addition to routine CT and MRI imaging, a variety of advanced quantitative techniques demonstrate promise in better characterizing the acute, sub-acute, and chronic phases of TBI; their use in TBI and concussion has been the subject of several recent review papers (Asken et al., 2018b; Irimia et al., 2012; Lindsey et al., 2021; Sharp et al., 2014). Diffusion imaging (including diffusion tensor imaging, diffusion spectrum imaging, and diffusion kurtosis imaging) and volumetric analysis of three-dimensional anatomic imaging may provide insight into gross and microstructural changes following a TBI. These findings may have value in both the short- and longer-term phases of recovery in patients with a broad range of severity of injuries and time since injury. Generally, studies report abnormalities in diffusion imaging metrics across several brain regions, with the most commonly reported findings in the corpus callosum, corona radiata, internal capsule, and cingulum bundle. In adults, there is some evidence of initially increased fractional anisotropy and decreased apparent diffusion coefficient or mean diffusivity in the acute to sub-acute recovery phase, although longitudinal studies do not necessarily suggest a consistent pattern over time (Asken et al., 2018b). Additional longitudinal studies and studies of mild TBI are needed to explore the precise trajectory of change in diffusion metrics over the course of recovery.

Magnetization transfer imaging (MTI) may also add sensitivity to the MRI evaluation of patients with all severities of TBI (Tu et al., 2017; Wilde et al., 2015). MTI examines the presence or absence of macromolecules, which include proteins and phospholipids that coat axonal membranes or myelin sheaths within the white matter. MTI has been used to infer the degree of myelin integrity and Wallerian degeneration, inflammation, and edema in various disease processes, including TBI. More recently, it has been evaluated in experimental models of TBI in relation to histologically verified myelin loss (Lehto et al., 2017). However, reports of utility in patients have been limited relative to other neuroimaging modalities.

Emerging research also suggests that a dysregulation of cerebral blood flow contributes to TBI pathophysiology and acute and chronic symptoms (Ellis et al., 2016). Change in cerebral blood flow following a vasoactive stimulus is defined as cerebrovascular reactivity and can be measured by imaging techniques including CT, MRI, positron emission tomography (PET)/single photon emission computed tomography (SPECT) perfusion techniques, and transcranial Doppler. Cerebrovascular reactivity (CVR) imaging is used for diagnosing and managing many cerebrovascular diseases but has only recently been studied in TBI (Zeiler et al., 2020a). Some work has begun to evaluate CVR as a neuroimaging biomarker of traumatic vascular injury in sports concussion (Churchill et al., 2019, 2020) and moderate to severe TBI (Zeiler et al., 2020b). The aim is to improve TBI management by improving diagnosis and prediction of recovery. Further work is required to determine whether this technique provides accurate, reliable, and reproducible neuroimaging-based measures of CVR and to correlate imaging measures with specific outcomes.

Another method using MRI technology is the use of functional MRI (fMRI). In this method, blood oxygen level dependent (BOLD)-based fMRI sequences are employed, and involve interpretations of neurological activity related to the oxygenation state of blood and hemodynamic response to the activity-related metabolic task. These sequences have been used to identify regions of brain activation that occur under both task-oriented and resting-state conditions that relate to brain injury, and possible symptoms and deficits related to the injury. Additionally, researchers have used fMRI to establish patterns of connectivity between brain regions and to describe how these connections are altered during both normal development and disease. In TBI, task-based fMRI studies have demonstrated alterations in brain activity across a number of cognitive tasks, including working memory, sustained attention, executive function, and language processing (Laatsch and Krisky, 2006; Palacios et al., 2013; Wu et al., 2020). Reduced connectivity has also been shown in brain regions following TBI (Kondziella et al., 2017; Wu et al., 2020).

PET is a neuroimaging technique that allows for detection and localization of radioisotopes associated with biologically active radiopharmaceuticals that aggregate within the brain following administration of the ligand. PET detects high-energy photons that result from positron decay. In TBI, much of the work conducted with PET to date has involved the imaging of glucose metabolism using 18F-fluorodeoxyglucose imaging. These studies have generally demonstrated reductions in glucose metabolism in multiple brain regions of TBI patients, which are reported to relate to the level of consciousness at the time of PET imaging (Shah et al., 2020). Radiopharmaceuticals revealing pathological correlates of long-term neurodegeneration in the setting of TBI have received considerable interest. Specifically, the amyloid imaging agent Pittsburgh Compound-B (11C-PiB) has been used in multiple studies of patients with TBI. Positive findings are demonstrated in some studies (Hong et al., 2014; Kawai et al., 2013), but findings are inconsistent in others (Ayubcha et al., 2021). Perhaps of greater interest are studies evaluating recently available radiopharmaceuticals targeting abnormally phosphorylated paired helical filament tau following TBI. Postmortem examinations of chronic traumatic encephalopathy (CTE) report the pathological finding of abnormal tau aggregation, and there has been significant interest in developing an in vivo biomarker for detecting this condition. To date, several studies have reported tauopathy in individuals with chronic TBI utilizing tau imaging agents (Gorgoraptis et al., 2019; Takahata et al., 2019).

Future Directions and Limits

Additional studies will be needed to determine the possible benefits of various imaging techniques in various TBI patient populations, injury severities, and care settings. Although advanced quantitative techniques have improved understanding of TBI, their role in diagnosis is still developing, and clinical platforms for utilizing these approaches in patient care remain forthcoming. Additional obstacles include difficulties in (1) directly comparing quantitative metrics derived from different scanners or acquisition parameters and (2) the lack of adequate normative data to enable harmonization across centers, the latter of which is being reconciled by the development of a Normative Neuroimaging Library.¹ Finally, while many of these imaging modalities demonstrate statistical relationships with symptom reports, cognitive functioning, and other outcomes, the validation of these biomarkers for diagnostic use in concussion and mild TBI is complicated by the lack of consistent diagnostic criteria and the existence of other indicators of injury. Additionally, anatomic heterogeneity in TBI

___________________

¹ See https://www.cohenveteransbioscience.org/programs/our-programs/normative-neuroimaging-library (accessed March 2, 2022).

complicates use of common regions of interest in which to focus imaging. It will also be important to determine the need for further development and optimization of informatics platforms to facilitate interpretation of imaging studies, including metrics on the character and degree of TBI pathologies.

BIOFLUID BIOMARKERS

Biomarkers within blood, cerebrospinal fluid (CSF), and saliva show promise in detecting underlying pathologies of TBI and risk for an incomplete recovery.

Current Options

A number of biomarkers are involved in recovery of tissue injury related to a TBI. Specifically, biomarkers include markers of inflammation and blood–brain barrier (BBB) integrity, as well axonal, neuronal, astroglial, and vascular injury. Initially, there is increased flow of immune cells into injured tissue, initiating a debriding process that is essential for tissue recovery. Yet, if this process is too extreme or prolonged, tissue injury can be substantial and can result in more injuries, termed the secondary injury process. Other TBI biomarkers indicate specific brain cell types, including astrocytes and microglia, showing that there is a cascade of biomarker activity that facilitates recovery. When these activities are insufficiently regulated, long-term risks to neurons can result, placing the individual at greater risk for compromise in function and increasing the risk for symptoms and chronic deficits.

The first combination biomarkers to receive Food and Drug Administration (FDA) approval in acute TBI are glial fibrillary acidic protein (GFAP) and ubiquitin carboxylterminal hydrolase L1 (UCH-L1) (Bazarian et al., 2018). GFAP is an astroglial intermediate filament structural cytoskeleton protein that is released on injury and cell death; after acute TBI, serum GFAP levels peak 20 hours after injury (Czeiter et al., 2020; Gill et al., 2018; Papa et al., 2019). UCH-L1, a neuron-enriched enzyme involved in ubiquitin turnover, is detectable as early as 1 hour after TBI, peaks at 8 hours, and declines slowly 48 hours after injury (Anderson et al., 2020; Czeiter et al., 2020; Papa et al., 2019). Research has shown high sensitivity and negative predictive value of the brain trauma indicator (BTI) test for predicting traumatic intracranial injuries on head CT scan acutely after TBI, and distinguishing CT-positive, more severely injured from CT-negative, mild TBI patients (Anderson et al., 2020). In more mild cohorts, including sports concussion, elevated GFAP and UCLH1 have been observed days following concussion, indicating that these biomarkers are accurate predictors in even the mildest of brain injuries (McCrea et al., 2020).

S100B, a marker of both astrocyte and oligodendrocyte activity, also shows promise as a biomarker of TBI, with higher levels being associated with greater TBI severity and poorer outcomes (Czeiter et al., 2020; Jones et al., 2020). A phase 1 cohort analysis (N = 1,409) of the ability of point-of-care GFAP and S100B levels to predict intracranial abnormalities at 24 hours after injury across the full TBI spectrum (Glasgow Coma Scale [GSC] 3–15) was conducted by the TRACK-TBI network. Receiver operator characteristic (ROC) curves for the prediction of CT-positive scans after injury had significantly higher area under the curve (AUC) for GFAP and S100B (0.85 and 0.67, respectively) (Okonkwo et al., 2020). The CENTER-TBI project included analyses of six serum biomarkers (S100B, NSE, GFAP, UCH-L1, neurofilament protein-light [NfL], and total tau), and found that GFAP achieved the highest discrimination for predicting CT abnormalities (AUC 0.89), better than any other single candidate protein or combination of biomarkers in the panel (Czeiter et al., 2020). Thus, ongoing studies continue to evaluate these biomarkers with the aim of ultimately

determining which biomarkers or combination of biomarkers provide the most diagnostic and prognostic value.

Other proteins may be of value for diagnosis and prediction following a TBI. Specifically, NfL is an intermediate filament that provides cytoskeleton support within neurons (Gaetani et al., 2019). Tau is an axonal protein that is associated with axonal damage, and has been implicated in both short- and long-term outcomes related to TBI (Mckee and Daneshvar, 2015). Serum levels of NfL correlate with CSF levels (Shahim et al., 2020), indicating that NfL has brain-specific activity that can be detected within samples of blood. Recently, it has been shown to be elevated in mild to severe injuries (McCrea et al., 2020; Thelin et al., 2019) and to be elevated years following a TBI, with the highest levels in individuals with more severe chronic symptoms (Guedes et al., 2020b; Shahim et al., 2020). Recent studies of plasma tau in acute sports concussion collected within the first 6–24 hours after injury suggest that higher levels of tau may be prognostic biomarkers of prolonged recovery (Pattinson et al., 2020). Lastly, concentrations of total tau in the blood have been linked to chronic symptoms suggestive of CTE in athletes with high numbers of concussions over their career (Mez et al., 2017).

Researchers are also studying exosomes, which are present in blood and carry biomarkers that may be useful in tracking TBI. Exosomes are lipid-membrane-bound extracellular vesicles whose cargo is rich in microRNA (miRNA) and protein, sequestered from the cytoplasm of the cell of origin. Exosomes are secreted by all cells and are known to have a range of biological functions, including cell-to-cell communication and signaling (Guedes et al., 2020a). Exosomes easily cross the BBB and are abundant in peripheral circulation, and their cell of origin can be identified by the proteins they carry on their membrane. Studies of exosomes show that they coordinate response from TBIs (Manek et al., 2018). Exosomal activity has been shown to remain dysregulated years following the injury and to relate to chronic symptoms (Goetzl et al., 2020; Guedes et al., 2020b). Preclinical studies show that exosomal content can promote improved recovery (Chen et al., 2020; Yang et al., 2019). Therefore, additional studies are needed to better understand the role of exosomes in recovery from TBI and their role in acute and chronic symptoms.

Future Directions and Limits

A challenge for the use of TBI biomarkers relates to both the heterogeneity of TBI pathologies and the broad spectrum of TBI severity (from concussion to coma). No single biomarker reflects all known pathophysiological mechanisms of TBI, particularly given their dynamic trajectories over time. In addition, the “majority of TBI biomarker research has focused on … diagnostic biomarkers of acute TBI within the first 24 hours after injury” (Kenney et al., 2021, p. 66), and few candidates have been identified for the diagnosis of sub-acute (up to 1 week postinjury) or chronic sequelae after TBI (3 months to years). Biomarker profiles between weeks and months will facilitate understanding TBI progression and are discussed further in the section on biomarker monitoring below.

Continued research is needed on the translation of biomarkers to clinical practice, including further validation of candidate biomarkers across diverse patient populations, to determine cutoffs for reliable detection of TBI and the possible impacts of comorbidities (e.g., polytrauma, posttraumatic stress disorder [PTSD], depression) on biomarkers. Further discovery of novel biomarkers that are more sensitive and specific continues to be needed to enable tracking of outcomes. For clinical utility, it will be important to determine the added value of fluid biomarkers over existing TBI clinical and neuroimaging evaluation methods. For widespread implementation within care and community settings, development and

validation of accessible, low-resource-burden, cost-effective point-of-care or other quick-delivery modalities for testing options for biomarkers will be critical.

NEUROPHYSIOLOGICAL BIOMARKERS

Although not typically considered “neuroimaging” procedures, a number of prototype technologies intended for clinical evaluation of patients with TBI are in development. These technologies include portable, quantitative electroencephalography (qEEG), eye movement tracking devices, pupilometers, infrared scanners, and others. Some of these technologies have gained clearance from FDA but have minimal penetration in clinical practice, particularly in hospital and trauma care settings.

An electroencephalogram (EEG) records the averaged excitatory and inhibitory postsynaptic potentials of cortical pyramidal neurons, which tend to oscillate at different frequency bands (e.g., delta, theta, alpha, beta, and gamma bands) and involve cortico-cortical and thalamocortical connections. Electrophysiology has the advantage of being inexpensive and easily transportable. This technique also has high temporal resolution and therefore provides information complementary to that derived with such technologies as MRI, which provide high spatial resolution. Several electrophysiological techniques hold promise for detecting mild TBI.

A clinical review of EEG typically involves subjective visual inspection of brain electrical activity, assessment of topography and frequencies, and detection of pathological features. Clinical EEG is generally used to monitor and diagnose epileptic seizures arising in patients with acute TBI. EEG dysfunctions (such as focal slowing) appear to be related to BBB breakdown (Korn et al., 2005; Tomkins et al., 2011). Changes observed in mild TBI clinical EEG have been reported as nonspecific, have low interrater agreement, and may be more useful when used with other approaches, such as qEEG and/or event-related potentials (Gaetz and Bernstein, 2001; Rapp et al., 2015).

qEEG is considered more robust than clinical EEG and given its digital form, involves statistical analyses of the raw signal in order to derive numerical results and relevant information on EEG data. qEEG changes appear to be sensitive to symptoms experienced from mild TBI, particularly balance instability (Thompson et al., 2005). Studies with large sample sizes have also found that qEEG is sensitive for detecting mild TBI (Rapp et al., 2015; Thatcher et al., 2001).

The capacity of EEG determinants—such as coherence, phase, and amplitude difference—to discriminate between mild and severe TBI during the post-acute period is high (sensitivity of 95 percent and specificity of 97 percent) (Thatcher et al., 2001). These results have been cross-validated in a sample of approximately 500 Department of Veterans Affairs (VA) patients (Rapp et al., 2015; Thatcher et al., 2001). Studies using power spectrum analyses have generally shown a decrease in alpha power and an increase in delta, beta, and theta bands. The findings vary in different studies, requiring standardization across sites to improve consistency (Rapp et al., 2015; Thatcher et al., 2001). Additional consideration is required to account appropriately for confounding factors and overlap in populations prone to present with psychiatric disorders (Thornton, 2003). Portable EEG devices have been developed for the assessment of mild TBI in sideline testing (for athletes); in the field (for military personnel); and in more traditional care settings, such as the emergency department. Finally, researchers recently combined transcranial magnetic stimulation (TMS) with EEG to study connectivity changes post-TBI, which may offer a promising avenue for investigating the neural substrates of connectivity dysfunction and reorganization after mild TBI (Coyle et al., 2018).

Event-related potentials (ERPs) allow researchers to understand cognitive processes using time-locked stimuli. P300 is related to attention and working memory. Its amplitude is often

related to the amount of attention required by a task, while its latency is related to the time required for stimulus categorization and discrimination (McCarthy and Donchin, 1981). Various studies have investigated ERPs and found altered control of thought processes and emotional processing in individuals with mild TBI and PTSD (Lew et al., 2005; Solbakk et al., 2005). P300 responses were found to be significantly delayed in latency and lower in amplitude in response to angry faces (Lew et al., 2005), suggesting difficulty in recognizing facial affect. Another study showed similar results in response to affective pictures and suggested reduced attentional resources and dysregulation of top-down processing (Lew et al., 2005; Solbakk et al., 2005). Reduced amplitude of P300 (around 40 percent of symptomatic athletes) and increased P300 latencies have also been observed in athletes with a history of concussion (Gaetz and Weinberg, 2000; Lavoie et al., 2004). P300 amplitudes correlate better with the severity of postconcussive symptoms relative to such factors as number of concussions, time since last concussion, severity of injury, or loss of consciousness (Dupuis et al., 2000; Gosselin et al., 2006).

Magnetoencephalography (MEG) records magnetic fields produced by electrical cortical activity and has better temporal resolution than EEG. MEG shows potential for the diagnosis of mild TBI and has revealed abnormal activity in the frontal, parietal, and temporal regions in patients with blast-related mild TBI (Mu et al., 2017). In a recent study, MEG demonstrated sensitivity in detection of changes in individuals with sub-acute/chronic mild TBI (identifying abnormal brain activity in 87 percent of mild TBI patients in delta wave [1–4 Hz]) (Huang et al., 2012). In general, however, MEG has been collected only in small cohorts, is very expensive to acquire, is not transportable, is available in only a few centers in the world, requires specialized expertise for analysis, and is sensitive to cortical but not subcortical changes.

When using electrophysiology as a diagnostic method for individuals with possible exposure to mild TBI, one should keep in mind the impact of common technical difficulties, such as electrical artifacts, electrode placement, skull defects, medication effects, and patient alertness, on interpreting data. For this reason, it may be more appropriate to use approaches such as qEEG and ERPs in conjunction with other techniques, such as neuroimaging and biofluid markers.

Vision and oculomotor assessment using eye tracking devices, saccadometers, and electrooculography has also been used to assess mild TBI and concussion (Cochrane et al., 2019; Ettenhofer et al., 2018, 2020; Hunfalvay et al. 2019; Kelly et al., 2019; Mani et al., 2018; Sussman et al., 2016). This method correlates with concussion symptoms in children and adults and has promising utility as a rapid, objective, and noninvasive aid for diagnosis. Several oculomotor measures, including metrics of fixation, smooth pursuit, saccades (Brooks et al., 2019), and convergence (Santo et al., 2020), have been investigated for use as potential biomarkers of altered brain function after TBI. One study demonstrated that oculomotor assessment may be an indicator of decreased integrity of frontal white matter tracts and of altered attention and working memory functioning (Maruta et al., 2010). Moreover, researchers have developed mobile eye tracking devices, which could benefit future clinical research by capturing eye movements remotely. However, firm consensus is as yet lacking regarding which visuomotor metric is most sensitive to TBI-related change.

Although currently limited, recent data suggest that TMS has prognostic value in detecting neurophysiological changes postconcussion (Major et al., 2015). At 1 to 5 years postconcussion, no differences were observed in the amplitude of motor-evoked potentials (MEP), but an increased motor threshold (i.e., the lowest stimulus intensity to produce a detectable MEP) was found compared with noninjured controls (De Beaumont et al., 2007; Tallus et al., 2012). Another study found a lengthened duration of the cortical silent period (cSP) (i.e., interruption of voluntary muscle contraction after TMS of the contralateral motor cortex) in

patients with concussion versus controls (De Beaumont et al., 2007, 2009; Tremblay et al., 2011). Despite finding no intracortical facilitation differences among single-concussion, multiple-concussions, and control groups, studies have found that patients with concussion have longer intracortical inhibition compared with controls (De Beaumont et al., 2009; Tremblay et al., 2011). Two studies found similar long-term neurophysiological changes (more than 5 years postconcussion), such as a lengthened cSP of shorter duration and a longer intracortical inhibition in retired athletes compared with controls (De Beaumont et al., 2009; Pearce et al., 2014). Therefore, long-term changes in intracortical inhibition, as well as increased stimulation threshold and slowed neurological conduction time, may be useful indicators when considering prognosis for mild TBI. Nevertheless, additional studies are needed to confirm the value of TMS as a prognostic tool.

USE OF BIOMARKERS FOR IMPROVED MONITORING AFTER TBI

An additional use of biomarkers is to monitor the changes to a person’s condition as it evolves over time. Biomarkers can also be used to monitor and assess treatment effectiveness by narrowly determining target engagement or broadly tracking progressive atrophy and neurodegeneration, reflecting brain cell injury or death. Biomarkers reflecting functional compromise and reversible injury would be powerful tools for monitoring patient status and injury severity in the acute and sub-acute periods, especially after mild TBI, when objective indicators of injury are lacking (e.g., absence of focal lesions); however, this is presently an underdeveloped area of research. Use of biomarker trajectories for monitoring is important for clinical assessment of a patient’s status, as well as for readouts of possible toxicity or side effects of an intervention in clinical TBI trials. The concept is to use biomarker profiles to monitor patient status or specific pathophysiological processes. Profiling biomarkers can also be used as predictive or pharmacodynamic indicators of specific progressive TBI mechanisms that are a target of a treatment (e.g., a measure of attenuated inflammation, reduced fiber tract atrophy or neuronal plasticity). Like diagnostic and prognostic biomarkers, monitoring biomarkers can be developed as trajectories of a single biomarker or of a panel of a variety of biomarker types (e.g., imaging, biofluid, physiologic). It is critical to determine biomarker profiles and half-lives in circulation in order to best use temporal profiles to assess severity and status of injury in a TBI patient.

Neuroimaging Biomarkers for Monitoring TBI

TBI can inflict acute irreversible brain damage in the form of neuronal and glial cell death, traumatic axonal injury, and vascular injury. TBI also results in neurological deficits due to metabolic depression, edema, excitotoxicity, and ionic dysregulation (Bergsneider et al., 2001; MacFarlane and Glenn, 2015; Vespa et al., 2005). These pathophysiological injury types occur secondary to the primary traumatic insult and likely compromise function of the neurovascular unit (NVU), which integrates the relationships among capillaries, neuronal networks, and glia (Bartnik-Olson et al., 2014). Knowledge of inflammation, metabolism, and functional homeostasis can provide assessment tools targeting neurological deficits rooted in NVU compromise. Imaging modalities such as MRI, CT, PET, SPECT, and transcranial Doppler (TCD) can show alterations in blood flow or hyperemia (Fatima et al., 2019; Van Horn et al., 2017) and may help identify perilesional or pericontusional “at-risk” tissues. MRI can be used as a tool for identifying regions of the brain that have incurred injury; however, it requires standardization and calibration of instruments to monitor brain health over time,

normalization across scanners and recording vocabulary, and deep machine learning algorithms to associate imaging features and clinical phenotypes.

Contrast-enhanced neuroimaging administered in TBI patients may help identify BBB disruption and associated vasogenic edema. Alternatively, diffusion-sensitive techniques, such as diffusion-weighted imaging with calculation of apparent diffusion coefficient maps, may identify restricted diffusion, identifying areas of cytotoxic edema and active necrosis. In the setting of contusion or in brain regions where blood flow is compromised, tissue around the central area of injury may be at risk of deteriorating over time; accordingly, it is called perifocal, pericontusional, or perilesional tissue with blood flow and energy metabolite changes (Vespa et al., 2007; Wu et al., 2013). Metabolic depression not only is present after severe TBI, but it also is a major endophenotype of mild TBI and concussion (Giza et al., 2017). The concept of metabolic vulnerability is broadly accepted as an important mechanism of TBI progression and is a contributor to exacerbated symptoms after repeated mild TBI (Greco et al., 2019). In a small cohort of football players with concussion, arterial spin labeling monitored decreased cerebral blood flow that recovered over various time periods associated with perseverance of psychological symptoms (Meier et al., 2015). Microdialysate measures document low oxygen extraction fraction, decreased oxygen/glucose ratio, and increased lactate/pyruvate ratio (Vespa et al., 2005). Scientists have recently adapted pH-weighted molecular MRI from prior use in brain tumor characterization to monitor metabolic vulnerability due to these secondary injury processes (Ellingson et al., 2019). Chemical exchange saturation transfer imaging monitors cerebral acidosis secondary to TBI and has shown promising correlations with Extended Glasgow Outcome Scale outcomes at 6 months postinjury. This pioneering work is hopeful as it bridges acute pathophysiology of at-risk tissue with accepted outcome measures in the field and provides images of abnormal brain physiology as a consequence of TBI.

Perilesional tissue is a prime target for TBI therapies as it is potentially salvageable; thus, imaging and metabolism-related biomarkers may be very useful as near-term predictive and pharmacodynamic biomarkers. Versions of nuclear spin magnetic resonance spectroscopy (H¹, P³¹, C¹³-MRS) are promising imaging modalities for tracking the energy state of the injured brain and metabolic recovery after concussion using such metabolites as N-acetyl aspartate and other compounds (Harris et al., 2015; Stovell et al., 2017; Vagnozzi et al., 2010); however, additional clinical validation and confirmation of selectivity are needed. Metabolic profiling is able to monitor impaired bioenergetic state after mild TBI using gas chromatography mass spectrometry (Wolahan et al., 2015; Yi et al., 2016). Yet, such metabolomic screens need to rigorously assess brain specificity for compounds to be developed into clinically useful monitoring tools. Protein biomarkers of metabolic compromise can be selected more easily for enriched treatment.

Astroglial metabolic biomarkers are prime candidates for tissue compromise, as astrocytes have key functions in the NVU. By wrapping around blood vessels, astrocytes are positioned for direct trauma release of biomarkers into circulation, and astrocyte injury is a known driver of metabolic crisis (Halford et al., 2017). For instance, the brain-specific isoform of the glycolytic enzyme aldolase, ALDOC, is rapidly released from membrane-wounded astrocytes in a human stretch-injury model, as well as in mouse, swine, and rat neurotrauma models in pericontused regions (Halford et al., 2017). Proteomic studies in these trauma models document that release of metabolic enzymes associate with cell wounding early postinjury and are a dominant protein class in CSF proteomes of patients with TBI (Halford et al., 2017; Levine et al., 2016). Profiling metabolic biomarkers could help in assessing the effectiveness of therapies aimed at boosting energy levels in vulnerable

brain areas after TBI and are among the most promising approaches currently investigated for clinical translation (Talley Watts et al., 2013).

Neurophysiological Biomarkers to Monitor TBI

Clinicians usually use EEG in the acute stage to monitor seizures in TBI patients but have also used this tool to detect changes in mild TBI over time (months to years) (Nuwer et al., 2005). Claassen and colleagues (2016) found that monitoring of brain activity using EEG can provide insight into the status of minimally conscious patients with TBI. This study showed that EEG measures of behavioral states provide distinctive signatures that complement behavioral assessments of patients with hemorrhage shortly after TBI. More research is needed to link these measures to patient outcomes.

Biofluid Markers to Monitor TBI

To employ biofluid markers successfully as TBI monitoring tools, it will be critical to gain more comprehensive insight on their trajectory, including release; presence in CSF; and, for noninvasive tracking, their appearance in blood, degradation, and clearance after injury. The temporal profiles of GFAP, UCH-L1, and S100B have been partially described, yet correlations with underlying pathophysiological processes leading to their temporal profiles are still largely elusive (Ercole et al., 2016; Papa et al., 2019; Thelin et al., 2017). A recent pilot study showed that percent change in serum UCH-L1 and S100B improved discrimination between athletes with and without concussion, versus analysis of these markers’ serum levels at any separate time points (Meier et al., 2017). Monitoring of percent change, accomplished by comparing individuals’ change to preinjury or preseason levels, reduces noise from large interindividual heterogeneity. Yet most emergency care providers do not know individual baseline biomarker levels of patients with mild TBI, and most rely on generalized reference values. In this case, carefully monitoring repeated measurements can provide a patient’s rate of change, allowing for individualized profiling during acute care.

Measuring S100B levels in serum (and more recently in saliva) has been shown to be a useful marker in assessing brain tissue after TBI (Asken et al., 2018a). This biomarker is already assisting clinicians and researchers, primarily in Europe (Calcagnile et al., 2012). However, its lack of specificity due to its presence following orthopedic trauma outside the brain has limited its use in polytrauma (Papa et al., 2014); thus, this biomarker has not been widely used in North America. However, clinicians have successfully implemented S100B monitoring to evaluate the need for head CT in patients with mild TBI and to detect secondary injury progression (Thelin et al., 2017). Persistently elevated total tau levels in preliminary studies of athletes were associated with persistent postconcussive symptoms compared with athletes with normal or only mildly elevated plasma tau measures whose symptoms resolved and who returned to full competition (Gill et al., 2017). Thus, tau may be a potential biomarker for monitoring recovery in athletes with TBI and could be used as a guide to determine when it is safe to return to play. Despite limited knowledge of biomarker kinetics, these are promising findings that provide approaches for the use of biomarkers to diagnose and monitor the progression of TBI sequelae, personalize individual patient care, and help predict outcomes.

REFERENCES

Ellingson, B. M., J. Yao, C. Raymond, A. Chakhoyan, K. Khatibi, N. Salamon, J. P. Villablanca, I. Wanner, C. R. Real, A. Laiwalla, D. L. McArthur, M. M. Monti, D. A. Hovda, and P. M. Vespa. 2019. pH-weighted molecular MRI in human traumatic brain injury (TBI) using amine proton chemical exchange saturation transfer echoplanar imaging (CEST EPI). NeuroImage: Clinical 22:101736.

Ellis, M. J., L. N. Ryner, O. Sobczyk, J. Fierstra, D. J. Mikulis, J. A. Fisher, J. Duffin, and W. A. Mutch. 2016. Neuroimaging assessment of cerebrovascular reactivity in concussion: Current concepts, methodological considerations, and review of the literature. Frontiers in Neurology 7:61.

Ercole, A., E. P. Thelin, A. Holst, B. M. Bellander, and D. W. Nelson. 2016. Kinetic modelling of serum S100b after traumatic brain injury. BMC Neurology 16:93.

Ettenhofer, M. L., J. N. Hershaw, J. R. Engle, and L. D. Hungerford. 2018. Saccadic impairment in chronic traumatic brain injury: Examining the influence of cognitive load and injury severity. Brain Injury 32(13-14):1740-1748.

Ettenhofer, M. L., S. I. Gimbel, and E. Cordero. 2020. Clinical validation of an optimized multimodal neurocognitive assessment of chronic mild TBI. Annals of Clinical and Translational Neurology 7(4):507-516.

Fatima, N., A. Shuaib, T. S. Chughtai, A. Ayyad, and M. Saqqur. 2019. The role of transcranial doppler in traumatic brain injury: A systemic review and meta-analysis. Asian Journal of Neurosurgery 14(3):626-633.

Freire-Aragón, M. D., A. Rodríguez-Rodríguez, and J. J. Egea-Guerrero. 2017. Update in mild traumatic brain injury. Actualización en el traumatismo craneoencefálico leve. Medicina Clinica 149(3):122-127.

Gaetani, L., K. Blennow, P. Calabresi, M. Di Filippo, L. Parnetti, and H. Zetterberg. 2019. Neurofilament light chain as a biomarker in neurological disorders. Journal of Neurology, Neurosurgery, and Psychiatry 90(8):870-881.

Gaetz, M., and D. Bernstein. 2001. The current status of electrophysiologic procedures for the assessment of mild traumatic brain injury. Journal of Head Trauma Rehabilitation 16(4):386-405.

Gaetz, M., and H. Weinberg. 2000. Electrophysiological indices of persistent post-concussion symptoms. Brain Injury 14(9):815-832.

Gill, J., K. Merchant-Borna, A. Jeromin, W. Livingston, and J. Bazarian. 2017. Acute plasma tau relates to prolonged return to play after concussion. Neurology 88(6):595-602.

Gill, J., L. Latour, R. Diaz-Arrastia, V. Motamedi, C. Turtzo, P. Shahim, S. Mondello, C. DeVoto, E. Veras, D. Hanlon, L. Song, and A. Jeromin. 2018. Glial fibrillary acidic protein elevations relate to neuroimaging abnormalities after mild TBI. Neurology 91(15):e1385-e1389.

Giza, C. C., M. L. Prins, and D. A. Hovda. 2017. It’s not all fun and games: Sports, concussions, and neuroscience. Neuron 94(6):1051-1055.

Goetzl, E. J., C. B. Peltz, M. Mustapic, D. Kapogiannis, and K. Yaffe. 2020. Neuron-derived plasma exosome proteins after remote traumatic brain injury. Journal of Neurotrauma 37(2):382-388.

Gorgoraptis, N., L. M. Li, A. Whittington, K. A. Zimmerman, L. M. Maclean, C. McLeod, E. Ross, A. Heslegrave, H. Zetterberg, J. Passchier, P. M. Matthews, R. N. Gunn, T. M. McMillan, and D. J. Sharp. 2019. In vivo detection of cerebral tau pathology in long-term survivors of traumatic brain injury. Science Translational Medicine 11(508):eaaw1993.

Gosselin, N., M. Thériault, S. Leclerc, J. Montplaisir, and M. Lassonde. 2006. Neurophysiological anomalies in symptomatic and asymptomatic concussed athletes. Neurosurgery 58(6):1151-1161.

Greco, T., L. Ferguson, C. Giza, and M. L. Prins. 2019. Mechanisms underlying vulnerabilities after repeat mild traumatic brain injuries. Experimental Neurology 317:206-213.

Guedes, V. A., C. Devoto, J. Leete, D. Sass, J. D. Acott, S. Mithani, and J. M. Gill. 2020a. Extracellular vesicle proteins and microRNAs as biomarkers for traumatic brain injury. Frontiers in Neurology 11:663.

Guedes, V. A., K. Kenney, P. Shahim, B. X. Qu, C. Lai, C. Devoto, W. C. Walker, T. Nolen, R. Diaz-Arrastia, J. M. Gill, and CENC Multisite Observational Study Investigators. 2020b. Exosomal neurofilament light: A prognostic biomarker for remote symptoms after mild traumatic brain injury? Neurology 94(23):e2412-e2423.

Halford, J., S. Shen, K. Itamura, J. Levine, A. C. Chong, G. Czerwieniec, T. C. Glenn, D. A. Hovda, P. Vespa, R. Bullock, W. D. Dietrich, S. Mondello, J. A. Loo, and I. B. Wanner. New astroglial injury-defined biomarkers for neurotrauma assessment. Journal of Cerebral Blood Flow and Metabolism 37(10):3278-3299.

Harris, J. L., I. Y. Choi, and W. M. Brooks. 2015. Probing astrocyte metabolism in vivo: Proton magnetic resonance spectroscopy in the injured and aging brain. Frontiers in Aging Neuroscience 7:202.

Hong, Y. T., T. Veenith, D. Dewar, J. G. Outtrim, V. Mani, C. Williams, S. Pimlott, P. J. Hutchinson, A. Tavares, R. Canales, C. A. Mathis, W. E. Klunk, F. I. Aigbirhio, J. P. Coles, J. C. Baron, J. D. Pickard, T. D. Fryer, W. Stewart, and D. K. Menon. 2014. Amyloid imaging with carbon 11-labeled Pittsburgh compound B for traumatic brain injury. JAMA Neurology 71(1):23-31.

Huang, M. X., S. Nichols, A. Robb, A. Angeles, A. Drake, M. Holland, S. Asmussen, J. D’Andrea, W. Chun, M. Levy, L. Cui, T. Song, D. G. Baker, P. Hammer, R. McLay, R. J. Theilmann, R. Coimbra, M. Diwakar, C. Boyd, J. Neff, T. T. Liu, J. Webb-Murphy, R. Farinpour, C. Cheung, D. L. Harrington, D. Heister, and R. R. Lee. 2012. An automatic MEG low-frequency source imaging approach for detecting injuries in mild and moderate TBI patients with blast and non-blast causes. Neuroimage 61(4):1067-1082.

Hunfalvay, M., C. M. Roberts, N. Murray, A. Tyagi, H. Kelly, and T. Bolte. 2019. Horizontal and vertical self-paced saccades as a diagnostic marker of traumatic brain injury. Concussion 4(1):CNC60.

Irimia, A., B. Wang, S. R. Aylward, M. W. Prastawa, D. F. Pace, G. Gerig, D. A. Hovda, R. Kikinis, P. M. Vespa, and J. D. Van Horn. 2012. Neuroimaging of structural pathology and connectomics in traumatic brain injury: Toward personalized outcome prediction. NeuroImage: Clinical 1(1):1-17.

Jones, C., C. Harmon, M. McCann, H. Gunyan, and J. J. Bazarian. 2020. S100B outperforms clinical decision rules for the identification of intracranial injury on head CT scan after mild traumatic brain injury. Brain Injury 34(3):407-414.

Kawai, N., M. Kawanishi, N. Kudomi, Y. Maeda, Y. Yamamoto, Y. Nishiyama, and T. Tamiya. 2013. Detection of brain amyloid ß deposition in patients with neuropsychological impairment after traumatic brain injury: PET evaluation using Pittsburgh Compound-B. Brain Injury 27(9):1026-1031.

Kelly, K. M., A. Kiderman, S. Akhavan, M. Quigley, E. Snell, E. Happy, A. Synowiec, E. Miller, M. Bauer, L. Oakes, Y. Eydelman, C. Gallagher, T. Dinehart, J. Schroeder, and R. Ashmore. 2019. Oculomotor, vestibular, and reaction time effects of sports-related concussion: video-oculography in assessing sports-related concussion. Journal of Head Trauma Rehabilitation 34(3):176-188.

Kenney, K., J. K. Werner, and J. M. Gill. 2021. 7-Proteomic, genetic, and epigenetic biomarkers in traumatic brain injury. In Brain Injury Medicine: Board Review, First Edition (pp. 66-70). Editors: B. C. Eapen and D. X. Cifu. Philadelphia, PA: Elsevier.

Kondziella, D., P. M. Fisher, V. A. Larsen, J. Hauerberg, M. Fabricius, K. Møller, and G. M. Knudsen. 2017. Functional MRI for Assessment of the Default Mode Network in Acute Brain Injury. Neurocritical Care 27(3):401-406.

Korn, A., H. Golan, I. Melamed, R. Pascual-Marqui, and A. Friedman. 2005. Focal cortical dysfunction and blood-brain barrier disruption in patients with postconcussion syndrome. Journal of Clinical Neurophysiology 22(1):1-9.

Laatsch, L., and C. Krisky. 2006. Changes in fMRI activation following rehabilitation of reading and visual processing deficits in subjects with traumatic brain injury. Brain Injury 20(13-14):1367-1375.

Lavoie, M. E., F. Dupuis, K. M. Johnston, S. Leclerc, and M. Lassonde. 2004. Visual p300 effects beyond symptoms in concussed college athletes. Journal of Clinical and Experimental Neuropsychology 26(1):55-73.

Lehto, L. J., A. Sierra, and O. Gröhn. 2017. Magnetization transfer SWIFT MRI consistently detects histologically verified myelin loss in the thalamocortical pathway after a traumatic brain injury in rat. NMR in Biomedicine 30(2):10.1002/nbm.3678.

Levine, J., E. Kwon, P. Paez, W. Yan, G. Czerwieniec, J. A. Loo, M. V. Sofroniew, and I. B. Wanner. 2016. Traumatically injured astrocytes release a proteomic signature modulated by STAT3-dependent cell survival. Glia 64(5):668-694.

Lew, H. L., J. H. Poole, J. Y. Chiang, E. H. Lee, E. S. Date, and D. Warden. 2005. Event-related potential in facial affect recognition: potential clinical utility in patients with traumatic brain injury. Journal of Rehabilitation Research and Development 42(1):29-34.

Lindsey, H. M., C. B. Hodges, K. M. Greer, E. A. Wilde, and T. L. Merkley. 2021. Diffusion-weighted imaging in mild traumatic brain injury: A systematic review of the literature. Neuropsychology Review. https://doi.org/10.1007/s11065-021-09485-5.

MacFarlane, M. P., and L. Glenn. 2015. Neurochemical cascade of concussion. Brain Injury 29(2):139-153.

Major, B. P., M. A. Rogers, and A. J. Pearce. 2015. Using transcranial magnetic stimulation to quantify electrophysiological changes following concussive brain injury: a systematic review. Clinical and Experimental Pharmacology & Physiology 42(4):394-405.

Manek, R., A. Moghieb, Z. Yang, D. Kumar, F. Kobessiy, G. A. Sarkis, V. Raghavan, and K. Wang. 2018. Protein biomarkers and neuroproteomics characterization of microvesicles/exosomes from human cerebrospinal fluid following traumatic brain injury. Molecular Neurobiology 55(7):6112-6128.

Mani, R., L. Asper, and S. K. Khuu. 2018. Deficits in saccades and smooth-pursuit eye movements in adults with traumatic brain injury: a systematic review and meta-analysis. Brain Injury 32(11):1315-1336.

Maruta, J., M. Suh, S. N. Niogi, P. Mukherjee, and J. Ghajar. 2010. Visual tracking synchronization as a metric for concussion screening. Journal of Head Trauma Rehabilitation 25(4):293-305.

McCarthy, G. and E. Donchin. 1981. A metric for thought: a comparison of P300 latency and reaction time. Science 211(4477):77-80.

McCrea, M., S. P. Broglio, T. W. McAllister, J. Gill, C. C. Giza, D. L. Huber, J. Harezlak, K. L. Cameron, M. N. Houston, G. McGinty, J. C. Jackson, K. Guskiewicz, J. Mihalik, M. A. Brooks, S. Duma, S. Rowson, L. D. Nelson, P. Pasquina, T. B. Meier, CARE Consortium Investigators, T. Foroud, B. P. Katz, A. J. Saykin, D. E. Campbell, S. J. Svoboda, J. Goldman, and J. DiFiori. 2020. Association of blood biomarkers with acute sport-related concussion in collegiate athletes: Findings from the NCAA and Department of Defense CARE Consortium. JAMA Network Open 3(1):e1919771.

Mckee, A. C., and D. H. Daneshvar. 2015. The neuropathology of traumatic brain injury. Handbook of Clinical Neurology 127:45-66.

Meier, T. B., P. S. Bellgowan, R. Singh, R. Kuplicki, D. W. Polanski, and A. R. Mayer. 2015. Recovery of cerebral blood flow following sports-related concussion. JAMA Neurology 72(5):530-538.

Meier, T. B., L. D. Nelson, D. L. Huber, J. J. Bazarian, R. L. Hayes, and M. A. McCrea. 2017. Prospective assessment of acute blood markers of brain injury in sport-related concussion. Journal of Neurotrauma 34(22):3134-3142.

Mez, J., D. H. Daneshvar, P. T. Kiernan, B. Abdolmohammadi, V. E. Alvarez, B. R. Huber, M. L. Alosco, T. M. Solomon, C. J. Nowinski, L. McHale, K. A. Cormier, C. A. Kubilus, B. M. Martin, L. Murphy, C. M. Baugh, P. H. Montenigro, C. E. Chaisson, Y. Tripodis, N. W. Kowall, J. Weuve, M. D. McClean, R. C. Cantu, L. E. Goldstein, D. I. Katz, R. A Stern, T. D. Stein, and A. C. McKee. 2017. Clinicopathological evaluation of chronic traumatic encephalopathy in players of American football. Journal of the American Medical Association 318(4):360-370.

Mu, W., E. Catenaccio, and M. L. Lipton. 2017. Neuroimaging in blast-related mild traumatic brain injury. Journal of Head Trauma Rehabilitation 32(1):55-69.

Nuwer, M. R., D. A. Hovda, L. M. Schrader, and P. M. Vespa. 2005. Routine and quantitative EEG in mild traumatic brain injury. Clinical Neurophysiology 116(9):2001-2025.

Okonkwo, D. O.,R. C. Puffer, A. M. Puccio, E. L. Yuh, J. K. Yue, R. Diaz-Arrastia, F. K. Korley, K. Wang, X. Sun, S. R. Taylor, P. Mukherjee, A. J. Markowitz, S. Jain, G. T. Manley, and Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) Investigators. 2020. Point-of-care platform blood biomarker testing of glial fibrillary acidic protein versus S100 calcium-binding protein B for prediction of traumatic brain injuries: A Transforming Research and Clinical Knowledge in Traumatic Brain Injury Study. Journal of Neurotrauma 37(23):2460-2467.

Palacios, E. M., SR. Sala-Llonch, C. Junque, D. Fernandez-Espejo, T. Roig, J. M. Tormos, N. Bargallo, and P. Vendrell. 2013. Long-term declarative memory deficits in diffuse TBI: Correlations with cortical thickness, white matter integrity and hippocampal volume. Cortex 49(3):646-657.

Papa, L., S. Silvestri, G. M. Brophy, P. Giordano, J. L. Falk, C. F. Braga, C. N. Tan, N. J. Ameli, J. A. Demery, N. K. Dixit, N. M. E. Mendes, R. L. Hayes, K. K. Wang, and C. S. Robertson. 2014. GFAP out-performs S100beta in detecting traumatic intracranial lesions on computed tomography in trauma patients with mild traumatic brain injury and those with extracranial lesions. Journal of Neurotrauma 31(22):1815-1822.

Papa, L., M. R. Zonfrillo, R. D. Welch, L. M. Lewis, C. F. Braga, C. N. Tan, N. J. Ameli, M. A. Lopez, C. A. Haeussler, D. Mendez Giordano, P. A. Giordano, J. Ramirez, and M. K. Mittal. 2019. Evaluating glial and neuronal blood biomarkers GFAP and UCH-L1 as gradients of brain injury in concussive, subconcussive and non-concussive trauma: a prospective cohort study. BMJ Paediatrics Open 3(1):e000473.

Pattinson, C. L., T. B. Meier, V. A. Guedes, C. Lai, C. Devoto, T. Haight, S. P. Broglio, T. McAllister, C. Giza, D. Huber, J. Harezlak, K. Cameron, G. McGinty, J. Jackson, K. Guskiewicz, J. Mihalik, A. Brooks, S. Duma, S. Rowson, L. D. Nelson, P. Pasquina, M. McCrea, J. M. Gill, and CARE Consortium Investigators. 2020. Plasma Biomarker concentrations associated with return to sport following sport-related concussion in collegiate athletes—A Concussion Assessment, Research, and Education (CARE) Consortium Study. JAMA Network Open 3(8):e2013191.

Pavlovic, D., S. Pekic, M. Stojanovic, and V. Popovic. 2019. Traumatic brain injury: Neuropathological, neurocognitive and neurobehavioral sequelae. Pituitary 22(3):270-282.

Pearce, A. J., K. Hoy, M. A. Rogers, D. T. Corp, J. J. Maller, H. G. Drury, and P. B. Fitzgerald. 2014. The long-term effects of sports concussion on retired Australian football players: A study using transcranial magnetic stimulation. Journal of Neurotrauma 31(13):1139-1145.

Rapp, P. E., D. O. Keyser, A. Albano, R. Hernandez, D. B. Gibson, R. A. Zambon, W. D. Hairston, J. D. Hughes, A. Krystal, and A. S. Nichols. 2015. Traumatic brain injury detection using electrophysiological methods. Frontiers in Human Neuroscience 9:11.

Santo, A. L., M. L. Race, and E. F. Teel. 2020. Near point of convergence deficits and treatment following concussion: A systematic review. Journal of Sport Rehabilitation 29(8):1179-1193.

Shah, S. A., R. J. Lowder, and A. Kuceyeski. 2020. Quantitative multimodal imaging in traumatic brain injuries producing impaired cognition. Current Opinion in Neurology 33(6):691-698.

Shahim, P., A. Politis, A. van der Merwe, B. Moore, V. Ekanayake, S. M. Lippa, Y. Y. Chou, D. L. Pham, J. A. But-man, R. Diaz-Arrastia, H. Zetterberg, K. Blennow, J. M. Gill, D. L. Brody, and L. Chan. 2020. Time course and diagnostic utility of NfL, tau, GFAP, and UCH-L1 in subacute and chronic TBI. Neurology 95(6):e623-e636.

Sharp, D. J., G. Scott, and R. Leech. 2014. Network dysfunction after traumatic brain injury. Nature Reviews: Neurology 10(3):156-166.

Solbakk, A. K., I. Reinvang, S. Svebak, C. S. Nielsen, and K. Sundet. 2005. Attention to affective pictures in closed head injury: Event-related brain potentials and cardiac responses. Journal of Clinical and Experimental Neuropsychology 27(2):205-223.

Stovell, M. G., J. L. Yan, A. Sleigh, M. O. Mada, T. A. Carpenter, P. Hutchinson, and K. Carpenter. 2017. Assessing metabolism and injury in acute human traumatic brain injury with magnetic resonance spectroscopy: Current and future applications. Frontiers in Neurology 8:426.

Sussman, E. S., A. L. Ho, A. V. Pendharkar, and J. Ghajar. 2016. Clinical evaluation of concussion: The evolving role of oculomotor assessments. Neurosurgical Focus 40(4):E7.

Takahata, K., Y. Kimura, N. Sahara, S. Koga, H. Shimada, M. Ichise, F. Saito, S. Moriguchi, S. Kitamura, M. Kubota, S. Umeda, F. Niwa, J. Mizushima, Y. Morimoto, M. Funayama, H. Tabuchi, K. F. Bieniek, K. Kawamura, M-R. Zhang, D. W. Dickson, M. Mimura, M. Kato, T. Suhara, and M. Higuchi. 2019. PET-detectable tau pathology correlates with long-term neuropsychiatric outcomes in patients with traumatic brain injury. Brain 142(10):3265-3279.

Talley Watts, L., S. Sprague, W. Zheng, R. J. Garling, D. Jimenez, M. Digicaylioglu, and J. Lechleiter. 2013. Purinergic 2Y1 receptor stimulation decreases cerebral edema and reactive gliosis in a traumatic brain injury model. Journal of Neurotrauma 30(1):55-66.

Tallus, J., P. Lioumis, H. Hämäläinen, S. Kähkönen, and O. Tenovuo. 2012. Long-lasting TMS motor threshold elevation in mild traumatic brain injury. Acta Neurologica Scandinavica 126(3):178-182.

Thatcher, R. W., D. M. North, R. T. Curtin, R. A. Walker, C. J. Biver, J. F. Gomez, and A. M. Salazar. 2001. An EEG severity index of traumatic brain injury. Journal of Neuropsychiatry and Clinical Neurosciences 13(1):77-87.

Thelin, E. P., D. W. Nelson, and B. M. Bellander. 2017. A review of the clinical utility of serum S100B protein levels in the assessment of traumatic brain injury. Acta Neurochirurgica1 59(2):209-225.

Thelin, E., F. Al Nimer, A. Frostell, H. Zetterberg, K. Blennow, H. Nyström, M. Svensson, B. M. Bellander, F. Piehl, and D. W. Nelson. 2019. A serum protein biomarker panel improves outcome prediction in human traumatic brain injury. Journal of Neurotrauma 36(20):2850-2862.

Thompson, J., W. Sebastianelli, and S. Slobounov. 2005. EEG and postural correlates of mild traumatic brain injury in athletes. Neuroscience Letters 377(3):158-163.

Thornton, K. 2003. The electrophysiological effects of a brain injury on auditory memory functioning. The qEEG correlates of impaired memory. Archives of Clinical Neuropsychology 18(4):363-378.

Tomkins, O., A. Feintuch, M. Benifla, A. Cohen, A. Friedman, and I. Shelef. 2011. Blood-brain barrier breakdown following traumatic brain injury: A possible role in posttraumatic epilepsy. Cardiovascular Psychiatry and Neurology 2011:765923. https://doi.org/10.1155/2011/765923.

Tremblay, S., L. de Beaumont, M. Lassonde, and H. Théoret. 2011. Evidence for the specificity of intracortical inhibitory dysfunction in asymptomatic concussed athletes. Journal of Neurotrauma 28(4):493-502.

Tu, T. W., J. D. Lescher, R. A. Williams, N. Jikaria, L. C. Turtzo, and J. A. Frank. 2017. Abnormal injury response in spontaneous mild ventriculomegaly wistar rat brains: A pathological correlation study of diffusion tensor and magnetization transfer imaging in mild traumatic brain injury. Journal of Neurotrauma 34(1):248-256.

Vagnozzi, R., S. Signoretti, L. Cristofori, F. Alessandrini, R. Floris, E. Isgrò, A. Ria, S. Marziali, G. Zoccatelli, B. Tavazzi, F. Del Bolgia, R. Sorge, S. P. Broglio, T. K. McIntosh, and G. Lazzarino. 2010. Assessment of metabolic brain damage and recovery following mild traumatic brain injury: A multicentre, proton magnetic resonance spectroscopic study in concussed patients. Brain 133(11):3232-3242.

Van Horn, J. D., A. Bhattrai, and A. Irimia. 2017. Multimodal imaging of neurometabolic pathology due to traumatic brain injury. Trends in Neurosciences 40(1):39-59.

Vespa, P., M. Bergsneider, N. Hattori, H. M. Wu, S. C. Huang, N. A. Martin, T. C. Glenn, D. L. McArthur, and D. A. Hovda. 2005. Metabolic crisis without brain ischemia is common after traumatic brain injury: A combined microdialysis and positron emission tomography study. Journal of Cerebral Blood Flow and Metabolism 25(6):763-774.

Vespa, P. M., K. O’Phelan, D. McArthur, C. Miller, M. Eliseo, D. Hirt, T. Glenn, and D. A. Hovda. 2007. Pericontusional brain tissue exhibits persistent elevation of lactate/pyruvate ratio independent of cerebral perfusion pressure. Critical Care Medicine 35(4):1153-1160.

Wilde, E. A., S. Bouix, D. F. Tate, A. P. Lin, M. R. Newsome, B. A. Taylor, J. R. Stone, J. Montier, S. E. Gandy, B. Biekman, M. E. Shenton, and G. York. 2015. Advanced neuroimaging applied to veterans and service personnel with traumatic brain injury: state of the art and potential benefits. Brain Imaging and Behavior 9(3):367-402.

Wolahan, S. M., D. Hirt, and T. C. Glenn. 2015. Translational metabolomics of head injury: Exploring dysfunctional cerebral metabolism with ex vivo NMR spectroscopy-based metabolite quantification. Brain Neurotrauma: Molecular, Neuropsychological, and Rehabilitation Aspects. Boca Raton, FL: F. H. Kobeissy.

Wu, H. M., S. C. Huang, P. Vespa, D. A. Hovda, and M. Bergsneider. 2013. Redefining the pericontusional penumbra following traumatic brain injury: Evidence of deteriorating metabolic derangements based on positron emission tomography. Journal of Neurotrauma 30(5):352-360.

Wu, S. J., L. M. Jenkins, A. C. Apple, J. Petersen, F. Xiao, L. Wang, and F. G. Yang. 2020. Longitudinal fMRI task reveals neural plasticity in default mode network with disrupted executive-default coupling and selective attention after traumatic brain injury. Brain Imaging and Behavior 14(5):1638-1650.

Yang, Y., Y. Ye, C. Kong, X. Su, X. Zhang, W. Bai, and X. He. 2019. MiR-124 enriched exosomes promoted the M2 polarization of microglia and enhanced hippocampus neurogenesis after traumatic brain injury by inhibiting TLR4 pathway. Neurochemical Research 44(4):811-828.

Yi, L., S. Shi, Y. Wang, W. Huang, Z. A. Xia, Z. Xing, W. Peng, and Z. Wang. 2016. Serum metabolic profiling reveals altered metabolic pathways in patients with post-traumatic cognitive impairments. Scientific Reports 6:21320.

Zeiler, F. A., M. Aries, M. Cabeleira, T. A. van Essen, N. Stocchetti, D. K. Menon, I. Timofeev, M. Czosnyka, P. Smielewski, P. Hutchinson, A. Ercole, and CENTER-TBI High Resolution ICU (HR ICU) Sub-Study Participants and Investigators. 2020a. Statistical cerebrovascular reactivity signal properties after secondary decompressive craniectomy in traumatic brain injury: A CENTER-TBI pilot analysis. Journal of Neurotrauma 37(11):1306-1314.

Zeiler, F. A., A. Ercole, M. Czosnyka, P. Smielewski, G. Hawryluk, P. Hutchinson, D. K. Menon, and M. Aries. 2020b. Continuous cerebrovascular reactivity monitoring in moderate/severe traumatic brain injury: A narrative review of advances in neurocritical care. British Journal of Anaesthesia S0007-0912(19):30966-3.